Bestowing structure upon the pores of a supramolecular network

Article abstract of DOI:10.1039/c4cc05934d

Trigonal molecules compartmentalise the pores of a honeycomb network of 3,4:9,10-tetracarboxylic diimide (PTCDI) and 1,3,5-triazine-2,4,6-triamine (melamine). Extending the 1,3,5-tri(phenylene-ethynylene)benzene core by a phenyl group allows for a well-defined accommodation of the molecule into two symmetry equivalent positions in the pore. The corresponding styryl or phenylene-ethynylene derivatives exceed the pore size and, thus, impede pore modification. This journal is

Full text of DOI:10.1039/c4cc05934d

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Bestowing structure upon the pores of a
supramolecular network†
Cite this: Chem. Commun., 2014,
50, 14175
Baharan Karamzadeh,a Thomas Eaton,b Izabela Cebula,‡a David Munoz Torres,b
˜
Markus Neuburger,b Marcel Mayor*bcd and Manfred Buck*a
Received 30th July 2014,
Accepted 19th September 2014
DOI: 10.1039/c4cc05934d
www.rsc.org/chemcomm
Trigonal molecules compartmentalise the pores of a honeycomb and approach the ultimate limit in controlling the position of
network of 3,4:9,10-tetracarboxylic diimide (PTCDI) and 1,3,5-triazine- individual molecular entities, a strategy was adopted which is
2,4,6-triamine (melamine). Extending the 1,3,5-tri(phenylene- based on the templated assembly of star shaped molecules.
ethynylene)benzene core by a phenyl group allows for a well-defined Going beyond the formation of homogeneous monolayers of
accommodation of the molecule into two symmetry equivalent posi- this type of molecules10,11 a concept for surface patterning was
tions in the pore. The corresponding styryl or phenylene–ethynylene explored which is complementary to a previously implemented
derivatives exceed the pore size and, thus, impede pore modification.
one.4 Instead of shaping the geometry of the pores of a supra-
molecular network by modification of a network component,
Supramolecular self-assembly at surfaces represents a conve- dividing a pore into segments by an additional molecule was
nient route to atomically precise nanostructures.1 In efforts investigated. Starting from a honeycomb network of 3,4:9,10-
towards higher complexity and functionality of molecular tetracarboxylic diimide (PTCDI) and 1,3,5-triazine-2,4,6-triamine
systems assembly strategies involving more than one or two (melamine), which has been studied both in ultrahigh vacuum
components2–5 can rely on either the simultaneous or sequential (UHV)12,13 and under ambient conditions,6,14,15 adsorption of a
assembly of components. Elegantly demonstrated for up to star-shaped molecule is expected to result in a well defined
four molecules at the liquid–solid interface,5 the former relies compartmentation of the network pores as illustrated in Fig. 1a.
on the complex interplay of geometrical and energetic factors.
To this end differently sized star shaped molecules were inves-
Its appeal is the simplicity of assembly in a single step but tigated consisting of a core of 1,3,5-tri(phenylene ethynylene)benzene
requires a careful choice of molecules and conditions. The latter (TPEB) and various extensions differing in length, namely phenyl-
approach adopted both for the liquid/solid interface6–8 and in ethynyl (PE), styryl (S) and phenyl moieties (P) (see Fig. 1b). For the
UHV2,9 is more involved but promises larger flexibility as, in latter both the unsubstituted (PH-TPEB) and hydroxyl substituted
particular, different schemes of assembly can be combined6 or molecules were investigated (POH-TPEB).
additional processes such as electrochemical deposition7 or
displacement8 be implemented.
From simple MM2 force field calculations (see Fig. S1, ESI†) it is
expected that PH/OH-TPEB fits easily into the pore in configurations
In our studies of hierarchical sequential assembly at the liquid/ schematically shown in Fig. 1a. For S-TPEB and PE-TPEB the
solid interface as a means to generate ultra-precise nanostructures
a EaStCHEM School Chemistry, University of St. Andrews, North Haugh,
St. Andrews, UK. E-mail: mb45@st-andrews.ac.uk
b Department of Chemistry, University of Basel, St. Johannsring 19, CH-4056 Basel,
Switzerland. E-mail: marcel.mayor@unibas.ch
c Institute for Nanotechnology, Karlsruhe Institute of Technology (KIT),
P.O. Box 3640, D-76021 Karlsruhe, Germany
d Lehn Institute of Functional Materials (LIFM), Sun Yat-Sen University (SYSU),
Xingang Rd. W., Guangzhou, China
† Electronic supplementary information (ESI) available: Experimental methods,
Fig. 1 (a) Melamine (blue triangle) and PTCDI (white rectangle) forming a
triple hydrogen bonded honeycomb network as shown in the STM image.
(b) Structures of the molecules investigated to sub-divide pores as illustrated
in (a) with the red stars representing energetically equivalent positions of the
molecules.
X-ray crystal structure of PH-TPEB and additional STM images of PTCDI–melamine
networks exposed to S-TPEB and PH-TPEB. CCDC 1000630. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c4cc05934d
‡ Permanent address: Institute of Experimental Physics, University of Wroclaw,
Pl. M. Borna 9, 50-204 Wroclaw, Poland.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 14175--14178 | 14175

View Article Online
Communication
ChemComm
situation is less clear. Even though the undistorted molecules
exceed the size of the pore, a fit is conceivable due to the consider-
able flexibility of the ethynylene moiety. While the ethynylene
moiety is widely perceived as a rigid linker, significant deviations
from linearity by up to 201 have been reported both in bulk and on
surfaces16,17 which provides additional degrees of freedom for the
molecules to be accommodated (see Fig. S1, ESI†). As regards
S-TPEB and POH-TPEB it is noted that these molecule are prochiral.
The conformers can exhibit either C3 symmetry, thus matching the
symmetry of the hexagonal pore, or C1 symmetry for which, in the
case of S-TPEB, one arm of the molecule would not point precisely
into the direction of a network vertex.
The synthesis of the star-shaped molecules is based on Pd
catalyzed cross-coupling chemistry and is sketched in Scheme 1.18
The 1,3,5-triethynylbenzene 3, as common precursor of S-TPEB and
PH-TPEB, was obtained from 1,3,4-tribromobenzene following a
Sonogashira protocol followed by deprotection of the TMS masked
ethynyl substituents. The benzylbromide 5 was converted into the
Wittig salt 6 by treatment with triphenylphosphine. The Wittig
reaction with benzaldehyde provided the iodo-functionalized trans-
stilbene 7 after exposure to reversible iodine addition/elimination
conditions. Sonogashira-type coupling between the triethynyl-
benzene 3 and the iodo-stilbene 7 provided the peripherally styryl
functionalized star S-TPEB as a white solid in 49% isolated yield
after column chromatography (cc). The 4-iodo-1,10-biphenyl 8 was
obtained in a one-pot Sandmeyer procedure from the commercial
4-amino-1,10-biphenyl.19 A similar cross-coupling protocol as
described above applied to 3 and 8 provided the peripherally
phenyl functionalized star PH-TPEB as a white crystalline solid
in 83% yield after column chromatography. The hydroxyl-substituted
star POH-TPEB was assembled in a divergent synthetic approach. Thus
the 1,3,5-triethynylbenzene 3 was further functionalized by applying
a Sonogashira protocol with 1-bromo-4-iodobenzene to provide the
extended start structure 4 in 79% isolated yield. A subsequent Suzuki
coupling with the boronic acid 9 provided the hydroxyl-substituted
star POH-TPEB which was isolated as hardly soluble white solid in a
poor yield of 10% after tedious purification by cc and recycling
gel-permeation-chromatography.
Scheme 1 Synthesis of the molecular stars S-TPEB, PH-TPEB and POH-TPEB.
Reagents and conditions: (a) TMSA, Pd(PPh3)2Cl2, CuI, THF, iPr2NH, 60 1C,
16 h, quant.; (b) K2CO3, THF, MeOH, r.t., 6 h, quant.; (c) IC6H4Br, Pd(PPh3)2Cl2,
CuI, THF, iPr2NH, r.t., 5 h, 79%; (d) PPh3, DMF, 85 1C, 1 h, 76%; (e) (1) toluene,
0 1C, tBuOK, C6H5CHO, r.t., 16 h, CC; (2) toluene, cat. I2, rfl., 3 h, 77%;
(f) Pd(PPh3)4, CuI, THF, iPr2NH, r.t., 16 h, 49% for S-TPEB, 83% for PH-TPEB;
(g) Pd(PPh3)4, K2CO3, toluene, EtOH, micro wave 120 1C, 45 min, 10%.
All star-type structures and precursors were characterized by
1H- and 13C-NMR spectroscopy and by mass spectrometry. The
identity of PH-TPEB was further corroborated by its solid state
structure. Slow evaporation of a diluted chloroform sample provided network has defects such as merged pores or irregularly arranged
single crystals suitable for X-ray analysis. The solid-state structure of PTCDI–melamine strands. There is a clear preference for the star
PH-TPEB, synthetic protocols of all new chemical structures and the molecules to adsorb in areas lacking the network structure. This is
description of the preparation of the PTCDI–melamine network and confirmed by images which show an intact network on an extended
its modification by the star molecules are provided in the ESI.†
scale (Fig. S3a, ESI†). The trigonal shape of the star molecules is only
In high resolution STM images (Fig. 2) of network modified discernible at defects (highlighted by the encircled areas in Fig. S3a,
Au(111) substrates exposed to either PE-TPEB or S-TPEB all compo- ESI†). A closer look at the images of Fig. 2 reveals that the
nents are straightforwardly identified. As for the native network, appearance of the star molecules varies substantially. While some
PTCDI and melamine molecules appear as bright elliptic protru- PE-TPEB (e.g. 1–3 in Fig. 2a) and S-TPEB (1 and 2 in Fig. 2b)
sions and dark vertices, respectively. The star molecules are even molecules exhibit their complete shape, i.e., seven protrusions
imaged at submolecular resolution as illustrated by those labelled arising from the benzene rings, others appear truncated (e.g. 4–7
by numbers (1–5 in Fig. 2a, 1 and 2 in Fig. 2b) which exhibit trigonal in Fig. 2a, 3–7 in Fig. 2b). Specifically, the PE-TPEB and S-TPEB
shapes composed of 7 protrusions, in accordance with the number molecules in an intact network pore (e.g. 6 and 7 in Fig. 2a, 5–7 in
of benzene rings in PE-TPEB and S-TPEB. In both images of Fig. 2 Fig. 2b) are always deviating from the threefold symmetry with one
areas are seen where the network structure is intact and where the arm shortened or even completely missing. While the long arms
14176 | Chem. Commun., 2014, 50, 14175--14178
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
Fig. 2 High resolution STM images in 3D representation of PTCDI–melamine
networks on Au(111)/mica after exposure to solutions of S-TPEB (a) and
PE-TPEB (b). To guide the eye some intact cells are marked by white
hexagons. Areas lacking the network structure are framed by dotted lines.
Some of the S-TPEB and PE-TPEB molecules discussed in more detail in the
text are labelled by numbers. (c) Models of PE-TPEB trapped in a network
pore. Scale bars: 2 nm.
Fig. 3 STM images of PTCDI–melamine networks on Au(111)/mica after
exposure to POH-TPEB. (a) As a guide to the eye a few pores with stars are
highlighted. Insets show Fourier transform of image (top right) and a
section of the Fourier filtered image (bottom right). (b) High resolution
image with two pores marked whose structural models are shown in (c).
case an average value of about 70 Å is measured, i.e., a larger 24 ꢀ O3
usually point towards melamine there is, in agreement with obser- structure compared to the 22 ꢀ O3 unit cell of the clean Au(111)
vations on other systems involving ethynylene moieties,17 significant surface. Furthermore, the length varies between 22 and 25 times the
flexibility in the geometry as in some cases the arms point to Au–Au distance. Interestingly, the periodicities of the reconstruction
melamines positioned at an angle of 1201 of the hexagonal pore and the network are not matching as detailed in the ESI† and
in other cases they point to those positioned at 1801.
illustrated by Fig. S3b. The relationship between network and substrate
One obvious interpretation for observing truncated stars is that periodicity observed here is different from the epitaxial one found in
these are incomplete molecules present as impurity in the solution. experiments under ultrahigh vacuum (UHV) reported earlier which
However, we exclude this as in adsorption experiments without the consists of a combination of hexagonal and parallelogram pore
network truncated molecules have never been observed. Instead, our geometries.13 The reason for this difference is not clear at present
explanation is that, consistent with the fact that most pores are but, given the variability of the gold reconstruction, it might be the
empty, the S-TPEB and PE-TPEB molecules are too large to be consequence of the rather different interfacial kinetics and energetics
accommodated by a pore. Due to the confined geometry the mole- of the liquid and UHV environments.
cules can adsorb onto the metal substrate with at most two arms and
For PH/OH-TPEB a very different result is obtained compared to
the third arm is dangling across the pore boundary as illustrated in PE-TPEB and S-TPEB which exceed the size of the network pore. The
Fig. 2c which depicts models of PE-TPEB in configurations as seen in complete star molecules are clearly seen in the STM images of Fig. 3
pores labelled 3 and 6 in Fig. 2b. Analogously, in the case of damaged (POH-TPEB) and Fig. S4 (ESI†) (PH-TPEB), thus, proving that these
network pores truncated shapes would be indicative of interfering molecules fit into the pores. There is, however, a decisive difference
star molecules with two arms crossing. Whether this lack of imaging between PH-TPEB and POH-TPEB. Comparison of the images of both
contrast can be explained by thermal motion of the weakly interacting molecules reveals that the fraction of network pores where star
arm or an electronic effect remains unclear at present.
molecules are seen is very low for the former (encircled pores in
Before turning to the smaller PH/OH-TPEB stars a few additional Fig. S4, ESI†), whereas for the latter, the stars are present in the
points are worth noting. Firstly, the highly defective networks as overwhelming majority of pores. However, this does not mean that
shown in Fig. 2 are not representative of the quality achievable by the there is a difference in coverage. Rather it arises from a difference in
solution based preparation of the supramolecular network. They were the dynamics of the molecule. As detailed in a forthcoming pub-
deliberately chosen to illustrate the adsorption behaviour of S-TPEB lication, PH-TPEB exhibits mobility on a timescale short compared
and PE-TPEB. As exemplified by Fig. 3a and discussed in detail to the time resolution of STM whereas for POH-TPEB motion is
elsewhere14 networks of high structural quality can be prepared slowed down due to the three OH groups which can be expected to
over extended areas. Secondly, the defects form during the synthesis result in stronger interactions with the network and/or the substrate
of the network and are not due to exposure of the network to the than PH-TPEB due to hydrogen bonding and/or dipole interactions.
star molecules. Thirdly, the Au reconstruction20 is not lifted upon
Despite the high degree of pore filling and the POH-TPEB mole-
formation of the network and subsequent exposure to the star cules fitting entirely into the pore, thus, maximising the interaction
molecules as evidenced by the stripe-like contrast variations seen in with the substrate, the Au surface, like for PE-TPEB and S-TPEB,
Fig. S3a (ESI†) and in agreement with UHV studies of the melamine– remains reconstructed as seen in Fig. 3a. This is in agreement with
PTCDI network.13 Compared to the adsorbate free Au(111) surface,13,21 other studies on extended aromatic systems for which the persistence
there is, however, a significant difference in the periodicity of the of the herringbone reconstruction has been observed as well.10,22,24
structure. As known from other studies22,23 the reconstructed surface
can be substantially altered by molecular adsorbates. In the present the Fourier transform (inset top right) confirm that the structural
Analysing the large scale image of Fig. 3a the sharp spots seen in
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 14175--14178 | 14177

View Article Online
Communication
ChemComm
quality of the supramolecular template is not affected by the Notes and references
adsorption of the star molecule. The Fourier filtered image (inset,
1 J. Elemans, S. B. Lei and S. De Feyter, Angew. Chem., Int. Ed., 2009,
48, 7298–7332; A. G. Slater, P. H. Beton and N. R. Champness, Chem.
Sci., 2011, 2, 1440–1448; L. Bartels, Nat. Chem., 2010, 2, 87–95;
J. V. Barth, Annu. Rev. Phys. Chem., 2007, 58, 375–407; T. R. Eaton,
D. M. Torres, M. Buck and M. Mayor, Chimia, 2013, 67, 222–226.
2 S. Stepanow, M. Lingenfelder, A. Dmitriev, H. Spillmann,
E. Delvigne, N. Lin, X. B. Deng, C. Z. Cai, J. V. Barth and K. Kern,
bottom right) yields a spot in the centre of the pore surrounded by a
sixfold pattern which is aligned with the melamine vertices of the
network. This pattern reflects the statistical occupation of the two
positions schematically shown in Fig. 1a which are energetically
equivalent due to the symmetry of the system.
ˇ
Nat. Mater., 2004, 3, 229–233; J. Cechal, C. S. Kley, T. Kumagai,
A close look at the high resolution images like the one depicted
in Fig. 3b, reveals additional features. Besides the dominating
adsorption geometry where the molecule retains it symmetric star
shape and all three arms point towards the melamine vertices
(e.g. pore 1), the geometry of the POH-TPEB molecule is distorted
in some of the pores with one arm pointing towards a PTCDI unit of
the network (pore 2). In all these cases there is a pronounced
additional protrusion which, based on the shape and imaging
contrast, is reasonable to assume to be a PTCDI molecule. While
a compartment defined by an undistorted star molecule is too small
to accommodate a PTCDI molecule, widening the angle between
two arms by deformation of the ethynylene units allows the PTCDI
to fit in as illustrated in the model of Fig. 3c. In this context it is
worth noting that for a network such as the one shown in Fig. 3b
which exhibits a high structural perfection we do not observe PTCDI
molecules trapped in the pores prior to the adsorption of the star
molecules. However, in the STM images the appearance of the pores
can vary to some extent. For a small fraction the measured height
difference between the PTCDI molecules of the network and the
pore area is different from the other pores or streaky features are
observed such as the one seen in the bottom of the STM image of
Fig. 1a. This suggests that residual PTCDI molecules are present in
some pores after the network preparation which, however, are too
mobile to be resolved by STM. Their presence is only revealed after
being locked into place by the adsorption of a star molecule.
The experiments described above identify a strategy and the
conditions for an iterative approach to ultraprecise two dimen-
sional structures. A primary template generated by bottom-up
self-assembly defines the positioning of another molecular
entity which further reduces dimensions through partitioning
of a pore. To achieve high yield and well defined compartmen-
tation the molecule must fully fit into the pore.
F. Schramm, M. Ruben, S. Stepanow and K. Kern, J. Phys. Chem. C,
2013, 117, 8871–8877.
3 A. Langner, S. L. Tait, N. Lin, C. Rajadurai, M. Ruben and K. Kern,
Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 17927–17930; P. A. Staniec,
˜
L. M. A. Perdigao, A. Saywell, N. R. Champness and P. H. Beton,
ChemPhysChem, 2007, 8, 2177–2181; K. Tahara, K. Inukai,
J. Adisoejoso, H. Yamaga, T. Balandina, M. O. Blunt, S. De Feyter
and Y. Tobe, Angew. Chem., Int. Ed., 2013, 52, 8373–8376; Y. B. Li,
Z. Ma, K. Deng, S. B. Lei, Q. D. Zeng, X. L. Fan, S. De Feyter,
W. Huang and C. Wang, Chem. – Eur. J., 2009, 15, 5418–5423.
¨ ¨
4 M. T. Raisanen, A. G. Slater, N. R. Champness and M. Buck, Chem.
Sci., 2012, 3, 84–92.
5 J. Adisoejoso, K. Tahara, S. Okuhata, S. Lei, Y. Tobe and S. De Feyter,
Angew. Chem., Int. Ed., 2009, 48, 7353–7357.
¨ ¨
6 R. Madueno, M. T. Raisanen, C. Silien and M. Buck, Nature, 2008,
454, 618–621.
¨ ¨
7 C. Silien, M. T. Raisanen and M. Buck, Angew. Chem., Int. Ed., 2009,
48, 3349–3352.
¨ ¨
8 C. Silien, M. T. Raisanen and M. Buck, Small, 2010, 6, 391–394.
9 M. Stohr, M. Wahl, H. Spillmann, L. H. Gade and T. A. Jung, Small,
2007, 3, 1336–1340.
10 S. D. Ha, B. R. Kaafarani, S. Barlow, S. R. Marder and A. Kahn,
J. Phys. Chem. C, 2007, 111, 10493–10497.
11 P. Szabelski, S. De Feyter, M. Drach and S. B. Lei, Langmuir, 2010, 26,
9506–9515; P. Szabelski, W. Rzysko, T. Panczyk, E. Ghijsens,
K. Tahara, Y. Tobe and S. De Feyter, RSC Adv., 2013, 3,
25159–25165; Z. Mu, L. Shu, H. Fuchs, M. Mayor and L. Chi,
J. Am. Chem. Soc., 2008, 130, 10840–10841.
12 J. A. Theobald, N. S. Oxtoby, M. A. Phillips, N. R. Champness and
P. H. Beton, Nature, 2003, 424, 1029–1031.
13 F. Silly, A. Q. Shaw, G. A. D. Briggs and M. R. Castell, Appl. Phys. Lett.,
2008, 92, 023102.
¨ ¨
14 I. Cebula, M. T. Raisanen, R. Madueno, B. Karamzadeh and
M. Buck, Phys. Chem. Chem. Phys., 2013, 15, 14126–14127.
15 V. V. Korolkov, N. Mullin, S. Allen, C. J. Roberts, J. K. Hobbs and
S. J. B. Tendler, Phys. Chem. Chem. Phys., 2012, 14, 15909–15916.
16 T. Takeda and Y. Tobe, Chem. Commun., 2012, 48, 7841–7843; S. Toyota,
T. Yamamori and T. Makino, Tetrahedron, 2001, 57, 3521–3528; G. Jeschke,
M. Sajid, M. Schulte, N. Ramezanian, A. Volkov, H. Zimmermann and
A. Godt, J. Am. Chem. Soc., 2010, 132, 10107–10117.
´
¨
17 D. Heim, D. Ecija, K. Seufert, W. Auwarter, C. Aurisicchio, C. Fabbro,
D. Bonifazi and J. V. Barth, J. Am. Chem. Soc., 2010, 132, 6783–6790;
Compared to a similar approach taken in a UHV experiment25
where partitioning of pores is accomplished by three non-covalently
interacting molecules, the strategy presented here using a single
molecule provides higher flexibility in the design. Notably, going
beyond the simple threefold symmetry of the star molecule by
modifying the geometry and/or introducing functionality and/or
chirality to subpores, which can be envisaged to be different for
each one, a new level of control at this small length scale opens up.
Furthermore, partitioning by a purely covalent structure is also
beneficial for the stability of the systems required in subsequent
steps of a hierarchical assembly process.
This work was supported in part by The Leverhulme Trust.
The synthetic team at the University of Basel acknowledges
financial support by the Swiss National Science Foundation
(SNF) and the Swiss Nanoscience Institute (SNI). BK is grateful
to EaStCHEM and the Funds for Women Graduates (FfWG) for
postgraduate studentships.
¨
D. Ecija, S. Vijayaraghavan, W. Auwarter, S. Joshi, K. Seufert, C. Aurisicchio,
D. Bonifazi and J. V. Barth, ACS Nano, 2012, 6, 4258–4265.
18 N. M. Jenny, M. Mayor and T. R. Eaton, Eur. J. Org. Chem., 2011,
4965–4983.
19 E. A. Krasnokutskaya, N. I. Semenischeva, V. D. Filimonov and
P. Knochel, Synthesis, 2007, 81–84.
¨
20 U. Harten, A. M. Lahee, J. P. Toennies and C. Woll, Phys. Rev. Lett.,
1985, 54, 2619–2622; A. R. Sandy, S. G. J. Mochrie, D. M. Zehner,
K. G. Huang and D. Gibbs, Phys. Rev. B, 1991, 43, 4667–4687.
21 C. Woll, S. Chiang, R. J. Wilson and P. H. Lippel, Phys. Rev. B, 1989,
¨
39, 7988–7991.
22 F. Rossel, P. Brodard, F. Patthey, N. V. Richardson and W.-D.
Schneider, Surf. Sci., 2008, 602, L115–L117.
23 J. T. Sun, L. Gao, X. B. He, Z. H. Cheng, Z. T. Deng, X. Lin, H. Hu,
S. X. Du, F. Liu and H. J. Gao, Phys. Rev. B, 2011, 83, 115419.
24 C. B. France, P. G. Schroeder, J. C. Forsythe and B. A. Parkinson,
Langmuir, 2003, 19, 1274–1281; F. Silly, A. Q. Shaw, M. R. Castell and
G. A. D. Briggs, Chem. Commun., 2008, 1907–1909; Y. Y. Lo,
J. H. Chang, G. Hoffmann, W. B. Su, C. I. Wu and C. S. Chang,
Jpn. J. Appl. Phys., 2013, 52, 101601–101606.
25 D. Ku¨hne, F. Klappenberger, W. Krenner, S. Klyatskaya, M. Ruben
and J. V. Barth, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 21332–21336.
14178 | Chem. Commun., 2014, 50, 14175--14178
This journal is ©The Royal Society of Chemistry 2014