Anisotropic growth of the thiophene-based layer on Si(111)-B

Article abstract of DOI:10.1039/c4cc01674b

The formation of large assemblies on the Si(111)-B surface is discussed with the help of STM simulations and DFT calculations. Although highly regular assemblies of DTB10B along the Si row direction are observed, the existence of two herringbone isomers introduces a lower periodicity within the 2D molecular network. The formation of herringbone units is explained by weak intermolecular interactions while the 1D assembling depends mainly on the interactions of the C₁₀ side chains with the Si(111)-B surface. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4cc01674b

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Anisotropic growth of the thiophene-based layer
on Si(111)–B†
Cite this: Chem. Commun., 2014,
50, 5484
A. Rochefort,*a Y. Makoudi,b A. Maillard,a J. Jeannoutot,b J. Blier,a F. Cheriouxb and
´
F. Palminob
Received 5th March 2014,
Accepted 30th March 2014
DOI: 10.1039/c4cc01674b
www.rsc.org/chemcomm
The formation of large assemblies on the Si(111)–B surface is
discussed with the help of STM simulations and DFT calculations.
Although highly regular assemblies of DTB10B along the Si row
direction are observed, the existence of two herringbone isomers
introduces a lower periodicity within the 2D molecular network.
The formation of herringbone units is explained by weak inter-
molecular interactions while the 1D assembling depends mainly on
the interactions of the C10 side chains with the Si(111)–B surface.
The adsorption of oligothiophenes on semiconducting surfaces
has not often been studied due to their high propensity to form
covalent bonds with surface atoms.1,2 On the other hand, the
use of highly doped Si-surfaces allows, to a certain degree,
controlling the reactivity of Si surface atoms.3 In contrast, the
number of studies on graphite or HOPG surfaces is much more
significant.4–6 On such weakly reactive surfaces, the structure of
adsorbed oligothiophene mostly depends on intermolecular
interactions, more especially on the nature and structure of
the functional groups attached to the thiophene unit.7,8 The
size of the molecular units used in these studies is usually quite
high, and such studies can rarely benefit from theoretical
Fig. 1 STM image of the organization of DTB10B molecules on the
Si(111)–B surface at high coverage.
support by STM simulations. Here, we used a Si(111)–B Such periodicity is not totally conserved in the molecular 2D
surface† to weaken molecule–substrate interaction in order to network, where the herringbone units are irregularly shifted
achieve the supramolecular network. We demonstrate that STM from one line to the other. From the size of this STM image, we
simulations are becoming practical for large-scale systems, and estimate that the molecular network domains may extend over
can be used to support and explain complex surface processes. 100 ꢀ 100 nm2. We want to emphasize that for a given direction
Fig. 1 shows that the adsorption of 1,4-bis(5-(2,20-dithienyl))- (see dashed lines), the herringbone units are without exception
2,5-bis(decyloxy)benzene† (DTB10B) on Si(111)–B can lead to a aligned in that unique direction.
large overlayer forming 1201 in-plane orientation (see orange
In order to understand the influence of the Si(111)–B sub-
and blue domains). In each domain, the adsorbed molecules strate on the overlayer growth, Fig. 2A shows an STM image
adopt a regular herringbone pattern along the indicated lines. near the edge where Si-adatoms can be easily identified in the
upper part of the image. We can clearly see that the alignment
a
´
Polytechnique Montreal, Engineering Physics Department and Regroupement
of the molecules observed in Fig. 1 follows exceptionally well
the Si-adatom corrugation of the Si(111)–B surface. In addition,
this figure also reveals that a series of well-defined Si-adatom
rows are separated by an angle of 601.
´ ´
´
´
quebecois sur les materiaux de pointe (RQMP), Montreal, Canada.
E-mail: alain.rochefort@polymtl.ca
b
´
´
Institut FEMTO-ST, Universite de Franche-Comte, CNRS, ENSMM,
32 Avenue de l’Observatoire, F-25044 Besancon, France
A more careful analysis of the STM features can be per-
formed in Fig. 2. According to the molecular lengths shown in
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4cc01674b
5484 | Chem. Commun., 2014, 50, 5484--5486
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Fig. 3 Simulated STM image (upper panel) obtained at the TH level
(V = ꢁ1.5 eV, I = 0.1 nA), and variation of the calculated tip height (lower)
along the A and B pathways reported in the upper panel.
Fig. 2 (A) STM image of the DTB10B/Si(111)–B near overlayer edge. Si
adatom rows are identified by white dashed lines. The two types of
herringbone dimers are indicated by the black lines. The inset gives the
structural parameters of the isolated DTB10B molecule. (B) Adsorption
model of a well-ordered DTB10B/Si(111)–B network.
introduced appropriate structural modifications in the adsorp-
tion model to reproduce experimental STM features. Since the
orientation of the C10 chains on the substrate is clear, we have
concentrated our effort on the geometry of the oligothiophene
core. We found an excellent agreement between experimental
the inset of Fig. 2A, the core of DTB10B containing the (Fig. 2) and simulated STM (see Fig. 3) images when the four
oligothiophene fragment corresponds to the brightest STM thiophene units are rigidly rotated by around 351 with respect
features composed of four protuberances well separated by a to the more central benzene ring that appears totally flat, such
lower contrast region in the middle. The aliphatic C10 side as the C10 chains, on the Si(111)–B surface.
chains are revealed in STM through less intense but linear
The description of the experimental STM image given above
features on both sides of the DTB10B core. More importantly, is totally consistent with our STM simulations shown in Fig. 3.
these side chains are well-aligned along the Si-adatom row The upper part of Fig. 3 shows a calculated STM image of a well-
direction (see white dashed lines) and are invariably located organized DTB10B overlayer where the C10 chains are located
between Si-adatom rows. This result strongly suggests that the between Si-adatom rows and aligned along the direction of
C10 chains play a major role in the orientation of the DTB10B these rows. As observed experimentally, the molecular core
units. From the STM contrasts in this well-organized region, an appears as four bright protrusions and the side chains located
adsorption model shown in Fig. 2B can be proposed.
on both sides of the core are less intense. As reported in the
Following the STM simulations approach developed by lower panel, we analyzed the profile of the STM image along
Boulanger-Lewandowski and Rochefort9 on large atomic the A and B straight lines. We distinguish the presence of the
models, we used a finite model containing about 6600 atoms protrusion associated with the DTB10B core and with the
as a starting point of our STM simulations.† The advantage of depletion of conducting states in the more central region.
this approach is the possibility to consider independent struc- Along the B pattern, the highest height near 2 Å is due to the
tural modifications within the assembled layer without impos- presence of oxygen atoms on which the C10 side chains are
ing a spatial periodicity in the calculations. For simplicity, we attached. Otherwise, the lengths of the calculated profiles are in
used a periodic DTB10B network, a Hamiltonian based on the excellent agreement with the size of the molecule reported in
extended Huckel method, and the STM image was obtained at the inset of Fig. 2. The geometry of the adsorbed phase
the Tersoff–Hamann level of theory. Instead of focusing on the proposed from STM simulations is also consistent with the
more stable adsorbed phase at 0 K resulting from a nearly results of DFT/B3LYP/6-31G calculations performed on the
impractical DFT calculation in the present case, we rather isolated DTB10B molecule.† Ground state geometry of DTB10B
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5484--5486 | 5485

View Article Online
ChemComm
Communication
herringbone dimers can be explained by weak interactions such
as CH–p attraction10 between DTB10B units. As the thiophene
units are rotated, isomers of the DTB10B dimer are proposed in
Fig. 4, where the black lines give the thiophene core position (see
also Fig. 2A). In principle, the proportion of both herringbone
isomers should be nearly equal, and once a dimer is assembled
within the 1D direction, the periodicity in the 2D direction will
only depend on the type of isomer that will adsorb in the next 1D
assembly. Although a more quantitative description of the
elementary adsorption steps would need additional experi-
mental and theoretical efforts, a simple analysis of gas phase
species and STM simulations can provide vital structural infor-
Fig. 4 Structural arrangement of DTB10B molecules forming two isomers mation of adsorbed phases.
of a herringbone dimer.
In conclusion, experiments reveal that adsorption of
DTB10B is dominated by the anisotropic interaction of the
C10 chains with Si(111)–B. Supported by STM simulations, we
showed that weak intermolecular interactions such as CH–p
bonding direct the orientation of DTB10B in the formation of a
herringbone dimer. Hence, the adsorption of DTB10B is highly
periodic in one direction but not in the 2D direction.
shows that C10 chains are not flat with respect to the oligothio-
phene core but are twisted by around 501. Since STM reveals
that the C10 chains lie on the Si(111)–B surface in the adsorbed
phase, including the more central benzene core, the thiophene
units need to rotate in order to minimize steric repulsion with
the C10 chains that become flatter. This forced planarization
should provoke the tilting of all thiophene units in a unique
direction. This geometry agrees with the profiles calculated in
Fig. 3, where the bright protrusions are well-aligned rather
than forming a zig-zag shape as it is the case when the thio-
phene units are independently rotated from one to the others
(see S4, ESI†).
´
This work was supported by NSERC, MDER, Nano-Quebec,
´
ANR (Nanokan, ANR-11-BS10-004) and by the Pays de Montbeliard
´
´
agglomeration. We are grateful to Calcul Quebec and Calcul
Canada for providing computational facilities.
References
1 Y. Cao, K. S. Yong, Z. H. Wang, J. F. Deng, Y. H. Lai and G. Q. Xu,
J. Chem. Phys., 2001, 115, 3287.
We now turn to the formation of the DTB10B network. A
systematic alignment of DTB10B dimers and C10 side chains
along the direction of Si-adatom rows is observed, but we do
not recover a similar periodicity in the 2D overlayer, which
shows several defects. For example, the two herringbone
arrangements identified by full and dashed black lines in
Fig. 2A contribute to break this 2D network periodicity. It is also
very interesting to note that molecules that form herringbones
are flipped by 1801 from one to the other. Statistically, the
adsorption of DTB10B should equally lead to two different
adsorbed species (N,M), one being the mirror image of the other.
Although the observed herringbone dimers have different
arrangements, they are strictly formed from an N–M pair. First,
the large proportion of herringbone dimers in the network
strongly suggests that DTB10B molecules form herringbone
dimers that diffuse, and assemble along the 1D direction of
Si-adatoms under the influence of the interaction of the C10
chains with the surface. Second, the presence of two different
2 N. Isobe, T. Shibayama, Y. Mori, K. Shobatake and K. Sawabe, Chem.
Phys. Lett., 2007, 443, 347.
3 (a) B. Baris, V. Luzet, E. Duverger, P. Sonnet, F. Palmino and
´
F. Cherioux, Angew. Chem., Int. Ed., 2011, 50, 4094; (b) B. Baris,
´
J. Jeannoutot, V. Luzet, F. Palmino, A. Rochefort and F. Cherioux,
ACS Nano, 2012, 6, 6905; (c) S. R. Wagner, R. R. Lunt and P. Zhang,
Phys. Rev. Lett., 2013, 110, 086107.
4 M. Linares, L. Scifo, R. Demadrille, P. Brocorens, D. Beljonne,
R. Lazzaroni and B. Grevin, J. Phys. Chem. C, 2008, 112, 6850.
¨
5 M. Bonini, L. Zalewski, E. Orgiu, T. Breiner, F. Dotz, M. Kastler and
P. Samori, J. Phys. Chem. C, 2011, 115, 9753.
6 R. Gutzler, C. Fu, A. Dadvand, Y. Hua, J. M. MacLeod, F. Rosei and
D. F. Perepichka, Nanoscale, 2012, 4, 5965.
7 I. F. Perepichka and D. F. Perepichka, Handbook of Thiophene-Based
Materials: Applications in Organic Electronics and Photonics, John
Wiley & Sons, 2009.
8 E. Mena-Osteritz, M. Urdanpilleta, E. El-Hosseiny, B. Koslowski,
¨
P. Ziemann and P. Bauerle, Beilstein J. Nanotechnol., 2011, 2, 802.
9 N. Boulanger-Lewandowski and A. Rochefort, Phys. Rev. B: Condens.
Matter Mater. Phys., 2011, 83, 115430.
10 E. R. T. Tiekink and J. Zukerman-Schpector, The Importance of
Pi-Interactions in Crystal Engineering: Frontiers in Crystal Engineering,
John Wiley & Sons, 2012.
5486 | Chem. Commun., 2014, 50, 5484--5486
This journal is ©The Royal Society of Chemistry 2014