- Email: [email protected]
Unusual self-assembly of chloroaluminium phthalocyanine on graphite
Accepted Manuscript
Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite Hai-yang Ma , Yan-ling Zhao , Myo Win Zaw , Jin-feng Jia , Rui-qin Zhang , Michel A. Van Hove PII: DOI: Reference:
S0039-6028(18)30495-3 https://doi.org/10.1016/j.susc.2018.11.010 SUSC 21382
To appear in:
Surface Science
Received date: Revised date: Accepted date:
10 June 2018 14 November 2018 15 November 2018
Please cite this article as: Hai-yang Ma , Yan-ling Zhao , Myo Win Zaw , Jin-feng Jia , Rui-qin Zhang , Michel A. Van Hove , Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite, Surface Science (2018), doi: https://doi.org/10.1016/j.susc.2018.11.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Highlights
AC
CE
PT
ED
M
AN US
CR IP T
The unusual ClAlPc molecular self-assembled layer on a buffer layer have large deformations compared with the common ClAlPc bilayer Almost freestanding environment is provided for the molecules when the influence from the substrate is shielded by the buffer layer The dominated intermolecular interaction is likely to be the dispersion forces via Reduced Density Gradient analysis
1 / 22
ACCEPTED MANUSCRIPT
Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite Hai-yang Ma a,b,*, Yan-ling Zhao b,*, Myo Win Zaw b, Jin-feng Jia a,# and Rui-qin Zhang b,#, Michel A. Van Hove c Key Laboratory of Artificial Structures and Quantum Control (Ministry of Education), School of
CR IP T
a
Physics and Astronomy, Shanghai Jiao Tong University, Shanghai 200240, China Department of Physics, City University of Hong Kong, Hong Kong, China
c
Institute of Computational and Theoretical Studies & Department of Physics, Hong Kong Baptist
AN US
b
University, Hong Kong, China
M
ABSTRACT
We report an unusual self-assembled layer structure of chloroaluminium phthalocyanine (ClAlPc)
ED
molecules on highly ordered pyrolytic graphite (HOPG), in which a close-packed well-ordered
PT
monolayer is separated from the substrate by a relatively disordered buffer layer, as revealed using scanning tunneling microscopy (STM). Our close-packed monolayer has a nearly rectangular lattice,
CE
instead of the distinctly different square lattice for the more commonly observed well-ordered bilayer
AC
structure. This may be due to the dominance of intermolecular interactions within the monolayer when the influence from the substrate is shielded by the buffer layer. Density Functional Theory (DFT) calculations and Reduced Density Gradient (RDG) analysis indicate that the dominant intermolecular interaction within the unusual layer is likely the London dispersion force. KEYWORDS: Weak interactions, dispersion forces, 2D self-assembly
2 / 22
ACCEPTED MANUSCRIPT
1. Introduction
Two-dimensional (2D) molecular self-assembled layers have been widely studied in recent years due to their potential applications in molecular electronics[1–4] and surface science[5,6]. Different
CR IP T
molecules have been studied on various substrates such as metals (Au, Ag, Cu)[7–10] and semiconductors (graphite, Si, BN)[11–13] with various methods[14–16]. Chemical bonding and physical adsorption are intrinsic driving forces for the formation of structures exhibiting long-range order[8,17,18], influenced by temperature[14], light[19], electric fields[4], etc. One may carefully
AN US
select both the substrate and the molecule to achieve particular results, such as two-dimensional quasi-crystal structures[20–22] or Sierpiński triangle fractals[23]. The substrate, which is essential to support the molecules and control the morphology of the surface layer, may also strongly alter their
M
physical and chemical properties[8,18]. One alternative way to reduce the impact of the substrate on
ED
the molecules is to introduce an “inert buffer layer”. Graphene[24] or intercalated layers[25] often play this role: on these supports nearly freestanding molecular layers are possible, but sometimes the
PT
binding within the molecular layer is too strong for the molecular layer to be stable[26].
CE
The chloroaluminium phthalocyanine molecule belongs to the phthalocyanine family[27] with the
AC
central metal ion being aluminum (Al) connected to a chlorine (Cl) atom pointing out from the dominant plane of the molecule. It is a polar molecule with permanent dipole moment perpendicular to the π-conjugation plane; charge transfer occurs between the molecule and some substrates after deposition[15,16,28]. Depending on the environment, the orientation of surface-adsorbed ClAlPc can be either Cl-up (Cl pointing to the vacuum) or reversed to Cl-down (Cl pointing to the substrate) [4,14]. This property may be applied to single-molecule-based storage[29]. ClAlPc may also be 3 / 22
ACCEPTED MANUSCRIPT
co-deposited with other molecules and interesting self-assembled structures can be obtained by controlling the relative amounts of different molecules[30,31]. In this work, we investigate the self-assembly of ClAlPc molecules on HOPG with a preformed buffer layer, by means of scanning tunneling microscopy (STM) and density functional theory (DFT)
CR IP T
calculations. Previous experimental work has extensively studied both weak and strong interactions (from hydrogen bonds to coordination bonds)[18] in all sorts of molecular self-assemblies. However, to our knowledge, very little attention has been paid to even weaker interactions, such as the
AN US
dispersion force. This may be due to two factors: (1) it is difficult to achieve an almost freestanding or isolated environment; and (2) weaker interactions may have little effect on the final structures. In our experiments, we find that ClAlPc molecules on a buffer layer fulfill the two conditions: we
M
observe large deformations caused by the dominant weak molecular interactions. We propose one possible model for the newly found self-assembly. Our theoretical and experimental results are
ED
highly consistent. Further theoretical analyses by the Reduced Density Gradient (RDG) method
PT
suggest that the dominant contributions to the interactions are the dispersion forces. It is remarkable that although the intermolecular attractions are weak (< 100 meV), they should have such significant
CE
effects on the self-assemblies.
AC
2. Experimental and theoretical methods
STM experiments were performed with an Omicron low temperature (liquid nitrogen, 78 K),
combined with a molecular beam epitaxy (MBE) system in a base pressure of about 5 × 10-10 mbar. Freshly cleaved HOPG was inserted to the vacuum chamber immediately, followed by degassing at ~703 K overnight. Commercially available chloroaluminium phthalocyanine (Sigma Adrich) was 4 / 22
ACCEPTED MANUSCRIPT
then thermally evaporated for 60 seconds at ~693 K onto the clean HOPG, which is kept at room temperature. Before deposition, the molecular source was degassed at an elevated temperature for ~3 hours, followed by pre-heating at growth temperature until the pressure approached ~2 × 10-9 mbar. This purifying procedure was similar to those described in previous reports and a purity higher than
CR IP T
99% was achieved by statistically analyzing many STM images[4,29,30]. Several samples have been prepared. All the STM topographs were recorded in constant-current mode. A mechanically sheared Pt-Ir tip was used.
AN US
Calculations were carried out utilizing the atomic orbital-based DFT method implemented in the SIESTA code[32]. To study the weak intermolecular interactions, we adopted the Van der Waals (VdW) functional[33,34], a universal nonlocal energy functional of the electron density, with a
M
double zeta plus polarization orbital basis set. Other parameters selected in the calculations include an energy cutoff of 300 Ry, a maximal force threshold of 0.02 eV/Å, a projected atomic orbital
ED
energy shift of 10 meV, and a maximal displacement tolerance of 1.0 × 10-4 Å. Note that SIESTA has
PT
been successfully applied in many simulations of 2D surfaces[35–40]. For the RDG[41] analysis the electron density was obtained from DFT calculations which were carried out at the level of
CE
ωB97XD/6-31+G* implemented in the Gaussian 09 package[42]. The spatial distribution of RDG
AC
was visualized by using a multifunctional wave-function analyzer[43].
3. Observations
An HOPG surface is relatively inactive compared with more widely used noble metal substrates like gold, silver and copper[44]. The relatively weak influence of HOPG on the molecules make this substrate an ideal candidate, allowing for molecular self-assembly via strong or weak intermolecular 5 / 22
ACCEPTED MANUSCRIPT
interactions[45]. When we deposit less than 0.1 monolayer of ClAlPc on the substrate (here one monolayer refers to one close-packed ClAlPc layer with π-plane parallel to the HOPG surface[29]), the molecules tend to form bilayer islands[28], as is also the case on graphene[14]. For these commonly observed bilayer islands, the molecules in the first layer, counting the layers from the bare
CR IP T
HOPG, lie flat on HOPG with close-packed structures, forming nearly square lattices, see refs. [4,29,31,46] and Fig. 1a. We observe regions with less ordering, both within the layer and around the edges, than reported in earlier work. These regions have monolayer thickness, cf. Fig. S1. In some
AN US
regions, this relatively disordered layer is “covered” by a more orderly outer layer. STM cannot directly probe the disordered layer underlying, but one can indirectly determine its existence though the uncovered region, such as the top area of Fig. 2(d). The more orderly outer layer exhibits a different periodic structure than the usual bilayer mentioned above, cf. Fig. 2(c) vs. 2(f). We shall
M
call this relatively disordered layer a “buffer layer” between the HOPG and the ordered outer layer
ED
an “unusual layer” in these “unusual islands”. The unusual layer can no longer be found after
PT
annealing the sample at ~423 K, while other layers are still there. Molecule-resolved STM
Fig. 1.
CE
topographs of monolayer ClAlPc on bare HOPG and the unusual layer on buffer layer are shown in
AC
With STM we can easily distinguish the unusual islands from the common bilayer islands. In Fig. 2(c), the square motif is characteristic of the surface morphologies of the bilayer islands[14,46], while in Fig. 2(f), a motif of “six-membered rings” is observed in the unusual islands. The differences are also clear in the Fourier Transform (FT) images of the molecule-resolved STM topographs as shown in insets of Fig. 2(c), (f). Furthermore, from the line profiles Fig. 2(b) and (e) we know that the thicknesses of the common bilayer and unusual islands are ~550 and ~700 pm 6 / 22
ACCEPTED MANUSCRIPT
respectively. A height difference of ~150 pm between the bilayer and unusual islands is obtained, while the monolayer and buffer layer have the same thickness of ~250 pm, cf. Fig. S1(c). The 2D lattice parameters are measured to be |a1| = 1.46 ± 0.02 nm and |a2| = 1.69 ± 0.02 nm with an angle γ = 92 ± 1° for the unusual islands, while for the bilayer islands, |a1| = |a2| = 1.50 ± 0.02 nm with γ =
CE
PT
ED
M
AN US
CR IP T
90 ± 1°, which is consistent with other reports[29,46].
AC
Fig. 1. Structures of a molecular monolayer of ClAlPc self-assembled on bare HOPG, (a) and (b), and on a buffer layer, (c) and (d). (a) Molecule-resolved STM topograph. Vsample = 2.0 V, Iset = 150 pA. The appearance of the dumbbells in only one direction may due to bias dependence[46] or tip effects. (b) Top view and side view of a model based on DFT optimizations. (c) and (d) As (a) and (b), but for the unusual layer on a buffer layer. The molecular models and dashed rhomboid in (c) 7 / 22
ACCEPTED MANUSCRIPT
indicate the unit cell. Scanning parameters for (c) are Vsample = 2.2 V, Iset = 33 pA. The optimized lattice parameters are square with |a1| = |a2| = 1.50 nm for model in (b) and |a1| = 1.40 nm, |a2| = 1.60
AC
CE
PT
ED
M
AN US
CR IP T
nm, the angle γ between a1 and a2 is 93° for model in (d). Image size for (a) and (c): 7.8 × 8.2 nm2.
Fig. 2. STM topographs of the two types of ClAlPc islands. (a), (b) and (c) show usual (“common”) 8 / 22
ACCEPTED MANUSCRIPT
bilayer islands. (d), (e) and (f) show unusual islands. (a) A bilayer island on HOPG, where the second layer, first layer and bare HOPG are seen from left to right. Image size: 78.1 × 81.7 nm2. Vsample = 3.0 V, Iset = 33 pA. (d) Unusual island on a buffer layer surrounded by bilayer islands. A rougher buffer layer is also visible at the top. Image size: 78.1 × 81.7 nm2. Vsample = 2.2 V, Iset = 33 pA. (b) and (e)
CR IP T
Corresponding line profiles along the solid lines in (a) and (d). (c) Molecule-resolved topograph of the bilayer islands on bare HOPG. The nearly square lattice is clear from the inset FT image. The bright spots are impurities while dark holes are vacancies. Image size: 31.2 × 32.7 nm2. Vsample = 1.8
AN US
V, Iset = 33 pA. (f) High-resolution topograph of an unusual island when a buffer layer is present. Blue hexagons specify the “six-membered ring” structures. In the inset FT images, missing spots which may due to glide symmetry are marked by orange dotted circles. The lattice vectors are
M
labeled as a1 and a2. Image size: 23.4 × 24.5 nm2. Vsample = 2.2 V, Iset = 33 pA. Some vertical stripes can be observed in (d) and (f)which may be instrumental (as they are systematically parallel to one
PT
ED
image side and have equal width in pixels regardless of image size, i.e. instrumental magnification).
Different contrasts are also observed when the unusual self-assembled islands are bounded by
CE
different surroundings, while the surface morphologies of the bilayer islands are quite uniform (Fig.
AC
2 and S2).For all the adjacent areas, we observe that the lattices of the two types of islands (Fig. S3(a)) are nearly aligned to each other (within about 5°), while the orientations (in-plane) of the molecules seen in Fig. S3 differ by 36 ± 1.2°. On the islands bounded by the bilayer islands, one may see density contrasts (Fig. 2(d) and 2(f)). When the islands are isolated, the contrasts become less pronounced and less sharp (Fig. S2(d)); they tend to be invisible for small bounded islands (Fig. S3(a)). 9 / 22
ACCEPTED MANUSCRIPT
4. Models
The above observations contain much information. First, in Fig. 1(c), we outline the unit cell of the unusual self-assembled islands by a yellow rhomboid. This can be done by identifying the
CR IP T
molecules along the boundaries of the two kinds of islands like in Fig. S3(a). Each corner of the rhomboid is surrounded by four bright spots which can be interpreted as the four phenyl groups of one ClAlPc molecule. We note that the center of the molecule at the rhomboid corner appears in STM to be identical to the “hole” between four molecules at the rhomboid center, suggesting that the
AN US
center of the molecule is not visibly imaged; this strongly suggests that the Cl atoms point down toward the substrate. Thus, the orientation of molecules on the buffer layer is most likely Cl-down (as we will also confirm later by comparing observations and calculations). By contrast, the bright
ED
HOPG are in Cl-up configuration.
M
spots in the center of the molecules in Fig. 1(a) suggest that monolayer ClAlPc molecules on bare
Second, in Fig. 2(f) we can interpret the “six-membered rings” to be rows of parallel arranged
PT
ClAlPc molecules: four of the six corners of the ring form a square, corresponding to one molecule.
CE
We circle some missing or very weak spots in the inset FT image with orange dotted circles which
AC
may be attributed to the apparent glide-plane symmetry of the STM image. Compared with the bilayer structure, the lattice of the unusual layer is compressed by ~3% in the a1 direction while it is stretched by ~13% in the a2 direction. The ClAlPc molecules appear to be intact, based on their size and appearance in STM. These considerable lattice deformations together with the changed orientations of the molecules indicate that HOPG and buffer layers have quite different effects on ClAlPc molecular self-assembly. Third, we note that the molecules on a buffer layer get closer 10 / 22
ACCEPTED MANUSCRIPT
together along one direction, which suggests that the influence of the buffer layer is weaker than that of HOPG. The tight molecular packing may bring about tilting of the molecules and may be the cause of the contrasts in STM topographs noted above. Meanwhile, the ClAlPc molecules can easily leave or move along the HOPG as well as the buffer layers, since no isolated molecule has been
CR IP T
observed on the surface; this implies that the binding of the molecules to the buffer layer is at least as weak as to HOPG. Additionally, due to the more random orientations of the molecules on the rougher buffer layer and the disorder within the buffer layer, we expect that on average the in-plane
AN US
interactions would be stronger than the out-of-plane interactions.
Building on the above observations and interpretations, we can propose a likely model in which the buffer layer provides a relatively freestanding environment for molecules on top of them, see Fig.
M
3(c), for our newly found unusual self-assembly of ClAlPc on HOPG.
Before providing additional support for our model, we analyze the possible driving forces that lead
ED
to the structures and describe the self-assembly processes with the above model in mind. Since no
PT
external effects and external chemical bonds are included here, we will only consider the intermolecular VdW forces. We discuss the three kinds of VdW forces:
CE
(1) Dipole-dipole. The hydrogen bonds should be excluded first since the intermolecular N-H
AC
atom pairs are too far apart (~3.5 Å). Furthermore, although this molecule has an intrinsic dipole moment perpendicular to the plane of the molecule, the molecules remain close to parallel to the layer surface. The parallel arrangements of the dipole moments in fact produce repulsive forces between the molecules, which reduces the tendency to form close-packed islands, contrary to our observations. (2) Dipole-induced dipole. The symmetry of this molecule implies no in-plane polarization, so 11 / 22
ACCEPTED MANUSCRIPT
that no dipole-induced dipole is expected. (3) Instantaneous dipoles. London dispersion forces are the attractive part of VdW forces[47,48]. These attractive forces can arise from structural deformations of molecules since unbalanced instantaneous dipoles may arise along some directions where symmetries were perturbed,
AC
CE
PT
ED
M
AN US
dispersion forces are the dominant interactions.
CR IP T
which is in line with the tilts we have proposed above. Therefore, we conclude that London
Fig. 3. Schematics of two types of self-assembly. (a) Bottom view and side view of a ClAlPc molecule. (b) The bilayer islands on HOPG with staggered stacking. (c) Unusual self-assembly of ClAlPc on a buffer layer over HOPG; the molecules may tilt due to the relatively strengthened intermolecular interactions and the corrugation of the buffer layer.
12 / 22
ACCEPTED MANUSCRIPT
5. Calculations and discussions
We perform theoretical calculations based on the previous assumption of a freestanding monolayer minimally affected by the buffer layer, i.e. no substrate or buffer layer is included. A 20 Å vacuum
CR IP T
layer is added along the c (out-of-plane) direction in a 3D periodic scheme. At first, we calculate the partial density of states (PDOS) of each element using standard DFT methods, from which we can see that the PDOS of carbon atoms dominates around the Fermi level (cf. Fig. S4 in Supplementary material). This indicates that the bright spots in Fig. 1(c) represent the phenyl rings (C atoms) since
AN US
the bias voltages applied are within the corresponding energy range. Combining with the STM observation of holes in the center of the molecules, we model the molecules with Cl-down configuration. It is clear in Fig. 2(c) that the molecules within the bilayer are also in Cl-down;
M
however, the reasons for this common orientation are different. For the unusual layers, the cause is
ED
the relatively larger corrugation (Fig. S1(c)) of the buffer layers. Ordered structures with ClAlPc molecules lying flat and in Cl-up are difficult to form if the substrate is not flat enough at the same
PT
time [49]. However, unevenness and disorder of the layers below does not guarantee the
CE
disappearance of ordered structures on top of them[50], ClAlPc molecules can adopt the Cl-down
AC
configuration to accommodate the corrugations. But for the bilayers, to balance the dipole moments, the second layer will stack itself onto the underlying layer with reversed dipole directions[14,15,46]. This staggered packing may be the reason why the heights of the bilayer islands are lower than the unusual ones. Next, we perform total energy calculations to determine the most reasonable lattice size and shape; the details are listed in table S1. At the same time, we also calculate the adsorption energy of a single 13 / 22
ACCEPTED MANUSCRIPT
molecule on graphene with the Cl-up configuration to give quantitative evaluations of the molecule-substrate interactions, cf. below. GGA/PBE calculations are also performed to show the existence of dispersion interaction in close-packed molecular models. First, among all the cases we have considered, VdW calculations show that a slightly tilted structure (< 5°) with lattice size 14 ×
CR IP T
16 Å2 and γ = 93° is the most stable. The optimized lattice size and the ratio |a1|/|a2| = 1.14 matches the experimental observations well, with only slight deviations < 5.4％ for all the parameters. The calculated molecular tilts confirm our previous suggestions. Second, in comparison with fixed
AN US
out-of-plane displacements within the molecule, the total energy lowering by the tilting is estimated to be ~90 meV per molecule. To evaluate the effects from HOPG, we also calculate the total energies using parameters from experiments, i.e. a square lattice with size 15 × 15 Å2 (while still excluding
M
the substrate; the local density approximation （ LDA ） functionals may underestimate the intermolecular distances, which are here ignored). The energy increases almost by the same amount
ED
as it is reduced by the tilt and both changes are much smaller than the binding energy of π-π
PT
interactions (~4 eV, as estimated by our VDW/DRSLL calculations) between the ClAlPc molecule and graphene, which shows that the intermolecular interactions are rather weak. Additionally, the
CE
relative rotation angle (about the surface normal) of the two self-assemblies in Fig. 1(b) and (d) is
AC
evaluated as 37.4°, which matches the experimental result well. These systematic theoretical results which are highly consistent with experiments show that HOPG does impose relatively strong constraints on the ClAlPc molecules and that these molecules will reassembly into a more stable pattern when the almost freestanding environment is provided by the buffer layers. Our model can completely describe the experimental observations. To further verify that the dispersion forces dominate the intermolecular interactions, we analyze 14 / 22
ACCEPTED MANUSCRIPT
the electron structure with the Reduced Density Gradient (RDG) method, where RDG is defined as RDG = 1/(2(3π2 )1/3 )|∇𝜌|/𝜌4/3 : this is a very powerful method to both identify and visualize weak interactions between molecules[41]. In Fig. 4(a) and (b) we plot RDG as a function of the electron density multiplied by the sign of the second Hessian eigenvalue (λ2) obtained for the optimized
CR IP T
structures from our DFT calculations; by multiplying with sign (λ2) we can distinguish the bonded (λ2 < 0) from unbonded interactions (λ2 > 0). A density cutoff ρ < 0.05 au was chosen to depict the
AC
CE
PT
ED
M
AN US
noncovalent region of interest. The two images are quite similar in the range 0.005 < ρ < 0.05 au,
Fig. 4. Reduced density gradient (RDG) analyses of ClAlPc molecular interactions using the structures obtained from our DFT calculations. (a) and (b) Graphs of RDG versus the electron density ρ multiplied by the sign of the second Hessian eigenvalue, sign(λ2), for usual lattice without 15 / 22
ACCEPTED MANUSCRIPT
tilt and the unusual tilted one. The pairs of numbers at the top right give the size of the unit cell in Å. Red rectangles highlight the areas of attractive VdW interaction. (c) and (d) Corresponding isovalue surfaces in real space taken at RDG = 0.8 au. The range of the signed electron density is designated by the colored scale bar. The distribution of the intermolecular dispersion forces (light green) is
CR IP T
pointed out by red circles; no hydrogen bonding (dark green) is observed around the N atoms.
which maps to relatively stronger noncovalent interactions including hydrogen bonding (λ2 < 0) or
AN US
steric clashes (λ2 > 0). However, in the regions that cover the attractive and weak London dispersion forces (ρ < 0.005 au and λ2 < 0)[51], more points appear for the tilted structures than the usual one, especially for the area closest to zero, as highlighted by the red rectangles. These changes can be
M
visualized directly from Fig. 4(c) and (d), showing the isovalue surfaces of the RDG distribution in real space. A clear increase of dispersion forces is seen in the intermolecular areas along the
ED
compressed direction, or a1 direction, and around the nearby intermolecular C and H atoms. Besides,
PT
no sign of hydrogen bonding (which would show up as dark green) is found around the N atoms, in
CE
agreement with the analysis in the last section.
AC
6. Conclusions
Based on molecule-resolved STM topographs and DFT calculations, we have investigated the
weak intermolecular interactions, possibly due to dispersion forces, which are induced in unusual self-assemblies of ClAlPc on HOPG. A buffer layer between the HOPG and the self-assembled unusual layer appears to nearly remove the periodic influence from the graphite and provide an almost freestanding environment for the self-assembled layers. Future experiments can aim to build 16 / 22
ACCEPTED MANUSCRIPT
well-designed buffer layers to free the molecules, thereby enabling the fragile intermolecular interactions to manifest their roles while finding additional interesting self-assembled structures and applications. Our ideas can also be extended to other fields where weak interactions must be taken
Corresponding Authors: Prof. Rui-qin Zhang, Email: [email protected]
#
Prof. Jin-feng Jia, Email: [email protected]
AN US
#
CR IP T
into consideration.
Author Contributions:
M
*These authors contributed equally to this work.
Conflict of interest
Acknowledgements
PT
ED
The authors declare no competing financial interest.
CE
This work was supported by the Collaborative Research Fund of Hong Kong Research Grants
AC
Council (Project No. C2014-15G). We also acknowledge the computing resources of the Tianhe2-JK cluster at the Beijing Computational Science Research Center and of the Tianhe2 cluster at the National Supercomputer Center in Guangzhou, China. ICTS is supported by the Institute of Creativity, which is sponsored by Hung Hin Shiu Charitable Foundation (孔憲紹慈善基金贊助).
References 17 / 22
ACCEPTED MANUSCRIPT
[7]
[8] [9]
[10]
[11]
AC
[12]
CR IP T
[6]
AN US
[5]
M
[4]
ED
[3]
PT
[2]
R. Otero, J.M. Gallego, A.L.V. de Parga, N. Martín, R. Miranda, Molecular Self-Assembly at Solid Surfaces, Adv. Mater. 23 (2011) 5148–5176. doi:10.1002/adma.201102022. A. Kumar, K. Banerjee, P. Liljeroth, Molecular Assembly on Two-Dimensional Materials, Nanotechnology. 28 (2016) 082001. doi:10.1088/1361-6528/aa564f. D. Xiang, X. Wang, C. Jia, T. Lee, X. Guo, Molecular-Scale Electronics: From Concept to Function, Chem. Rev. 116 (2016) 4318–4440. doi:10.1021/acs.chemrev.5b00680. T. Niu, Electric-Field-Induced Molecular Switch of Single Dipolar Phthalocyanine on Cu(111): A Scanning Tunneling Microscopy Study, J. Phys. Chem. C. 119 (2015) 19802– 19807. doi:10.1021/acs.jpcc.5b04318. L. Dong, P.N. Liu, N. Lin, Surface-Activated Coupling Reactions Confined on a Surface, Acc. Chem. Res. 48 (2015) 2765–2774. doi:10.1021/acs.accounts.5b00160. Q. Fan, J.M. Gottfried, J. Zhu, Surface-Catalyzed C-C Covalent Coupling Strategies toward the Synthesis of Low-Dimensional Carbon-Based Nanostructures, Acc. Chem. Res. 48 (2015) 2484–2494. doi:10.1021/acs.accounts.5b00168. M. Bieri, M.T. Nguyen, O. Gröning, J. Cai, M. Treier, K. Aït-Mansour, P. Ruffieux, C.A. Pignedoli, D. Passerone, M. Kastler, K. Müllen, R. Fasel, Two-dimensional polymer formation on surfaces: Insight into the roles of precursor mobility and reactivity, J. Am. Chem. Soc. 132 (2010) 16669–16676. doi:10.1021/ja107947z. T. Niu, Surface strain mediated dipole alignment of ClAlPc on Au(111), Appl. Phys. Lett. 106 (2015) 161601. doi:10.1063/1.4918989. D. Écija, J.I. Urgel, A.C. Papageorgiou, S. Joshi, W. Auwärter, A.P. Seitsonen, S. Klyatskaya, M. Ruben, S. Fischer, S. Vijayaraghavan, J. Reichert, J. V. Barth, Five-vertex Archimedean surface tessellation by lanthanide-directed molecular self-assembly, Proc. Natl. Acad. Sci. 110 (2013) 6678–6681. doi:10.1073/pnas.1222713110. A. Wäckerlin, S. Fatayer, T. Nijs, S. Nowakowska, S.F. Mousavi, O. Popova, A. Ahsan, T.A. Jung, C. Wäckerlin, Molecular Chessboard Assemblies Sorted by Site-Specific Interactions of Out-of-Plane d-Orbitals with a Semimetal Template, Nano Lett. 17 (2017) 1956–1962. doi:10.1021/acs.nanolett.6b05344. P. Järvinen, S.K. Hämäläinen, K. Banerjee, P. Häkkinen, M. Ijäs, A. Harju, P. Liljeroth, Molecular Self-Assembly on Graphene on SiO 2 and h-BN Substrates, Nano Lett. 13 (2013) 3199–3204. doi:10.1021/nl401265f. T. Chao Niu, Y. Li Huang, J. Tao Sun, S. Kera, N. Ueno, A. Thye Shen Wee, W. Chen, Tunable two-dimensional molecular dipole dot arrays on graphite, Appl. Phys. Lett. 99 (2011) 14–17. doi:10.1063/1.3647563. Y. Makoudi, M. Beyer, J. Jeannoutot, F. Picaud, F. Palmino, F. Cherioux, Supramolecular self-assembly of brominated molecules on a silicon surface, Chem. Commun. 50 (2014) 5714–5716. doi:Doi 10.1039/C4cc01158a. Y. Ogawa, T. Niu, S.L. Wong, M. Tsuji, A.T.S. Wee, W. Chen, H. Ago, Self-assembly of polar phthalocyanine molecules on graphene grown by chemical vapor deposition, J. Phys. Chem. C. 117 (2013) 21849–21855. doi:10.1021/jp406681j. Y.L. Huang, W. Chen, F. Bussolotti, T.C. Niu, A.T.S. Wee, N. Ueno, S. Kera, Impact of molecule-dipole orientation on energy level alignment at the submolecular scale, Phys. Rev. B - Condens. Matter Mater. Phys. 87 (2013) 1–6. doi:10.1103/PhysRevB.87.085205. T. Hosokai, H. MacHida, A. Gerlach, S. Kera, F. Schreiber, N. Ueno, Impact of structural
CE
[1]
[13]
[14]
[15]
[16]
18 / 22
ACCEPTED MANUSCRIPT
[22]
[23]
[24] [25]
[26]
CR IP T
AC
[27]
AN US
[21]
M
[20]
ED
[19]
PT
[18]
CE
[17]
imperfections on the energy-level alignment in organic films, Phys. Rev. B - Condens. Matter Mater. Phys. 83 (2011) 1–7. doi:10.1103/PhysRevB.83.195310. P. Zalake, S. Ghosh, S. Narasimhan, K. George Thomas, Descriptor-Based Rational Design of Two-Dimensional Self-Assembled Nanoarchitectures Stabilized by Hydrogen Bonds, Chem. Mater. 29 (2017) 7170–7182. doi:10.1021/acs.chemmater.7b01183. L. Cai, Q. Sun, M. Bao, H. Ma, C. Yuan, W. Xu, Competition between Hydrogen Bonds and Coordination Bonds Steered by the Surface Molecular Coverage, ACS Nano. 11 (2017) 3727– 3732. doi:10.1021/acsnano.6b08374. M.J. Comstock, N. Levy, A. Kirakosian, J. Cho, F. Lauterwasser, J.H. Harvey, D.A. Strubbe, J.M.J. Fréchet, D. Trauner, S.G. Louie, M.F. Crommie, Reversible photomechanical switching of individual engineered molecules at a metallic surface, Phys. Rev. Lett. 99 (2007) 1–4. doi:10.1103/PhysRevLett.99.038301. N.A. Wasio, R.C. Quardokus, R.P. Forrest, C.S. Lent, S.A. Corcelli, J.A. Christie, K.W. Henderson, S.A. Kandel, Self-assembly of hydrogen-bonded two-dimensional quasicrystals, Nature. 507 (2014) 86–89. doi:10.1038/nature12993. X. Ye, J. Chen, M. Eric Irrgang, M. Engel, A. Dong, S.C. Glotzer, C.B. Murray, Quasicrystalline nanocrystal superlattice with partial matching rules, Nat. Mater. 16 (2017) 214–219. doi:10.1038/nmat4759. J.I. Urgel, D. Écija, G. Lyu, R. Zhang, C. Palma, W. Auwärter, N. Lin, J. V Barth, Quasicrystallinity expressed in two-dimensional coordination networks, Nat. Chem. 8 (2016) 657–662. doi:10.1038/nchem.2507. J. Shang, Y. Wang, M. Chen, J. Dai, X. Zhou, J. Kuttner, G. Hilt, X. Shao, J.M. Gottfried, K. Wu, Assembling molecular Sierpiński triangle fractals, Nat. Chem. 7 (2015) 389–393. doi:10.1038/nchem.2211. J.M. Macleod, F. Rosei, Molecular self-assembly on graphene, Small. 10 (2014) 1038–1049. doi:10.1002/smll.201301982. C. Riedl, C. Coletti, T. Iwasaki, A.A. Zakharov, U. Starke, Quasi-free-standing epitaxial graphene on SiC obtained by hydrogen intercalation, Phys. Rev. Lett. 103 (2009) 1–4. doi:10.1103/PhysRevLett.103.246804. P. Ruffieux, S. Wang, B. Yang, C. Sanchez, J. Liu, T. Dienel, L. Talirz, P. Shinde, C.A. Pignedoli, D. Passerone, T. Dumslaff, X. Feng, K. Muellen, R. Fasel, On-surface synthesis of graphene nanoribbons with zigzag edge topology, Nature. 531 (2015) 15. doi:10.1038/nature17151. J.M. Gottfried, Surface chemistry of porphyrins and phthalocyanines, Surf. Sci. Rep. 70 (2015) 259–379. doi:10.1016/j.surfrep.2015.04.001. H. Fukagawa, S. Hosoumi, H. Yamane, S. Kera, N. Ueno, Dielectric properties of polar-phthalocyanine monolayer systems with repulsive dipole interaction, Phys. Rev. B Condens. Matter Mater. Phys. 83 (2011) 1–8. doi:10.1103/PhysRevB.83.085304. Y.L. Huang, Y. Lu, T.C. Niu, H. Huang, S. Kera, N. Ueno, A.T.S. Wee, W. Chen, Reversible single-molecule switching in an ordered monolayer molecular dipole array, Small. 8 (2012) 1423–1428. doi:10.1002/smll.201101967. J.L. Zhang, T.C. Niu, A.T.S. Wee, W. Chen, Self-assembly of binary molecular nanostructure arrays on graphite, Phys. Chem. Chem. Phys. 15 (2013) 12414. doi:10.1039/c3cp00023k. J.L. Zhang, J.L. Xu, T.C. Niu, Y.H. Lu, L. Liu, W. Chen, Reversible switching of a
[28]
[29]
[30] [31]
19 / 22
ACCEPTED MANUSCRIPT
[38]
[39]
[40]
[41]
AC
[42] [43]
CR IP T
[37]
AN US
[36]
M
[35]
ED
[34]
PT
[33]
CE
[32]
single-dipole molecule imbedded in two-dimensional hydrogen-bonded binary molecular networks, J. Phys. Chem. C. 118 (2014) 1712–1718. doi:10.1021/jp408890k. J.M. Soler, E. Artacho, J.D. Gale, A. Garcia, J. Junquera, P. Ordejon, D. Sanchez-Portal, The SIESTA method for ab initio order-N materials simulation, J. Phys. Condens. Matter. 14 (2001) 2745. doi:10.1088/0953-8984/14/11/302. G. Román-Pérez, J.M. Soler, Efficient implementation of a van der waals density functional: Application to double-wall carbon nanotubes, Phys. Rev. Lett. 103 (2009) 1–4. doi:10.1103/PhysRevLett.103.096102. M. Dion, H. Rydberg, E. Schröder, D.C. Langreth, B.I. Lundqvist, Van der Waals density functional for general geometries, Phys. Rev. Lett. 92 (2004) 22–25. doi:10.1103/PhysRevLett.92.246401. R. Coquet, G.J. Hutchings, S.H. Taylor, D.J. Willock, Calculations on the adsorption of Au to MgO surfaces using SIESTA, J. Mater. Chem. 16 (2006) 1978. doi:10.1039/b601213b. A.S. Foster, M.A. Gosálvez, T. Hynninen, R.M. Nieminen, K. Sato, First-principles calculations of Cu adsorption on an H-terminated Si surface, Phys. Rev. B - Condens. Matter Mater. Phys. 76 (2007) 1–8. doi:10.1103/PhysRevB.76.075315. A. Hauschild, K. Karki, B.C.C. Cowie, M. Rohlfing, F.S. Tautz, M. Sokolowski, Molecular distortions and chemical bonding of a large π-conjugated molecule on a metal surface, Phys. Rev. Lett. 94 (2005) 1–4. doi:10.1103/PhysRevLett.94.036106. Y.L. Zhao, W. Wang, F. Qi, J.F. Li, G. Kuang, R.Q. Zhang, N. Lin, M.A. Van Hove, Donor/acceptor properties of aromatic molecules in complex metal-molecule interfaces, Langmuir. 33 (2017) 451–458. doi:10.1021/acs.langmuir.6b02649. M. Huang, Y.L. Zhao, W. Xiong, S. V. Kershaw, Y. Yu, W. Li, T. Dudka, R.Q. Zhang, Collaborative enhancement of photon harvesting and charge carrier dynamics in carbon nitride photoelectrode, Appl. Catal. B Environ. 237 (2018) 783–790. doi:10.1016/j.apcatb.2018.06.046. Y.L. Zhao, R. Zhao, F. Qi, R.Q. Zhang, M.A. Van Hove, Substrate-mediated and temperature-modulated long-range interactions between bromine adatom stripes on Cu(1 1 1), Appl. Surf. Sci. 463 (2019) 253–260. doi:10.1016/j.apsusc.2018.08.168. E.R. Johnson, S. Keinan, P. Mori Sánchez, J. Contreras García, A.J. Cohen, W. Yang, NCI : revealing non-covalent interactions, J. Am. Chem. Soc. 132 (2010) 6498–6506. doi:10.1021/ja100936w. M.J. et al Frisch, Gaussian 09, Revision C.01, Gaussian 09, Revis. C.01. (2009). T. Lu, F. Chen, Multiwfn: A multifunctional wavefunction analyzer, J. Comput. Chem. 33 (2012) 580–592. doi:10.1002/jcc.22885. T. Niu, J. Zhang, W. Chen, Molecular ordering and dipole alignment of vanadyl phthalocyanine monolayer on metals: The effects of interfacial interactions, J. Phys. Chem. C. 118 (2014) 4151–4159. doi:10.1021/jp4101653. A.T.S. Wee, W. Chen, Molecular interactions on epitaxial graphene, Phys. Scr. T146 (2012) 014007. doi:10.1088/0031-8949/2012/T146/014007. Y.L. Huang, R. Wang, T.C. Niu, S. Kera, N. Ueno, J. Pflaum, A.T.S. Wee, W. Chen, One dimensional molecular dipole chain arrays on graphite via nanoscale phase separation, Chem. Commun. 46 (2010) 9040. doi:10.1039/c0cc03251d. J.P. Wagner, P.R. Schreiner, London Dispersion in Molecular Chemistry - Reconsidering
[44]
[45] [46]
[47]
20 / 22
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Steric Effects, Angew. Chemie - Int. Ed. 54 (2015) 12274–12296. doi:10.1002/anie.201503476. [48] F. London, The general theory o f molecular forces., Trans. Faraday Soc. 33 (1937) 8–26. doi:10.1039/tf937330008b. [49] S. Zhong, J.Q. Zhong, X.Z. Wang, M.Y. Huang, D.C. Qi, Z.K. Chen, W. Chen, Investigation of interface properties for ClAlPc/C 60 heterojunction-based inverted organic solar cell, J. Phys. Chem. C. 116 (2012) 2521–2526. doi:10.1021/jp210533n. [50] H. Huang, S.L. Wong, J. Sun, W. Chen, A.T.S. Wee, Trapping single polar molecules in SiC nanomesh via out-of-plane dipoles, ACS Nano. 6 (2012) 2774–2778. doi:10.1021/nn300258b. [51] M. Alonso, T. Woller, F.J. Martín-Martínez, J. Contreras-García, P. Geerlings, F. De Proft, Understanding the fundamental role of dispersion interactions in shaping carbon-based materials, Chem. - A Eur. J. 20 (2014) 4931–4941. doi:10.1002/chem.201400107.
21 / 22
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Graphic Abstract
22 / 22
Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite Hai-yang Ma , Yan-ling Zhao , Myo Win Zaw , Jin-feng Jia , Rui-qin Zhang , Michel A. Van Hove PII: DOI: Reference:
S0039-6028(18)30495-3 https://doi.org/10.1016/j.susc.2018.11.010 SUSC 21382
To appear in:
Surface Science
Received date: Revised date: Accepted date:
10 June 2018 14 November 2018 15 November 2018
Please cite this article as: Hai-yang Ma , Yan-ling Zhao , Myo Win Zaw , Jin-feng Jia , Rui-qin Zhang , Michel A. Van Hove , Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite, Surface Science (2018), doi: https://doi.org/10.1016/j.susc.2018.11.010
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Highlights
AC
CE
PT
ED
M
AN US
CR IP T
The unusual ClAlPc molecular self-assembled layer on a buffer layer have large deformations compared with the common ClAlPc bilayer Almost freestanding environment is provided for the molecules when the influence from the substrate is shielded by the buffer layer The dominated intermolecular interaction is likely to be the dispersion forces via Reduced Density Gradient analysis
1 / 22
ACCEPTED MANUSCRIPT
Unusual Self-assembly of Chloroaluminium Phthalocyanine on Graphite Hai-yang Ma a,b,*, Yan-ling Zhao b,*, Myo Win Zaw b, Jin-feng Jia a,# and Rui-qin Zhang b,#, Michel A. Van Hove c Key Laboratory of Artificial Structures and Quantum Control (Ministry of Education), School of
CR IP T
a
Physics and Astronomy, Shanghai Jiao Tong University, Shanghai 200240, China Department of Physics, City University of Hong Kong, Hong Kong, China
c
Institute of Computational and Theoretical Studies & Department of Physics, Hong Kong Baptist
AN US
b
University, Hong Kong, China
M
ABSTRACT
We report an unusual self-assembled layer structure of chloroaluminium phthalocyanine (ClAlPc)
ED
molecules on highly ordered pyrolytic graphite (HOPG), in which a close-packed well-ordered
PT
monolayer is separated from the substrate by a relatively disordered buffer layer, as revealed using scanning tunneling microscopy (STM). Our close-packed monolayer has a nearly rectangular lattice,
CE
instead of the distinctly different square lattice for the more commonly observed well-ordered bilayer
AC
structure. This may be due to the dominance of intermolecular interactions within the monolayer when the influence from the substrate is shielded by the buffer layer. Density Functional Theory (DFT) calculations and Reduced Density Gradient (RDG) analysis indicate that the dominant intermolecular interaction within the unusual layer is likely the London dispersion force. KEYWORDS: Weak interactions, dispersion forces, 2D self-assembly
2 / 22
ACCEPTED MANUSCRIPT
1. Introduction
Two-dimensional (2D) molecular self-assembled layers have been widely studied in recent years due to their potential applications in molecular electronics[1–4] and surface science[5,6]. Different
CR IP T
molecules have been studied on various substrates such as metals (Au, Ag, Cu)[7–10] and semiconductors (graphite, Si, BN)[11–13] with various methods[14–16]. Chemical bonding and physical adsorption are intrinsic driving forces for the formation of structures exhibiting long-range order[8,17,18], influenced by temperature[14], light[19], electric fields[4], etc. One may carefully
AN US
select both the substrate and the molecule to achieve particular results, such as two-dimensional quasi-crystal structures[20–22] or Sierpiński triangle fractals[23]. The substrate, which is essential to support the molecules and control the morphology of the surface layer, may also strongly alter their
M
physical and chemical properties[8,18]. One alternative way to reduce the impact of the substrate on
ED
the molecules is to introduce an “inert buffer layer”. Graphene[24] or intercalated layers[25] often play this role: on these supports nearly freestanding molecular layers are possible, but sometimes the
PT
binding within the molecular layer is too strong for the molecular layer to be stable[26].
CE
The chloroaluminium phthalocyanine molecule belongs to the phthalocyanine family[27] with the
AC
central metal ion being aluminum (Al) connected to a chlorine (Cl) atom pointing out from the dominant plane of the molecule. It is a polar molecule with permanent dipole moment perpendicular to the π-conjugation plane; charge transfer occurs between the molecule and some substrates after deposition[15,16,28]. Depending on the environment, the orientation of surface-adsorbed ClAlPc can be either Cl-up (Cl pointing to the vacuum) or reversed to Cl-down (Cl pointing to the substrate) [4,14]. This property may be applied to single-molecule-based storage[29]. ClAlPc may also be 3 / 22
ACCEPTED MANUSCRIPT
co-deposited with other molecules and interesting self-assembled structures can be obtained by controlling the relative amounts of different molecules[30,31]. In this work, we investigate the self-assembly of ClAlPc molecules on HOPG with a preformed buffer layer, by means of scanning tunneling microscopy (STM) and density functional theory (DFT)
CR IP T
calculations. Previous experimental work has extensively studied both weak and strong interactions (from hydrogen bonds to coordination bonds)[18] in all sorts of molecular self-assemblies. However, to our knowledge, very little attention has been paid to even weaker interactions, such as the
AN US
dispersion force. This may be due to two factors: (1) it is difficult to achieve an almost freestanding or isolated environment; and (2) weaker interactions may have little effect on the final structures. In our experiments, we find that ClAlPc molecules on a buffer layer fulfill the two conditions: we
M
observe large deformations caused by the dominant weak molecular interactions. We propose one possible model for the newly found self-assembly. Our theoretical and experimental results are
ED
highly consistent. Further theoretical analyses by the Reduced Density Gradient (RDG) method
PT
suggest that the dominant contributions to the interactions are the dispersion forces. It is remarkable that although the intermolecular attractions are weak (< 100 meV), they should have such significant
CE
effects on the self-assemblies.
AC
2. Experimental and theoretical methods
STM experiments were performed with an Omicron low temperature (liquid nitrogen, 78 K),
combined with a molecular beam epitaxy (MBE) system in a base pressure of about 5 × 10-10 mbar. Freshly cleaved HOPG was inserted to the vacuum chamber immediately, followed by degassing at ~703 K overnight. Commercially available chloroaluminium phthalocyanine (Sigma Adrich) was 4 / 22
ACCEPTED MANUSCRIPT
then thermally evaporated for 60 seconds at ~693 K onto the clean HOPG, which is kept at room temperature. Before deposition, the molecular source was degassed at an elevated temperature for ~3 hours, followed by pre-heating at growth temperature until the pressure approached ~2 × 10-9 mbar. This purifying procedure was similar to those described in previous reports and a purity higher than
CR IP T
99% was achieved by statistically analyzing many STM images[4,29,30]. Several samples have been prepared. All the STM topographs were recorded in constant-current mode. A mechanically sheared Pt-Ir tip was used.
AN US
Calculations were carried out utilizing the atomic orbital-based DFT method implemented in the SIESTA code[32]. To study the weak intermolecular interactions, we adopted the Van der Waals (VdW) functional[33,34], a universal nonlocal energy functional of the electron density, with a
M
double zeta plus polarization orbital basis set. Other parameters selected in the calculations include an energy cutoff of 300 Ry, a maximal force threshold of 0.02 eV/Å, a projected atomic orbital
ED
energy shift of 10 meV, and a maximal displacement tolerance of 1.0 × 10-4 Å. Note that SIESTA has
PT
been successfully applied in many simulations of 2D surfaces[35–40]. For the RDG[41] analysis the electron density was obtained from DFT calculations which were carried out at the level of
CE
ωB97XD/6-31+G* implemented in the Gaussian 09 package[42]. The spatial distribution of RDG
AC
was visualized by using a multifunctional wave-function analyzer[43].
3. Observations
An HOPG surface is relatively inactive compared with more widely used noble metal substrates like gold, silver and copper[44]. The relatively weak influence of HOPG on the molecules make this substrate an ideal candidate, allowing for molecular self-assembly via strong or weak intermolecular 5 / 22
ACCEPTED MANUSCRIPT
interactions[45]. When we deposit less than 0.1 monolayer of ClAlPc on the substrate (here one monolayer refers to one close-packed ClAlPc layer with π-plane parallel to the HOPG surface[29]), the molecules tend to form bilayer islands[28], as is also the case on graphene[14]. For these commonly observed bilayer islands, the molecules in the first layer, counting the layers from the bare
CR IP T
HOPG, lie flat on HOPG with close-packed structures, forming nearly square lattices, see refs. [4,29,31,46] and Fig. 1a. We observe regions with less ordering, both within the layer and around the edges, than reported in earlier work. These regions have monolayer thickness, cf. Fig. S1. In some
AN US
regions, this relatively disordered layer is “covered” by a more orderly outer layer. STM cannot directly probe the disordered layer underlying, but one can indirectly determine its existence though the uncovered region, such as the top area of Fig. 2(d). The more orderly outer layer exhibits a different periodic structure than the usual bilayer mentioned above, cf. Fig. 2(c) vs. 2(f). We shall
M
call this relatively disordered layer a “buffer layer” between the HOPG and the ordered outer layer
ED
an “unusual layer” in these “unusual islands”. The unusual layer can no longer be found after
PT
annealing the sample at ~423 K, while other layers are still there. Molecule-resolved STM
Fig. 1.
CE
topographs of monolayer ClAlPc on bare HOPG and the unusual layer on buffer layer are shown in
AC
With STM we can easily distinguish the unusual islands from the common bilayer islands. In Fig. 2(c), the square motif is characteristic of the surface morphologies of the bilayer islands[14,46], while in Fig. 2(f), a motif of “six-membered rings” is observed in the unusual islands. The differences are also clear in the Fourier Transform (FT) images of the molecule-resolved STM topographs as shown in insets of Fig. 2(c), (f). Furthermore, from the line profiles Fig. 2(b) and (e) we know that the thicknesses of the common bilayer and unusual islands are ~550 and ~700 pm 6 / 22
ACCEPTED MANUSCRIPT
respectively. A height difference of ~150 pm between the bilayer and unusual islands is obtained, while the monolayer and buffer layer have the same thickness of ~250 pm, cf. Fig. S1(c). The 2D lattice parameters are measured to be |a1| = 1.46 ± 0.02 nm and |a2| = 1.69 ± 0.02 nm with an angle γ = 92 ± 1° for the unusual islands, while for the bilayer islands, |a1| = |a2| = 1.50 ± 0.02 nm with γ =
CE
PT
ED
M
AN US
CR IP T
90 ± 1°, which is consistent with other reports[29,46].
AC
Fig. 1. Structures of a molecular monolayer of ClAlPc self-assembled on bare HOPG, (a) and (b), and on a buffer layer, (c) and (d). (a) Molecule-resolved STM topograph. Vsample = 2.0 V, Iset = 150 pA. The appearance of the dumbbells in only one direction may due to bias dependence[46] or tip effects. (b) Top view and side view of a model based on DFT optimizations. (c) and (d) As (a) and (b), but for the unusual layer on a buffer layer. The molecular models and dashed rhomboid in (c) 7 / 22
ACCEPTED MANUSCRIPT
indicate the unit cell. Scanning parameters for (c) are Vsample = 2.2 V, Iset = 33 pA. The optimized lattice parameters are square with |a1| = |a2| = 1.50 nm for model in (b) and |a1| = 1.40 nm, |a2| = 1.60
AC
CE
PT
ED
M
AN US
CR IP T
nm, the angle γ between a1 and a2 is 93° for model in (d). Image size for (a) and (c): 7.8 × 8.2 nm2.
Fig. 2. STM topographs of the two types of ClAlPc islands. (a), (b) and (c) show usual (“common”) 8 / 22
ACCEPTED MANUSCRIPT
bilayer islands. (d), (e) and (f) show unusual islands. (a) A bilayer island on HOPG, where the second layer, first layer and bare HOPG are seen from left to right. Image size: 78.1 × 81.7 nm2. Vsample = 3.0 V, Iset = 33 pA. (d) Unusual island on a buffer layer surrounded by bilayer islands. A rougher buffer layer is also visible at the top. Image size: 78.1 × 81.7 nm2. Vsample = 2.2 V, Iset = 33 pA. (b) and (e)
CR IP T
Corresponding line profiles along the solid lines in (a) and (d). (c) Molecule-resolved topograph of the bilayer islands on bare HOPG. The nearly square lattice is clear from the inset FT image. The bright spots are impurities while dark holes are vacancies. Image size: 31.2 × 32.7 nm2. Vsample = 1.8
AN US
V, Iset = 33 pA. (f) High-resolution topograph of an unusual island when a buffer layer is present. Blue hexagons specify the “six-membered ring” structures. In the inset FT images, missing spots which may due to glide symmetry are marked by orange dotted circles. The lattice vectors are
M
labeled as a1 and a2. Image size: 23.4 × 24.5 nm2. Vsample = 2.2 V, Iset = 33 pA. Some vertical stripes can be observed in (d) and (f)which may be instrumental (as they are systematically parallel to one
PT
ED
image side and have equal width in pixels regardless of image size, i.e. instrumental magnification).
Different contrasts are also observed when the unusual self-assembled islands are bounded by
CE
different surroundings, while the surface morphologies of the bilayer islands are quite uniform (Fig.
AC
2 and S2).For all the adjacent areas, we observe that the lattices of the two types of islands (Fig. S3(a)) are nearly aligned to each other (within about 5°), while the orientations (in-plane) of the molecules seen in Fig. S3 differ by 36 ± 1.2°. On the islands bounded by the bilayer islands, one may see density contrasts (Fig. 2(d) and 2(f)). When the islands are isolated, the contrasts become less pronounced and less sharp (Fig. S2(d)); they tend to be invisible for small bounded islands (Fig. S3(a)). 9 / 22
ACCEPTED MANUSCRIPT
4. Models
The above observations contain much information. First, in Fig. 1(c), we outline the unit cell of the unusual self-assembled islands by a yellow rhomboid. This can be done by identifying the
CR IP T
molecules along the boundaries of the two kinds of islands like in Fig. S3(a). Each corner of the rhomboid is surrounded by four bright spots which can be interpreted as the four phenyl groups of one ClAlPc molecule. We note that the center of the molecule at the rhomboid corner appears in STM to be identical to the “hole” between four molecules at the rhomboid center, suggesting that the
AN US
center of the molecule is not visibly imaged; this strongly suggests that the Cl atoms point down toward the substrate. Thus, the orientation of molecules on the buffer layer is most likely Cl-down (as we will also confirm later by comparing observations and calculations). By contrast, the bright
ED
HOPG are in Cl-up configuration.
M
spots in the center of the molecules in Fig. 1(a) suggest that monolayer ClAlPc molecules on bare
Second, in Fig. 2(f) we can interpret the “six-membered rings” to be rows of parallel arranged
PT
ClAlPc molecules: four of the six corners of the ring form a square, corresponding to one molecule.
CE
We circle some missing or very weak spots in the inset FT image with orange dotted circles which
AC
may be attributed to the apparent glide-plane symmetry of the STM image. Compared with the bilayer structure, the lattice of the unusual layer is compressed by ~3% in the a1 direction while it is stretched by ~13% in the a2 direction. The ClAlPc molecules appear to be intact, based on their size and appearance in STM. These considerable lattice deformations together with the changed orientations of the molecules indicate that HOPG and buffer layers have quite different effects on ClAlPc molecular self-assembly. Third, we note that the molecules on a buffer layer get closer 10 / 22
ACCEPTED MANUSCRIPT
together along one direction, which suggests that the influence of the buffer layer is weaker than that of HOPG. The tight molecular packing may bring about tilting of the molecules and may be the cause of the contrasts in STM topographs noted above. Meanwhile, the ClAlPc molecules can easily leave or move along the HOPG as well as the buffer layers, since no isolated molecule has been
CR IP T
observed on the surface; this implies that the binding of the molecules to the buffer layer is at least as weak as to HOPG. Additionally, due to the more random orientations of the molecules on the rougher buffer layer and the disorder within the buffer layer, we expect that on average the in-plane
AN US
interactions would be stronger than the out-of-plane interactions.
Building on the above observations and interpretations, we can propose a likely model in which the buffer layer provides a relatively freestanding environment for molecules on top of them, see Fig.
M
3(c), for our newly found unusual self-assembly of ClAlPc on HOPG.
Before providing additional support for our model, we analyze the possible driving forces that lead
ED
to the structures and describe the self-assembly processes with the above model in mind. Since no
PT
external effects and external chemical bonds are included here, we will only consider the intermolecular VdW forces. We discuss the three kinds of VdW forces:
CE
(1) Dipole-dipole. The hydrogen bonds should be excluded first since the intermolecular N-H
AC
atom pairs are too far apart (~3.5 Å). Furthermore, although this molecule has an intrinsic dipole moment perpendicular to the plane of the molecule, the molecules remain close to parallel to the layer surface. The parallel arrangements of the dipole moments in fact produce repulsive forces between the molecules, which reduces the tendency to form close-packed islands, contrary to our observations. (2) Dipole-induced dipole. The symmetry of this molecule implies no in-plane polarization, so 11 / 22
ACCEPTED MANUSCRIPT
that no dipole-induced dipole is expected. (3) Instantaneous dipoles. London dispersion forces are the attractive part of VdW forces[47,48]. These attractive forces can arise from structural deformations of molecules since unbalanced instantaneous dipoles may arise along some directions where symmetries were perturbed,
AC
CE
PT
ED
M
AN US
dispersion forces are the dominant interactions.
CR IP T
which is in line with the tilts we have proposed above. Therefore, we conclude that London
Fig. 3. Schematics of two types of self-assembly. (a) Bottom view and side view of a ClAlPc molecule. (b) The bilayer islands on HOPG with staggered stacking. (c) Unusual self-assembly of ClAlPc on a buffer layer over HOPG; the molecules may tilt due to the relatively strengthened intermolecular interactions and the corrugation of the buffer layer.
12 / 22
ACCEPTED MANUSCRIPT
5. Calculations and discussions
We perform theoretical calculations based on the previous assumption of a freestanding monolayer minimally affected by the buffer layer, i.e. no substrate or buffer layer is included. A 20 Å vacuum
CR IP T
layer is added along the c (out-of-plane) direction in a 3D periodic scheme. At first, we calculate the partial density of states (PDOS) of each element using standard DFT methods, from which we can see that the PDOS of carbon atoms dominates around the Fermi level (cf. Fig. S4 in Supplementary material). This indicates that the bright spots in Fig. 1(c) represent the phenyl rings (C atoms) since
AN US
the bias voltages applied are within the corresponding energy range. Combining with the STM observation of holes in the center of the molecules, we model the molecules with Cl-down configuration. It is clear in Fig. 2(c) that the molecules within the bilayer are also in Cl-down;
M
however, the reasons for this common orientation are different. For the unusual layers, the cause is
ED
the relatively larger corrugation (Fig. S1(c)) of the buffer layers. Ordered structures with ClAlPc molecules lying flat and in Cl-up are difficult to form if the substrate is not flat enough at the same
PT
time [49]. However, unevenness and disorder of the layers below does not guarantee the
CE
disappearance of ordered structures on top of them[50], ClAlPc molecules can adopt the Cl-down
AC
configuration to accommodate the corrugations. But for the bilayers, to balance the dipole moments, the second layer will stack itself onto the underlying layer with reversed dipole directions[14,15,46]. This staggered packing may be the reason why the heights of the bilayer islands are lower than the unusual ones. Next, we perform total energy calculations to determine the most reasonable lattice size and shape; the details are listed in table S1. At the same time, we also calculate the adsorption energy of a single 13 / 22
ACCEPTED MANUSCRIPT
molecule on graphene with the Cl-up configuration to give quantitative evaluations of the molecule-substrate interactions, cf. below. GGA/PBE calculations are also performed to show the existence of dispersion interaction in close-packed molecular models. First, among all the cases we have considered, VdW calculations show that a slightly tilted structure (< 5°) with lattice size 14 ×
CR IP T
16 Å2 and γ = 93° is the most stable. The optimized lattice size and the ratio |a1|/|a2| = 1.14 matches the experimental observations well, with only slight deviations < 5.4％ for all the parameters. The calculated molecular tilts confirm our previous suggestions. Second, in comparison with fixed
AN US
out-of-plane displacements within the molecule, the total energy lowering by the tilting is estimated to be ~90 meV per molecule. To evaluate the effects from HOPG, we also calculate the total energies using parameters from experiments, i.e. a square lattice with size 15 × 15 Å2 (while still excluding
M
the substrate; the local density approximation （ LDA ） functionals may underestimate the intermolecular distances, which are here ignored). The energy increases almost by the same amount
ED
as it is reduced by the tilt and both changes are much smaller than the binding energy of π-π
PT
interactions (~4 eV, as estimated by our VDW/DRSLL calculations) between the ClAlPc molecule and graphene, which shows that the intermolecular interactions are rather weak. Additionally, the
CE
relative rotation angle (about the surface normal) of the two self-assemblies in Fig. 1(b) and (d) is
AC
evaluated as 37.4°, which matches the experimental result well. These systematic theoretical results which are highly consistent with experiments show that HOPG does impose relatively strong constraints on the ClAlPc molecules and that these molecules will reassembly into a more stable pattern when the almost freestanding environment is provided by the buffer layers. Our model can completely describe the experimental observations. To further verify that the dispersion forces dominate the intermolecular interactions, we analyze 14 / 22
ACCEPTED MANUSCRIPT
the electron structure with the Reduced Density Gradient (RDG) method, where RDG is defined as RDG = 1/(2(3π2 )1/3 )|∇𝜌|/𝜌4/3 : this is a very powerful method to both identify and visualize weak interactions between molecules[41]. In Fig. 4(a) and (b) we plot RDG as a function of the electron density multiplied by the sign of the second Hessian eigenvalue (λ2) obtained for the optimized
CR IP T
structures from our DFT calculations; by multiplying with sign (λ2) we can distinguish the bonded (λ2 < 0) from unbonded interactions (λ2 > 0). A density cutoff ρ < 0.05 au was chosen to depict the
AC
CE
PT
ED
M
AN US
noncovalent region of interest. The two images are quite similar in the range 0.005 < ρ < 0.05 au,
Fig. 4. Reduced density gradient (RDG) analyses of ClAlPc molecular interactions using the structures obtained from our DFT calculations. (a) and (b) Graphs of RDG versus the electron density ρ multiplied by the sign of the second Hessian eigenvalue, sign(λ2), for usual lattice without 15 / 22
ACCEPTED MANUSCRIPT
tilt and the unusual tilted one. The pairs of numbers at the top right give the size of the unit cell in Å. Red rectangles highlight the areas of attractive VdW interaction. (c) and (d) Corresponding isovalue surfaces in real space taken at RDG = 0.8 au. The range of the signed electron density is designated by the colored scale bar. The distribution of the intermolecular dispersion forces (light green) is
CR IP T
pointed out by red circles; no hydrogen bonding (dark green) is observed around the N atoms.
which maps to relatively stronger noncovalent interactions including hydrogen bonding (λ2 < 0) or
AN US
steric clashes (λ2 > 0). However, in the regions that cover the attractive and weak London dispersion forces (ρ < 0.005 au and λ2 < 0)[51], more points appear for the tilted structures than the usual one, especially for the area closest to zero, as highlighted by the red rectangles. These changes can be
M
visualized directly from Fig. 4(c) and (d), showing the isovalue surfaces of the RDG distribution in real space. A clear increase of dispersion forces is seen in the intermolecular areas along the
ED
compressed direction, or a1 direction, and around the nearby intermolecular C and H atoms. Besides,
PT
no sign of hydrogen bonding (which would show up as dark green) is found around the N atoms, in
CE
agreement with the analysis in the last section.
AC
6. Conclusions
Based on molecule-resolved STM topographs and DFT calculations, we have investigated the
weak intermolecular interactions, possibly due to dispersion forces, which are induced in unusual self-assemblies of ClAlPc on HOPG. A buffer layer between the HOPG and the self-assembled unusual layer appears to nearly remove the periodic influence from the graphite and provide an almost freestanding environment for the self-assembled layers. Future experiments can aim to build 16 / 22
ACCEPTED MANUSCRIPT
well-designed buffer layers to free the molecules, thereby enabling the fragile intermolecular interactions to manifest their roles while finding additional interesting self-assembled structures and applications. Our ideas can also be extended to other fields where weak interactions must be taken
Corresponding Authors: Prof. Rui-qin Zhang, Email: [email protected]
#
Prof. Jin-feng Jia, Email: [email protected]
AN US
#
CR IP T
into consideration.
Author Contributions:
M
*These authors contributed equally to this work.
Conflict of interest
Acknowledgements
PT
ED
The authors declare no competing financial interest.
CE
This work was supported by the Collaborative Research Fund of Hong Kong Research Grants
AC
Council (Project No. C2014-15G). We also acknowledge the computing resources of the Tianhe2-JK cluster at the Beijing Computational Science Research Center and of the Tianhe2 cluster at the National Supercomputer Center in Guangzhou, China. ICTS is supported by the Institute of Creativity, which is sponsored by Hung Hin Shiu Charitable Foundation (孔憲紹慈善基金贊助).
References 17 / 22
ACCEPTED MANUSCRIPT
[7]
[8] [9]
[10]
[11]
AC
[12]
CR IP T
[6]
AN US
[5]
M
[4]
ED
[3]
PT
[2]
R. Otero, J.M. Gallego, A.L.V. de Parga, N. Martín, R. Miranda, Molecular Self-Assembly at Solid Surfaces, Adv. Mater. 23 (2011) 5148–5176. doi:10.1002/adma.201102022. A. Kumar, K. Banerjee, P. Liljeroth, Molecular Assembly on Two-Dimensional Materials, Nanotechnology. 28 (2016) 082001. doi:10.1088/1361-6528/aa564f. D. Xiang, X. Wang, C. Jia, T. Lee, X. Guo, Molecular-Scale Electronics: From Concept to Function, Chem. Rev. 116 (2016) 4318–4440. doi:10.1021/acs.chemrev.5b00680. T. Niu, Electric-Field-Induced Molecular Switch of Single Dipolar Phthalocyanine on Cu(111): A Scanning Tunneling Microscopy Study, J. Phys. Chem. C. 119 (2015) 19802– 19807. doi:10.1021/acs.jpcc.5b04318. L. Dong, P.N. Liu, N. Lin, Surface-Activated Coupling Reactions Confined on a Surface, Acc. Chem. Res. 48 (2015) 2765–2774. doi:10.1021/acs.accounts.5b00160. Q. Fan, J.M. Gottfried, J. Zhu, Surface-Catalyzed C-C Covalent Coupling Strategies toward the Synthesis of Low-Dimensional Carbon-Based Nanostructures, Acc. Chem. Res. 48 (2015) 2484–2494. doi:10.1021/acs.accounts.5b00168. M. Bieri, M.T. Nguyen, O. Gröning, J. Cai, M. Treier, K. Aït-Mansour, P. Ruffieux, C.A. Pignedoli, D. Passerone, M. Kastler, K. Müllen, R. Fasel, Two-dimensional polymer formation on surfaces: Insight into the roles of precursor mobility and reactivity, J. Am. Chem. Soc. 132 (2010) 16669–16676. doi:10.1021/ja107947z. T. Niu, Surface strain mediated dipole alignment of ClAlPc on Au(111), Appl. Phys. Lett. 106 (2015) 161601. doi:10.1063/1.4918989. D. Écija, J.I. Urgel, A.C. Papageorgiou, S. Joshi, W. Auwärter, A.P. Seitsonen, S. Klyatskaya, M. Ruben, S. Fischer, S. Vijayaraghavan, J. Reichert, J. V. Barth, Five-vertex Archimedean surface tessellation by lanthanide-directed molecular self-assembly, Proc. Natl. Acad. Sci. 110 (2013) 6678–6681. doi:10.1073/pnas.1222713110. A. Wäckerlin, S. Fatayer, T. Nijs, S. Nowakowska, S.F. Mousavi, O. Popova, A. Ahsan, T.A. Jung, C. Wäckerlin, Molecular Chessboard Assemblies Sorted by Site-Specific Interactions of Out-of-Plane d-Orbitals with a Semimetal Template, Nano Lett. 17 (2017) 1956–1962. doi:10.1021/acs.nanolett.6b05344. P. Järvinen, S.K. Hämäläinen, K. Banerjee, P. Häkkinen, M. Ijäs, A. Harju, P. Liljeroth, Molecular Self-Assembly on Graphene on SiO 2 and h-BN Substrates, Nano Lett. 13 (2013) 3199–3204. doi:10.1021/nl401265f. T. Chao Niu, Y. Li Huang, J. Tao Sun, S. Kera, N. Ueno, A. Thye Shen Wee, W. Chen, Tunable two-dimensional molecular dipole dot arrays on graphite, Appl. Phys. Lett. 99 (2011) 14–17. doi:10.1063/1.3647563. Y. Makoudi, M. Beyer, J. Jeannoutot, F. Picaud, F. Palmino, F. Cherioux, Supramolecular self-assembly of brominated molecules on a silicon surface, Chem. Commun. 50 (2014) 5714–5716. doi:Doi 10.1039/C4cc01158a. Y. Ogawa, T. Niu, S.L. Wong, M. Tsuji, A.T.S. Wee, W. Chen, H. Ago, Self-assembly of polar phthalocyanine molecules on graphene grown by chemical vapor deposition, J. Phys. Chem. C. 117 (2013) 21849–21855. doi:10.1021/jp406681j. Y.L. Huang, W. Chen, F. Bussolotti, T.C. Niu, A.T.S. Wee, N. Ueno, S. Kera, Impact of molecule-dipole orientation on energy level alignment at the submolecular scale, Phys. Rev. B - Condens. Matter Mater. Phys. 87 (2013) 1–6. doi:10.1103/PhysRevB.87.085205. T. Hosokai, H. MacHida, A. Gerlach, S. Kera, F. Schreiber, N. Ueno, Impact of structural
CE
[1]
[13]
[14]
[15]
[16]
18 / 22
ACCEPTED MANUSCRIPT
[22]
[23]
[24] [25]
[26]
CR IP T
AC
[27]
AN US
[21]
M
[20]
ED
[19]
PT
[18]
CE
[17]
imperfections on the energy-level alignment in organic films, Phys. Rev. B - Condens. Matter Mater. Phys. 83 (2011) 1–7. doi:10.1103/PhysRevB.83.195310. P. Zalake, S. Ghosh, S. Narasimhan, K. George Thomas, Descriptor-Based Rational Design of Two-Dimensional Self-Assembled Nanoarchitectures Stabilized by Hydrogen Bonds, Chem. Mater. 29 (2017) 7170–7182. doi:10.1021/acs.chemmater.7b01183. L. Cai, Q. Sun, M. Bao, H. Ma, C. Yuan, W. Xu, Competition between Hydrogen Bonds and Coordination Bonds Steered by the Surface Molecular Coverage, ACS Nano. 11 (2017) 3727– 3732. doi:10.1021/acsnano.6b08374. M.J. Comstock, N. Levy, A. Kirakosian, J. Cho, F. Lauterwasser, J.H. Harvey, D.A. Strubbe, J.M.J. Fréchet, D. Trauner, S.G. Louie, M.F. Crommie, Reversible photomechanical switching of individual engineered molecules at a metallic surface, Phys. Rev. Lett. 99 (2007) 1–4. doi:10.1103/PhysRevLett.99.038301. N.A. Wasio, R.C. Quardokus, R.P. Forrest, C.S. Lent, S.A. Corcelli, J.A. Christie, K.W. Henderson, S.A. Kandel, Self-assembly of hydrogen-bonded two-dimensional quasicrystals, Nature. 507 (2014) 86–89. doi:10.1038/nature12993. X. Ye, J. Chen, M. Eric Irrgang, M. Engel, A. Dong, S.C. Glotzer, C.B. Murray, Quasicrystalline nanocrystal superlattice with partial matching rules, Nat. Mater. 16 (2017) 214–219. doi:10.1038/nmat4759. J.I. Urgel, D. Écija, G. Lyu, R. Zhang, C. Palma, W. Auwärter, N. Lin, J. V Barth, Quasicrystallinity expressed in two-dimensional coordination networks, Nat. Chem. 8 (2016) 657–662. doi:10.1038/nchem.2507. J. Shang, Y. Wang, M. Chen, J. Dai, X. Zhou, J. Kuttner, G. Hilt, X. Shao, J.M. Gottfried, K. Wu, Assembling molecular Sierpiński triangle fractals, Nat. Chem. 7 (2015) 389–393. doi:10.1038/nchem.2211. J.M. Macleod, F. Rosei, Molecular self-assembly on graphene, Small. 10 (2014) 1038–1049. doi:10.1002/smll.201301982. C. Riedl, C. Coletti, T. Iwasaki, A.A. Zakharov, U. Starke, Quasi-free-standing epitaxial graphene on SiC obtained by hydrogen intercalation, Phys. Rev. Lett. 103 (2009) 1–4. doi:10.1103/PhysRevLett.103.246804. P. Ruffieux, S. Wang, B. Yang, C. Sanchez, J. Liu, T. Dienel, L. Talirz, P. Shinde, C.A. Pignedoli, D. Passerone, T. Dumslaff, X. Feng, K. Muellen, R. Fasel, On-surface synthesis of graphene nanoribbons with zigzag edge topology, Nature. 531 (2015) 15. doi:10.1038/nature17151. J.M. Gottfried, Surface chemistry of porphyrins and phthalocyanines, Surf. Sci. Rep. 70 (2015) 259–379. doi:10.1016/j.surfrep.2015.04.001. H. Fukagawa, S. Hosoumi, H. Yamane, S. Kera, N. Ueno, Dielectric properties of polar-phthalocyanine monolayer systems with repulsive dipole interaction, Phys. Rev. B Condens. Matter Mater. Phys. 83 (2011) 1–8. doi:10.1103/PhysRevB.83.085304. Y.L. Huang, Y. Lu, T.C. Niu, H. Huang, S. Kera, N. Ueno, A.T.S. Wee, W. Chen, Reversible single-molecule switching in an ordered monolayer molecular dipole array, Small. 8 (2012) 1423–1428. doi:10.1002/smll.201101967. J.L. Zhang, T.C. Niu, A.T.S. Wee, W. Chen, Self-assembly of binary molecular nanostructure arrays on graphite, Phys. Chem. Chem. Phys. 15 (2013) 12414. doi:10.1039/c3cp00023k. J.L. Zhang, J.L. Xu, T.C. Niu, Y.H. Lu, L. Liu, W. Chen, Reversible switching of a
[28]
[29]
[30] [31]
19 / 22
ACCEPTED MANUSCRIPT
[38]
[39]
[40]
[41]
AC
[42] [43]
CR IP T
[37]
AN US
[36]
M
[35]
ED
[34]
PT
[33]
CE
[32]
single-dipole molecule imbedded in two-dimensional hydrogen-bonded binary molecular networks, J. Phys. Chem. C. 118 (2014) 1712–1718. doi:10.1021/jp408890k. J.M. Soler, E. Artacho, J.D. Gale, A. Garcia, J. Junquera, P. Ordejon, D. Sanchez-Portal, The SIESTA method for ab initio order-N materials simulation, J. Phys. Condens. Matter. 14 (2001) 2745. doi:10.1088/0953-8984/14/11/302. G. Román-Pérez, J.M. Soler, Efficient implementation of a van der waals density functional: Application to double-wall carbon nanotubes, Phys. Rev. Lett. 103 (2009) 1–4. doi:10.1103/PhysRevLett.103.096102. M. Dion, H. Rydberg, E. Schröder, D.C. Langreth, B.I. Lundqvist, Van der Waals density functional for general geometries, Phys. Rev. Lett. 92 (2004) 22–25. doi:10.1103/PhysRevLett.92.246401. R. Coquet, G.J. Hutchings, S.H. Taylor, D.J. Willock, Calculations on the adsorption of Au to MgO surfaces using SIESTA, J. Mater. Chem. 16 (2006) 1978. doi:10.1039/b601213b. A.S. Foster, M.A. Gosálvez, T. Hynninen, R.M. Nieminen, K. Sato, First-principles calculations of Cu adsorption on an H-terminated Si surface, Phys. Rev. B - Condens. Matter Mater. Phys. 76 (2007) 1–8. doi:10.1103/PhysRevB.76.075315. A. Hauschild, K. Karki, B.C.C. Cowie, M. Rohlfing, F.S. Tautz, M. Sokolowski, Molecular distortions and chemical bonding of a large π-conjugated molecule on a metal surface, Phys. Rev. Lett. 94 (2005) 1–4. doi:10.1103/PhysRevLett.94.036106. Y.L. Zhao, W. Wang, F. Qi, J.F. Li, G. Kuang, R.Q. Zhang, N. Lin, M.A. Van Hove, Donor/acceptor properties of aromatic molecules in complex metal-molecule interfaces, Langmuir. 33 (2017) 451–458. doi:10.1021/acs.langmuir.6b02649. M. Huang, Y.L. Zhao, W. Xiong, S. V. Kershaw, Y. Yu, W. Li, T. Dudka, R.Q. Zhang, Collaborative enhancement of photon harvesting and charge carrier dynamics in carbon nitride photoelectrode, Appl. Catal. B Environ. 237 (2018) 783–790. doi:10.1016/j.apcatb.2018.06.046. Y.L. Zhao, R. Zhao, F. Qi, R.Q. Zhang, M.A. Van Hove, Substrate-mediated and temperature-modulated long-range interactions between bromine adatom stripes on Cu(1 1 1), Appl. Surf. Sci. 463 (2019) 253–260. doi:10.1016/j.apsusc.2018.08.168. E.R. Johnson, S. Keinan, P. Mori Sánchez, J. Contreras García, A.J. Cohen, W. Yang, NCI : revealing non-covalent interactions, J. Am. Chem. Soc. 132 (2010) 6498–6506. doi:10.1021/ja100936w. M.J. et al Frisch, Gaussian 09, Revision C.01, Gaussian 09, Revis. C.01. (2009). T. Lu, F. Chen, Multiwfn: A multifunctional wavefunction analyzer, J. Comput. Chem. 33 (2012) 580–592. doi:10.1002/jcc.22885. T. Niu, J. Zhang, W. Chen, Molecular ordering and dipole alignment of vanadyl phthalocyanine monolayer on metals: The effects of interfacial interactions, J. Phys. Chem. C. 118 (2014) 4151–4159. doi:10.1021/jp4101653. A.T.S. Wee, W. Chen, Molecular interactions on epitaxial graphene, Phys. Scr. T146 (2012) 014007. doi:10.1088/0031-8949/2012/T146/014007. Y.L. Huang, R. Wang, T.C. Niu, S. Kera, N. Ueno, J. Pflaum, A.T.S. Wee, W. Chen, One dimensional molecular dipole chain arrays on graphite via nanoscale phase separation, Chem. Commun. 46 (2010) 9040. doi:10.1039/c0cc03251d. J.P. Wagner, P.R. Schreiner, London Dispersion in Molecular Chemistry - Reconsidering
[44]
[45] [46]
[47]
20 / 22
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Steric Effects, Angew. Chemie - Int. Ed. 54 (2015) 12274–12296. doi:10.1002/anie.201503476. [48] F. London, The general theory o f molecular forces., Trans. Faraday Soc. 33 (1937) 8–26. doi:10.1039/tf937330008b. [49] S. Zhong, J.Q. Zhong, X.Z. Wang, M.Y. Huang, D.C. Qi, Z.K. Chen, W. Chen, Investigation of interface properties for ClAlPc/C 60 heterojunction-based inverted organic solar cell, J. Phys. Chem. C. 116 (2012) 2521–2526. doi:10.1021/jp210533n. [50] H. Huang, S.L. Wong, J. Sun, W. Chen, A.T.S. Wee, Trapping single polar molecules in SiC nanomesh via out-of-plane dipoles, ACS Nano. 6 (2012) 2774–2778. doi:10.1021/nn300258b. [51] M. Alonso, T. Woller, F.J. Martín-Martínez, J. Contreras-García, P. Geerlings, F. De Proft, Understanding the fundamental role of dispersion interactions in shaping carbon-based materials, Chem. - A Eur. J. 20 (2014) 4931–4941. doi:10.1002/chem.201400107.
21 / 22
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
M
AN US
CR IP T
Graphic Abstract
22 / 22