Linear dislocation tunes chirality : STM study of chiral transition and amplification in a molecular assembly on an HOPG surface

Abstract

Molecular arrangement and transition in the domain boundary of a chiral two-dimensional assembly is clearly revealed by high-resolution STM images on an HOPG surface and a linear dislocation formed by molecular trimers and located at opposite chiral domains is found to directly reverse the chirality on DTCD self-assembly.

---

Chirality is a natural and spontaneous phenomenon and a fundamental scientific issue due to its importance from life genesis and cosmogony to surface nanostructure, enantioselective heterogeneous catalysis, nonlinear optics, etc. To date, scientists have made significant progress in the determination of absolute chirality , the separation of racemic mixtures, and the fabrication of chiral assemblies in two-dimensional (2D) solid surfaces. In particular, with the development of scanning tunnelling microscopy (STM),¹ direct visualization of chirality at atomic/submolecular resolution in the ultrahigh vacuum,² ambient³ and solution^4,5 environments has been successfully realized, providing important information on molecular features, arrangement and relationships in different chiral domains. For example, a racemic mixture was found to spontaneously separate into chiral domains on a solid surface.^4,6,7 Molecular dimers with opposite handedness were observed in the deposition of a cysteine racemic mixture onto a gold surface. The molecular dimers are either of the LL form or of the DD form.⁸ Besides chiral molecules, adsorption-induced chiral assembly can also be observed on achiral adsorbates.⁹ With respect to chiral phase transition, Böhringer et al. reported the results of a coverage-driven chiral phase transition from a conglomerate to a racemate at low and high molecular coverages, respectively.^10a Moreover, a chiral phase transition in 2D supramolecular assemblies with prochiral molecules was reported.^10b As a result of the studies, it is known that the shifting of the subtle balance between the weak lateral coupling, substrate bonding, and the packing requirements encountered with the increased molecular coverage should be responsible for the chiral phase transition. Recently, the formation and expression of chirality in self-assembled monolayers on solid surfaces has been reviewed in several reports.¹¹ The study of chirality is still of great interest in fields such as surface science, chemistry and materials science. However, there are few reports on the molecular arrangement and transition in chiral domain boundaries, which is related to the mechanism of formation and crystallography of 2D chirality . The chiral origination, amplification and transition are still of great interest and a challenge in molecular science.

In this communication, we report a detailed study of molecular arrangement at a chiral domain boundary. The results are of significance in the origination, transition and amplification of chirality in a 2D self-assembly of an achiral molecule of 1,4-di[4-N-(trihydroxymethyl)methylcarbamoylphenyl]-2,5-didodecyloxybenzene (DTCD, see Scheme 1 for its chemical structure) on a highly oriented pyrolytic graphite (HOPG ) surface. The molecular arrangements at the domain boundaries as well as in the domains are clearly revealed by high-resolution STM images. It was found that the molecules adsorb on the HOPG surface and self-organize into windmill-like molecular tetramers with different chiralities. Moreover, a well-defined linear dislocation formed through molecular tetramers to trimers and located at opposite chiral domains is found to directly reverse the chirality on DTCD self-assembly.

Scheme 1 Molecular structure of DTCD molecule.

Fig. 1a is a typical STM image of the DTCD assembly. Several molecular domains ranging from a few tens to several hundreds of nanometres are observed and denoted with R and S in the image. The domain boundaries are formed by merging neighboring domains together with irregular borderlines. The molecular arrangement in domains R and S is shown in Fig. 1b and c, respectively. It is clear that the images are composed of ordered bright rods. From the molecular structure and theoretical simulation, each rod is assigned as a DTCD molecule. Intriguingly, four DTCD molecules form a tetramer in a windmill-like fashion with a dark depression in the center. After careful observation, it is found that there are two types of tetramers, clockwise fashion shown in Fig. 1b and anticlockwise fashion shown in Fig. 1c. Therefore, two chiral domains are formed in the molecular assembly in which the chiral tetramer is an essential chiral unit but with different rotation. On the basis of the molecular arrangement and chirality , structural models are proposed in Fig. 1d and e for R and Schiral assemblies, respectively. The pink open-circle arrows indicate the propagation direction of windmills on the HOPG surface. From the chemical structure of DTCD and the molecular arrangement with theoretical simulation, hydrogen bonds are believed to exist between the DTCD molecules and should be responsible for the formation of the chiral tetramers. The possible hydrogen bonds in a molecular cluster are proposed in the insets of Fig. 1d and e for chiral domains R and S, respectively. From this model, it can be clearly seen that the hydrogen bonds between four molecules result in a tetramer and stabilize the self-assembled monolayer. Furthermore, it is the sequence and position of the hydrogen bonds that determine the rotation direction of the windmills and thus the origination of chirality . On the basis of STM observations, the process of chiral assembly can be proposed. Once a windmill nucleus is formed, the windmill will spread and amplify its chirality into the whole domain. The domain preserves the handedness originating in its nucleus and epitaxially grows on the surface. The domain size is mainly dominated by the number of nuclei and surface status. When neighboring domains meet together, a domain boundary is formed. Owing to the random growth, the borderline is usually irregular. In general, there are three boundaries according to the chiral relationship between the neighboring domains. In the present study, three kinds of domain boundaries of R–R, R–S, and S–S were observed.

Fig. 1 (a) A large-scale STM image of the self-assembled DTCD monolayer on HOPG . High-resolution STM images of the chiral DTCD assemblies with (b) clockwise and (c) anticlockwise fashion. (d) and (e) are proposed structural models for (b) and (c) respectively. Tunnelling conditions: (a) E_bias = 893 mV, I = 529 pA; (b) E_bias = 939 mV, I = 386 pA; (c) E_bias = 972 mV, I = 476 pA.

The above description illustrates a general origination and amplification process of chiral assembly by achiral molecules.^4–6,8–12 The domain boundary is formed when the neighboring domains meet together. The domain boundary usually appears in irregular borderline. However, on careful inspection, a new type of domain boundary is found as indicated by the dashed red line and red arrow in Fig. 2. Fig. 2 is an STM image acquired on a DTCD adlayer. It can be seen that there are two types of domain boundaries in Fig. 2. One is shown by a white line, which is a typical domain boundary formed by merging neighboring domains together with an irregular borderline where the molecules exist in a disordered and random structure. Another boundary is a well-defined linear dislocation illustrated by the dashed red line. Around the dislocation, the chirality is totally different. Fig. 3a is a high-resolution STM image showing the well defined domain boundary from Fig. 2. A distinct feature in the image is a linear dislocation denoted by a blue dashed line. This line is the domain boundary of these neighboring R and S domains. The essential units, i.e. the windmill-like tetramers in the neighboring domains, are outlined with squares in dashed white lines. It is clear that the tetramers in the neighboring domains possess opposite handedness , suggesting an R–Schirality . Comparing the position of molecular rows of these domains (denoted by red lines 1 and 1′ for R and S domains, respectively), a position displacement can be seen between these two domains. Fig. 3b is a structural model for the molecular arrangement at the domain boundary. From this model, it can be seen that molecular trimers (indicated by the red triangle) are produced at the boundary, instead of tetramers in the domains. Between these trimers and tetramers a well-defined linear dislocation is formed. Through the dislocation the chirality of the adlayer is directly changed. Therefore, the linear dislocation is the domain boundary and also the origination of the chiral phase transition between the neighboring domains. The formation of these trimers interrupt the amplification of the tetramer domain and result in a chiral phase transition. Fig. 3c is a proposed structural model describing the chiral phase transition. Two circles indicate the molecular arrangement in the trimers. Details from a tetramer to a trimer are illustrated in Fig. 3d.

Fig. 2 STM image showing different domain boundaries with a dashed red line (a linear dislocation) and white line. Tunnelling conditions: E_bias = 845 mV, I = 529 pA.

Fig. 3 (a) High-resolution STM image showing the origin and phase transition of chirality by well-defined linear dislocation marked with a dashed blue line. Tunnelling conditions: E_bias = 845 mV, I = 529 pA. Different chiral domains (R and S) are induced on the two sides of the linear dislocation. (b) Structural model for (a). The molecular trimer in the fault position is emphasized. (c) and (d) Schematic illustration of the fault effect on chirality .

The molecular assembly is formed in a thermodynamic process. During the assembly formation/growth, molecular trimers occasionally appear and result in a well-defined linear dislocation. Without nucleus formation, the mechanism of the linear dislocation is totally different from the formation of the domain boundary in Fig. 1 where a nucleus formation and growth process is proposed with irregular borderlines. After the dislocation, molecules still form tetramers due to their higher stability. Therefore, the linear dislocation with trimers plays a role in the chiral phase transition and the domain boundary is different from that formed by merging the neighboring domains together. The transition procedure is schematically illustrated in Scheme 2. From the scheme, it is clear that the linear dislocation with molecular trimers breaks the amplification of the chiral domain and results in a chiral phase transition. From the dislocation, a new chirality in the assembly is directly induced.

Scheme 2 Schematic illustration of the chiral transition and amplification on 2D DTCD adlayer from a linear dislocation formed through tetramers to trimers. The pink windmills are clockwise and blue windmills anticlockwise.

However, it is found that not all dislocations could change the assembly chirality . A molecular dislocation formed by molecular dimers was experimentally observed (see Fig. S1 in ESI† ). This dislocation interrupts the continuity of the molecular row with a position displacement, while no chiral phase transition can be observed. The chiral domain is amplified across the fault and over the surface without a change in handedness . The results indicate that in this research the transition from tetramer to trimer with different positions and sequences of the hydrogen bonds is a dominant factor for chiral change. However, a study of other possibilities which may influence the chiral transition is still in progress.

In summary, the molecular arrangements at the chiral domain boundary as well as in the domain are clearly revealed by high resolution STM images. DTCD molecules form two types of windmill-like clusters, positioned clockwise and anticlockwise in a chiral fashion. A windmill is composed of four DTCD molecules connected by hydrogen bonds as a tetramer and is the essential unit in the chiral assembly. It is found that through a well-defined linear dislocation formed by molecular trimers the chirality can be directly reversed in the assembly. The linear dislocation plays a role in chiral initiation and as the domain boundary between R and Schirality . The linear boundary is different from that formed by merging the neighboring domains together. This finding will be significant in the understanding of the origination and transition of 2D surface chirality .

The authors thank the financial supports from National Natural Science Foundation of China (Grants 20575070, 20673121, and 20121301), National Key Project on Basic Research (Grant 2006CB806100), and the Chinese Academy of Sciences.

Notes and references

1. (a) K. Itaya, Prog. Surf. Sci., 1998, 58, 121; (b) R. Otero, F. Rosei and F. Besenbacher, Annu. Rev. Phys. Chem., 2006, 57, 497; (c) M. Montano, D. C. Tang and G. A. Somorjai, Catal. Lett., 2006, 107, 131; (d) M. Wang, S. Bugarski and U. Stimming, J. Phys. Chem. C, 2008, 112, 5165.
2. (a) P. Messina, A. Dmitriev, N. Lin, H. Spillmann, M. Abel, J. V. Barth and K. Kern, J. Am. Chem. Soc., 2002, 124, 14000; (b) R. Fasel, M. Parschau and K. H. Ernst, Nature, 2006, 439, 449.
3. (a) B. Ohtani, A. Shintani and K. Uosaki, J. Am. Chem. Soc., 1999, 121, 6515; (b) S. De Feyter, P. C. M. Grim, M. Rucker, P. Vanoppen, C. Meiners, M. Sieffert, S. Valiyaveettil, K. Mullen and F. C. De Schryver, Angew. Chem., Int. Ed., 1998, 37, 1223; (c) A. Gesquiere, P. Jonkheijm, F. J. M. Hoeben, A. Schenning, S. De Feyter, F. C. De Schryver and E. W. Meijer, Nano Lett., 2004, 4, 1175.
4. Q. M. Xu, D. Wang, L. J. Wan, C. Wang, C. L. Bai, G. Q. Feng and M. X. Wang, Angew. Chem., Int. Ed., 2002, 41, 3408.
5. (a) Q. M. Xu, D. Wang, L. J. Wan, C. L. Bai and Y. Wang, J. Am. Chem. Soc., 2002, 124, 14300; (b) M. Kunitake, T. Hattori, S. Miyano and K. Itaya, Langmuir, 2005, 21, 9206.
6. Q. H. Yuan, C. J. Yan, H. J. Yan, L. J. Wan, B. H. Northrop, H. Jude and P. J. Stang, J. Am. Chem. Soc., 2008, 130, 8878.
7. T. Huang, Z. P. Hu, A. D. Zhao, H. Q. Wang, B. Wang, J. L. Yang and J. G. Hou, J. Am. Chem. Soc., 2007, 129, 3857.
8. A. Kuhnle, T. R. Linderoth, B. Hammer and F. Besenbacher, Nature, 2002, 415, 891.
9. (a) N. V. Richardson, New J. Phys., 2007, 9, 16; (b) Y. G. Cai and S. L. Bernasek, J. Am. Chem. Soc., 2004, 126, 14234.
10. (a) M. Böhringer, W.-D. Schneider and R. Berndt, Angew. Chem., Int. Ed., 2000, 39, 792; (b) F. Vidal, E. Delvigne, S. Stepanow, N. Lin, J. V. Barth and K. Kern, J. Am. Chem. Soc., 2005, 127, 10101.
11. (a) V. Humblot, S. M. Barlow and R. Raval, Prog. Surf. Sci., 2004, 76, 1; (b) K. H. Ernst, Top. Curr. Chem., 2006, 265, 209; (c) N. Katsonis, E. Lacaze and B. L. Feringa, J. Mater. Chem., 2008, 18, 2065; (d) D. B. Amabilino, S. De Feyter, R. Lazzaroni, E. Gomar-Nadal, J. Veciana, C. Rovira, M. M. Abdel-Mottaleb, W. Mamdouh, P. Iavicoli, K. Psychogyiopoulou, M. Linares, A. Minoia, H. Xu and J. P. Luis, J. Phys.: Condens. Matter, 2008, 20, 184003.
12. (a) S. De Feyter and F. C. De Schryver, Chem. Soc. Rev., 2003, 32, 139; (b) L. C. Giancarlo and G. W. Flynn, Acc. Chem. Res., 2000, 33, 491.

---

Footnote

† Electronic supplementary information (ESI) available: Methods section and an STM image. See DOI: 10.1039/b817525j

---