Growth of covalently bonded Sierpiński triangles up to the second generation

Abstract

Growth of covalently bonded Sierpiński triangles (CB-STs) on metal surfaces was investigated by scanning tunneling microscopy (STM). Three synthetic strategies (namely, dehydration condensation, cyclotrimerization coupling and Schiff-base reactions) were used to fabricate CB-STs. Second-generation CB-STs were obtained at the solid–vacuum interface utilizing the Schiff-base reaction between 4,4′′-dialdehyde-1,1′:3′,1′′-terphenyl (TPDAL) and 1,3,5-tris(4-aminophenyl)benzene (TAPB) on Au(111). The CB-ST patterns persist at annealing temperatures as high as 500 K. Homotactic three-fold motifs, insufficient migration and irreversible covalent reaction are the main limitations for growing higher-generation STs. The present results provide new insights on the growth of STs on metal surfaces.

---

Introduction

Self-similar fractal patterns are of importance in aesthetics, mathematics, science and engineering due to their complicated yet fascinating architectures. There are many examples of self-similar fractal structures in nature, such as trees, snowflakes, lightning and coastlines. To understand the mechanism of their growth process, many theoretical and experimental studies have been performed on molecular fractals in the last two decades,^1–8 with the molecular Sierpiński triangle (ST) as a model system for self-similar fractal patterns.^9,10

We successfully fabricated the defect-free molecular STs on Ag(111) by weak cyclic halogen bonds¹¹ or O⋯H–O hydrogen-bonds.¹² Subsequently, metal–organic STs (MOSTs) bonded through coordination interaction were prepared on Au(111) by us and Xu's group.^13,14 Both of these two MOSTs are stable at room temperature (RT). To further improve the stability of STs, the synthesis of covalently bonded STs was studied here.

Three synthetic pathways for constructing stable covalently bonded STs on metal surfaces were carried out: namely, dehydration condensation reaction of 4,4′′-diamino-1,1′:3′,1′′-terphenyl (TPDAM) and 1,3,5-tris(4-carboxyphenyl)benzene (TCPB) (Scheme 1a), cyclotrimerization coupling reaction of 4,4′′-diacetyl-1, 1′:3′,1′′-terphenyl (TPDAC) (Scheme 1b) and Schiff-base reaction of 4,4′′-dialdehyde-1,1′:3′,1′′-terphenyl (TPDAL) and 1,3,5-tris(4-aminophenyl)benzene (TAPB) (Scheme 1c).^15–17 Second-generation covalently bonded STs were successfully realized for the third reaction. Several critical factors that are detrimental to the formation of higher-generation STs can be identified. It is concluded that homotactic three-fold motifs, insufficient migration and irreversible covalent reaction are the main limitations for growing higher-generation STs.

Scheme 1 Schemes of the reaction mechanisms of covalent Sierpiński triangles. (a) Dehydration–condensation reaction of 4,4′′-diamino-1,1′:3′,1′′-terphenyl (TPDAM) and 1,3,5-tris(4-carboxyphenyl)benzene (TCPB) . (b) Cyclotrimerization coupling reaction of 4,4′′-diacetyl-1,1′:3′,1′′-terphenyl (TPDAC) . (c) Schiff-base reaction of 4,4′′-dialdehyde-1,1′:3′,1′′-terphenyl (TPDAL) and 1,3,5-tris(4-aminophenyl)benzene (TAPB) . The homotactic windmill node in the upper right corner stands for homotactic product, which prevents STs growing larger.

Experimental

The experiments were carried out with a Unisoku scanning tunneling microscope (STM) with a base pressure of 10⁻¹⁰ Torr. The single-crystalline Au(111) and Cu(111) surfaces were cleaned by repetitive cycles of Ar ion sputtering and annealing at 400 °C. The polycrystalline Pt/Ir tip was annealed first in the preparation chamber and treated by gently dipping into Au(111) or Cu(111). Molecules were thermally deposited on the substrates held at RT from a Ta boat heated by direct current. All STM images were acquired at liquid helium temperature and processed using software WSxM.¹⁸ We define 1 monolayer (ML) as the coverage of the most dense structure observed.

Results and discussion

As shown in Fig. 1a, TPDAM molecules are dispersedly adsorbed on Au(111) at a low coverage of ∼0.25 ML, as a result of a long-range repulsive interaction induced by molecule-substrate charge transfer¹⁹ or substrate-mediated interaction via quantum interference of surface electrons.²⁰ In contrast, TCPB molecules tend to assemble into 2D hexagonal networks with extensive hydrogen bonds (Fig. 1b), demonstrating dominant intermolecular interactions. Similar hexagonal networks have also been observed for TCPB on liquid–solid interface.²¹ When TPDAM were added to the as-prepared TCPB/Au(111) sample at RT, TPDAM molecules were adsorbed into the hexagonal holes of the TCPB network, with the latter thereby functioning as a template (Fig. 1c). Annealing of the sample was performed at temperatures ranging from RT to 380 K. However, only a small number of reactive products were formed through dehydration–condensation reaction (Fig. 1d). Gradually increasing the annealing temperature to 420 K resulted in a significant desorption of both TCPB and TPDAM molecules. It looks like that chemical reaction with lower reactive temperature is beneficial to prepare CB-ST structures.

Fig. 1 (a) Constant current (CC) STM image of TPDAM molecules (designated as ) adsorbed on Au(111) (9 × 9 nm², V_b = 50 mV, I_t = 60 pA). White circle encloses a single TPDAM molecule. (b) Constant height (CH) STM image of TCPB molecules (designated as ) adsorbed on Au(111) (9 × 9 nm², V_b = 20 mV). White circle encloses a single TCPB molecule. Overlaid molecular model shows TCPB hexagonal network is hydrogen-bonded. (c) STM image of coadsorption structure of TPDAM and TCPB on Au(111) (9 × 9 nm², V_b = 100 mV, I_t = 20 pA, CC). (d) Coadsorption structure of TPDAM and TCPB on Au(111) with post-annealing at 380 K for 10 min (10 × 10 nm², V_b = 1 V, I_t = 20 pA, CC). Arrows in all images indicate [11−2] crystal direction of Au(111). Their length corresponds to 2 nm.

Inspired by the synthesis of three-fold nodal motif via cyclotrimerization of acetyls,¹⁶ we deposited TPDAC molecules on Cu(111). This is the second reaction scheme (Scheme 1b) in this paper. Here Cu(111) was chosen as the surface since it is known to have a stronger molecule–substrate interaction and higher desorption temperature in comparison to the Au(111) surface. When deposited at RT, TPDAC molecules form chain-like structures (Fig. 2a). Cyclotrimerization coupling reaction occurred after annealing the TPDAC/Cu(111) sample at 380 K for half an hour. First-generation CB-ST structure and sporadic ST nodes were observed on Cu(111) (Fig. 2b). However, higher-generation ST structures have not been obtained. The strong molecule–surface interaction might restrict free diffusion of reactive products on Cu(111) which prevents CB-STs growing larger. When changing the substrate to less reactive Au(111), TPDAC molecules desorbed from the substrate before reaction. So, weak molecule–surface interaction is as important as low reactive temperature in growing CB-STs.

Fig. 2 (a) Self-assembled structure of TPDAC molecules (designated as ) deposited on Cu(111) at RT (8 × 8 nm², V_b = 100 mV, I_t = 40 pA, CC). White circle encloses a single TPDAC molecule. (b) A first-generation of Sierpiński triangle formed by cyclotrimerization coupling reaction of TPDAC molecules under 380 K annealing (8.6 × 8.6 nm², V_b = 10 mV, CH). Arrows in all images indicate the [11−2] crystal direction of Cu(111). Their length corresponds to 2 nm.

The third reaction scheme is the Schiff-base reaction (Scheme 1c) between TPDAL and TAPB on Au(111).^22,23 First, the two molecules were deposited separately to investigate their adsorption behaviors. At a coverage ∼1 ML, TPDAL molecules were found to form braid-like structures through weak hydrogen bonds (Fig. 3a).¹² At low coverages, single clover-like TAPB molecules dispersedly adsorbed on Au(111) were observed (designated as in Fig. 3b). When TPDAL was subsequently added to the as-prepared TAPB/Au(111) sample at RT, it was observed that each TAPB molecule is surrounded by several TPDAL molecules with hydrogen bond interaction (Fig. 3c). For the circled structures, each petal of the clover-shaped TAPB molecule forms two “N–H⋯O=C” hydrogen bonds with two adjacent TPDAL molecules, as indicated in Fig. 3d.

Fig. 3 (a) Self-assembled structure of TPDAL molecules (designated as ) on Au(111). White circle shows a single TPDAL molecule (6.6 × 6.6 nm², V_b = 10 mV, CH). (b) STM image of TAPB molecules (designated as ) on Au(111). White circle shows a single TAPB molecule (12 × 12 nm², V_b = 1 V, I_t = 20 pA, CC). (c) Dihydrogen bond interaction of 1-amino and 2-aldehyde (6.6 × 6.6 nm², V_b = 200 mV, I_t = 40 pA, CC). (d) Model of the TPDAL and TAPB molecules in the white circle in (c). All arrows in STM images indicate the [11−2] crystal direction of Au(111) and their length corresponds to 1 nm (a), 2 nm (b) and 1 nm (c), respectively.

The Schiff-base reaction between TPDAL and TAPB molecules was activated by annealing the sample to 380 K. Fig. 4 presents the STM images (a, c, e) and models (b, d, f) of the molecular configurations, in which the TPDAL and TAPB molecules form three-fold nodes through the formation of three C=N double bonds. Stepwise growth of these configurations produces the successive generations of the canonical Sierpiński triangles. They are denoted here as CB-ST-n with n equal to 0 (a), 1 (c) and 2 (e). As demonstrated in the figure by blue triangles, the creation of the CB-ST-n is an iterative procedure that starts from an equilateral triangle (Fig. 4a, top-right corner). The blue triangles in (b), (d), (f) overlaid on the molecular models illustrate the correspondence between the basic structural units of the molecular self-assembly and geometric construction. It was found that the structures could bear annealing temperature as high as 500 K, revealing a better stability of CB-STs than other STs. However, the CB-ST-2 configuration is the largest Sierpiński triangle observed in this system. In comparison, 4-genarations of cyclic halogen bond and metal–organic STs and 3-genarations of cyclic hydrogen bond STs have been reported.^11–13

Fig. 4 (a, c and e) STM images of CB-ST-0 to CB-ST-2 structure. Imaging parameters: (a) 3.3 × 3.3 nm², V_b = 10 mV, CH; (b) 5.5 × 5.5 nm², V_b = 1 V, I_t = 20 pA, CC; (c) 9 × 9 nm², V_b = 1 V, I_t = 20 pA, CC. (b, d and f) Models of CB-ST-0, CB-ST-1 and CB-ST-2, respectively. All arrows in STM images indicate the [11−2] crystal direction of Au(111) and their length corresponds to 0.5 nm (a), 1 nm (c) and 2 nm (e), respectively.

A large scale STM image (Fig. 5a) shows that many other motifs coexist with CB-STs on Au(111). One typical structure is the homotactic node (Fig. 5b and c). From the statistic, 20% of three-fold nodes are homotactic. This ratio is much larger than that reported in the previous results (5%).^10–12 It is known that only heterotactic nodes are present in STs and homotactic nodes can hamper the formation of higher-generation STs.

Fig. 5 (a) STM image of large-scale network formed by annealing of TPDAL and TAPB (90 × 90 nm², V_b = 1 V, I_t = 20 pA, CC). The arrow indicates the [11-2] crystal direction and its length corresponds to 20 nm. The circle encloses an intact CB-ST-2. (b) High-resolution image (4.3 × 4.3 nm², V_b = 1 V, I_t = 20 pA, CC) of a homotactic node and its chemical structure (c).

Unlike the self-correction process during the halogen bonded or metal–organic bonded STs growth, the absence of H₂O molecules in vacuum decreased the reversibility of the aldehyde-amine coupling reaction. As a result, the “wrong” homotactic nodes prevent the CB-STs growing larger. If adding a certain amount of water to the reaction,¹⁷ a node changing from homotactic to heterotactic could be realized by regulating the thermodynamic equilibrium of the covalent bond formation. In consequence, elimination of homotactic nodes might lead to enhanced formation of larger undisturbed triangles. This work is in progress.

Conclusions

In summary, we have investigated three synthetic schemes (dehydration–condensation, cyclotrimerization coupling and Schiff-base reactions) in order to construct CB-STs. By Schiff-base reaction of TPDAL and TAPB on Au(111), we prepared the second-generation CB-STs at vacuum–solid interfaces for the first time. It is concluded that homotactic three-fold motifs, insufficient migration and irreversible covalent reaction are the main limitations for growing higher-generation STs. The present results provide new insights on the growth of STs on metal surfaces.

Acknowledgements

This work was supported by National Natural Science Foundation of China (No. 21522301, 21373020, 21403008, 61321001, 21433011, 61271050), the Ministry of Science and Technology (No. 2014CB239302 and 2013CB933404) and the Research Fund for the Doctoral Program of Higher Education (No. 20130001110029).

References

1. B. B. Mandelbrot, The Fractal Geometry of Nature, Henry Holt and Company, New York, 1982.
2. K. I. Sugiura, H. Tanaka, T. Matsumoto, T. Kawai and Y. Sakata, Chem. Lett., 1999, 28, 1193.
3. G. R. Newkome, P. Wang, C. N. Moorefield, T. J. Cho, P. P. Mohapatra, S. Li, S. H. Hwang, O. Lukoyanova, L. Echegoyen, J. A. Palagallo, V. Iancu and S. W. Hla, Science, 2006, 312, 1782.
4. K. Fujibayashi, R. Hariadi, S. H. Park, E. Winfree and S. Murata, Nano Lett., 2008, 8, 1791.
5. G. R. Newkome and C. Shreiner, Chem. Rev., 2010, 110, 6338.
6. D. Nieckarz and P. Szabelski, Chem. Commun., 2014, 50, 6843.
7. R. Sarkar, K. Guo, C. N. Moorefield, M. J. Saunders, C. Wesdemiotis and G. R. Newkome, Angew. Chem., Int. Ed., 2014, 53, 12182.
8. M. Wang, C. Wang, X. Q. Hao, J. J. Liu, X. H. Li, C. L. Xu, A. Lopez, L. Sun, M. P. Song, H. B. Yang and X. P. Li, J. Am. Chem. Soc., 2014, 136, 6664.
9. S. L. Tait, Nat. Chem., 2015, 7, 370.
10. G. C. Gu, N. Li, X. Zhang, S. M. Hou, Y. F. Wang and K. Wu, Acta Phys.–Chim. Sin., 2016, 32, 195.
11. J. Shang, Y. F. Wang, M. Chen, J. X. Dai, X. Zhou, J. L. Kuttner, G. Hilt, X. Shao, J. M. Gottfried and K. Wu, Nat. Chem., 2015, 7, 389.
12. X. Zhang, N. Li, G. C. Gu, H. Wang, D. Nieckarz, P. Szabelski, Y. He, Y. F. Wang, C. Xie, Z. Y. Shen, J. T. Lü, H. Tang, L. M. Peng, S. M. Hou, K. Wu and Y. F. Wang, ACS Nano, 2015, 9, 11909.
13. N. Li, X. Zhang, G. C. Gu, H. Wang, D. Nieckarz, P. Szabelski, Y. He, Y. F. Wang, J. T. Lü, H. Tang, L. M. Peng, S. M. Hou, K. Wu and Y. F. Wang, Chin. Chem. Lett., 2015, 26, 1198.
14. Q. Sun, L. L. Cai, H. H. Ma, C. X. Yuan and W. Xu, Chem. Commun., 2015, 51, 14164.
15. R. Tanoue, R. Higuchi, N. Enoki, Y. Miyasato, S. Uemura, N. Kimizuka, A. Z. Stieg, J. K. Gimzewski and M. Kunitake, ACS Nano, 2011, 5, 3923.
16. B. Yang, J. Björk, H. P. Lin, X. Q. Zhang, H. M. Zhang, Y. Y. Li, J. Fan, Q. Li and L. F. Chi, J. Am. Chem. Soc., 2015, 137, 4904.
17. X. H. Liu, C. Z. Guan, D. Wang and L. J. Wan, Adv. Mater., 2014, 26, 6912.
18. I. Horcas, R. Fernandez, J. M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero and A. M. Baro, Rev. Sci. Instrum., 2007, 78, 013705.
19. I. Fernandez-Torrente, S. Monturet, K. Franke, J. Fraxedas, N. Lorente and J. Pascual, Phys. Rev. Lett., 2007, 99, 176103.
20. Y. F. Wang, X. Ge, C. Manzano, J. Kröger, R. Berndt, W. A. Hofer, H. Tang and J. Cerda, J. Am. Chem. Soc., 2009, 131, 10400.
21. R. Gutzler, T. Sirtl, J. F. Dienstmaier, K. Mahata, W. M. Heckl, M. Schmittel and M. Lackinger, J. Am. Chem. Soc., 2010, 132, 5084.
22. L. R. Xu, X. Zhou, Y. X. Yu, W. Q. Tian, J. Ma and S. B. Lei, ACS Nano, 2013, 7, 8066.
23. L. Jiang, A. C. Papageorgiou, S. C. Oh, Ö. Sağlam, J. Reichert, D. A. Duncan, Y. Q. Zhang, F. Klappenberger, Y. Y. Guo, F. Allegretti, S. More, R. Bhosale, A. Mateo-Alonso and J. V. Barth, ACS Nano, 2016, 10, 1033.

---

Footnote

† These two authors made an equal contribution to this work.

---