Stabilizing a Molecular Switch at Solid Surfaces: A Density-Functional Theory Study of Azobenzene at Cu(111), Ag(111), and Au(111)
📝 Abstract
We present a density-functional theory trend study addressing the binding of the trans-cis conformational switch azobenzene (C6H5-N=N-C6H5) at three coinage metal surfaces. From the reported detailed energetic, geometric, and electronic structure data we conclude that the governing factor for the molecule-surface interaction is a competition between covalent bonding of the central azo (-N=N-) bridge on the one hand and the surface interaction of the two closed-shell phenyl (-C6H5) rings on the other. With respect to this factor the cis conformer exhibits a more favorable gas-phase geometric structure and is thus more stabilized at the studied surfaces. With the overall binding still rather weak the relative stability of the two isomers is thereby reduced at Ag(111) and Au(111). This is significantly different at Cu(111), where the cis bonding is strong enough to even reverse the gas-phase energetic order at the level of the employed semi-local electronic exchange and correlation (xc) functional. While this actual reversal may well be affected by the deficiencies due to the approximate xc treatment, we critically discuss that the rationalization of the general effect of the surface on the meta-stable molecular states is quite robust. This should equally hold for the presented analysis of recent tip-manipulation and photo-excitation isomerization experiments from the view point of the derived bonding mechanism.
💡 Analysis
We present a density-functional theory trend study addressing the binding of the trans-cis conformational switch azobenzene (C6H5-N=N-C6H5) at three coinage metal surfaces. From the reported detailed energetic, geometric, and electronic structure data we conclude that the governing factor for the molecule-surface interaction is a competition between covalent bonding of the central azo (-N=N-) bridge on the one hand and the surface interaction of the two closed-shell phenyl (-C6H5) rings on the other. With respect to this factor the cis conformer exhibits a more favorable gas-phase geometric structure and is thus more stabilized at the studied surfaces. With the overall binding still rather weak the relative stability of the two isomers is thereby reduced at Ag(111) and Au(111). This is significantly different at Cu(111), where the cis bonding is strong enough to even reverse the gas-phase energetic order at the level of the employed semi-local electronic exchange and correlation (xc) functional. While this actual reversal may well be affected by the deficiencies due to the approximate xc treatment, we critically discuss that the rationalization of the general effect of the surface on the meta-stable molecular states is quite robust. This should equally hold for the presented analysis of recent tip-manipulation and photo-excitation isomerization experiments from the view point of the derived bonding mechanism.
📄 Content
In view of the rapidly advancing miniaturization in microelectronics and sensing, molecules are envisioned as fundamental building blocks in a future “molecular nanotechnology”. Since controlled switching between defined states is a crucial basis component for storage and logic, molecules offering this functionality (e.g. through externally induced changes between conformational isomers) attain a central importance. Considering contacting and defined integration into a larger framework, it is more precisely the function of the molecule when stabilized at a solid surface that is of key interest. While a large variety of molecules can be controllably switched in gas-phase or solution, still little is known about their function in such an adsorbed state. Suppression of the switching capability e.g. due to steric hindrance is a possible scenario, but completely new properties under the influence of the solid surface are equally conceivable. The atomic-scale understanding necessary for a technological exploitation of such effects builds on a detailed structural and electronic characterization of the adsorbed molecular switch, as well as its response to external stimuli like fields, forces or external currents. 1 As a prototypical conformational switch of modest complexity azobenzene (C 6 H 5 -N=N-C 6 H 5 ) has been on the research agenda for many years. 2 In solution it exhibits a reversible photo-isomerization between an energetically more stable planar trans geometry and a threedimensional cis configuration, in which the planes of the two phenyl-rings are tilted with respect to each other. A common theme in existing experimental studies addressing the function in surface mounted geometries is that a too strong substrate interaction is suspected to cause ad-verse effects on the switching efficiency. This has either led to the use of ligands like alkanethiol chains to decouple the azobenzene moiety in vertical geometries 3,4 or to a focus on unreactive materials like close-packed surfaces of coinage metals and there in particular on Au(111). On the latter surface bare azobenzene could be successfully isomerized using tip-manipulation techniques 5 , but even in this alleged weak physisorption limit switching with light could not be achieved 6 . Reasoned as an ultrafast quenching of the electronic excitation azobenzenederivatives that are further lifted up from the surface appear as a viable alternative and reversible photomechanical switching has indeed been reported for the functionalized molecule tetra-tert-butyl azobenzene (TBA) 6,7 . The intricate role played even then by the metallic substrate is nevertheless exemplified by the fact that exactly for the same azobenzene-derivative, TBA, light-induced switching could not be achieved at Ag(111). 8 As already stated, the characterization of the stable (or meta-stable, long-lived) molecular states at the surface is a necessary prerequisite to a detailed understanding of these experimental findings and of the actual excitation mechanism. While already less ambitious than a comprehensive first-principles treatment of the entire switching process, corresponding static electronic structure theory calculations still pose a considerable challenge. The large system sizes resulting from the sheer extension of the azobenzene molecule and the periodic supercell geometries dictated by the metallic bandstructure, can at present only be tackled by density-functional theory (DFT) with local-or semi-local exchange and correlation (xc) functionals. While this is the state of the art and the lowest level of theory at which one can still at least hope for the aspired quantitative and predictive-quality modeling, it is clear from the start that the specificities of the molecule-surface binding directly challenge two wellknown deficiencies of the named functionals: the spurious self-interaction and the lack of long-range dispersive (van der Waals) attraction. With several frontier orbitals of different symmetry (vide infra) presumably involved in the bonding, the detrimental effect of self-interaction blurred orbital energies on fundamental quantities like the preferred adsorption site has already been demonstrated for much simpler adsorbates 9,10,11,12,13 . Similarly well known is the important role played by long-range van der Waals interactions when aromatic π-like orbitals participate in the molecule-surface interaction 14,15,16 .
The present trend study addressing the binding of azobenzene at the three coinage metal surfaces, Cu(111), Ag(111) and Au(111) within large-scale DFT calculations correspondingly has a twofold focus. On the one hand, we critically discuss the obtained detailed energetic, geometric, and electronic structure data in the context of the sketched limitations of the employed semi-local xc functional, namely the generalized gradient approximation (GGA) functional due to Perdew, Burke and Ernzerhof (PBE) 17 . With respect to the lack of long-range dispersive
This content is AI-processed based on ArXiv data.