Molecular structure formation at surfaces and interfaces

Our research concentrates on the formation of molecular structures at surfaces and interfaces. We study the physico-chemical mechanisms responsible for molecular self-assembly and on-surface reactions on the surfaces of bulk insulators and we investigate solvation structures at solid-liquid interfaces with atomic and molecular resolution.

We employ dynamic atomic force microscopy (AFM) as our main measurement tool. We use and improve this method (by developing both hardware and software modifications as well as analysis tools) for high-resolution imaging both in the ultra-high vacuum environment and at the solid-liquid interface.

I. Ultra-High Vacuum Environment

Method Development

Photograph of one of our UHV chambers
Photograph of one of our UHV chambers.

Real-space imaging of insulating surfaces at the atomic and molecular level requires state-of-the-art dynamic atomic force microscopy. Special measurement protocols need to be developed when aiming for investigating processes at room temperature, e.g., for compensating thermal drift.

We constantly improve the signal-to-noise ratio of our instruments and have implemented new methods such as the atom-tacking technique [1, 2]. This method effectively compensates thermal drift, thus enabling three-dimensional force mapping even at room temperature. Moreover, the high-resolution capability of our instruments allows for collecting Kelvin-probe force microscopy (KPFM) data with high resolution, which gives valuable insights into the charge distribution on surfaces.

[1] P. Rahe et al., Rev. Sci. Instrum. 82 (2011) 063704
[2] P. Rahe et al., Rev. Sci. Instrum. 85 (2014) 043707

Understanding and Controlling Self-Assembly on Insulating Surfaces

Molecular stripes on a calcite surface in ultra-high vacuum (images of different size ranging from a 1 x 1 µm overview image on the left to a high-resolution 10 x 10 nm image on the right)
Molecular stripes on a calcite surface in ultra-high vacuum (images of different size ranging from a 1 x 1 µm overview image on the left to a high-resolution 10 x 10 nm image on the right) [7].

Controlling molecular self-assembly on surfaces requires an in-depth understanding of both the intermolecular as well as the molecule-surface interactions. Compared to metal surfaces, rather little is known about the molecule-surface interaction of insulating substrates. This is mainly due to the fact that many standard surface science techniques rely on conducting surfaces and are, therefore, not suited to study insulating materials.

Often, molecular self-assembly on insulating surfaces is hampered by the comparably weak molecule-surface interaction. Thus, strategies need to be developed for enhancing and precisely tuning the influence of the substrate in molecular self-assembly on insulting surfaces [3-5]. We have systematically changed the functionality of molecular building blocks for, e.g., varying the intermolecular interaction while at the same time maintaining a given interaction strength with the supporting substrate [6].

As an complementary approach to our real-space imaging capability, we have recently established a setup for temperature-programmed desorption (TPD), which allows for gaining insight into the binding of the molecules towards the surface.

[3] P. Rahe et al., Adv. Mater. 25 (2013) 3948
[4] M. Kittelmann et al., J. Phys.: Condens. Matter 24 (2012) 354007
[5] M. Körner et al., Phys. Rev. Lett. 107 (2011) 016101
[6] C. Hauke et al., ACS Nano 7 (2013) 5491
[7] J. Neff et al., J. Phys. Chem. C 119 (2015) 24927

Kinetics and Phase Transitions of Molecular Assemblies on Surfaces

 Stripe-like adsorbates appear and vanish over the course of time
Stripe-like adsorbates appear and vanish over the course of time[10].

For gaining a comprehensive understanding of molecular self-assembly on surfaces, it is important to investigate the kinetics of fundamental processes at surfaces and to quantify relevant energy barriers. A very important parameter for self-assembly is, e.g., the diffusion barrier. Yet, this fundamental parameter has almost exclusively been studied on metals. We have quantified the diffusion barrier for C60 on CaF2(111), again demonstrating the rather weak influence of the insulating substrate on the molecular self-assembly [8].

We have followed the deprotonation kinetics of a benzoic acid on calcite by collecting NC-AFM movies (> 10 h) [9] and currently work on identifying the molecular arrangement within the different phases in a collaboration with theorists. Recently, we studies the dynamic stripe growth observed for benzoic acids on calcite [7, 10].

[8] F. Loske et al., Phys. Rev. B 82 (2010) 155428
[9] M. Kittelmann et al., ACS Nano 6 (2012) 7406
[10] J. L. Neff et al., Faraday Discussion 204 (2017) 419

 

On-Surface Synthesis

Diacetylene reaction on calcite
Diacetylene reaction on calcite [14].

On-surface synthesis provides a promising strategy for creating functional molecular structures on surfaces that go beyond molecular self-assembly. Especially when aiming for stable structures that can be brought outside the ultra-high vacuum environment, covalently-linked structures are desired. Moreover, covalently-linked conjugated molecules allow for efficient electron transport and are, thus, particularly interesting for future molecular electronics applications. When having these applications in mind, insulating substrates are mandatory to provide sufficient decoupling of the molecular structure from the substrate surface. However, on-surface synthesis had been achieved only on metallic substrates.

We have demonstrated the first proof-of-principle for successful covalent linking of a benzoic acid derivative directly on a bulk insulator [11]. As a next step, we have extended this concept by a sequential and site-specific linkage, providing the possibility for hierarchical structure formation [12]. Recently, we have presented the first example for photochemical initiation of an on-surface reaction on a bulk insulator. In the latter example, the substrate was found to guide the reaction direction, which constitutes an interesting strategy for controlled fabrication of oriented structures on surfaces [13].

This work is funded by an EU large-scale integrating project in FET Proactive — Planar Atomic and Molecular Scale devices (PAMS) — which is one out of only five projects funded in this area.

[11] M. Kittelmann et al., ACS Nano 5 (2011) 8420
[12] M. Kittelmann et al., ACS Nano 7 (2013) 5614
[13] R. Lindner et al., Angew. Chem. Int. Ed. 53 (2014) 7952
[14] Phys. Chem. Chem. Phys. 19 (2017) 15172-15176

 

II. Solid-Liquid Interfaces

High-Resolution Imaging and 3D Mapping at Solid-Liquid Interfaces

Photograph of one of our AFMs for high-resolution imaging in liquids.
Photograph of one of our AFMs for high-resolution imaging in liquids.

High-resolution imaging of the solid-liquid interface requires the identification and reduction of all possible noise sources in order to increase the signal-to-noise ratio as much as possible. With this approach, we have modified commercial instruments for enabling dynamic AFM operation in liquids [15]. Additionally, we have implemented a photothermal excitation method that allows for a controlled cantilever excitation, thus allowing the quantitative analysis of our AFM data [16]. Recently, we developed custom routines for high-resolution three-dimensional scanning at interfaces [17, 18]. The resulting volumetric data sets allow us to investigate the solvation structure at solid-liquid interfaces.

AFM images showing the atomic resolution of calcite (left) and gypsum (right) in water
AFM images showing the atomic resolution of calcite (left) and gypsum (right) in water

[15] S. Rode et al., Rev. Sci. Instrum. 82 (2011) 073703
[16] H. Adam et al., Rev. Sci. Instrum. 85 (2014) 023703
[17] C. Marutschke et al., Nanotechnology 25 (2014) 335703
[18] H. Söngen et al., Rev. Sci. Instrum. 87 (2016) 063704

Solvation Layer Mapping

Solvation layer above a calcite surface [19]. The figure shows a vertical slice (perpendicular to the surface, which is at the bottom). The left panel shows the simulated water density (obtained by molecular dynamics simulation). The panel on the right shows our experimental data.
Solvation layer above a calcite surface [19]. The figure shows a vertical slice (perpendicular to the surface, which is at the bottom). The left panel shows the simulated water density (obtained by molecular dynamics simulation). The panel on the right shows our experimental data.

For a comprehensive understanding of structure formation at the solid-liquid interface, the influence of the solvent molecules is pivotal. We have developed a routine that allows for solvation layer mapping on surfaces using dynamic AFM [17, 18]. We successfully used this technique to identify different calcium from magnesium cations on the surface of dolomite [CaMg(CO3)2] [19]. Here, we collaborated with the group of A. S. Foster, Aalto University, Helsinki, funded by a DAAD project.

Recently, we extended our focus to non-flat terraces of surfaces by investigating the solvation structure around defect sites [20] and step edges.

 

Solvation across a step edge at the gypsum-water interface.
Solvation across a step edge at the gypsum-water interface.

[19] Söngen et al., Langmuir 33 (2017) 125-129
[20] Söngen et al., Phys. Rev. Lett. 120 (2018) 116101

Molecular Self-Assembly at the Solid-Liquid Interface

The presence of the molecule "Congo Red" on calcite [22] causes both a restructuring of the surface as well as the formation of self-assembled structures.
The presence of the molecule "Congo Red" on calcite [22] causes both a restructuring of the surface as well as the formation of self-assembled structures.

The structure formation of organic molecules at the solid-liquid interface is a key aspect of understanding many application-relevant processes (such as scale inhibition or mimicking biomineralization). We have studied the pH-dependent self-assembly of organic molecules on calcite [21] and compared the surface structures with the macroscopic effect on crystallization.

[21] M. Schreiber et al., Soft Matter 9 (2013) 7145
[22] R. Momper et al., Langmuir 31 (2015) 7283