AVS2017 Session SS+HC-TuM: Controlling Mechanisms of Surface Chemical Reactions

Adsorption of multifunctional molecules onto semiconductor surfaces provides a pathway to functionalize the surface while leaving unreacted functional groups available for reaction. These hybrid interfaces are important for applications in biosensors, microelectronics, and energy storage. These applications require homogeneous and controllable adsorption of these complex molecules. In this presentation, we will discuss two studies that help elucidate the driving forces governing the overall product distribution during adsorption of multifunctional molecules.

The first example describes the adsorption of homobifunctional pyrazine on the Ge(100)-2x1 surface. Pyrazine (C₄H₄N₂) is a six-membered ring molecule with two nitrogen atoms in the para position. The electron-rich nitrogen possesses a lone pair that can easily react with the surface through a dative bond. While benzene is not known to react with germanium, the addition of nitrogen to the aromatic ring opens up additional reaction channels. Previous studies of pyridine (C₅H₅N) on Ge(100) showed that reaction through the nitrogen can yield a mix of adsorption products. Our X-ray photoelectron spectroscopy (XPS) results suggest that pyrazine reacts with Ge(100)-2x1 to form a mix of reaction products. Analysis of the N(1s) and C(1s) spectra indicates that reaction of pyrazine occurs through the N-dative bond and via N=C and C=C cycloaddition reactions. The product distribution was found to be a function of temperature and coverage. This dependence suggests differences in the reaction channels as well as cooperative effects between adsorbate molecules.

The second example describes the adsorption of homotrifunctional 1,2,3-benzenetriol on Ge(100)-2x1 surface. 1,2,3-benzenetriol (C₆H₆O₃) is a trihydroxy phenol with three neighboring hydroxyl groups. The proximity between OH-groups promotes intra and inter-molecular hydrogen bonding. XPS and Fourier transform infrared spectroscopy results demonstrated that the reaction occurs through an OH-dissociative pathway. Coverage dependent shows that the extent of surface reactions per molecule decreases as the surface becomes crowded, and we predict this behavior is due to a combination of sterics and intermolecular hydrogen bonding. It has been previously shown that intermolecular hydrogen bonding can lower the reactivity of an OH group and prevent further reactions. Our theoretical analysis confirms that adsorbed molecules will preferentially interact with each other if unreacted OH groups are in close proximity. We expect that these interactions contribute to lowering the saturation coverage, as well as decrease the amount of dissociation events per molecule.

The bottom-up construction of low-dimensional macromolecular nanostructures directly on a surface is a promising approach for future application in molecular electronics and integrated circuit production. However, challenges still remain in how to control the reaction pathways toward the formation of targeted nanostructures or dimensions. In this presentation, I will report our recent studies on the on-surface synthesis of low dimensional organic nanostructures on different substrate surfaces. These studies were performed under ultra-high vacuum (UHV) conditions using a combination of scanning tunnelling microscopy (STM), low energy electron diffraction (LEED) and high-resolution X-ray photoelectron spectroscopy (HR-XPS). Several examples will be shown to demonstrate that by employing different substrate templates,^1-4 special high-dilution synthesis condition⁵ or hydrogen bond protection with dedicated precursor molecules on surfaces⁶, surface reactions can be driven with different pathways towards the desired products. For example, on Cu(111) and Cu(110), after depositing the same precursor molecule, 4,4’’-dibromo-meta-terphenyl (DMTP), at elevated temperatures, Ullmann coupling reaction can proceed via different pathways to form different nanostructures.^1-3 While on the Cu(110)-(2x1)O surface, by controlling the width of the Cu stripes, the spatial confinement effect can steer the reaction of DMTP to form either 1D zigzag organometallic oligomeric chains with different lengths or organometallic macrocycles with different widths.⁴ More interestingly, when recently we transplanted the (pseudo-) high dilution method in solution to the conditions of on-surface synthesis in UHV, we found that the reaction of DMTP on Ag(111) can be tailored to specially form cyclic hyperbenzene with high-yield (84%) in contrast to the commonly formed zig-zag open-chain polymers.⁵ Except for the presentation of these different structures on surfaces, we will also discuss the principles and mechanisms behind.

This work is supported by the National Natural Science Foundation of China (21473178) and the National Basic Research Program of China (2013CB834605)

References:

(1) Fan, Q.; Wang, C.; et al., Angew. Chem. Int. Ed.2013, 52, 4668.

(2) Fan, Q.; Wang, C.; et al., ACS Nano2014, 8, 709.

(3) Dai, J.; Fan, Q.; et al., Phys. Chem. Chem. Phys.2016, 18, 20627.

(4) Fan, Q.; Dai, J.; et al., ACS Nano2016, 10, 3747.

(5) Fan, Q.; Wang, T.; et al., ACS Nano2017, DOI: 10.1021/acsnano.7b01870.

(6) Wang, T.; Lv, H.; et al., Angew. Chem. Int. Ed. 2017, 56, 4762.

The rapidly emerging field of “single-atom catalysis” aims to drastically reduce the amount of precious metal required to catalyze chemical reactions by replacing nanoparticles with single-atom active sites. Although there are now many reports of active single-atom catalysts [1], the concept itself remains controversial because it is challenging to characterize real catalysts and determine the reaction mechanism. In our work, we study fundamental properties of supported single metal atoms using a surface science approach. We employ the Fe₃O₄ (001) surface as a model support, because it can stabilize dense arrays of single metal atoms to temperatures as high as 700 K [2,3]. In this contribution, we address the adsorption behavior and reactivity of the Pt and Rh adatoms, dimers, and small clusters using a combination of atomically resolved STM and non-contact AFM, high-resolution spectroscopy, and density functional theory. We conclude that Pt adatoms are inactive because CO adsorption results in mobility, and rapid sintering into Pt₂ dimers [5]. Pt dimers, on the other hand, are stable, and highly efficient CO oxidation catalysts. By isotopically labelling the oxide surface with ¹⁸O, we unambiguously show that a Mars van Krevelen mechanism is responsible for the catalytic activity. Rh adatoms interact more strongly with the oxide, and do not sinter upon adsorption of CO. As a result, we find that Rh adatoms do catalyze CO oxidation, also via an MvK mechanism.

[1] – Acc. Chem. Res. 46(8), pp.1740-1748.

[2] – Phys.Rev.Lett.108(2012): 216103

[3] – Science 346 (2014): 1215-1218.

[4] – Angew. Chem. Int. Ed. 54.47 (2015): 13999-14002.

[5] – PNAS 113.32 (2016): 8921-8926.

Platinum supported on oxides, such as TiO₂, are widely studied catalysts to drive oxidation reactions. Even though there are fundamental studies that have been done on single crystal Pt and TiO₂ to understand the reactivity and the mechanism, there lies a sizable knowledge gap due to the complexity of the real catalytic systems compared to the single crystal studies. We studied CO oxidation on a unique model system where Pt nanoparticles (NPs) are deposited on TiO₂ NPs supported on an inert HOPG surface. Our study takes the complexity of the material a step forward. In-operando Ambient Pressure X-ray photoelectron spectroscopy (AP-XPS) was used to study the oxidation states of Pt, Ti, and O during the reaction to understand the role of different oxidation states of the elements on the reaction mechanism. Ex-situ prepared model catalyst which mostly contains a mixture of Pt(4) and Pt(2) were first heated to obtain a mixture of Pt(0) and Pt(2). During the reaction, the TiO₂ remains stoichiometric with no indication of any change in the oxidation state. At 400 K, CO is adsorbed on Pt resulting in a decrease of the Pt(2)/Pt(0) ratio. O1s spectra show the formation of Pt-O bond at 450 K. A rapid decrease of O1s (Pt-O) and a decrease of Pt(2)/Pt(0) ratio were observed simultaneously with CO₂ production at 500 K. In conclusion, we were able to study CO oxidation on a more complex model system of Pt/TiO₂ and followed the reaction mechanism. CO first adsorb on Pt and reacts with the oxygen that is dissociated on Pt sites which agree with the existing Langmuir–Hinshelwood (LH) mechanism. Furthermore, we found out that CO gets adsorbed on Pt(2) rather than on Pt(0) when there is a mixture of Pt(2) and Pt(0).

Nitrogen containing carbon materials have been reported to show catalytic activities such as an oxygen reduction reaction (ORR) in fuel cells. Among several types of nitrogen species in carbon materials, pyridinic nitrogen (nitrogen atom bound to two C atoms) was found to create ORR active sites in our previous work [1]. We then try to prepare catalytically active carbon surfaces covered with pyridinic nitrogen-containing aromatic molecules with high density. Here we report model catalyst studies using HOPG (highly oriented pyrolytic graphite) electrode covered with pyridinic nitrogen-containing aromatic molecules (dibenz[a,c] acridine (DA) molecule). The DA molecules were deposited on HOPG with different coverage by simply dropping solutions of the DA molecules at room temperature. Scanning tunneling microscopy (STM) measurements revealed that a well-ordered two-dimensional structure of DA monolayer is formed on HOPG surfaces with high densities via π-π interaction, rather than aggregates to form three-dimensional clusters. The nitrogen concentration of the DA-covered HOPG surfaces was estimated to be 0.5~1.5 at.% by XPS. The DA-covered HOPG model catalysts reveled activities of ORR. The specific activity per pyridinic nitrogen atom was estimated to be 0.08 (e sec^-1 pyriN^-1) at 0.3 eV, which is comparable to that for pyridinic nitrogen incorporated graphene sheets (0.07 ~ 0.14 (e sec^-1 pyriN^-1))[1]. The current densities at 0.1, 0.2, and 0.3 V vs RHE were in proportional to the surface coverage of DA molecules, indicating that the ORR active site was created by DA molecule adsorbed on HOPG. The present studies clearly show that fixing nitrogen-containing aromatic molecules on graphitic carbon materials is one of promising approaches to prepare active ORR carbon catalysts.

References

[1] D.Guo, R.Shibuya, T.Kondo, J.Nakamura, et al., Science, 351 (2016), 361-365.

Filtration systems for absorption and decomposition of chemical warfare agents (CWAs) are the first line of defense against exposure to these toxic compounds. Composite materials (such as ASZM-TEDA) commonly used in filtration systems consist of high-surface-area carbon supports impregnated with various metal oxides, including CuO_x. Despite decades of work to develop highly effective and versatile filtration materials, little is known about the mechanisms of CWA degradation by material surfaces and filter deactivation/poisoning, in part due to the challenges involved with spectroscopic characterization of filtration material surfaces under operating conditions. Enabling the rational design of more advanced filtration and decomposition materials for broad-spectrum protection against CWAs and other toxic industrial compounds requires a sophisticated understanding of the chemical mechanisms behind CWA sorption and degradation on the molecular scale.

We will present work detailing the mechanism of adsorption and decomposition of dimethyl methylphosphonate (DMMP), a CWA simulant, on polycrystalline copper oxide surfaces. Ambient-pressure XPS enables examination of these surfaces and adsorbed species upon exposure to DMMP and other common atmospheric gases, such as water vapor and NO_x. Multiple decomposition products are observed on CuO_x surfaces, and the oxidation state of the metal appears to influence the mechanistic pathway. Complementary density functional theory (DFT) and in situ FTIR studies corroborate our experimental findings and are used to propose likely decomposition pathways. Exposure of the CuO_x surfaces to water vapor or NO_x prior to introducing DMMP affects the decomposition product distribution, but does not appear to significantly inhibit the initial DMMP adsorption.

The surface deposition of metal-organic frameworks (MOFs) opens the opportunity to use these highly designable materials for applications in electronics, data storage, sensing or biointerfaces. Several protocols have been reported to grow or attach MOFs onto a variety of surfaces, but nevertheless, the bandwidth of successfully deposited MOFs is still extremely limited: Besides the archetypical HKUST-1 and ZIF-5, only a family of tetragonal layer systems (with or without pillars) could be grown into well-defined SURMOF layers [1]. Although with some of these systems impressive results for different devices, such as photovoltaic cells, have been reported, the limitation to these classes of MOFs is unsatisfactory.

In this talk, we wish to discuss that the lack of success with other systems is related to the fact that most projects relied solely on the well-established surface-template effects, in which the coordination of surface-bound, functional groups to the metal SBU defines not only the binding mode but the orientation of the whole film. We will show that the rules on the surface differ from the ones in solution by a reduced number of trajectories, which in turn hamper certain reactions [2]. It is also of importance to keep in mind the different surface energies of the crystal surfaces, which can dictate a reorientation of the preformed system by Ostwald ripening during the deposition process [3]. All these processes involve different activation energies, the careful control of which opens the opportunity to grow well-oriented films of MOF systems, which have behaved notoriously difficult up to now.

References

[1] Zhuang, J.-L.; Terfort, A.; Wöll, C. Coord. Chem. Rev. 307 (2016) 391-424.

[2] Zhuang, J.-L.; Kind, M.; Grytz, C. M.; Farr, F.; Diefenbach,M.; Tussupbayev, S.; Holthausen, M. C.; Terfort, A. J. Am. Chem. Soc. 137 (2015) 8237-8243.

[3] Yu, X.-J.; Zhuang, J.-L.; Scherr,J.; Abu-Husein,T.; Terfort, A. Angew. Chem. Int. Ed. 29 (2016) 8348-8352.