AVS2012 Session IS+AS+SS+EN-TuM: In Situ Spectroscopic Studies of Catalysis and Gas-Solid Reactions

Hydrogen fuel generation by solar water splitting in photoelectrochemical cells (PEC) is one of the first steps in artificial photosynthesis and an essential part of the holy grail of solar energy conversion. Iron oxide, literally "rust", is an interesting PEC photoanode material because of its affordability, good stability, good spectral match of the solar spectrum, and yet controversial because of its poor electronic structure. At present, iron oxide is taking center stage as prospective PEC anode material.

PEC electrodes are typically semiconducting metal oxides to form electron-hole pairs when strucked by light. In the photoanodes such as hematite, the generated holes must diffuse to the iron oxide surface where they can oxidize water to oxygen. However, the electronic structure of iron oxide is such that the photogenerated holes tend to recombine and annihilate with the electrons before reaching the surface and performing the required chemical work on water splitting. Currently, researchers worldwide try to understand the peculiarities of iron oxide so as to invent strategies to improve this material.

The Advanced Light Source produces soft X-rays which are optimally suited to study the electronic structure of electrode materials and which can detect electron holes. But the holes needed for solar water splitting by iron oxide require an anodic electric bias plus the illumination. Moreover, the holes are transitional and quite elusive. Also, soft X-rays cannot easily peek into a PEC cell. The unique design of the in-situ cell at the ALS has overcome the burden [1-3]. Recently the experiment has been performed for studying, under in-situ and operando conditions, the hole generation in a specifically designed photoelectrochemical cell. The oxygen valence band signature was recorded while tuning the PEC relevant parameters, two different types of holes in the valence band near the Fermi energy are discovered [4].

References:

[1] "X-ray Emission Spectroscopy of Hydrogen Bonding and Electronic Structure of Liquid Water", J.-H. Guo et al., Phys. Rev. Lett. 89, 137402 (2002).

[2] “Electronic Structure of Cobalt Nanocrystals Suspended in Liquid”, H. Liu et al., Nano Lett. 7, 1919 (2007).

[3] “In situ soft X-ray absorption spectroscopy investigation of electrochemical corrosion of copper in aqueous NaHCO₃ solution”, P. Jiang et al., Electrochem. Comm. 12, 820 (2010).

[4] “Direct Observation of Two Electron Holes in a Hematite Photoanode during Photoelectrochemical Water Splitting”, A. Braun et al., J. Phys. Chem. C 116, 16870 (2012).

Heterogeneous catalysts consisting of small particles having a high concentration of structural defects and under-coordinated sites make up the majority of the catalytic processes in industrial chemistry. One important recent example of interest shows that the interface-confined coordinatively unsaturated ferrous (CUF) sites together with the metal supports (FeO_1-x/Pt(111)) are active for dioxygen activation which causes the ensemble to be highly efficient for CO oxidation at low temperature under typical operating conditions of a proton-exchange membrane fuel cell.[1-2] In this work, we report another spectroscopic evidence to further confirm an enhanced reactivity at the edges of small catalyst particles. The system in interest is partially oxidized SnOx (Sn²⁺) nanoislands supported on Pt(111) for ethanol oxidation reaction (EOR), an electrode material in a direct alcohol fuel cell (DAFC). Our findings suggested that SnO_x/Pt(111) inverse catalysts have improved activity for EOR in acidic media as compared to a bare Pt(111) surface.[3] We also found that the most active surface had a small coverage of SnO_x (0.3- 0.4 ML). Water activation at low potentials is currently attributed to be the promoting effect of SnOx nanoparticles, since this enhances the oxidation of chemisorbed CO formed on Pt sites during the EOR.[4] To better understand this increased activity, we performed study with the goal to indicate the actual state of Sn in SnO_x nanoislands before and after the SnO_x /Pt(111) is used in EOR showing the increased activity: purely oxide Sn or mixed Sn alloy and Sn oxide, using a combination of APXPS and XANES techniques. BE shift in the XPS core-line spectra of Sn and O, soft X-ray XANES spectra (Sn M4,5-edge, O K-edge) will be collected and compared to the corresponding XPS spectra (Sn 3d, O 1s) to explain the actual state of Sn before and after the SnO_x/Pt(111) is used in the EOR.

1. Fu, Q., et al., Interface-Confined Ferrous Centers for Catalytic Oxidation. Science, 2010. 328: p. 1141.

2. Deng, X., et al., Reactivity Differences of Nanocrystals and Continuous Films of α-Fe2O3 on Au(111) Studied with In Situ X-ray Photoelectron Spectroscopy. J. Phys. Chem. C, 2010. 114: p. 22619.

3. Zhou, W.P., et al., Enhancement in Ethanol Electro-Oxidation by SnOx Nanoislands Grown on Pt(111): Effect of Metal Oxide-Metal Interface Sites. Journal of Physical Chemistry C, 2011. 115: p. 16247.

4. Axnanda, S., W.P. Zhou, and M.G. White, CO Oxidation on Nanostructured SnOx/Pt(111) surfaces: Unique Properties of Reduced SnOx. Phys. Chem. Chem. Phys, 2012. Submitted.

X-ray Photoelectron Spectroscopy (XPS) has been the main surface analysis method for determining the chemical environment and coordination of nitrogen and transition metal (TM) in the non-precious group metal oxygen reduction reaction (ORR) electrocatalysts. Even though there is an agreement that Me-N_x serve as one of the possible active sites in ORR, the distribution of Me-N₂ vs Me-N₄ centers and their specific role still remains unresolved. XPS which heavily relies on use of reference spectra in accurate identification of species cannot address this issue directly as no reference compounds with Me-N₂ moieties are available. The assignment of peaks and nitrogen coordination is not straightforward due to overlapping peaks that appear within a narrow energy window of 2.5-eV and the full width half maximum (fwhm) for individual species is on the order of 1.2-1.5-eV. Being able to calculate binding energy shifts based on molecular structure can be very important tool for assisting in this task. We will report on BE shifts that have been calculated at the DFT level and their comparison to experimentally obtained values for metal-less and metal-containing porphyrins. Information obtained from the DFT calculations will be used as input into curve-fitting XPS spectra for various model N-Me containing compounds as well as from electrocatalsyst. We will compare chemical information derived from conventional XPS as well as in-situ ambient-pressure XPS using variable energy synchrotron source.

Catalysts are used to facilitate the important industrial chemical processes, leading to products valued in the trillions of dollars annually just in the U.S and most catalysts used in large-scale processes are solids. To maximize the number of sites available for reaction, catalysts are typically comprised of metallic/metal oxide nanoparticles dispersed on high surface area supports. The activity and selectivity of metallic nanoparticles strongly depend on their size, shape and composition [1-8]. In order to design more active and selective catalysts, it is essential to identify the catalytically active sites and understanding their geometric and electronic properties which requires: (1) synthesis of well-defined catalyst structures and (2) the ability to correlate individual reaction pathway(s) with the type of active site(s) available on the catalyst surface under reaction conditions.

1. Pd nanoparticles synthesis in solution: Understanding the nucleation and growth mechanisms.
2. Supported Pt, PtRe and PtNi nanoparticles under aqueous phase reaction condition: Correlating the structural and electronic properties with the catalytic activity and selectivity.

[1] Boudart, M. Adv. Catal. 1969, 20, 153.

[2] Boudart, M. Journal of Molecular Catalysis 1985, 30, 27.

[ 3] Ichikawa, S.; Poppa, H.; Boudart, M. Journal of Catalysis 1985, 91, 1.

[4] Somorjai, G. A.; Carrazza, J. Industrial & Engineering Chemistry Fundamentals 1986, 25, 63 [ 5] Liu, Z.; Hu, J. E.; Wang, Q.; Gaskell, K.; Frenkel, A. I.; Jackson, G. S.; Eichhorn, B. Journal of the American Chemical Society 2009, 131, 6924.

[6] Alayoglu, S.; Nilekar, A. U.; Mavrikakis, M.; Eichhorn, B. Nature Materials 2008, 7, 333.

[7] Sinfelt, J. H. Journal of Catalysis 1973, 29, 308.

The platinum group metals are known to be excellent catalysts for the oxidation of carbon monoxide, and the reaction mechanisms over the surfaces of these metals have been studied for a long time. Nevertheless, only during recent years a new picture has emerged which suggests that the catalytically active phase often is formed first under reaction conditions – which implies realistic pressures rather than ultrahigh vacuum (UHV) – and that it is different from the adsorption structures known from UHV experiments. In the case of the Pt(111) surface a very oxygen-rich chemisorbed phase has been suggested as the catalytically active phase [1], whereas a surface oxide have been suggested for the Ru(0001) surface [2]. Thus, for each different surface different phases and mechanisms might be at play, and, moreover, the phase might depend quite strongly on the conditions (pressure and temperature) used.

With this in mind we have studied the CO oxidation reaction over the Ir(111) surface and the related adsorption systems of CO and oxygen on Ir(111) using a combination of in situ Ambient pressure x-ray photoelectron spectroscopy (APXPS) – carried out at the new APXPS instrument at the Swedish synchrotron radiation facility MAX IV Laboratory – and ex situ Scanning tunnelling microscopy (STM) and x-ray photoelectron spectroscopy (XPS) measurements performed in UHV.

A recent surface x-ray diffraction study reports different oxygen phases for the Ir(111) surface at oxygen pressures from 10^-6 to 100 mbar – chemisorbed oxygen, a trilayer, a multilayer oxide, and a bulklike oxide [3]. Concentrating on pressures at around 1 mbar, we find a variety of oxygen-rich structures. The corresponding CO adsorption phase formed at 1 mbar CO pressure is an assembly of separated CO₁₆ clusters with the CO molecules sitting in on-top sites [4].

The reactivity at 1 mbar total pressure (O₂:CO ratio 9:1) and at different temperatures was studied by APXPS and simultaneous monitoring of the gas composition. We find that the phase with the highest activity for the oxidation of CO is a surface phase which contains both CO and oxygen. By comparing with the measured adsorption structures of oxygen we find that the oxygen structure is quite similar to the p(2x1)-O structure formed on Ir(111) under UHV conditions. This contrasts with was is found for other platinum group metals such as the Pt(111) surface [2], for which CO oxidation is favoured over oxygen rich phases.

[1] A. L. Gerrard, J. F. Weaver, J. Chem. Phys. 123 (2005) 224703.

[2] H. Over et al.,, Science 287 (2000) 1474.

[3] Y. B. He et al., J. Phys. Chem. 112 (2008) 11946.

[4] L.-M. Yang, S.-L. Yau, J. Phys. Chem. B. 104 (2000) 1769.