Symposium ES06—Atomic-Level Understanding of Materials in Fuel Cells and Electrolyzers

Interfaces play an essential role in nearly all aspects of life and are critical for electrochemistry. Electrochemical systems ranging from high-temperature solid oxide fuel cells (SOFC) to batteries to capacitors have a wide range of important interfaces between solids, liquids, and gases which play a pivotal role in how energy is stored, transferred, and/or converted. This talk will focus on our use of ambient pressure XPS (APXPS) to directly probe the solid/gas and solid/liquid electrochemical interface. In particular, I will discuss our progress to probe the solid/liquid interface across a range of metal/aqueous systems and the role halide ions play on the stability of the interfacial properties. Additionally, I will highlight some of our recent investigations into CO₂ adsorption phenomena on various metals as we move closer to developing the capabilities for studying CO₂ reduction under electrochemical conditions. Information gained from these studies will aid in the guided design and control of future electrochemical interfaces.

Understanding of the in-reaction structure and the functionality of a catalyst are closely connected. Detailed structural information can be revealed through vibrational techniques; however, a number of challenges need to be overcome. For instance, low scattering rates for Raman spectroscopy have made in-situ Raman of catalysis difficult. In this contribution, we demonstrate stimulated Raman spectroscopy (SRS) as an in-situ probe to address the challenge of low scattering rates in Raman. As a nonlinear spectroscopic technique, SRS holds several advantages including efficiency. Leveraging the improved efficiency, we apply SRS to measure the vibrational structure of an amorphous cobalt oxide catalysis during the oxygen evolution reaction (OER). Additional experiments using isotope labelling are used to identify the nature of the vibration structure. These results present the potential of SRS as a useful tool to investigate the much sought-after catalyst surface in-situ. The spectroscopic techniques, methods of analysis, and the experimental results will be presented.

The dramatic change in electrochemical behavior of nickel (oxy)hydroxide films upon incorporation of Fe ions provides an opportunity to establish effective electrocatalyst design principles. It is well established that the addition of miniscule amounts of Fe(III) ions to nickel (oxy)hydroxide results in significant improvements in kinetics for the electrocatalytic water oxidation, but the chemistry underlying these gains remains a matter of debate. This presentation will introduce and discuss an asymmetric geometric distortion that was revealed by structure-property analysis of a series of disordered Fe–Ni (oxy)hydroxides.[1] These distortions were revealed by first identifying prominent structural motifs in the composition series by X-ray absorption spectroscopy, then correlating their structural descriptors to electrochemical and structural data acquired by in-situ UV-visible spectroelectrochemical data. Structural analysis indicated the presence of di-μ-hydroxo Ni-Ni and di-μ-hydroxo Ni-Fe motifs, with relative proportions reliant on Fe-content. Both motifs exhibit trigonal distortions away from ideal octahedral coordination environments, imparting a local D_3d symmetry that is known in crystalline β-Ni(OH)₂. The distortion of Ni-Ni motifs in both the oxidized and reduced phases of the disordered materials is comparable to that observed in β-Ni(OH)₂, with O-Ni-O bond angles of 82 degrees. The Ni-Fe motif contrasts with this, exhibiting a relaxed distortion in the reduced phase (84-degree bond angle) and an increased distortion (<80-degree bond angle) in the oxidized phase. Spectroelectrochemical experiments reveal a previously unreported change in optical absorbance at ca. 1.5 V vs. RHE in Fe-containing samples, which is assigned as the oxidation of Ni(II) ions in the distorted di-μ-hydroxo Ni-Fe motifs. A mechanism is proposed wherein the pre-catalytic redox process is assigned to oxidation of Ni-Ni motifs, while the catalytic process is decoupled from this process and assigned to the electrochemical oxidation of Ni sites in the Ni-Fe motifs. The asymmetric distortion is attributed to the redox-inactivity of the Fe(III) ions in the material and the favored Fe(III)-O bond length being located between that of Ni(II)-O and Ni(IV)-O.

[1] Smith, R. D. L. et al. Energy Environ. Sci. 2018, 11, 2476–2485.

Water electrolyzers and fuel cells are promising technologies that can play a role in a sustainable future. Catalyst development is needed to advance these systems, to improve performance and to lower costs such that the systems can be cost-competitive with respect to conventional technology. This talk aims to describe catalyst development challenges and opportunities within this area. The focus will be on catalyzing interconversions involving H₂, O₂, H₂O, and H₂O₂, with an emphasis on low- or zero-precious metal content catalyst materials. Examples will be given for catalyst design from first principles, the development of materials based on those principles, and their integration into functional devices.

Oxide materials are important electrocatalysts for the oxygen reduction and evolution reactions in fuel cells and electrolyzers. Electrochemical studies of well-defined surfaces grown by pulsed laser deposition or molecular beam epitaxy establish the intrinsic activity of oxide catalysts in a way that cannot be realized with polydisperse nanoparticle systems, and can also reveal how different terminations¹ and structures² affect the energetics and kinetics. Furthermore, the surface speciation of epitaxial films in an aqueous environment can be quantitatively assessed in situ using ambient pressure X-ray photoelectron spectroscopy. Understanding oxide electronic structure in an aqueous environment is also critical for promoting charge transfer reactions. To this end, we investigate changes in the electronic structure as a function of the oxygen and water chemical potential, enabling comparison with the metal redox potential³ and catalytic activity.⁴ The resultant molecular-level understanding of interfacial interactions developed from epitaxial surfaces can guide the rational design of high-surface-area oxide catalysts for technical applications.

References
[1] K.A. Stoerzinger, R. Comes, S.R. Spurgeon, S. Thevuthasan, K. Ihm, E.J. Crumlin, S.A. Chambers. “Influence of LaFeO₃ Surface Termination on Water Reactivity”. J. Phys. Chem. Lett. 8, (2017) 1038-1043.
[2] Stoerzinger, K.A. Choi, W.S. Jeen, H. Lee, H.N. and Shao-Horn, Y. “Role of Strain and Conductivity in Oxygen Electrocatalysis on LaCoO₃ Thin Films”. J. Phys. Chem. Lett. 6 (2015) 487-492.
[3] Stoerzinger, K.A. Du, Y. Ihm, K. Zhang, K.H.L. Cai, J. Diulus, J.T. Frederick, R.T. Herman, G.S. Crumlin, E.J. Chambers, S.A. “Impact of Sr-Incorporation on Cr Oxidation and Water Dissociation in La_(1-x)Sr_xCrO₃”. Adv. Mater. Interfaces 5 (2018) 1701363.
[4] Stoerzinger, K.A. Hong, W.T. Wang, X. Rao, R.R. Subramanyam, S.P.B. Li, C. Ariando, Venkatesan, T. Liu, Q. Crumlin, E.J. Varanasi, K.K. Shao-Horn, Y. “Decreasing the Hydroxylation Affinity of La_(1-x)Sr_xMnO₃ Perovskites To Promote Oxygen Reduction Electrocatalysis”. Chem. Mater. 29 (2017) 9990–9997.

We present an approach that couples electrochemistry, surface science, and density functional theory (DFT) calculations to precisely identify structure property relationships in electrocatalytic oxygen evolution reaction (OER) catalysts. The OER is an essential component in electrochemical CO₂ conversion systems, and mixtures of nickel, cobalt, and iron have shown remarkable alkaline OER activity that exceeds state-of-the-art IrO₂. However, atomic-level understanding of how OER activity evolves with the catalyst shape, size, composition, and support are still lacking in the literature. Our approach grows well-defined transition metal electrocatalysts with precisely controlled morphology, composition, and loading. Scanning tunneling microscopy (STM) and X-ray photoelectron spectroscopy quantify the catalyst loading (mol/cm²), number density (particles/cm²), and morphology with unrivaled spatial resolution. Electrochemical testing then correlates mass activity, turnover frequency, and specific current density with the catalyst structure and composition. We find clear experimental and computational evidence for edge-enhanced OER at pure Fe₂O₃ structures, and this insight was translated into the synthesis of a nanoparticulate analog for demonstration in alkaline electrolyzer devices. On the other hand, mixed metal Ni/Fe systems showed a more complex evolution of structure-dependent activity that did not directly depend on the relative density of under-coordinated edge sites. Our approach provides atomic-level structure-property relationships for emerging OER catalysts that can be used to guide scalable synthetic strategies and expedite the deployment of more energy efficient electrocatalyst systems.

RuO₂ and IrO₂ are among the best catalysts for the oxygen evolution reaction (OER), an electrochemical reaction that limits the efficiency of water splitting. The high activity of these rutile oxides has been attributed to their ability to stabilize the OER intermediates via electroadsorption which involves related elementary proton and electron transfer steps. However, the rates (kinetics) of the individual proton and electron transfer steps related to the oxygen electroadsorption are still unclear. In this presentation, we present our measurements of these kinetics on well-defined RuO₂ and IrO₂ surfaces. We use rate-dependent cyclic voltammograms to extract the rates of the oxygen electroadsorption on RuO₂ and IrO₂. Well-defined surfaces are used to minimize convolution from different facets. We further identify the pathway and related fundamental rate laws by investigating the influence of different electrolytes on kinetics. Our results provide insights into the kinetics of oxygen electroadsorption on oxides and the fundamental origin of the sluggishness in the OER kinetics.

Oxygen reduction reaction (ORR) cathodes account for the largest efficiency loss and one of the largest cost portions of PEM fuel cells. There has been intensive research effort on ORR electrocatalysts from precious metals to nonprecious metals (also called platinum group metal (PGM)-free catalysts). Developing highly efficient precious metal electrocatalysts and highly active PGM-free catalysts still remains a grand challenge, while making them stable and durable in the harsh PEM fuel cell environment is even more challenging. In this talk, we will discuss our research to understand the instability of the catalysts from materials innovation to advanced characterization. We will introduce our environmental TEM study of PGM catalysts, PGM-free catalysts instability and the proposed solutions.

Recent advances of polymer electrolyte membrane fuel cell (PEMFC) technology has been substantial and successfully led to the commercialization of fuel cell electric vehicles (FCEVs) such as Toyota Mirai. However, the continuation of FCEV deployment is facing several hurdles, of which one major issue is related to the high cost of fuel cell system as compared to conventional engines. Economic analysis has shown that catalyst is by far the most expensive material component for high production volume. ^[1] Current state-of-art PEMFCs use platinum-based catalysts to speed up hydrogen oxidation reaction (HOR) and oxygen reduction reaction (ORR) on the anode and the cathode, respectively. While the HOR proceeds very fast on Pt surface, the ORR has a sluggish kinetics, as it involves sequential steps of proton coupled electron transfer processes and a “scaling” issue makes it difficult to optimize every elemental reaction step simultaneously. ^[2]
Many previous wisdoms have focused mainly on ORR catalyst material design, for example, alloy, core-shell, shape control and so on. Engineering catalyst-electrolyte interface properties may offer another way to overcoming the problem. One recent effort involves the modification of Pt catalyst surface with organic ligands and has demonstrated some successes in enhancing catalyst activity. ^[3-7] The improvement was mostly attributed to either the change of surface electronic structure of the catalyst, or the protection of Pt active sites from poisoning by strong anion adsorption. ^[3-5] In this study, we systematically examined a series of organic ligands with different structures. Combining material characterization with electrochemical analyses allowed us to assess not only the catalyst activity and durability, but also the ligand stability. Such a comparative study is expected to provide additional insights on ligand structural design for further improvement of catalyst performance.

References
(1) DOE Fuel Cell Technologies Office Record 17007: Fuel Cell System Cost – 2017
(2) M. T. M. Koper, et al., J. of Electroanal. Chem., 2011, 660, 254-260
(3) Y. H. Chung, et al., J. Phys. Chem., Lett. 2013, 4, 1304-1309
(4) Z. Y. Zhou, et al., J. Phys. Chem., C, 2012, 116, 10592-10598
(5) D. Strmcnik, et al., Nature Chem., 2010, 2, 771
(6) K. Saikawa, et al., Electrochem. Com., 2018, 87, 5–8
(7) M. Kobayashi, et al., ECS Abstract MA2018-01 2360

Polymer Electrolyte Fuel cell (PEFC) is attracted attention as one of the next generation power sources because it exhausts theoretically only water. However, its high cost and short service life are inhibiting the wide-spread commercialization. It is thought as one reason that this is attributing to using the platinum-based electrocatalyst. To reduce cost and lengthen service life radically, the non-platinum catalyst such as oxide, carbon-alloy, and the enzyme has been developing. The oxide-based catalyst has the high possibility as the alternative the platinum-based catalyst due to possessing the high stability under PEFC operation conditions. Several researchers had reported that a part of oxide, such as group 4- and 5-metal oxide and precious metal oxide, showed the high activity of oxygen reduction reaction (ORR). However, these ORR activity does not achieve that of the conventional platinum catalyst. One reason for this situation is that the active site on oxide surface is not fully understanding as the common recognition. Thus, it is hard to advance about the new concept to increase the active site density. The situation inside the catalyst layer during the ORR progress is too much complex since it needs to take consideration with the diffusion of oxygen, proton, electron, and water in catalyst layer other than the ORR mechanism on the oxide surface.
Therefore, we focused on the mono-layer molecule electrode forming by the metal oxide nanosheet in order to exclude a variety of the diffusion resistance on an electrode. This study focused on the role of the oxygen vacancy site on titanium oxide nanosheet (TiO₂ns) and titanium-niobium oxide nanosheet (TiNbO₂ns) and the redox site on ruthenium oxide nanosheet (RuO₂ns) for the ORR.
A mono-layer molecule electrode was prepared by Layer by Layer method combining with the poly-cation copolymer and each transition-metal oxide nanosheet. Various types of transition-metal oxide nanosheet were synthesized along previous literature. The deposition condition of transition-metal oxide nanosheet on the substrate was observed by AFM. The electrochemical measurement was conducted by the three electrode cell composing with a reversible hydrogen electrode as the reference and a carbon rod as the counter electrode. At that time, 0.1 M HClO₄ at room temperature was employed as the electrolyte. ORR activity was evaluated by the potential at -0.05 mA cm^-2 of difference current between oxygen flow conditions and nitrogen flow conditions.
The ORR activity of RuO₂ns mono-layer electrode showed about 0.6 V which was almost agreed with that of evaluation used powder sample. Addition, the redox peaks of RuO₂ns were clearly observed also in a mono-layer electrode. Thus, it is thought that a mono-layer electrode, which was prepared by Layer by Layer method, was also possible to evaluate accurately the electrochemical behavior including of ORR activity. In the case of TiO₂ns without oxygen vacancy or Nb doping, the ORR activity was lower than that of RuO₂ns. Thus, it is thought that ORR activity depends on the kind of transition metal.

Oxygen can only be reduced at the quadruple phase boundary (catalyst, carbon support, ionomer and oxygen) of the cathode catalyst layer with the non-conducting electrocatalyst. To maximize the quadruple phase boundary sites is crucial to increase the peak power density of each membrane electrode assembly. The quadruple phase boundary is depending on the ratio of catalyst, carbon support and ionomer. The loading of the catalyst layer is also crucial to the fuel cell performance. In this study, the non-stoichiometric α-MnO₂ manganese dioxide nanorod material has been synthesized and the ratios of carbon, ionomer, and catalyst loadings were optimized in alkaline membrane fuel cell. In total, ten membrane electrode assemblies have been manufactured and tested. Taguchi design method has been applied in order to understand the effect of each parameter. The conclusion finds out the ionomer has more influence (68% factor effect) on the alkaline membrane fuel cell peak power performance than carbon and loading.

Complexity of the fuel cell technology drives researchers to study different model systems in order to get insight into the processes during Solid Oxide Fuel Cell (SOFC) operation. One of such model systems is a Pt/YSZ where platinum plays role of a cathode on yttria-stabilized zirconia (YSZ) electrolyte. Different phenomena may lead to undesirable electrode deactivation. Along with irreversible morphological changes (e.g. film dewetting, blistering), another process, observed as an additional resistivity during anodic polarization, hinders oxygen ion transport in our model-electrode system. This process is assigned to the formation of Pt "phase" oxide at the Pt/YSZ interface and triple-phase boundaries. PtO_x is predicted to be responsible for passivation of the working electrode in Pt/YSZ model systems. Indirect but convincing evidences of electrochemically driven Pt oxidation on well-defined Pt electrodes are present in several studies. Although there is still a debate on the exact mechanism of PtO_x formation and its structure (PtO, PtO₂ or Pt₃O₄), there is no doubt it influences oxygen exchange kinetics, which decreases electrocatalytic activity of an electrode in Pt/YSZ model system.
Employing X-ray diffraction (XRD) combined with electrochemistry measurements at beamline P08, Petra III Synchrotron, DESY, Hamburg, we provide first direct evidence of the existence of PtO_x phase at the Pt(111)/YSZ(111) interface after anodic polarization, which decomposes once the potential is reversed. In situ/operando surface XRD on electrode/electrolyte interface reveals its structure and evolution on atomic scale. Our results show feasibility of applying synchrotron based surface XRD to study fuel cell related electrode/electrolyte model interfaces operando and contribute to better understanding of a complex multistep oxygen reduction and evolution reactions in electrochemical devices, such as SOFCs and oxygen lambda sensors.

Introduction of zirconia into ceria is known to enhance the reducibility of the latter, despite the fixed tetravalent state of zirconium. Here we use a host of techniques to explore the origin of this behavior, its relevance to surface oxidation state, and its impact on kinetic properties. We find, for example, from in situ XANES (X-ray Absorption Near Edge Spectroscopy) that while the surface of ceria is reduced relative to the bulk whether or not zirconium is present, the extent to which the surface and bulk regions differ depends strongly on the Zr concentration. Specifically, with increasing Zr, the differential between the two regions diminishes. In parallel, using electrical conductivity relaxation methods, it is observed that the bulk chemical diffusion of oxygen in Ce_0.8Zr_0.2O_2-δ is lower than that in undoped ceria. The connections between these and related observations is discussed.

Ceramic oxides are used for a wide variety of technologically relevant applications from electrochemical devices, novel resistive switching devices and oxygen sensors. Applications such as these typically rely upon the ability of oxides to conduct ions efficiently through the lattice. For oxides, such as CeO₂ (ceria), it is common to dope with aliovalent elements to increase the number of available charge carriers. While the methods developed for bulk-doping have been successful, they have provided little insight when optimizing the grain boundary (GB) ionic conductivity in polycrystalline ceramic oxides. Recent nanoscale compositional characterization has shown different nominal concentrations could result in orders of magnitude increase in GB ionic conductivity relative to the undoped samples [1]. This effect has been attributed to increases in local dopant/solute concentrations at the GB. This study aims to predict the optimal dopants that, when present in high concentrations, increase the ionic conductivity across the GB.

To show how the various dopants modulate the ionic conductivity across the GB, computational modeling is employed using molecular statics to optimize the GB interfacial structure for three symmetric tilt and one twist GB in CeO2. The GB structures are then doped using a high concentration profile obtained from ref 1. Molecular dynamics simulations are performed to correlate migration energetics with changes in conductivity for the various dopants. The energetic and structural properties of various oxygen migration pathways are compared for each dopant species to ascertain how nanoscale composition and local atomic structure can modulate ionic conductivity across GBs. The trends obtained using molecular dynamics are validated using density functional theory calculations. This study further develops our understanding of high solute GB composition and structure, enabling the development of methods such as selective doping to improve macroscopic ionic conductivity for both the grain and GB.

[1] Bowman, W. J. et al. (2017) Nanoscale, DOI: 10.1039/C7NR06941C
[2] We gratefully acknowledge ASU’s HPC staff for support and assistance with computing resources, NSF grant DMR-1308085, and ASU’s John M. Cowley Center for High Resolution Electron Microscopy.

Interfaces play important roles in many functional thin film devices. However, the atomic configurations at interfaces between two phases with different crystal structures are not easily determined experimentally, yet they are the key for understand the properties induced. Vertically aligned nanocomposites (VAN) typically have pillar-matrix nanostructures combining two distinct phases. Up to two orders of magnitude increase in out-of-plane ionic conductivity was observed using Sm-doped CeO₂ (SDC) /SrTiO₃ (STO) VAN films grown on STO substrates. To investigate the roles of vertical interfaces, we used random structure searching to explore the atomic configurations at the vertical interfaces. The structures were found to not contain any features that promoted interfacial oxygen diffusion. Instead, they suggested that vacancy movement may be suppressed near the interface due to the existence of trapping sites. In addition, motivated by the controversial claims of enhanced ionic conductivity at yttria stabilized zirconia (YSZ) /STO planar interfaces, we performed fully first-principles random structure searching at a model CeO₂/STO (001) interface. We found previously unknown stable interfaces structures that differ from the bulk phases. Vacancy formation energies and diffusion pathways were studied and their effects on ionic conductivity will be discussed.

Recent experimental studies have shown that the electrochemical reduction of CO₂ depends on many factors, including the composition and structure of the electrocatalyst, the composition of the electrolyte, and the effects of mass transfer between the bulk electrolyte and the catalyst. The choice of electrocatalyst affects the principal products produced. For example Ag would be the catalyst of choice to reduce CO₂ to CO, whereas Cu would be the preferred catalyst for the production of C₂₊ hydrocarbons and alcohols. The choice of alkali metal cation used for the electrolyte also influences catalyst activity and product selectivity. For both Ag and Cu, the action and the selectivity towards forming carbon -containing products versus hydrogen is strongly enhance by using Cs⁺ rather than Li⁺. Mass transfer of from the electrolyte to the catalyst and of OH^- anions from the catalyst to the bulk electrolyte also affect the product distribution. It will be shown that manner in which mass transfer affects the distribution of product formed during the electrochemical reduction of CO₂ is best understood from simulations of a complete electrochemical cell. This point will be illustrated based upon simulations of CO₂ reduction occurring in an aqueous electrolyte containing planar electrodes.

The development of effective electrocatalysts to convert carbon dioxide into value-added products such as methane and ethylene requires a detailed atomistic understanding of the reaction pathway. While it is well known that copper (100) surfaces exhibit a higher selectivity towards ethylene production, both electrochemical and non-electrochemical pathways have been proposed to describe this process. In both cases, the presence of adsorbed CO plays a crucial role and the pathways are observed to be active under different conditions. In this work, we study the role of the electrochemical environment in stabilizing CO across different copper surface sites and evaluate the energy barriers of key steps in the carbon dioxide electroreduction pathway. Using the recently developed Effective Screening Medium – Reference Interaction Site Method (ESM-RISM), we provide a detailed assessment of the roles of both applied voltages and electrolyte composition effects. Our results demonstrate that these two factors are synergistic, leading to behavior that qualitatively deviates from conventional models.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

Electrochemical reduction of emitted CO₂ to fuels and complex hydrocarbons represents a promising solution to efficiently store intermittent, renewable energies-generated electricity in the form of chemical energy of carbon-based fuels. Copper is one of the most interesting CO₂ reduction electrocatalysts, thanks to its ability to reduce carbon dioxide to complex hydrocarbons with good efficiency, however when in bulk polycrystalline form it has low activity and requires a high overpotential for the reduction reaction.
Oxide-derived Cu catalysts are one of the most efficient way to increase the intrinsic activity of Cu: this class of materials have higher selectivity for complex hydrocarbons and longer stability. Moreover, their CO₂ reduction activity and selectivity are strongly affected by the morphology, which controls the mass transport of the reactants to the active sites and influences the local pH.
In this contribution, the CO₂ reduction activity and selectivity of Cu catalysts derived from the reduction of Copper Oxide (CuO_x) hierarchical nanostructures with different structural features are investigated.
We propose an approach that couples two aspects: the first one is employing an amorphous CuO_x material, which presents a higher availability of undercoordinated surface sites which, coupled with the high surface area, increases the activity. The second aspect is controlling the morphology of the CuO_x nanostructures down to the nanoscale by exploiting Pulsed Laser Deposition (PLD) as synthesis method: this technique, in fact, allows for a fine control over the deposit morphology by finely tuning the deposition parameters. In this way, the chemical composition and thickness of the hierarchical quasi-1D nanostructures are kept constant and it is possible to modify their aspect ratios, their nanostructuring and their micro- and nanopores diameters. With this approach, it is possible to evaluate the effect of the nanoscale morphological features on the overall catalytic activity – that is, on a smaller length scale than what is usually performed for Oxide-derived Copper catalysts.
Characterization of the PLD-synthesized nanostructures shows an amorphous matrix with embedded crystalline seeds of CuO and Cu₂O, differently from other reported oxide-derived Cu catalysts. CO₂ reduction performance of the oxide-derived Cu nanostructures is assessed through chronoamperometric measurements in aqueous electrolyte, showing good current densities up to -60 mA cm^-2 and a good stability for at least one hour of prolonged operation. Liquid products analysis, performed via ¹H NMR spectroscopy, shows a high productivity of ethanol at limited potentials applied (V=-0.4 V_RHE).
This contribution shows how coupling an amorphous, oxide-derived Cu catalyst with a fine control over morphology can be an efficient way to increase the activity of CO₂ reduction catalysts.

The development of highly selective and earth-abundant electrocatalysts for CO₂ reduction is becoming increasingly important for renewable energy applications. The challenge here is the strong competition from water reduction, as well as the low selectivity towards a desired product. In this talk I will introduce the rational tuning of electronic properties of catalytic materials for their significantly improved CO₂ reduction performance. By dispersing transition metals into isolated single atoms with electronic structures significantly different from their bulk counterparts, we can dramatically suppress the competing hydrogen evolution and deliver an ultra-high CO₂ reduction selectivity of more than 95% under ambient conditions in water.

The electrooxidation of urea continues to attract considerable interest as an alternative to the oxygen evolution reaction (OER) as the anodic reaction in the electrochemical generation of hydrogen due to the lower potential required to drive the reaction and the abundance of urea available in waste streams. Herein we investigate the effect of Sr substitution in a series of La_2-xSr_xNiO_4+δ Ruddlesden-Popper catalysts on the electrooxidations of urea, methanol, and ethanol and conclude that activities toward the urea oxidation reaction increase with increasing Ni oxidation state. The 75% Sr-substituted La_0.5Sr_1.5NiO_4+δ catalyst exhibits an exceptionally high mass activity of 588 mA and 7.85 A for the electrooxidation of urea in 1 M KOH containing 0.33 M urea, demonstrating the potential applications of Ni-based Ruddlesden-Popper materials for direct urea fuel cells and low-cost hydrogen production. Additionally, we find the same correlations between Ni oxidation state and activities for the electrooxidations of methanol and ethanol, as well as identify processes that result in catalyst deactivation for all three oxidations. Most importantly, this is the first demonstration of how systematically increasing Ni – O bond covalency by pushing the formal oxidation state of Ni above +3 serves to increase catalyst activity for these reactions and will act as a governing principle for the rational design of catalysts for the electrooxidation of urea and other small molecules going forward.

Strategies for closing the anthropogenic carbon cycle using the chemical conversion of CO₂ into useful molecules have become a scientific and technological priority. Of all chemical ways to convert CO₂ into CO, oxygenates or hydrocarbons, the direct electrochemical coupling of water electrolysis and CO₂ reduction appears attractive, as it is a one-pot process step, occurs at ambient temperature and pressures, involves solely water and CO₂ as reactants, and may involve renewable electricity from wind, solar or hydro power plants.

The CO₂ electroreduction is by and large carried out on the surface of metal catalysts. At industrially relevant current densities and electrode potentials, however, these catalysts typically suffer from low faradaic C₁ product selectivities due to the competing, very fast reduction of water, i.e. the hydrogen evolution reaction. New catalyst concepts with tunable selectivity for hydrogen evolution versus CO₂ reduction are needed.

In this contribution, we share recent comparative experimental and computational mechanistic insight in the electroreduction of CO₂ on metallic multi-site versus molecularly inspired metal ion-N modified high surface area carbon single-site catalysts. We provide new insight into CO selectivity trends and put particular emphasis on the competing reaction pathways to methane and methanol. We show that carbon-based single site catalysts are an intriguing cost-effective alternative to their metallic rivals at industrial current densities.

Supported single-atom catalysts provide an ideal system to study reaction mechanisms on a molecular level and to bridge heterogeneous and homogeneous catalysis. Single atoms have shown remarkably different activity and selectivity for several reactions, however for low temperature CO oxidation the reaction mechanism is not well understood and there is debate whether single atoms have higher intrinsic activity than nanoparticles. In this talk, the mechanism of CO oxidation on atomically dispersed Ir/MgAl₂O₄ catalysts will be presented and compared to that on nanoparticles. Using in-situ and in-operando DRIFTS and HERFD-XANES complemented by DFT calculations, we identified the active Ir single-atom complex, and resting state of the catalyst for low temperature CO oxidation. We provide evidence that strong CO adsorption on single atoms leads to an Eley-Rideal mechanism where Ir(CO) is the active complex (i.e. a CO ligand is present during the catalytic cycle). The results show that due to the ability of single atoms to coordinate with more than one ligand, strong adsorption by CO does not necessarily lead to catalyst poisoning. Additionally, we demonstrate that the unusual kinetics on single atoms can be used as a facile surface sensitive characterization tool for quantifying the number of single atom sites in catalysts containing a mixture of single atoms and nanoparticles. The results provide a new perspective for understanding reaction mechanisms and interpreting spectroscopic and catalytic activity of supported metal single atoms.

New semiconductor metal oxides capable of driving water-splitting reactions by solar irradiation alone are required for sustainable hydrogen production. Whereas most metal oxides only marginally deliver the photochemical energy to split water molecules, uranium oxides are efficient photoelectrocatalysts due to their absorption properties (E_g ~ 2.0 - 2.6 eV) and easy valence switching among uranium centers that additionally augment the photocatalytic efficiency. Although considered a scarce resource, the abundance of uranium compounds in the environment is manifested in the huge quantities of stored UF₆ gas, produced as waste streams in the nuclear fuel enrichment process. Here we demonstrate that thin films of depleted uranium oxide (U₃O₈) and their bilayers with hematite (a-Fe₂O₃) are high activity water oxidation catalysts due to electronically coupled phase boundaries. The electronic structure of uranium oxides showed an optimal band edge alignment in U₃O₈//Fe₂O₃ bilayers (DFT calculations) resulting in improved charge-transfer at the heterojunction as supported by TAS and XAS measurements. The enhanced photocurrent density of the heterostructures with respect to well-known hematite offers unexplored potential of uranium oxide in artificial photosynthesis.

The abundance of methane and its low cost makes it an optimal raw material for chemical precursors and energy-dense fuels. Conventional synthesis of methanol involve a multistep process, which requires high energy input and high cost [1]. Alternatively, electrochemical methane oxidation using a catalyst is a promising single-step approach to achieve a direct conversion of methane to methanol at lower temperatures and cost [2].

An efficient catalyst needs to be developed for this energetically challenging process. The catalyst should be active for C-H bond activation, thereby enabling oxidative hydroxylation of methane, and simultaneously inhibit further methanol oxidation. To date, a variety of the catalysts, including supported metals (Pd, Ru, Au, Ag) and metal oxides (V₂O₅, Fe₂O₃, CoO, Mn₂O₃, MoO₃, CrO), have been tested in electrochemical cells and shown promise for this approach [2]. However, further systematic studies are essential for understanding the mechanism for methane oxidation and thereby enabling the rational design of the catalysts. In this regard, porous metal oxides and supported single-atom catalysts (SAC) are particularly interesting [3]. These catalysts show potential for high methane conversion and are ideal model catalysts to identify active sites and facilitate understanding of the reaction process at a molecular level. In spite of great interest; however, it still remains challenging to achieve atomically dispersed metals in high loadings for efficient catalysis.

In this study, a design of first-row transition metal oxides and MOF-derived supported single-atom catalysts will be presented. The catalysts are synthesized with wet-chemistry, and subsequently, characterized for electrochemical methane-to-methanol oxidation. Structural and chemical analyses of catalysts using a combination of various microscopic and spectroscopy techniques are used to establish structure/property relationship. These insights provide valuable basis for a scientific-guided approach toward new metal oxide and supported single-atom catalysts for this challenging process.


1. Ravi, M., et al, Angew. Chem. Int. Ed., 2017, 56, 26464.
2. Tomita, A., et al, Angew. Chem. Int. Ed., 2008, 47, 1462.
3. Starokon, E. V., Phys Chem. C., 2011, 115, 2155.

Stabilizing transition metals (M) in MPt alloy under acidic conditions is challenging, yet crucial to advance Pt catalysis towards oxygen reduction reaction (ORR) in fuel cells. In this work, MPt nanoparticles (NPs) with intermetallic L1₀ (tetragonal) structured core and atomic Pt shell were found to be highly stable and active ORR catalysts in fuel cells. At 60°C in acid, the tetragonal L1₀-MPt better stabilizes M than its cubic A1-MPt counterpart. Monodisperse MPt NPs with ~9 nm in size were synthesized via organic solution phase method. The as-synthesized L1₀-MPt NPs were treated wth acid, followed by a repairing annealing treatment to obtain the core/shell L1₀-MPt/Pt NPs. The Pt shell is epitaxially linked to the L1₀-MPt core, leading to a farvored compressive strain in the Pt shell. L1₀-FePt/Pt showed superior activity to commercial Pt/C catalyst in the membrane electrode assembly (MEA) and almost no actvitiy drop after 30,000 voltage cycles at 80°C. In the L1₀-CoPt/Pt the Pt shell was more compressive than in the L1₀-FePt/Pt, resulting in extraordinary mass activities (MA) of 0.56 A/mg_Pt before durability test and 0.45 A/mg_Pt after 30,000 voltage cycles in the MEA at 80°C, exceeding the DOE 2020 targets on Pt activity and durability (0.44 A/mg_Pt in MA and <40% loss in MA after 30,000 cycling). Density functional theory study suggests that the ligand effect of Co and the more compressive strain brought by Co compared to Fe better tunes the binding of oxygenated species to the 2~3 Pt shell, thus leading to enhanced ORR performance in fuel cells.

Anion exchange membrane (AEM) electrolyzer is based on PEM electrolyzer combined with an anion exchange membrane (AEM) in the middle of cell. AEM electrolysis provides benefits of both PEM membrane (acidic) and traditional liquid alkaline (KOH) electrolysis with efficiency likely in between these two existing technologies. Although the most important benefit of AEM is possibility of adapting non-PGM catalysts, low activity of those catalysts remain still big challeges.
In this presentation, we prepared three-dimensional honeycomb-like Cu_0.81Co_2.19O₄ nanosheet arrays supported by Ni foam via electrochemical co-deposition of Co and Cu hydroxides on Ni foam followed by thermal oxidation. The co-deposition with Cu changes the morphology of the Co hydroxide deposit to form honeycomb-like nanostructures, significantly decreasing the onset potential for oxygen evolution. The Cu_0.81Co_2.19O₄ anode displays an exceptionally low overpotential of 290 mV at a current density of 10 mA cm⁻² in 1 M KOH, and an anion exchange membrane water electrolysis cell employing the above anode achieves a current density of 100 mA cm⁻² at 1.68 V in 0.1 M KOH.

Transition Metal Chalcogenides (TMD_s) recently gained attention as non-precious Hydrogen Evolution Reaction (HER) electrocatalysts that could substitute Platinum, the state-of-the-art HER catalyst, thanks to their abundance and outstanding catalytic activity. More importantly, when crystalline, their structure and morphology strongly influence the HER performance. This relationship has been exploited as design rule to obtain TMD_s catalysts with high efficiency.
When in amorphous form (a-TMD_s), these materials have even higher activity, thanks to their disordered structure with a high density of under-coordinated sites. However, the structure-performance relationship is unknown and traditional synthesis methods do not grant precise control over the deposited morphology, usually resulting in unsatisfactory HER performances for self-supported a-TMD_s.

In this contribution, we explore a HER performance/structure correlation for self-supported amorphous TMD_s catalysts, focusing on amorphous Molybdenum Sulphide (a-MoS_x), Molybdenum Selenide (a-MoSe_x) and Tungsten Selenide (a-WSe_x). We couple the intrinsic high activity of amorphous materials with a precise control over the morphology of the synthesized catalysts, obtained by utilizing Pulsed Laser Deposition (PLD) as synthesis method. In this way it is possible to control down to the nanoscale the structure and the morphological features of the materials, obtaining structures that range from compact films to hierarchical nanostructured morphologies.

We firstly analyse the shared electrochemical activation process, occurring at the beginning of catalysis, where the pristine a-TMD_s are transformed into the actual HER catalysts. Pre/post characterization is employed to evaluate how the activation modifies the composition and the catalytic performance. We find that the effect of the activation process on the electrochemical performance is related to the different transition metal (i.e. Mo or W) in the a-TMD.

By tuning the morphology of the a-TMD_s, we maximize the Tafel slope value and the -10 mA cm^-2 overpotential (η₁₀). The optimized morphologies reach HER performances among the state-of-art for TMD_s materials, regardless of their support or crystalline structure. For a-MoS_x, a Tafel slope of 35 mV dec^-1 – close to the Pt benchmark of 30 mV dec^-1 – and a -10 mA cm^-2 overpotential (η₁₀) of 126 mV are obtained, while for a-WSe_x a η₁₀ of 196 mV is registered.

In conclusion, by controlling during the synthesis process the a-TMD_s nanostructures’ morphology we show the correlation between structure and HER performance, that is reflected by the relation between morphological features and intrinsic HER catalytic parameters, such as charge transport efficiency and onset potential; moreover, we exploit this correlation on materials with intrinsic high activity, obtaining catalysts that with their performances provides a promising alternative to precious catalysts.

Water electrolysis has been emerging as a strategic approach to produce hydrogen, especially with the clean and renewable energy sources like solar and wind. To this end, the design, development, and fabrication of facile, active, stable earth abundant electrocatalysts for highly efficient hydrogen evolution reaction are essential. In this presentation, we will show a bunch of electrochemical catalysts we recently developed for the hydrogen generation. We experimentally demonstrate the facet-dependent HER activity and the interfacial dipole enhanced HER activity in chalcogenide based catalysts. With the assistance of synchrotron radiation characterization techniques, we were able to track and identity the changes in the local atomic environments, and reveal and establish structure/function relationship in a catalyst. This could be coordinated with the HER performance and provide fundamental understanding on the reaction mechanism.

Lack of a highly active, stable, earth-abundant electrocatalyst for carrying out hydrogen evolution and oxidation reactions (HER and HOR) in strongly acidic conditions is a major technical challenge for developing economical polymer electrolyte membrane (PEM)-based electrolyzers and fuel cells. We have found that processing Hf oxide with an atmospheric N₂ plasma forms an acid-insoluble hafnium oxynitride material. We propose that under electrochemical environments this material is transformed into an active oxynitride hydroxide that demonstrates unprecedented high catalytic activity and stability for both HER and HOR in strong acidic media for earth-abundant materials. The zero onset potentials and high current densities demonstrate that this material is a promising alternative to Pt. This material demonstrates excellent HER and HOR activities in acids using non-platinum group metal (PGM) catalysts, opening new opportunities to develop technologically and economically viable unitized regenerative fuel cell (URFC) systems and cost-effective PEM electrolyzers. Furthermore, these results have broad implications for using nitrogen incorporation (e.g. by N₂ plasma treatment) to activate other non-conductive compounds and films to form new active electrocatalysts.

Water electrolysis produces hydrogen gas with high purity and no greenhouse gas emission. Large-scale applications of water electrolyzers require low cost, efficient, and stable electrocatalysts. In this presentation, I will introduce our recent efforts on nonprecious electrocatalysts for both cathodic hydrogen evolution reaction (HER) and anodic oxygen evolution reaction (OER) of water electrolysis. Our best electrolyzer can reach current densities of 500 and 1000 mA cm−2 at 1.586 and 1.657 V, respectively. To understand and better design the electrocatalysts, we employed density functional theory calculations to reveal active sites and their roles in water activation and hydrogen adsorption.

The search for low-cost electrocatalysts to drive the oxygen reduction reaction (ORR) in polymer electrolyte membrane fuel cells has intensified with the recent deployment of commercial fuel cell electric vehicles. Platinum group metal-free (PGM-free) catalysts may offer a solution to using standard Pt-based catalysts if their activity and durability can be significantly improved. Breakthroughs in atomically-dispersed transition metal (Fe, Co, Mn) catalysts derived from metal organic frameworks (MOFs) have demonstrated improved performance and durability in fuel cell tests. It is expected that an increased understanding of the atomic-level structure–property relationships in PGM-free catalyst systems will enable further enhancement of these recent gains.

To this end, we have utilized in situ low-voltage, aberration-corrected scanning transmission electron microscopy (ac-STEM) coupled with quantitative analytical techniques such as electron energy loss spectroscopy (EELS) and energy dispersive X-ray spectroscopy (EDS) to study the structural evolution of MOF-derived catalysts at various stages of synthesis and fuel cell testing. The high spatial resolution (sub-Å) of ac-STEM allows for direction visualization of the catalytically active sites, which take the form of metal-nitrogen moieties within or at the edges of the meso-graphitic carbon support particles. We will present a quantitative analysis of the evolution in composition and support graphitization during high temperature pyrolysis (up to 1100^oC) for catalysts containing Fe, Co, and Mn. The effects of metal loading and the negative impact of metal clustering on performance will be demonstrated. The critical role of support graphitization in the formation of the highly activity metal-nitrogen sites and electrode durability will also be explored. High-resolution microscopy results will be correlated with other advanced characterization methods, including synchrotron-based scattering techniques, and discussed within the context of fuel cell performance and durability tests to provide a more complete view of the materials-based mechanisms that control the electrochemical performance of PGM-free catalysts.

This research is sponsored by the Fuel Cell Technologies Office, Office of Energy Efficiency and Renewable Energy, U.S. Department of Energy, as part of the ElectroCat Consortium, which is part of the Energy Materials Network. Microscopy performed as part of a user project at ORNL’s Center for Nanophase Materials Sciences (CNMS), which is a DOE Office of Science User Facility.

Recent advances in detector technology for transmission electron microscopy (TEM)¹ have made it possible to acquire high-quality images of nanoparticles with atomic scale spatial resolution, and a time resolution on the order of milliseconds. Here, we present a new method that we have developed to quantify atomic behaviour in TEM by tracking the position of atomic columns in a time-series of images using a two-dimensional elliptical gaussian fitting procedure implemented in Matlab. Our method, which we refer to as Time-Resolved Atomic Column Tracking (TRACT) allows us to determine the centre-position of cation columns to picoscale accuracy, and track the displacements of the columns between different frames. TRACT provides an experimental technique to probe the relationship between exchange and diffusion processes and dynamic changes in atomic structure at specific sites. This could have a significant impact on materials design for solid oxide fuel cells, photocatalysts, and a number of other technologies.
We have applied TRACT to analyze time-resolved atomic resolution image series of ceria nanoparticles, in which surface cerium cation columns are observed to undergo displacements between individual frames. Comparison of experimental images with simulations has indicated that these cationic displacements are caused by oxygen vacancy creation and annihilation events^2-3. We use TRACT to relate the frequency of cation displacements to activation energy for oxygen vacancy creation at specific atomic sites. Comparison of the data to multislice image simulations allows us to build a pseudo-three dimensional model of the particle surface, which aids interpretation.
Additionally, we have applied TRACT to compare images of an anatase nanoparticle in vacuum, and in a 0.01 Torr H₂O atmosphere. This allows us to quantify cationic displacements associated with H₂O intercalation into the anatase lattice.

References:
1. McMullan, G. Faruqi, A.R., Clare, D. & Henderson, R. Ultramicroscopy, 147, 156-163 (2014)
2. E.L. Lawrence, B.D.A Levin, T.M. Boland, S.L.Y. Chang, P.A. Crozier, Microsc. Microanal. 24(S1), 54-55 (2018).
3. T. Boland, E.L. Lawrence, B.D.A. Levin, P. Rez, P.A. Crozier, Microsc. Microanal. 24(S1), 144-145 (2018)
We gratefully acknowledge support from NSF DMR-1308085 and ASU’s John M. Cowley Center for High Resolution Electron Microscopy. We acknowledge Pierre Stadelmann of the École Polytechnique Fédérale de Lausanne for assistance with image simulation.

Titanium dioxide, or titania, is an important material being researched for energy and environmental applications, chiefly photocatalysis. Titania’s photocatalytic properties allow it to be used in water-splitting, detoxification, and photovoltaics_[1]. Titania exists in multiple polymorphic structures, of which the most common are rutile and anatase. We focused on anatase for the purposes of this research, due to its promising results for hydrolysis_[3].

Anatase is a non-stoichiometric oxide (TiO_2-x) which enables it to perform redox reactions by absorbing and releasing oxygen into its environment. When this happens, oxygen vacancies form within the anatase lattice. In situ electron microscopy can provide a window into these structural changes taking place in anatase during photocatalysis. However, to correctly interpret vacancy induced structures and evolution from TEM imaging, it is necessary to compare the experimental images with simulated structural models.

Here we model the oxygen vacancy structures in anatase, using a combination of molecular dynamics and density functional theory. Using a forcefield designed for the simulation of titania-water interactions_[2], different vacancy structures were relaxed and analyzed. To compare the theoretical models with experimental data, TEM image simulation was performed. We investigate a series of different vacancy configuration and generated TEM fingerprints for comparison with the experiment. This comparison yielded results which demonstrate a proof of concept for a technique suggesting that it should be possible to identify oxygen vacancy structures directly from the TEM image contrast.

The effect of surfaces on vacancy structure, and the interaction of water with the surface of anatase will also be examined. This research aims to satisfy an unfilled niche in our atomic-level understanding of oxides, by providing the methodology for analysis of vacancy formation from very subtle phenomena in TEM images.

References:
1) De Angelis, Filippo. “Theoretical Studies on Anatase and Less Common TiO2 Phases: Bulk, Surfaces, and Nanomaterials.” Chem. Rev., 2014, 114 (19), pp 9708–9753.
2) Kim, Sung-Yup, et. al. “Development of a ReaxFF Reactive Force Field for Titanium Dioxide/Water Systems.” Langmuir, 2013, 29 (25), pg 7838-7846.
3) Luttrell, Tim, et. al. “Why is anatase a better photocatalyst than rutile? - Model studies on epitaxial TiO2 films.” Scientific Reports, 2014, Volume 4 (4043). nature.com.

Continuum models of solvation have played a crucial role in quantum chemistry simulations and are now starting to be popular for the computational characterization of solvated, possibly electrified, interfaces. Some recent advances in the field have opened the possibility of modeling heterogeneous catalysis and electrochemistry in a first-principles-based framework, where the multiscale nature of the developed approaches provides a significant reduction of the computational burden while retaining a good accuracy. Nonetheless, extending continuum approaches to condensed-matter simulations present some non-trivial issues, related to the complexity of the electrostatic problem in charged two-dimensional interfaces and to the open structure of many crystalline substrates. Here we will present some of our recently proposed approaches to overcome these limitations, in particular focusing on a hierarchy of approaches and algorithms to describe the electrochemical diffuse layer. Moreover, handling environment (solvent and electrolyte) effects through the continuum embedding allows us to exploit a more rigorous grand canonical approach to study the thermodynamic properties of electrochemical interfaces, thus overcoming some possible limitations of the computational-hydrogen electrode (CHE) technique. Applications to noble metal (electro-)catalysis and beyond will be presented.

Despite significant promise as a clean and sustainable energy carrier solution, today’s low-cost hydrogen is conventionally produced from fossil fuels. A far more attractive alternative involves the use of abundant renewable solar and wind energy, which provide avenues for producing hydrogen from water, either indirectly through electricity generation and electrolysis, or else directly through solar photochemical and photoelectrochemical processes. However, these new avenues introduce significant materials challenges, prompting the use of computational methods for aiding discovery, understanding, and optimization. To this end, I will review some of our recent simulation activities within the U.S. Department of Energy HydroGEN Advanced Water Splitting Materials Consortium, which aims to accelerate the development of materials for renewable hydrogen production from water. Focusing on the use of first-principles computational methods, I will provide examples of how atomic-level understanding of chemical bonding in 2D materials has been applied towards the discovery of new hydrogen evolution electrocatalysts. I will also show how simulations have been combined with in situ experiments to understand atomistic mechanisms of water splitting at photoelectrochemical interfaces, leading to novel approaches for improving device lifetime and performance. These successes demonstrate the power of computational methodologies for translating atomic-level understanding into practical solutions for renewable hydrogen production.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.

Despite an extensive interest in developing advanced direct electrochemical reduction of CO₂ into value-added fuels and specialty chemicals using renewable energy as an input, its application is still constrained partly due to the fundamental challenges related to activity and selectivity toward the particular product of interest. Detailed first principles calculations are invaluable in complementing the experimental investigations to unravelling important aspects of the electrochemical reduction reaction mechanisms for various type of catalysts. Here, we present advances in the understanding of reactivity and selectivity in the electrocatalysis of converting CO₂ into methanol using density functional theory and thermodynamics. We will also discuss the role of protons and electrons transfer, and possible processes through different surface-bound reaction intermediates in order to steer catalyst selectivity among the vast number of possible carbon-based products.

MXene compounds are drawing attention due to their advantageous characteristics, including the large surface area of their two-dimensional structure, which may enable high energy density for electrochemical storage. These compounds have the generic formula M_{n + 1}X_nT_x (n = 1–3), where M, X, T_x represent the transition-metal cation, the carbon or nitrogen atom, and the surface termination of MXene layers, respectively. Since the discovery of MXene in the laboratories of Barsoum and Gogotsi, a number of studies have shown applications of their catalytic, electronic, mechanical, and optical properties [1,2]. In parallel, first-principles simulations have delivered quantitative information about their electrochemical response [3,4,5]; however, there is still debate over the charge-transfer mechanisms at the origins of electrochemical storage in MXene-based pseudocapacitors. In particular, it is still not fully understood how charge transfer is affected by transition-metal chemistry, stoichiometry, and surface termination. In the present study, we performed first-principles calculations on MXene compounds with 12 different transition metals (Sc–Cr, Y–Mo, La, and Hf–W), three different composition ratios (M₂XT_x, M₃X₂T_x, and M₄X₃T_x), and two different terminal groups (oxygen and fluorine). It was found that the electronic structures, formation energies and the amount of charge transferred from the cations to the MXene electrode is dependent on composition and that their charge is stored in the bond between the MXene and adsorbing cations. Importantly, the Bader atomic charge analysis revealed that the changes in valence number of the M-elements and X-elements are rather small, while the terminal groups play an important role in the pseudocapacitive charge storage of MXene compounds. Additionally, we carried out grand canonical simulations using a dataset of first-principles free energies, taking into account solvation effects using implicit solvent models based on the self-consistent continuum solvation (SCCS) formalism and including configurational entropy contributions via finite-temperature Monte Carlo sampling in an effort to clarify the charge storage mechanisms underlying the operation of the pseudocapacitive electrode [6]. We provide a detailed description of the sequence of desorption steps at the MXene surface as the applied voltage increases. Moreover, changes in the valence state and projected density of states were observed during the cation desorption reaction. These results provide design principles to improve the performance of MXene as an electrode material for electrochemical energy storage and to assess the electrochemical stability of MXene-based electrocatalyst under applied voltage and controlled pH.

(References)
[1] M. Naguib, J. Come, B. Dyatkin, V. Presser, P. Taberna, P. Simon, M. W. Barsoum, and Y. Gogotsi, Electrochemistry Communications 16, 61–64 (2012).
[2] B. Anasori, M. R. Lukatskaya, and Y. Gogotsi, Nature Reviews Materials 2, 16098 (2017).
[3] M. Naguib, V. N. Mochalin, M. W. Barsoum, and Y. Gogotsi, Advanced Materials 26, 992–1005 (2014).
[4] B. Anasori, Y. Xie, M. Beidaghi, J. Lu, B. C. Hosler, L. Hultman, P. R. C. Kent, Y. Gogotsi, and M.W. Barsoum, ACS Nano 9, 9507–9516 (2015).
[5] C. Zhan, M. Naguib, M. Lukatskaya, P. R. C. Kent, Y. Gogotsi, and D. Jiang, Journal of Physical Chemistry Letters 9, 1223−1228 (2018).
[6] N. Keilbart, Y. Okada, A. Feehan, S. Higai, and I. Dabo, Physical Review B 95, 115423 (2017).

We have demonstrated the use of metal phosphide nanoparticles (MPN) as catalysts for key processes (oxygen reduction, formic acid oxidation, methanol oxidation) in low temperature fuel cells and in the related applications of CO₂-reduction and hydrogen peroxide production via oxygen reduction. At a basic level, we show that MPNs analogs of well-known metallic fuel cell catalysts such as Pd show significant improvements in stability and activity per mg of catalyst. Further, the P atoms can act to tune the metal atom charge density as demonstrated for AgP₂ catalyzed CO₂ reduction, where experiment and density functional theory both show a large decrease in overpotential for CO production compared to Ag, again with improved stability.

However, the strongest improvements are observed for systems in which the dilution of the metal atom on the surface completely alters the coordination of the adsorbed species. This is demonstrated in our recent work developing Co₂P, CoP and CoP₂ as HER and ORR catalysts, where increased P content decreases the overpotential for HER to near the level of the Pt benchmark. DFT shows that sites with low absolute free energies of H adsorption are associated with single coordinating binding or highly strained bridge sites. Of interest is the extent to which MPNs can act as analogs to single atom catalysts, with initial results suggesting oxygen reduction by PtP₂ NPs generates hydrogen peroxide, similar to single atom Pt catalysts and unlike traditional Pt NPs.

Development of materials that can feasibly solve the growing dichotomy of simultaneously increasing energy demands and carbon emissions is an imperative that has become uniquely important for current research efforts. An emerging avenue in this regard is the conversion of vastly abundant renewable energy sources that can be harnessed and directed towards synthesis of traditionally fossil fuel-based products from atmospheric feedstocks like CO₂. In order to translate such an energy conversion process into a viable technology, it is critical that design principles of selective electrocatalyst materials be elucidated in order to mitigate operating costs and improve overall process efficiency.

Results that will be presented herein, focuses on structure—function correlations across three versatile classes of transition metal chalcogenide materials, including metal dichalcogenides (MX₂s; M = Mo, W; X = S, Se) Chevrel-phase sulfides (CPs) (M_xMo₆S₈; M = transition or alkali metal; x = 0-4), and thiospinels (AB₂X₄; A = Cu, Mg; B = Ti, V, Cr; X = S, Se). These chalcogenides were found in our preliminary work to display exceptional promise as CO₂R catalysts. As such, the compositional, structural, and dimensional tunability of these classes of compounds will allow for development of structure—function correlations and design rules. Furthermore, we have identified CPs as selective towards the electrochemical reduction of CO2 and CO to methanol (major liquid product) under an applied potential of -0.4 V vs RHE. Moreover, CPs such as Cu2Mo6S8 exhibit an interesting electronic structure-function correlation founded on the basis of X-ray absorption spectroscopy. We correlate the selective electrochemical reduction of CO2 and CO to methanol due to the potentially increased reactivity of the chalcogen species to the filling of 3p orbitals by metal promoters such as Cu, Ni and Cr. It is suspected that CO hydrogenation will be the rate-limiting step in the CO₂R to liquid fuels, hence being able to surmount the energy barrier associated with this step will rely strongly on the ability of the catalyst to facilitate hydrogenation. These reactive chalcogens neighbor suspected active sites (Mo) for CO binding, and so the overall reduction of CO₂ and CO to methanol through a CO hydrogenation mechanism is likely enhanced by the presence of an electron-donating species like a transition or alkali metal.

The electrochemical carbon dioxide reduction (CO₂R) has been received large attention in converting CO₂ into valuable chemical species. For practical application of CO₂R, a catalyst with low overpotential and selectivity for the desired product is required. In general, C₁ species such as CO and HCOOH are mainly produced on the catalyst at the cathode. Production of multi-carbon (C_N) species such as ethanol and n-propanol can achieve higher economic value when produced with optimized current density, Faradaic efficiency (FE), and overpotential. Recently, Francis et al. demonstrated that MoS₂ converts CO₂ into 1-propanol as a major product for CO₂R at onset potential of −0.5 V vs. RHE.[1] It is also reported that the current density decreases when increasing the portion of edge sites, which implicates that the active site places on the basal plane rather than the edges. Despite the fact that MoS₂ produces 1-propanol, which is a highly valuable product, practical utilization of MoS₂ is limited because FE of CO₂R is only about 5% because of the poor selectivity compared to hydrogen evolution reaction (HER).
In this work, we first study the mechanism of CO₂R at sulfur vacancy (V_S) of MoS₂ by calculating free energies through the density functional theory calculations and computational hydrogen electrode model. We find several pathways leading to C₁ products such as HCOOH, HCHO, CH₃OH and CH₄. C₂ and C₃ pathways are discovered considering C-C coupling among C₁ intermediates. We want to note that HCHO is found to be a key intermediate for C₂ and C₃ pathways by coupling of sp² carbons. The computationally predicted reaction products and corresponding onset potentials are in good agreement with the experiment, which leads to the conclusion that V_S is the main active site of CO₂R on MoS₂. To discover a material producing C_N products with better selectivity than MoS₂, computational screening of the catalytic performances of CO₂R is performed for anion vacancies of various transition-metal dichalcogenides (TMDs). As a result, TaSe₂ and SnSe₂ are suggested as good candidates that can convert CO₂ into CO and HCHO, respectively, with much higher selectivity compared to MoS₂. Especially, we expect that production of HCHO at Se-vacancy of SnSe₂ possibly leads to the production of C_N products.
[1] Francis, S. A. et al. Reduction of aqueous CO₂ to 1-propanol at MoS₂ electrodes. Chem. Mater., 30, 4092-4908 (2018)