From the chemical bond to the chemical reactor: Computational and Materials challenges in gas conversion technologies
Location: Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore, India
The development of new catalytic materials will be crucial to tackle the impending energy and environmental grand challenges. In the short term, catalytic processes will help us to make the most of existing resources, sustaining the demands of energy and chemical products of a growing world population, providing some useful time to develop new technologies for power generation and energy storage (e.g. solar cells, biofuels of second generation). For example, great technological and financial opportunities are offered by the catalytic conversion of natural gas into chemical feedstock and liquid fuels for transportation. Natural gas is in fact abundant and cheaper than conventional oil, the gas-oil differential price being expected to broad in the future. Catalysts are also at the heart of many technologies employed to control pollutant levels in air and water (sulfur, CO2, nitrates), in both the power generation and chemical manufacturing sectors.
Understanding conversion processes at the atomistic level is the first important step to design new catalysts with better performance. The synergy of in situ experimental techniques and atomistic simulations nowadays allows us to study catalytic reactions with accuracy unthinkable just a decade ago. At the academic level, over the years, experimental characterization techniques for catalysts and catalytic reactions have improved substantially, proving that a nanoscale approach to catalysis is indeed possible and instrumental to catalyst engineering. In parallel, also computational modeling has experienced important progresses; however the field is limited by important problems. Just to name a few: The simulation of rare events at the surface, the lack of accuracy of current methods for describing transition metals and oxide interfaces, the inclusion of finite temperature and pressure effects, difficulties of transferring reliably the calculated properties across time and length scales.
Deploying this fundamental knowledge to a large industrial scale is the other step for a successful catalyst development, an area where several industries proved to be successful, often with limited contacts with the academic community. In many cases, the synthesis of a catalyst in large quantities (e.g., tons) and its use in an industrial process may be not possible, even if the material has offered good performance in the discovery phase. Now more than ever, a collaborative effort between industrial R&D and academic institutions is therefore crucial for developing and deploying new successful conversion processes.
These are precisely the objectives of the present workshop, namely i) to bring together computational modeling experts that focus on different time/length scales, and ii) to confront them with the best available experimental characterization techniques and, most notably, with the industrial R&D needs. We want to bring together scientists from both academic and industrial R&D laboratories, to help identify challenges in materials and techniques/approaches. The main field of application will be, but will not be limited to, new gas conversion processes. The goal is not a miniconference, but a workshop capable to deliver a set of priorities that need to be solved in the near future and, possibly, to agree on how to solve them. The workshop is stimulated and partially supported by private companies, which will help to reach and attract the industrial R&D community working in the field. This community is often not fully connected with the scientists usually attending CECAM workshops. Strategically, we propose to hold the conference in India (Bangalore). In the last two decades Bangalore has became an important technological hub. Several international companies have set their R&D centers, leveraging on a local academic community with a strong background and tradition.
India and South East Asian countries will play a key role in developing technologies for energy conversion and production. New and more efficient gas conversion technologies will be crucial for their fast developing economies and growing populations. We believe this will be an opportunity for EU researchers and CECAM in particular.
The catalytic processes that underpin the mass production of chemicals via gas-to-liquid transformations, present materials challenges spanning across length and time scales, literally ranging from the chemical bond to the industrial reactor. The structure and hence the functionality of a catalyst strongly depend on the conditions (pressure, temperature, time on stream, etc.) in which a catalyst operates. The field is realizing that an ‘active phase’ may only be present under reaction conditions while the catalyst functions. The characterization of such active phases in real conditions, and of the related chemical reactions that are promoted, is of paramount importance for technological advances. This is the central theme of the workshop and is at the forefront of research for academic and industrial research. The main limitations common to experiment and simulation are the difficulties in studying the material in operando. From the experimental side, this requires in situ experimental techniques, enabling the simultaneous access to spectroscopic and kinetic measurements while the catalyst is at work. From the computational modeling side this requires a truly multiscale approach capable of combining the quantum mechanical description at the molecular level with statistical approaches, coarse graining methods and finite-element engineering techniques.
Computational modeling based on quantum mechanical approaches has proved very successful in characterizing fundamental materials aspects and chemical processes of catalysts at the atomic/molecular level. Some examples relevant in the context of the workshop include methanol synthesis [Grabrow2011], hydrogenation reactions [Armbrüster2012] and ethylene epoxydation [Piccinin2010]. Although bringing highly valuable information useful for the characterization and design of industrial catalysts, academic studies often address model systems (flat and crystalline surfaces, UHV conditions, ultra small nanoparticles, …) that are quite far from the real conditions of industrial reactors (high temperature and pressure, complex dynamical surface nanostructures, …). Traditionally, there have been two principal approaches to fill this materials gap: i) ab-initio thermodynamics to predict the active phases forming on catalyst surfaces as a function of T and p [Reuter2001]; ii) The identification and use of reaction descriptors which allows one to connect the microscopic properties of the catalyst to the reaction kinetics [Rassmussen2012]. The latter approach is used for identifying trends in reactivity and is currently at the basis of catalyst genome and materials informatics projects aimed at the computational assisted discovery of better catalysts [Nørskov2013]. The reliable application of Density Functional Theoretical approaches to reaction kinetics strongly relies on the most recent developments for calculating reaction and activation free energies that go beyond the Nudged Elastic Band approach, such as Stochastic Methods based on surface walkers [Shang2013]. Another application concerns the use of DFT energetics to parameterize microkinetic models whose predictions can then be directly compared to experimental reactivity measurements [Sutton2013].
One of the current challenges is identifying innovative ways of bringing information from these well established approaches, which are still based on an atomistic description of the catalyst, to the meso and micro scale. As an example, this requires the development of kinetic Monte Carlo (kMC) methods capable of addressing the complexity of competing surface reactions as well as the catalyst composition, surface structure, lateral interactions, and operating conditions. These factors can then be correlated to turnover frequency. It would be desirable to extend the existing kMC approaches to study large chemistries on complex catalytic structures [Stamatakis2012].
Modeling catalysis at the scale of a chemical reactor further requires accounting for mass and heat transport. This is typically done using a continuum approach where Navier-Stokes equations are solved for a given reactor geometry, using as boundary conditions the information extracted from microkinetic models. Often, in chemical engineering approaches, these microkinetic models are based on phenomenological mean-ﬁeld equations. Only recently research moved towards an integration of first-principles based kMC with the continuum description of computational fluid dynamics [Matera2010, Maestri2013].
On the experimental side, early UHV-based surface science approaches to heterogeneous catalysis have been quite successful in unraveling the elementary steps of surface reactions, but the experimental conditions have typically been very different from those in technological processes. Therefore, substantial efforts have recently been devoted to developing a number of surface-sensitive methods that allow us to examine the evolution and transformation of active model catalyst phases under (near) ambient pressure.
Photon-in electron-out techniques like X-ray photoelectron spectroscopy (XPS), are intrinsically surface sensitive due to the strong interaction of (low energy) electrons with matter. Pressures in the range of mbar can nowadays be reached in so-called near ambient pressure XPS measurements [Bluhm2007]. The tunable photon energy and the high energy resolution of synchrotron sources, in particular, enabled this technique to yield detailed element-specific information of the chemical environment at the surface of catalysts. This approach was recently used, for example, to monitor the dynamics of various oxygen species on silver during the epoxidation of ethylene [Rocha2012].
Recent progress in the application of surface vibrational spectroscopy at near ambient pressure (in particular polarization modulation infrared reﬂection absorption spectroscopy (PM-IRAS) and sum frequency generation (SFG) spectroscopy [Rupprechter2008]) allows researchers to monitor heterogeneous catalytic reactions under conditions approaching those of technological catalysis. Using this technique it was possible to follow the evolution of the structure of supported catalyst particles for oxidation reactions and its impact on the catalyst performance [Rupprechter2008].
Several industries have already deployed gas to liquid technologies on large scale plants. For instance, Sasol’s slurry phase distillate process (Oryx plant in Qatar) and Shell’s Heavy Paraffins Synthesis (Pearl plant, Qatar) are current examples of the result of decades of research covering both fundamental and technical aspects of the catalytic process. On the other hand, the sharp drop of the price of the natural gas in North America has created new opportunities for new players in the energy and chemicals industries, trying to compete with innovative conversion process and small scale plants.
Shobhana Narasimhan (JNCASR, Bangalore, India) - Organiser
Leonardo Spanu (Shell Technology Center, Bangalore) - Organiser
Stefano Fabris (CNR-IOM DEMOCRITOS Simulation Center and SISSA, Trieste) - Organiser
Simone Piccinin (CNR-IOM (Trieste)) - Organiser