Modern Applications of High Throughput R&D in Heterogeneous Catalysis

High Throughput Experimentation (HTE) has over the past 15 years become a mature technology platform that has found entry in almost all disciplines in catalyst and process research and development. The drivers that have led to this massive aspiration within academia and industry are clearly associated to HTE´s largest promise: a dramatic increase in efficiency. This increase in efficiency, if exploited appropriately can lead to two effects that explain the appreciation from the academic and industrial point of view. Firstly with increased efficiency HTE offers the perspective to take a look at catalyst or process options outside of “conventional” corridors. Secondly HTE holds promise to pave the path for shortened experimental periods to gain systematic insights, offering the perspective of faster time-to-market. From the point of view of the technology platforms which are available today it can clearly be seen that market demands have pushed the technology providers from initial smart and unconventional designs for accelerated screening to elaborated platforms that can map catalyst behavior and display response patterns to process conditions within technically relevant reaction corridors with unexpected precision. This clearly uprates HTE utilized in catalyst and process development over its roots in the pharmaceutical industry where it is still mainly used for screening purposes. Our intention is to give the reader an entry and an integrative overview into the mindset, solution approaches and challenges of HTE and enable her/him to critically formulate the right questions to ask and how to judge aspects of feasibility. The core aspect in this chapter will be centered on workflow development for certain applications narrated as case studies that are considered prototypic for a whole technology segment. All the aspects in the “Design- Make-Test-Model”-cycle will be discussed in these case studies. In our conclusions we will give an outlook and try to get a glimpse of the future.

The recent development of solid oxide fuel cells (SOFC) capable of using liquefied petroleum gas (LPG) as a feedstock has created a viable high energy density technology for replacing the diesel generators and batteries used to create power in remote or mobile applications. Due to the unsurpassed volumetric energy density of liquid hydrocarbon fuels and the existing infrastructure for their distribution, it is preferable to transport and store diesel and jet fuel and catalytically crack it on-site to LPG. The ideal cracking catalyst must be capable of operating without excessively coking or being poisoned by sulfur present in the feedstock, and must be capable of being regenerated at temperatures similar to the reaction temperature, preferably using only air. These stringent requirements required rapidly identifying a series of base catalysts with high activity towards cracking, an additional series of potential additives to mitigate coking and promote regeneration, and the final verification of their combined properties to ensure optimal performance. Here we demonstrate a comprehensive and iterative high-throughput (HT) methodology for identifying such novel catalyst formulations with high activity towards cracking to LPG, which are capable of being operated for extended periods of time and regenerated in air multiple times without degradation in their activity. The approach combines a rapid, qualitative primary optical screen via thin-film techniques, a series of quantitative screens using mg powder quantities, and a final verification of the best samples using a single sample reactor. This versatile approach permitted the systematic study of a large phase-space of potential catalysts and additives combinations, ranging from noble metal promoted simple oxide catalysts to zeolite based catalysts. Initial studies focused on the identification of a catalyst with suitable activity, in excess of 20% conversion, using a 16-well reactor. This portion of the work focused on providing a comprehensive quantification of the product distribution of the different catalysts using industrially available catalyst formulations. Subsequently, thin-film samples with a spread of compositions deposited as nanoparticles on the surface of the most active catalysts were employed to screen bi-metallic additives for their relative coking and regeneration rates. Several additives with the highest figure of merit were recommended for a second round of screening of bi-metallic promoted zeolites in the 16-well reactor, where the emphasis shifted to evaluating the effect of both the additives and multiple regenerations on the product distribution. Final verification of the best additivemodified catalyst was obtained in a single reactor in which the catalyst was run under a highly concentrated feed for multiple days and demonstrated negligible degradation.

In this contribution we give an overview of the developments published during the last decade on truly parallel or fast sequential fixed bed microreactors for primary or secondary high-throughput screening of heterogeneous catalysts. After a review part we focus on our own parallel reactor setups developed for the discovery and optimization of new Deacon catalysts. In hydrogen chloride recycling this oxidation reaction is energetically highly attractive, but due to extreme corrosivity of reactant / product mixtures it poses a formidable challenge on the materials of the screening reactor. The problem was solved by constructing a new parallel microreactor setup built from corrosion resistant Nibased alloys together with gas flow restriction by adjustable capillaries for a uniform gas distribution, and back pressure valves for elimination of cross contamination between the parallel reactors. Apart from catalyst screening, the setups have been used to perform ageing studies of the catalysts identified as active for the Deacon reaction. In chlorine production either by electrolysis or by the Deacon reaction, both catalyst systems contain RuO2 as essential component. RuO2 is a metallic conductor, chemically very stable in corrosive environments, but due to its low abundance its market value is subject to pronounced speculative variations. Thus, one objective of our investigations has been the reduction or even complete replacement of expensive Ru ideally accompanied by a simultaneous activity increase. To discover new Ru-free catalysts for the Deacon reaction a basic search in a wide parameter space was launched. Ru reduction was achieved by doping RuO2 with various elements. The hit and lead compositions identified were characterized by physisorption measurements as well as powder X-ray diffraction together with Rietveld refinements of the diffraction data.

The development of a high throughput research (HTR) infrastructure and experimentation capability has been a major focus for The Dow Chemical Company. This R&D effort was built on the vision that high throughput experimentation can broadly impact the effectiveness of our businesses. Over the past ten years this has resulted in the construction of HTR infrastructures covering a wide diversity of technologies, such as liquid formulations, coatings and adhesives, thermoplastics and thermosets, organic synthesis, and both homogeneous and heterogeneous catalyses. These workflows have greatly improved Dow’s R&D effectiveness and enhanced the rate of development of commercially applicable technology. To illustrate the HTR methodology and workflow in place, a case study on propane oxidative dehydrogenation (ODH) is discussed in this chapter. It is shown that an effective workflow consists of a series of interlocking elements that include experimental design, catalyst synthesis, materials characterization, and catalytic reactivity screening. These elements are connected together through data and materials handling.

High-throughput experimentation for catalysis testing has become an established practice in catalytic research, yet obtaining accurate and relevant results is still considered an art. Here, the engineering concepts of the Flowrence parallel smallscale reactor systems are discussed. The influence of catalyst particle size, flow patterns, pressure drop and temperature profiles on the quality of catalytic results is exemplified by several case studies on Fischer-Tropsch, Oxidative Coupling of Methane and Hydrotreating to obtain Ultra Low Sulfur Diesel.

The Interrogative Kinetics (IK) methodology is presented as a new direction in catalyst development that combines elements of high-throughput methodology with detailed, intrinsic kinetic characterization comparable to molecular beam surface science experiments. Similar to high-throughput techniques, the IK approach uses real materials (as opposed to model systems) and studies the kinetic function of the material as its chemical composition is incrementally changed. In the case of metal oxides, both oxygen and metals can be varied. Using the TAP (Temporal Analysis of Products) pulse response technique the surface composition can be manipulated with respect to oxygen while capturing detailed kinetic information [1-3].

To extend the IK methodology to metals, new instrumentation is described here that will incorporate with TAP kinetic characterization the incremental adjustment of surface metals composition in submonolayer amounts using atomic beam deposition. Recently developed instrumentation to transport catalyst particles entrained in an inert gas flow and microreactor carousels are presented that streamline the comparison of multiple samples and accelerate the development process. In contrast to traditional kinetic methods that attempt to obtain performance parameters from a well-defined ‘steady-state’ of a catalyst surface, Interrogative Kinetics attempts to systematically probe the catalyst as the surface composition changes, providing fundamental information describing the composition-activity relationship of technical (non-ideal) materials. This incremental titration of the surface can be thought of as a ‘chemical calculus’. Using technical materials in this manner one can study complexity and better understand how emergent properties of the system arise.

Catalyst characterization plays an important role in developing catalytic materials because it helps to identify active catalytic phases or centers and to elucidate reaction mechanisms. Several characterization methods have been parallelized and successfully applied in the high-throughput workflow of catalyst development. We present characterization equipment developed at the Leibniz Institute for Catalysis and show applications in three research areas: (i) in catalyst synthesis for identifying suitable preparation conditions, (ii) in understanding factors governing catalytic performance for identifying structure-reactivity relationships, and (iii) in understanding catalyst deactivation for elucidating the deactivation mechanisms. 6-fold and 8-fold reactors are used for temperature programmed reduction (TPR), oxidation (TPO), desorption (TPD) and chemical titration experiments. TPR has been successfully applied for studying the reducibility of supported Ni-containing catalysts for CO2 reforming of CH4 and for investigating the interaction between metal oxide species and support materials. Chemisorption of H2 and CO2 has provided information about Ni dispersion and basic properties of said catalysts, and with TPO of spent catalysts coke formation was monitored. To prepare catalysts for visible-light induced water splitting, we studied the formation of nitrides and oxynitrides from oxides upon their temperature programmed treatment under a flow of ammonia. We also used 6-fold and 36-fold parallel reactors utilizing UV/vis catalyst analysis in reflectance mode to study changes in valence state and coordination of supported VOx during oxidative dehydrogenation of propane, and to follow coke formation during catalyst deactivation in non-oxidative dehydrogenation of propane.

Rennovia is employing cutting-edge technology to adapt chemical catalytic processes, already proven to be scalable and efficient in the refining and chemical industries, for the conversion of biorenewable feedstocks to existing major-market chemicals. This chapter outlines Rennovia’s general approach to the cost-advantaged production of renewable chemicals, and describes the scale-up of Rennovia’s bio-based adipic acid process and how it offers the potential for significant commercial and environmental advantages over the current petrochemical process.

Global challenges in chemicals and energies comprise:

• Increasing constraints in supply from fossils and recently nuclear power.

• Increasing constraints by regulations (CO2 emission, global warming, …).

• Increasing demand e.g. spurred by the dynamic growths in the BRICS countries.

• By 2020 China is projected to consume more than 50% of the global energy.

Corresponding R+D organizations need to cope with the challenges typically with the same amount of resources. The only way out of this “catch 22” is to standardize and accelerate materials R+D via enabling, automated solutions for e.g.:

• Catalyst preparation by incipient wetness impregnation, excessive liquid impregnation, (co-)precipitation (mixed oxides), hydrothermal treatment (zeolites) and screening in e.g. hydrogenation, alkylation, isomerization, oxidation, cracking.

• Ligand and organometallic catalyst synthesis and screening for e.g. polyolefins, rubbers, fine and specialty chemicals.

• Catalyst testing / optimization in batch, semi-continuous, continuous mode.

• Alternative energy solutions (e.g. solar cell, battery, storage, fuel cell materials).

• Renewable chemicals (e.g. from biomass, CO2).

Case studies and platform approaches from different research groups (section 9.1 to 9.5) on fuel cell, biorefinery, fine and specialty chemicals, basic chemicals, exhaust gas and syn gas catalysis provide examples about benefits, implementation scenarios, and stateof- the-art research platforms / approaches in catalysis R+D. The contributions of the different research groups are presented in correspondingly autonomous sections.

With the need to meet the ever increasing energy demand from non-fossil fuel sources, studies into PEM (proton exchange membrane) Fuel Cells become an important area of research today. ACAL Energy’s patented FlowCath technology provides a low-cost and radical alternative to current technology using a liquid catalyst (catholyte) to replace up to 80% of the solid Pt found in standard PEM Fuel Cells. In this work, we incorporate high-throughput methods to improve the productivity and quality of research-enabling the development of new liquid catholytes. The utilization of high-throughput method produced in excess of two hundred formulations which were tested against the current catholyte system. Here, we show a selection of post-synthetic formulations which resulted in improved performance compared to the standard catholyte system. Two new formulations were developed with improved anodic current as well as no decrease in the standard rate constant for electron transfer (k0) while an increase in k0 and an increase in cathodic current were also evident, respectively. The development of these high-throughput methods will allow for rapid development of the FlowCath technology and aid in the synthesis of the next generation of catholyte.

REALCAT, ‘Advanced High-Throughput Technologies Platform for Biorefineries Catalysts Design’, is a novel high-throughput platform primarily dedicated to the development of all the forms of catalytic reactions (chemical catalysis, namely heterogeneous and homogeneous, as well as biotechnologies) and their novel innovative combinations (hybrid catalysis). It includes the preparation, the characterization of (i) compounds with biological activity and (ii) solids & molecular compounds, as well as subsequent advanced catalytic performance tests. It thus includes top level highthroughput screening equipment, composed of tools for rapid characterization, a series of parallel reactors, and fast/ultra-fast analytical devices. This project received € 8.7 million funding from the French Government, managed by the National Research Agency (ANR) within the Future Investments program (PIA) under the contract ANR-11- EQPX-0037 attributed at the end of 2011. This innovative platform will be fully operational in early 2014. The platform is highly versatile, and, in addition to catalysts for biorefineries, it can handle projects for many different kinds of homogeneous, heterogeneous and biotechnological catalytic applications in gas, liquid or gas-liquid phases, involving a large variety of reactions (synthesis, functionalization, polymerization, etc…) for a wide spectrum of applications (solvents, monomers, polymers, fine chemicals for industrial, food, feed, pharmaceutical, cosmetics, etc.). It can even run projects dedicated to the synthesis of materials, namely organic, inorganic or hybrid solids (e.g., metal organic frameworks, MOFs) for non-catalytic applications.

Parallel experimentation is a key tool in the fine chemicals industry to develop cost-efficient processes in a short time. This is especially true for hydrogenation reactions where a vast number of commercial catalysts are available with significantly different properties and reactivities. The ability to quickly cover a good part of experimental space and identify the optimal catalyst and reaction conditions would not be possible without parallel experimentation. This article gives an overview of how such experiments are performed at DSM and selected examples of successful projects.

Chemspeed’s AUTOPLANT ZEO Robotic Platform together with the Workflow Management Software provides faster and better zeolite synthesis. Reproducibilities over 20 fully automated 100mL reactors per run are within ±5%. The physico-chemical properties of zeolites produced proved to be scalable to 2-Liter stainless-steel autoclave. Varying a suite of raw materials with different physicochemical properties, the ratios of n(SDA)/n(SiO2) and the ratios of n(H2O)/n(SiO2), the productivity could be increased by a factor of 10. Zeolites were reproducibly synthesized in high solid-content (viscous) regimes.

Both in exhaust gas after-treatment and energy-related catalysis, such as the conversion of synthesis gas to chemicals and fuels, more efficient and new catalysts need to be designed. Usually, rather established catalyst preparation methods like impregnation, (co)-precipitation, hydrothermal synthesis of zeolites and ion exchange are used. For this purpose, a high output platform for catalyst synthesis (HOCATS) has recently been established at KIT. With two case studies in bulk and environmental chemistry we discuss that robot-controlled catalyst synthesis at medium throughput is an attractive and promising approach if it is combined with efficient testing and characterization. For the latter preferentially also parallel or automated tools are required that still need to be further developed. In addition, deeper characterization of the catalysts by advanced spectroscopic methods is required since heterogeneous catalysts are typically very dynamic under reaction conditions. Finally, we emphasize that an alternative approach to high throughput screening is the use of computational screening using scaling relations and Brønsted-Evans-Polanyi relations.