Ruthenium and Cobalt Catalysts for Hydrogen Production from Ammonia

In the global energy transition, green ammonia can serve as a carbon-free energy vector and is suited to store renewable energy long-term in its chemical bonds. Ammonia can then be partially cracked for direct use as a fuel or fully cracked for use in a hydrogen fuel cell. In both cases, catalytic ammonia cracking is paramount to release hydrogen on demand and avoid NOx formation. Currently, the benchmark catalyst for ammonia cracking is ruthenium supported on carbon nanotubes (CNT). Future catalyst development requires a departure from trial-and-error catalyst discovery. This thesis presents a new method for the synthesis of catalysts with precise nanoparticle sizes to unlock fundamental reaction knowledge, redefining the way we design and optimise heterogeneous catalysts. Ruthenium nanoparticles (Ru NPs) are synthesised in the absence of capping ligands in a bespoke continuous microreactor, enabling the control of average nanoparticle size with narrow size distributions. A method is developed to immobilise the Ru NPs onto various supports while maintaining size via electrostatic stabilisation, bridging the gap between colloidal science and heterogeneous catalysis. This reveals for the first time unique size- and support-activity relationships. For a Ru NP size of ~2.5-3 nm, CeO2 and ZrO2 are identified as superior supports compared to the benchmark CNT, attributed to the electron-donating properties of CeO2 and the ability of CeO2 and ZrO2 to spillover hydrogen into their bulk lattices. The morphology of CeO2 is seen to affect catalytic activity, with nanoparticles and nanocubes outperforming nanorods. The correlation between Ru NP size and activity is demonstrated experimentally, in agreement with theoretical studies that predict a maximum density of ‘B5’ active sites for ~2-3 nm hemispherical Ru. Cobalt (Co) catalysts are also investigated to replace Ru with a more abundant metal for the large-scale implementation of ammonia as an energy carrier. Catalytic activity improves by a factor of 2.8 via the continuous synthesis and immobilisation of colloidal 11 nm Co NPs on alumina, compared to ~60 nm Co from traditional impregnation. Strategies from Fischer-Tropsch synthesis on the promotion and reduction of cobalt catalysts are applied for the first time to ammonia cracking, demonstrating the transfer of knowledge between these catalysis fields. This reveals that a balance is required to achieve complete cobalt oxide reduction while maintaining the more active hexagonal close-packed phase of Co at elevated ammonia cracking temperatures. This shows the importance of considering both NP size and crystal structure during catalyst development and reaction conditions. Overall, this thesis demonstrates that the development of catalysts using the continuous synthesis of colloidal active sites provides an accelerated way of decoupling size, promotion, and support effects, contributing guidelines to the future design of catalysts for structurally sensitive reactions.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2023

Author dc:creator

• El-Kadi, Joseph

Advisor dc:contributor.advisor

• Torrente Murciano, Laura

• Ammonia
• Catalysis
• Flow Synthesis
• Microfluidics
• Nanoparticles
• Sol-immobilisation

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/674a4379-30cc-437b-91be-b3bfec9d2579/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.101650

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/357422