> ## Content Index > Fetch the complete content index at: https://research-pop.com/llms.txt > Use this file to discover other available public pages before exploring further. # Shandong University, UCL, Hunan University and Nankai University teams stabilize ammonia-decomposition catalysts with amorphous BaO–Al₂O₃ - URL: https://research-pop.com/shandong-university-ucl-hunan-university-and-nankai-university-teams-stabilize-ammonia-decomposition-catalysts-with-amorphous-bao-al2o3/ - Published: 2026-10-01T11:57:53.000Z - Updated: 2026-10-01T12:01:15.000Z - Author: ResearchPOP ![](https://storage.ghost.io/c/7c/5e/7c5e0911-739b-4a35-94eb-3e810bf78ba7/content/images/2026/10/Screenshot-2026-10-01-at-12.37.14.png) Source: [https://doi.org/10.1038/s41467-026-78117-4](https://doi.org/10.1038/s41467-026-78117-4?ref=research-pop.com) ## At a glance A team from Shandong University, University College London, Hunan University and Nankai University has developed a template-free route to confine cobalt or nickel nanoparticles within an amorphous BaO–Al₂O₃ environment. Instead of allowing the solid-state reaction to proceed fully to crystalline BaAl₂O₄, the researchers arrest a disordered intermediate under ammonia-decomposition conditions. This amorphous environment performs several connected functions. It limits the movement and growth of the metal nanoparticles, provides sites for NH₃ adsorption, modifies the electronic environment around the metal and leaves pathways through which reactants and products can move. The optimized 25Co-AB-2R-600 catalyst produced 1,358 mmol H₂ gmetal⁻¹ min⁻¹ at 550 °C and a gas hourly space velocity (GHSV) of 600,000 mL g⁻¹ h⁻¹. In a separate durability test at 500 °C and GHSV 60,000 mL g⁻¹ h⁻¹, its NH₃ conversion declined by less than 5% over 480 hours. The two results describe different operating conditions and should be read separately. ## Background Ammonia contains hydrogen at a high volumetric density and can be transported using established storage infrastructure. When decomposed according to 2NH₃ → N₂ + 3H₂, it releases hydrogen without producing carbon-containing gases at the reaction stage. This makes ammonia decomposition relevant to hydrogen transport and distributed hydrogen generation, provided that unreacted ammonia can be controlled and the required heat can be supplied efficiently. The reaction remains kinetically demanding. Breaking N–H bonds and recombining surface nitrogen into N₂ generally require elevated temperatures. Ruthenium is highly active, but its scarcity and cost encourage the development of catalysts based on more abundant metals such as cobalt and nickel. At high temperature, small metal particles tend to migrate, merge and grow. This process, known as sintering, reduces exposed metal surface area and can gradually lower catalytic activity. Encapsulating the particles within an oxide or porous host can slow that movement, but the surrounding material must strike a balance. If it is too open, it may provide little protection. If it is too dense, NH₃ cannot readily reach the metal and N₂ and H₂ cannot leave. Conventional encapsulation methods can also require sacrificial templates, expensive precursors or several synthesis and removal steps. The authors explored a different route based on solid-state chemistry. During reaction between BaCO₃ and Al₂O₃-derived species, an amorphous BaO–Al₂O₃ intermediate forms before the system crystallizes as BaAl₂O₄. They asked whether that intermediate could be retained around active metal nanoparticles and used directly as a functional catalytic environment. ## Research question Can a kinetically arrested BaO–Al₂O₃ intermediate form a durable but transport-accessible environment around non-noble metal nanoparticles during ammonia decomposition? The study also asks whether the roles of its components can be separated experimentally. In the authors’ model, Al–O regions contribute NH₃ adsorption, Ba–O regions modify electron density and assist bond activation and product release, and the interconnected amorphous structure constrains the metal without forming an impermeable crystalline shell. ## Inside the study ![](https://storage.ghost.io/c/7c/5e/7c5e0911-739b-4a35-94eb-3e810bf78ba7/content/images/2026/10/Screenshot-2026-10-01-at-12.37.27.png) The researchers prepared a catalyst with designed loadings of 25 wt% cobalt and 10 wt% aluminium using a commercial BaCO₃-derived system. Introducing the Co and Al precursors fragmented the initially micrometre-scale BaCO₃ particles and increased contact among the components. This closer contact created the conditions for an interfacial solid-state reaction during activation. The thermal programme proved central to the final structure. The selected catalyst was heated in pure NH₃ at 2 °C min⁻¹ to 600 °C and held for one hour. Under these conditions, the Ba-, Al- and metal-containing components reorganized into a strongly amorphous structure while crystallization to BaAl₂O₄ remained limited. Faster heating, activation at 700–800 °C or longer dwell times shifted the material toward residual carbonate or more crystalline products. The result is best understood as kinetic control. BaAl₂O₄ is not simply replaced by a different equilibrium phase. Instead, the reaction is interrupted within a processing window in which the BaO–Al₂O₃ environment has formed around the metal but has not completed long-range crystallization. Electron microscopy showed metallic cobalt particles averaging approximately 9.2 nm within continuous amorphous regions. At the particle edges, the surrounding layer was about 1.2 nm thick. The structure is thus not a conventional thick, separately synthesized shell. It is an interconnected disordered environment formed in situ as the oxide and carbonate components react around the metal. ![](https://storage.ghost.io/c/7c/5e/7c5e0911-739b-4a35-94eb-3e810bf78ba7/content/images/2026/10/Screenshot-2026-10-01-at-12.37.35.png) Identifying the local phases required measurements that limited exposure to air. BaO readily reacts with atmospheric CO₂ to form carbonate, which can obscure the state present during catalysis. The team collected Ba K-edge X-ray absorption spectra in situ at 600 °C under 75% H₂ and 25% N₂. The spectrum matched BaO. Quasi-in situ Al 2p X-ray photoelectron spectroscopy gave a binding energy of 73.7 eV, supporting a non-crystalline Al–O environment. Co K-edge extended X-ray absorption fine structure gave an average Co–Co coordination number of approximately 8.5, consistent with nanoscale metallic cobalt rather than bulk cobalt or isolated Co ions. Raman spectroscopy provided the complementary long-range comparison: faster heating left stronger carbonate signatures, while activation at 800 °C favoured crystalline BaAl₂O₄. Catalytic performance followed these structural changes under matched testing conditions. At 500 °C and GHSV 30,000 mL g⁻¹ h⁻¹, reducing the activation heating rate from 10 to 2 °C min⁻¹ increased NH₃ conversion from 65.0% to 88.7%. A sample activated at 800 °C reached 77.8%, while a crystalline BaAl₂O₄ control reached 63.2%. These comparisons connect the more strongly amorphous structure with higher activity without relying only on samples tested under different flow conditions. ![](https://storage.ghost.io/c/7c/5e/7c5e0911-739b-4a35-94eb-3e810bf78ba7/content/images/2026/10/Screenshot-2026-10-01-at-12.37.52.png) The highest reported production rate came from a more demanding throughput test. At 550 °C and GHSV 600,000 mL g⁻¹ h⁻¹, 25Co-AB-2R-600 generated H₂ at 1,358 mmol gmetal⁻¹ min⁻¹. This value describes the amount of hydrogen produced relative to the mass of active metal under the stated high-flow condition. Long-term stability was measured separately. At 500 °C and GHSV 60,000 mL g⁻¹ h⁻¹, NH₃ conversion decreased by less than 5% over 480 hours. Keeping these conditions attached to the result is important: the durability experiment does not show 480-hour operation at the GHSV used for the maximum production-rate measurement. The mechanistic experiments examined why the amorphous environment affects more than particle growth. Temperature-programmed ammonia measurements showed 2.9 times as many medium-strength NH₃ adsorption sites as on the Co/BaCO₃ control. The optimized catalyst also had an NH₃ reaction order of 0.17\. After reaction, its surface showed no detectable accumulation in the N 1s region, suggesting that strongly retained nitrogen-containing species did not build up to a measurable level under the examined conditions. ![](https://storage.ghost.io/c/7c/5e/7c5e0911-739b-4a35-94eb-3e810bf78ba7/content/images/2026/10/Screenshot-2026-10-01-at-12.38.17.png) NH₃–D₂ temperature-programmed surface-reaction experiments followed hydrogen–deuterium exchange and product release. NH₂D appeared from 121.5 °C on the optimized catalyst, compared with 262.4 °C on the Co/BaCO₃ control. N₂ and H₂ were also released more readily. These measurements support easier activation and turnover of adsorbed ammonia-derived species, although they do not directly image every bond-breaking and recombination event during steady-state catalysis. The authors assign complementary roles to the surrounding components. Amorphous Al₂O₃-like regions provide the principal sites for NH₃ adsorption. BaO donates electron density and changes the neighbouring metal environment, assisting N–H cleavage and the release of N₂ and H₂. The continuous but disordered BaO–Al₂O₃ network restrains cobalt movement while preserving transport pathways between the gas phase and the encapsulated particles. This interpretation explains why simply forming a crystalline barium aluminate phase does not reproduce the same behaviour. Long-range disorder is not treated as an absence of structure. It creates a distribution of local Ba–O, Al–O and metal–oxide interfaces that can participate in adsorption, electronic regulation and transport while accommodating structural change at high temperature. The team also applied the approach to nickel and nickel–cobalt materials. These systems formed related amorphous encapsulation structures, showing that the strategy is not limited to cobalt. The useful processing window remained composition-dependent, however. The active metal, barium and aluminium components had to react together under suitable heating and atmosphere conditions to generate the intended structure. ## Takeaways and outlook The study turns a normally transient solid-state intermediate into a working catalyst architecture. Rather than synthesizing a finished crystalline support and then depositing metal onto it, the researchers allow the metal and oxide components to reorganize together and stop the process before BaAl₂O₄ crystallization is complete. The evidence connects structure and function across several levels. Microscopy locates cobalt nanoparticles within amorphous regions. In situ and quasi-in situ spectroscopy identifies BaO, non-crystalline Al–O coordination and nanoscale metallic cobalt under protected conditions. Controlled activation comparisons link the processing window to NH₃ conversion, while adsorption, isotope-exchange and kinetic experiments support the proposed chemical roles of the shell. The 480-hour test then examines whether that structure can retain performance during extended operation. The broader design idea is that an encapsulating environment need not be chemically passive or perfectly crystalline. Here, the disordered BaO–Al₂O₃ phase helps restrain the metal, adsorb ammonia, modify electron density and maintain molecular transport. Future work can test how broadly this arrested-intermediate strategy applies to other metals, supports and high-temperature reactions, and how reliably the narrow structural window can be reproduced at larger preparation scales. ## About the researchers Yuan-Yi Zhang (Shandong University) is the first author. Jin-Cheng Liu (Nankai University), Feng Ryan Wang (University College London) and Chun-Jiang Jia (Shandong University) are the corresponding authors. The other authors are Kai Xu and Wei-Wei Wang of Shandong University, Hao-Xin Liu of University College London, and Chao Ma of Hunan University. ## Original research Yuan-Yi Zhang; Kai Xu; Wei-Wei Wang; Hao-Xin Liu; Chao Ma; Jin-Cheng Liu; Feng Ryan Wang; Chun-Jiang Jia. “Amorphous BaO–Al₂O₃ Encapsulation Enables Durable Metal Catalysts for Ammonia Decomposition.” *Nature Communications* (2026). DOI: [10.1038/s41467-026-78117-4](https://doi.org/10.1038/s41467-026-78117-4?ref=research-pop.com). Open access. --- ## Research POP Notes This article reflects the independent interpretation of the Research POP team and does not represent the views of the authors, their institutions or the journal. If you identify any inaccuracies or have concerns regarding the content, figures or attribution, please contact us at [team.researchpop@gmail.com](mailto:team.researchpop@gmail.com). We will review the matter promptly and make corrections or remove the relevant material where appropriate.