472372 Catalytic Ammonia Decomposition to Hydrogen and Nitrogen on Supported Nanosized and Single-Site Cobalt Catalysts: Effects of Atomicity, Ligand Environment and Support

Thursday, November 17, 2016: 10:30 AM
Franciscan B (Hilton San Francisco Union Square)
Konstantin V. Khivantsev, Fahad Almalki and Miao Yu, Chemical Engineering, University of South Carolina, Columbia, SC

Hydrogen is one of the most widely used chemicals in the world. Petroleum refining and ammonia production are the major fields with the largest demand for hydrogen whereas automotive fuel is an emerging sector with the hydrogen demand growing incredibly fast. The total expected value of the whole hydrogen market is 152 billion USD by 2021. The main disadvantage of using pure hydrogen is the difficulty associated with its storage and transport.

Ammonia, on the other hand, is a hydrogen rich fuel (17.65 wt% hydrogen) that can be liquefied, stored and transported easily. As a liquid it has a volumetric hydrogen density about 45% higher than liquid hydrogen. To use this hydrogen, ammonia has to be decomposed over a catalyst to produce the desired clean fuel - hydrogen (H2) along with nitrogen (N2) a non-toxic, non-greenhouse gas that has no contribution to carbon emissions.

The most active catalysts used nowadays for ammonia decomposition are carbon nanotube-supported ruthenium nanoparticles. They are expensive and not always straightforward to produce. Therefore, finding new catalysts for ammonia decomposition that contain Earth-abundant metals utilized with atomic efficiency could be a viable alternative to expensive ruthenium catalysts. Moreover, gaining fundamental knowledge about metal-catalyzed ammonia decomposition, starting from the single-atom presents an important challenge that would allow us to establish more precise structure-catalytic property relationships, as well as create a pathway to produce more active/stable catalysts.

Herein, we demonstrate for the first time that the single-site silica-supported cobalt catalyst is an effective and stable catalyst for ammonia decomposition to nitrogen and hydrogen with activity starting at 480 °C and high TOF at elevated temperatures. The catalyst retains its single-site nature up until 630 °C, as evidenced by XPS, Raman and Diffuse-Reflectance UV-VIS spectroscopic measurements. Reduction of the catalyst takes place above 630 °C (with accompanying color change from blue to black) producing a reduced supported cobalt or cobalt nitride CoxNphase, which is even more active in catalytic NH3 decomposition, with activity above 390 °C. We discuss the proposed mechanism of NH3 decomposition by single-site Co(II)/SiO2 catalyst: even though the catalyst itself is very robust and it cannot be reduced or oxidized below 630 °C, it has unique activity towards nitrogen-containing molecules like ammonia and pyridine; those molecules chemisorb on Co(II) with the concomitant change of distorted tetrahedral environment of CoO4 unit to square pyramidal and octahedral as evidenced by Diffuse Reflectance UV-VIS. The catalytic activity of Co(II)/silica catalyst is assessed considering the previous report by A. Hock et al which indicates Co(II)/SiO2 is an active and selective catalyst for propane dehydrogenation at temperatures close to 600 °C. Based on the similarity of N-H and C-H bond strengths, the ability of Co(II)/SiO2 to break strong N-H bonds therefore signifies a general trend of this catalyst to break strong X-H bonds. However, the single-site cobalt catalyst is less active than extended metal or metal nitride surface, exemplifying the important fundamental and mechanistic difference between the regimes of operation for the single-site and nanosized supported Co catalysts for NH3 decomposition under given conditions.

Moreover, we show that the ligand environment around cobalt center can be modified on a single-atom scale: from CoO4 to CoN4 by incorporating single Co(II) atoms into mesoporous carbon nitride C3N4. This change in the ligand environment is not favourable for ammonia decomposition, with activity starting at much higher temperature compared to analogous Co(II)/SiOcatalyst.

Additionally, we demonstrate that the same dehydrogenation principle can be applied to ethylamine C2H5NH2, with the Co(II)/SiO2 catalyst capable of selectively dehydrogenating the mixed RCH2-NH2 bond of ethylamine to produce acetonitrile CH3CN.

To get a more general and fundamental understanding of the catalytic behavior of mononuclear Co(II) species embedded into silica, we tested it in two bench mark reactions for which nanosized supported cobalt and cobalt oxide particles are predominantly used: CO and CO2 hydrogenation. CO hydrogenation activity starts at temperatures above 380-400 °C with the main product being methane (~90%) and minor C2 products at 2 MPa CO/Hpressure; CO2 hydrogenation of CO2/H2 mixture at 1 bar leads to reverse water-gas shift reaction and formation of CO as the main COreduction product, detectable at temperatures above 300 °C.

In conclusion, Co(II)/SiO2 was shown to be catalytically active for ammonia decomposition as well as carbon monoxide/carbon dioxide hydrogenation. The direct comparison of two catalysts with atomically dispersed and nanodispersed supported metal atoms/particles once again exemplifies the fundamental difference between catalysis by single metal-atoms (cationic species) and traditional nanosized supported catalysts.

Extended Abstract: File Not Uploaded
See more of this Session: Catalytic Hydrogen Generation
See more of this Group/Topical: Catalysis and Reaction Engineering Division