No abstract

Current methods of hydrogen production from methane generate more than 5 kg of CO2 for every 1 kg of hydrogen. Methane pyrolysis on conventional solid heterogeneous catalysts produces hydrogen without CO2, but the carbon coproduct poisons the catalyst. This can be avoided by using a molten metal alloy catalyst. We present here a study of methane pyrolysis using mixtures of molten Cu–Bi alloys as the catalyst. We find that molten Cu–Bi is an active catalyst, even though pure molten Bi and Cu are not. Surface tension measurements and constant-temperature ab initio molecular dynamics simulations indicate that the surface is enriched in Bi and that the catalytic activity is correlated with the concentration of Bi at the surface. Bader charge analysis indicates that bismuth donates charge to copper. In the most stable configuration of dissociated methane on these liquid surfaces, CH3 binds to a bismuth surface atom and H to Cu. The energy barriers for the dissociative adsorption of methane, calculated using the nudged elastic band (NEB) method, are between 2.5 and 2.6 eV, depending on the binding site on the surface of the Cu45Bi55 alloy. The computed barriers are in rough agreement with the experimental apparent activation energy of 2.3 eV.

Mixtures of molten iron–sodium-potassium chloride salts are found to be catalytic for methane pyrolysis. In a differential bubble column reactor, the apparent activation energy of the molten salt decreases from 301 kJ/mol for the eutectic NaCl-KCl to 171 kJ/mol for 3 wt % of iron-added as FeCl3. The solid carbon produced in the iron-containing salt mixture has a graphitic structure which is distinct from the more disordered carbon produced in the iron-free eutectic, suggesting a different solid carbon formation pathway. Results from H–D exchange investigations are consistent with a different reaction pathway for methane pyrolysis in the iron-containing NaCl-KCl melt than in the melt without Fe. The activity of the salt mixture was stable for over 50 h, producing molecular hydrogen and separable solid carbon. It is likely that the activity is due to the presence of Fe in molecular ions stabilized in the NaCl-KCl melt that facilitate the C–H bond activation in methane.

The catalytic decomposition of methane, propane, benzene, and crude petroleum was investigated between 900 and 1000 °C in molten metal bubble column reactors. The conversion to gas phase products and solid carbon was measured after introducing the gas phase reactants into a bubble column reactor containing a catalytic molten mixture of 27 mol % Ni and 73 mol % Bi. The conversions of propane, benzene, and crude oil are 100% at temperatures >950 °C at a reactor residence time of ∼1 s. Equilibrium selectivity of 100% H 2 and carbon was not achieved in the short residence time, but can be achieved at longer residence times. The solid carbon products obtained from methane pyrolysis were more graphitic than those produced from the other, highermolecular weight reactants; the latter were more amorphous, as measured by Raman spectroscopy and electron microscopy and resembled carbon black. A model is proposed for carbon formation in bubble column reactors, in which amorphous carbon products are derived from the gas-phase decomposition and graphitic carbon products are formed from dissolution and reprecipitation of carbon into and out of the molten metal.

Molten salts have received renewed attention as potential reaction media for methane pyrolysis, in which CO2-free hydrogen gas can be produced and the solid carbon can be continuously removed and...

The economic potential for commodity chemical production using sunlight is examined using a general comparative analysis method. The market values and feedstock prices together with basic thermodynamic constraints are used to evaluate the solar-to-chemical conversion potential for selected products using four different solar conversion pathways. Potential products are compared using several metrics including the net value of products per unit of energy input [$/kWh] and annual value of products per unit of solar exposed area [$/m 2 -year]. Low-volume, high-value chemical products such as iodine and tellurium would provide the greatest economic potential for a solar conversion process whereas high-volume, low-value products including hydrogen and methane would have tremendous challenges in generating sufficient revenue to pay for the required capital costs. Artificial photosynthesis research might be better served if the focus were to be first finding some (or any) reasonably valuable product that could be produced economically using sunlight before attempts are supported to make low-value chemical products such as hydrogen and hydrocarbons.

Methane pyrolysis on solid catalysts begins with C−H bond dissociation. In the formation of solid carbon, the initial C−C bond formation is thought to occur on/in the surface of most transition-metal catalysts and not in the gas phase. This has been shown to be true for Ni surfaces. In CH 4 −D 2 exchange reactions on polycrystalline Ni, CD 4 is the major product observed, which is consistent with surface-mediated dehydrogenation. However, on Cu surfaces, CH 3 D is the dominant exchange product observed, suggesting that deep dehydrogenation is unfavorable. Further, in a methane flow cell reactor, the carbon density and crystal size of the graphene formed from pyrolysis on Cu surfaces depend on the distance downstream in the flow field, whereas no such dependence is observed for Ni. Density functional theory calculations support an entropically favorable pathway, whereby on Cu after dissociative chemisorption of methane, the surface adsorbed CH 3 °desorbs rather than undergoing further dehydrogenation; on Ni, complete dehydrogenation is favorable. Reactive methyl radicals from Cu surfaces would participate in gas-phase pyrolysis upon desorption, forming the initial C−C bond in the gas phase with subsequent readsorption and surface modification of gas-phase generated oligomers.

The partial oxidation of methane to carbon and steam was investigated in molten salts for a process to produce CO 2 -free electrical power and solid carbon. Lithium iodide and lithium bromide catalysts were used in a bubble column where insoluble carbon accumulates on the melt surface and could be continuously removed. The salt acts as a heat transfer medium and reacts with oxygen to produce halogens and consume hydrogen halides in a chemical looping cycle. The halogens react with methane in gas-phase bubbles and form hydrogen halides and carbon. Hydrogen halides are then neutralized by an oxide and form steam and a halide salt. The halide salt reacts with oxygen, forming an oxide and closing both the halogen and the salt chemical looping cycles in a single vessel. Selectivities to carbon of 90% were measured for 56% methane conversion in a 12 cm bubble column reactor. The carbon was characterized by scanning electron microscopy, energy dispersive X-ray spectroscopy, and Raman spectroscopy. Iodide and bromide salts were investigated along with the behavior of iodine, bromine, methyl iodide, and methyl bromide intermediates.