DOE OSTI2019
The initial aim of this project was to develop a low-cost compact, modularly scalable, point of use, on demand production system for the reforming of natural gas (NG) into products such as acetylene, hydrogen and other related hydrocarbons. Based on a pulsed dielectric barrier discharge (DBD), the fast rising voltage pulses produced by the system causes the dielectric breakdown of the gas to lag behind the rapid voltage increase so that a voltage greater than the static breakdown voltage of the gas, called an overvoltage, is briefly attained. The century long commercial use of the DBD discharge to push the highly endothermic oxygen to ozone reaction served as an additional motivation. The project meets the recommendations of the National Research Council 2007 report on plasma science for a greater investment in low temperature plasma science and engineering. An important finding is that the method investigated has a high specific energy requirement (SER) of the order of 1,000 to 5,000 kWh per kg of hydrogen, and a corresponding low efficiency of only ~0.1 to 0.5% of converting electrical into chemical (ΔH, ΔG) energy. This compares unfavorably with 52 kWh/kgH 2 required for NG thermal plasma reforming, or a pulsed spark discharge, Lotfalipour et. al, and Gao et. al. (For comparison electrolysis of water into hydrogen and oxygen gas requires only 45 to 80kWh/kgH 2 ). This project garnered data with different reactor configurations, materials, gases and mixtures including NG, methane, hydrogen, propane, argon, nitrogen and air, gap sizes ranging from 0.5 to 4.1 mm, electrode areas from 5 to 460 cm, voltages to 27 kV, pulse frequencies to 20 kHz and peak currents into the hundreds of amperes, temperatures to 625 °C and pressures from 15 mbar to 2 bar. An SRS 300 residual gas analyzer (RGA) yielded quadrupole mass spectrometry (QMS) data on the reformate. To analyze the data we extended the public domain software OpenChrom®, Wenig, to QMS data, yielding accurate identification and quantitative data on reformate products. This extension of OpenChrom® is applicable beyond this project and is a significant contribution to the general community of researchers. The documentation is available on the internet, see Whitlock, and a copy has been appended to this report. The main product of methane and natural gas plasma reforming is hydrogen, constituting ~90% of the products, followed by 1% to 10% ethane, propane, ethylene, acetylene and isobutene. Neopentane, propene, butane and pentane constitute less than 1 % of product. Adding a nickel catalyst increases the selectivity for hydrogen. Carbon nanosheets (graphene), a product of commercial interest, were synthesized by plasma reforming of acetylene at low pressure. The reformate analysis shows hat saturated and branched hydrocarbons are favored compared with unsaturated and linear hydrocarbons. For example, ethane, propane, iso-butane, 2-methyl butane, and 2-methyl pentane are produced in higher amounts compared with unsaturated and linear components. At low pressure acetylene and ethylene production rate is higher and propane production lower. We developed a recirculating system in which hydrogen is removed through a permeable membrane as fast as it is being produced by the reactor and the remaining products are recirculated at a high flow rate of several liters per minute. This keeps the concentration of H 2 in the reactor low, while achieving a total output concentration of 79%, high enough for pressure swing adsorption to be used for further purification. The recirculating system could be used with other gas reforming technologies. We report on four different reactors,1) a spark discharge reactor, 2) a standard DBD cylindrical reactor, 3) a small DBD reactor where catalyst powder can be quickly loaded onto a carrier to investigate the effect of the plasma on the catalyst and the effectiveness of the of the catalyst in promoting gas reformation and 4) a configurable DBD reactor to study the performance effects of dielectric selection, electrode gap distance, and electrode/dielectric interface on plasma reactor operation. This project achieved a commercializable 1kW plasma system including reactors a plasma pulse generator and a gas supply module.