The production of acetylene in a high-intensity electric arc
Thesis DisciplineChemical Engineering
Degree GrantorUniversity of Canterbury
Degree NameDoctor of Philosophy
The aim of this work was to develop further a process for producing acetylene in a high-intensity arc first suggested by Abrahamson. This process consisted of ablating a graphite anode into an atmosphere of hydrogen and then rapidly quenching the resultant hot gas mixture so as to preserve the acetylene which is the major hydrocarbon species in the carbon/hydrogen system at high temperatures. The reactor designed by Abrahamson had a throughput of up to 70 gm hr-¹ and an energy consumption of up to 3 kW. This reactor was scaled up in the -1 present work to have a throughput of up to 500 gm hr-¹ and an energy consumption of up to 30 kW. The reactor design was modified from the original to improve its operating capability and this also simplified the design of the reactor surrounds. The number of cathodes used was reduced from three to one, the cathode material changed from tungsten to carbon, and the design modified so that carbon did not deposit on the cathode tip during operation, a factor which had severely limited the length of operation possible in the small-scale plant. Abrahamson noted consistent deficits in the carbon balance over his reactor and hypothesised that this was due to the splitting off of graphite particles from the anode during ablation which were sufficiently small to escape collection in filters and hence in the carbon balance. This hypothesis was investigated at length in this work by accurate mass flow measurement and product gas analysis. No deficiencies were found in the carbon balances in this work. However, significant quantities of carbon compounds other than acetylene were found, most notably carbon monoxide and methane, and it is considered that the failure to monitor these by Abrahamson gave rise to the deficiency in his carbon balance which he explained by what is considered in the light of this work to be both a doubtful and unnecessary hypothesis. Further evidence against the prolonged existence of small carbon particles is given by calculations that showed that the majority of such particles would be evaporated within about 3 mm of the anode face. The results from this work are compared with those obtained by Abrahamson and with those predicted by his theory of “supercooled” equilibrium which hypothesised the attainment of gas phase only equilibrium prior to quenching and the supercooling of the gas stream with respect to solid carbon during quenching. It was found at first sight that the current results did not appear to agree with “supercooled” equilibrium. This disagreement is explained in terms of the quench rate obtained in the water-cooled tube used to quench the product and, in fact, although no positive support is given, the results in this work may not be at variance with Abrahamson's theory and this was adopted for subsequent calculations. Using the “supercooled” equilibrium theory, the optimum pre-quench conditions for operation for operation were found to be a mass carbon to hydrogen ratio of 2.5 and a specific enthalpy of 3.5 kcal gm-¹. Using these conditions, the best results obtained in this work (conversion of carbon to acetylene of 90%, an acetylene yeild yield of 11.8 mol % and an energy consumption of 45 kWh kg-¹ acetylene) and the best results obtained by Abrahamson (conversion 55%, yield 8 mol % and energy consumption 130 kWh kg-¹ acetylene), a procedure has been developed by which the performance of larger reactors can be predicted. It is predicted that although increasing the scale of the reactor will reduce the energy requirement per unit of acetylene this cannot be reduced to an economic level without some means of recovering waste heat. The optimum pre-quench enthalpy of 3.5 kcal gm-¹ can be obtained with the heat imparted to the carbon on ablation of the anode (less than 40% of the total heat input) so that for optimum conditions to prevail large amounts of the input energy to the arc are wasted. The presentation of the reactor heat flows in block diagram form enabled the energetic feasibility of this process and similar ones to be studied simply, but precisely and as a result it is shown that this low efficiency of energy usage given above may be able to be overcome by using a secondary reagent feed of methane following the arc. Under these conditions, the energy requirement is shown to be potentially as low as 7 kWh kg-¹, a figure which is economical when compared with other processes (see page 79 for a table summarising these) and is especially important in the light of the discovery of major natural gas deposits in New Zealand during the process of this work. It is concluded that further development in this direction would be worthwhile.