An Electrochemical Ammonia Synthesis Process

Abstract

One of the greatest concerns of this century is climate change due to rising greenhouse gas emissions. The ammonia industry is responsible for 1.4% of the global CO2 emissions, thereby having a negative climate impact. However, ammonia is an essential ingredient in nitrogen fertilisers and is also considered as a potential energy carrier. Ammonia is currently produced in the energy intensive Haber-Bosch process, where natural gas, coal or oil are used as the hydrogen and energy source. For these reasons, it is necessary to investigate more sustainable alternatives. One of the alternatives is electrochemical ammonia synthesis, where ammonia is formed out of nitrogen and water, and renewable energy sources fuel the process. The goal of this thesis is to give insight into the required performance metrics of the electrolyser and its production process in order for this technology to become technologically feasible and competitive with a Haber-Bosch process. This is done by investigating different options for the pre-treatment, electrochemical ammonia synthesis and separation steps, and comparing their energy requirements. The final product of this thesis is a process design for an electrochemical ammonia synthesis plant with a production capacity of 1,500 tonnes per day. Cryogenic distillation and pressure swing adsorption are researched and modeled as nitrogen generators. An adsorption column for the pressure swing adsorption unit is modeled in Matlab, while Aspen is used for modeling of the distillation column. Subsequently, an alkaline and a proton exchange membrane electrolyser are considered as ammonia synthesisers. The electrochemical cells are modeled as black boxes, operating at 353 K and 1 bar and 30 bar respectively. Next, a distillation column and a flash drum are modeled in Aspen as ammonia separators. Finally, four different process diagrams are created, two that are based on an alkaline electrolyser and two based on a proton exchange membrane electrolyser. Their overall energy consumptions are analysed and for both types of electrolysers the most optimal route for ammonia production is found. The results of this thesis point out that cryogenic distillation is preferred over adsorption for the generation of nitrogen, with an energy consumption of 0.56 kWh/kg nitrogen for compression, cooling and distillation. Adsorption can be competitive with cryogenic distillation when the nitrogen recovery rate is increased, or at lower production capacities. In a reasonable case for future electrolysers, operating at a cell voltage of 1.77 V and a Faradaic efficiency of 100%, the energy consumption for an alkaline electrolyser amounts to 12.00 kWh/kg ammonia and to 11.95 kWh/kg ammonia for a proton exchange membrane electrolyser. A distillation column was considered as utility for the separation of ammonia from the KOH solute product stream from an alkaline electrolyser. For the cathodic product stream from a proton exchange membrane electrolyser, containing only ammonia, hydrogen and nitrogen, flash separation was determined to be the best separation technology. For both separation options, it was found that an ammonia concentration of at least 10 mol% in the cell’s product stream is required for an efficient separation. Ultimately, with electrolysers operating at 1.77 V and a Faradaic efficiency of 70%, the total energy consumption of an alkaline based process is equal to 17.30 kWh/ kg ammonia and 15.45 kWh/kg ammonia for a proton exchange membrane based process, with overall energy efficiencies of 30% and 33% respectively. Based on the energy consumption of their respective pre¬treatment and separation steps, a proton exchange membrane electrolyser is favoured over an alkaline electrolyser. In order for an alkaline based process to be advantageous, its electrochemical cell should consume at least 1.80 kWh/kg ammonia less than a proton exchange membrane electrolyser. Finally, currently it is not possible for AEL or PEMEL based ammonia synthesis processes to be competitive with the Haber-Bosch production process in terms of energy consumption. However, with Faradaic efficiencies of 100% and minimal overpotentials, a PEMEL based process does come close to reaching this objective.