Process design and cost evaluation for H2 and ethylene production from bioethanol

1. Introduction The steam reforming of bioethanol gained growing attention as a pathway towards renewable hydrogen production. Bioethanol is being produced mainly from the fermentation of “first generation” raw materials, e.g. sugar cane and corn, which compete with the feed and food chain. However, also second generation production processes are becoming available, producing bioethanol from lignocellulosic biomass. During the last 10 years our group focused on process development for the production of hydrogen from second generation bioethanol, which was kindly supplied by the Biochemtex group through the Proesa technology. More recently we also turned our attention to ethylene production through ethanol dehydration under unconventional reaction conditions1. For both applications we focused on the use of diluted bioethanol solutions, i.e. obtainable after simple flash separation of excess water and fermentation residua (in which ethanol concentration is ca. 50 wt%). This substrate is particularly interesting for steam reforming since it already contains the water amount that should be co-fed in the steam reformer. On the other hand heavy dehydration of ethanol (commonly accomplished after a flash unit with a rectification columns and molecular sieves) is an energy waste in this case. For ethylene production cofeeding water may inhibit the thermodynamic conversion, but it helps improving catalyst life preventing coking. So, for both applications, the possibility to exploit diluted bioethanol solutions is intriguing. Therefore, in this work we explored the effect of possible impurities contained in raw bioethanol (S-containing compounds, acids and higher alcohols) on catalyst durability for both reactions. Then we designed and simulated the performance of a fully integrated and optimized plant for hydrogen production. One layout was dedicated to centralized hydrogen production, sized to transform 40kton/year bioethanol (Biochemtex plant capacity installed in Crescentino (VC), Italy). For this plant we performed a full economic assessment of the plant. The other layout was designed for distributed micro-cogeneration, so that a fuel processor with 6 integrated reactors for hydrogen production and purification was coupled with a fuel cell. The plant capacity was set to achieve 5 kWelectrical + 5 kWthermal energy output2. 2. Experimental Home prepared catalysts were tested for bioethanol steam reforming (ESR) and ethylene production using diluted bioethanol solutions: 50 wt% and 90 wt%, compared with 99.9 vol% ethanol. The catalysts were 10 wt% Ni/ZrO2/9 wt% K2O, prepared by flame pyrolysis in our lab, while for ethanol dehydration we used an H-BEA zeolite, with tuned Si/Al = 17 atomic ratio, kindly supplied by Prof. S. Dzwigaj (Pierre et Marie Curie University, Paris). ESR was carried out at atmospheric pressure, temperature between 300 and 625°C and with overall water/ethanol ratio = 3 mol/mol. Ethanol dehydration reaction was performed at atmospheric pressure, temperature ranging from 300 to 500 °C and water/ethanol ratio from 0 to 3 mol/mol. The materials were characterized by N2 adsorption-desorption, X-ray Diffraction (XRD), Scanning Electron Microscopy (SEM) and temperature programmed reduction/oxidation (TPR/TPO) and FT-IR of adsorbed pyridine for acidity. Spent samples were analysed through TPO, microRaman, FESEM and TEM. Process simulation was carried out using the Aspen Plus process simulator, under steady state conditions. The economic evaluation was performed through the Aspen Economic Evaluator tool. 3. Results and discussion The effect of impurities concentration for both reactions was negligible. This allowed to exploit the potential of 50 wt% second generation bioethanol as poorly expensive raw material than absolute ethanol. Based on detailed kinetic modelling on our samples we set up a full flowsheet for centralized hydrogen production and for residential size distributed cogeneration through fuel cells. Pelectrical output ranged from 3.3 to 6.0 kW, whereas the total heat output Pthermal ranged from 3.9 to 7.2 kW. The highest overall energy output corresponded to Pelectrical = 4.8 kW, PThermal = 7.2 kW, for a total 12 kW energy output. This was achieved by feeding a mixture with water/ethanol ratio = 11 (mol/mol), irrespectively of the catalyst mass, and setting the fuel consumption to the burner (which heats the steam reformer using part of the reformate) so to have an average temperature of 635°C in the ESR reactor. As for the centralised hydrogen production the following plant was set up, operating at 20 bar. The burner was analysed in different scenarios, i.e. using natural gas, azeotropic ethanol (the diluted one is unsuitable) or part of the reformate. This latter option, though adapt for distributed cogeneration, induces to an unprofitable plant. This economic assessment also evidenced that the plant is particularly sensitive to operating costs, especially related to raw materials (account for ca. 80%). Thus the economic feasibility of hydrogen production through this route is strictly dependent on the availability of a cheap bioethanol source. 4. Conclusions The valorization of second generation bioethanol to produce hydrogen and ethylene has been explored deepening the possibility to exploit diluted (less expensive) solutions, such as 50 wt%. The possible impurities present in raw ethanol showed negligible effect on activity for both reactions. Based on simulation and economic evaluation of two different sizes of plants for hydrogen production, the system proved very dependent on feeding cost, confirming that the search for adapt and cheaper raw materials is the key for exploitation of these technologies. References 1. I. Rossetti, J. Lasso, M. Compagnoni, E. Finocchio, G. Ramis, A. Di Michele, S. Dzwigaj, Appl. Catal. B, 210 (2017) 407 2. A. Tripodi, M. Compagnoni, G. Ramis, I. Rossetti, Int. J. Hydrogen Energy, in press.