A preliminary physicochemical, exergetic and economic study of hydrogen production utilizing glycerol

Abstract

This work has as objective to demonstrate technical and economic viability of hydrogen production utilizing glycerol. The volume of this substance, which was initially produced by synthetic ways (from oil-derived products), has increased dramatically due mainly to biodiesel production through transesterification process which has glycerol as main residue. The surplus amount of glycerol has been generally utilized to feed poultry or as fuel in boilers, beyond other applications such as production of soaps, chemical products for food industry, explosives, and others. The difficulty to allocate this additional amount of glycerol has become it in an enormous environment problem, in contrary to the objective of biodiesel chain, which is to diminish environmental impact substituting oil and its derivatives, which release more emissions than biofuels, do not contribute to CO2-cycle and are not renewable sources. Beyond to utilize glycerol in combustion processes, this material could be utilized for hydrogen production. However, a small quantity of works (theoretical and experimental) and reports concerning this theme could be encountered. Firstly, the produced glycerol must be purified since non-reacted amounts of materials, inclusively catalysts, contribute to deactivate catalysts utilized in hydrogen production processes. The volume of non-reacted reactants and non-utilized catalysts during transesterification process could be reutilized. Various technologies of thermochemical generation of hydrogen that utilizes glycerol (and other fuels) were evaluated and the greatest performances and their conditions are encountered as soon as the most efficient technology of hydrogen production. Firstly, a physicochemical analysis must be performed. This step has as objective to evaluate the necessary amount of reactants to produce a determined volume of hydrogen and determine thermodynamic conditions (such as temperature and pressure) where the major performances of hydrogen production could be encountered. The calculations are based on the process where advance degrees are found and hence, fractions of products (especially hydrogen, however, CO2, CO, CH4 and solid carbon could be also encountered) are calculated. To produce 1 Nm3/h of gaseous hydrogen (necessary for a PEMFC - Proton Exchange Membrane Fuel Cell - containing an electric efficiency of about 40%, to generate 1 kWh), 0,558 kg/h of glycerol is necessary in global steam reforming, 0,978 kg/h of glycerol in partial oxidation and cracking processes, and 0,782 kg/h of glycerol in autothermal reforming process. The dry reforming process could not be performed to produce hydrogen utilizing glycerol, in contrary to the utilization of methane, ethanol, and other hydrocarbons. In this study, steam reforming process was preferred due mainly to higher efficiencies of production and the need of minor amount of glycerol as cited above. In the global steam reforming of glycerine, for one mole of glycerol, three moles of water are necessary to produce three moles of CO2 and seven moles of H2. The response reactions process was utilized to predict steam reforming process more accurately. In this mean, the production of solid carbon, CO, and CH4, beyond CO2 and hydrogen was predicted. However, traces of acetaldehyde (C2H2), ethylene (C2H4), ethylene glycol, acetone, and others were encountered in some experimental studies. The rates of determined products obviously depend on the adopted catalysts (and its physical and chemical properties) and thermodynamic conditions of hydrogen production. Eight reactions of steam reforming and cracking were predicted considering only the determined products. In the case of steam reforming at 600°C, the advance degree of this reactor could attain its maximum value, i.e., overall volume of reactants could be obtained whether this reaction is maintained at 1 atm. As soon as temperature of this reaction increases the advance degree also increase, in contrary to the pressure, where advance degree decrease as soon as pressure increase. The fact of temperature of reforming is relatively small, lower costs of installation could be attained, especially cheaper thermocouples and smaller amount of thermo insulators and materials for its assembling. Utilizing the response reactions process in steam reforming, the predicted volumes of products, for the production of 1 Nm3/h of H2 and thermodynamic conditions as cited previously, were 0,264 kg/h of CO (13% of molar fraction of reaction products), 0,038 kg/h of CH4 (3% of molar fraction), 0,028 kg/h of C (3% of molar fraction), and 0,623 kg/h of CO2 (20% of molar fraction). Through process of water-gas shift reactions (WGSR) an additional amount of hydrogen could be produced utilizing mainly the volumes of produced CO and CH4. The overall results (steam reforming plus WGSR) could be similar to global steam reforming. An attention must to be taking into account due to the possibility to produce an additional amount of CH4 (through methanation process) and solid carbon (through Boudouard process). The production of solid carbon must to be avoided because this reactant diminishes (filling the pores) and even deactivate active area of catalysts. To avoid solid carbon production, an additional amount of water is suggested. This method could be also utilized to diminish the volume of CO (through WGSR process) since this product is prejudicial for the activity of low temperature fuel cells (such as PEMFC). In some works, more three or even six moles of water are suggested. A net energy balance of studied hydrogen production processes (at 1 atm only) was developed. In this balance, low heat value of reactant and products and utilized energy for the process (heat supply) were cited. In the case of steam reforming utilizing response reactions, global steam reforming, and cracking processes, the maximum net energy was detected at 700°C. Partial oxidation and autothermal reforming obtained negative net energy in all cited temperatures despite to be exothermic reactions. For global steam reforming, the major value was 114 kJ/h. In the case of steam reforming, the highest value of net energy was detected in this temperature (-170 kJ/h). The major values were detected in the cracking process (up to 2586 kJ/h). The exergetic analysis has as objective, associated with physicochemical analysis, to determine conditions where reactions could be performed at higher efficiencies with lower losses. This study was performed through calculations of exergetic and rational efficiencies, and irreversibilities. In this analysis, as in the previously performed physicochemical analysis, conditions such as temperature of 600°C and pressure of 1 atm for global steam reforming process were suggested due to lower irreversibility and higher efficiencies. Subsequently, higher irreversibilities and lower efficiencies were detected in autothermal reforming, partial oxidation and cracking process. Comparing global reaction of steam reforming with more-accurate steam reforming, it was verified that efficiencies were diminished and irreversibilities were increased. These results could be altered with introduction of WGSR process. An economic analysis could be performed to evaluate the cost of generated hydrogen and determine means to diminish the costs. This analysis suggests an annual period of operation between 5000-7000 hours, interest rates of up to 20% per annum (considering Brazilian conditions), and pay-back of up to 20 years. Another considerations must to be take into account such as tariffs of utilized glycerol and electricity (to be utilized as heat source and (or) for own process as pumps, lamps, valves, and other devices), installation (estimated as US$ 15.000 for a plant of 1 Nm3/h) and maintenance cost. The adoption of emission trading schemes such as carbon credits could be performed since this is a process with potential of mitigates environment impact. Not considering credit carbons, the minor cost of calculated H2 was 0,16288 US$/kWh if glycerol is also utilized as heat sources and 0,17677 US$/kWh if electricity is utilized as heat sources. The range of considered tariff of glycerol was 0-0,1 US$/kWh (taking as basis LHV of H2) and the tariff of electricity is US$ 0,0867 US$/kWh, with demand cost of 12,49 US$/kW. The costs of electricity were obtained by Companhia Bandeirante, localized in São Paulo State. The differences among costs of hydrogen production utilizing glycerol and electricity as heat source was in a range between 0,3-5,8%. This technology in this moment is not mature. However, it allows the employment generation with the additional utilization of glycerol, especially with plants associated with biodiesel plants. The produced hydrogen and electricity could be utilized in own process, increasing its final performance.