Removal of Acid Gases from Biomass-to-Liquid Process Syngas Used as Raw Materials for Fischer-Tropsch Technology

Transcription

1 Journal of the Japan Institute of Energy, 93, (014) 17 Removal of Acid Gases from Biomass-to-Liquid Process Syngas Used as Raw Materials for Fischer-Tropsch Technology Kreangkrai MANEEINTR 1, Thanaphat LUEMUNKONG, and Tawatchai CHARINPANITKUL (Received January 7, 014) Due to the high energy consumption and decreasing oil supply, the alternative energy sources such as biomass and agricultural waste can be considered as the main sources of energy in the future. Currently, one of the technologies to convert such biomass to more valuable products is biomass-to-liquid process. However, acid gases like CO and H S are produced as intermediates causing the decrease in product heating values, operational problems, corrosion, and environmental concern. Therefore, the objective of this work is to simulate the absorption process of acid-gas cleaning by using various solutions at different operating conditions in order to meet the requirement. The results show that MDEA has higher performance to remove both gases compared to other chemicals. Furthermore, at 0.9 MPa and 5 C, the treated gas provides 3.19 vol% CO and 7.7 ppb of H S for MEA,.13 % CO and 0.0 ppb of H S for MDEA and 7.45 % CO and 1.1*10-6 ppb of H S for Selexol. Moreover, the liquid flow rate, solution concentration and effects of temperature and pressure are also investigated for the optimal design of the absorption column. Key Words Biomass, Acid gas cleaning, Absorption, Alkanolamine, Physical solvent 1. Introduction Due to the high energy consumption and decreasing oil supply, the alternative energy sources such as biomass including agricultural waste can be considered as the main sources of energy in the future 1). Based on Thailand s infrastructure, biomass including agricultural waste such as palm oil, straw, corn and so forth, can be used as energy sources. Consequently, technologies to convert this biomass to liquid hydrocarbon have been developed for this application. Currently, one of the technologies to convert such biomass to more valuable products is Biomass-To- Liquid (BTL) process producing syngas from biomass by using gasification. Later, Fischer-Tropsch (FT) process will be applied to convert this gas into liquid hydrocarbon ). Nevertheless, from BTL technology, acid gases such as carbon dioxide (CO ) and hydrogen sulfide (H S) can be formed as intermediates causing the decrease in product 1 Department of Mining and Petroleum Engineering, Faculty of Engineering, Chulalongkorn University Bangkok 10330, Thailand Department of Chemical Engineering, Faculty of Engineering, Chulalongkorn University Bangkok 10330, Thailand heating values, operational problems, corrosion, catalyst poisoning 3) and environmental concern such as climate change and global warming 4). Therefore, these acid gases are required to be removed from the process by using acid gas treatment to prevent the operational problems and improve product values prior to sending to FT unit. The requirement for the minimum level of contaminants in the gas cleaning of CO and H S concentrations from many criteria are less than 5 % and 10 ppb, respectively 1). Hanaoka et al. 5), has studied experimentally the bench-scale production of liquid fuel from biomass by using gasification. From the study, the composition of impurities in product gas from gasification process is 11 % CO and 9.1 ppm H S. Furthermore, the conditions of gas cleaning such as pressure and temperature are at 0.9 MPa and room temperature. Nowadays the technologies used for acid gas removal are absorption, adsorption, membrane separation and cryogenics. However, with the characteristics and conditions of the impurities in the gas, the effective technology to remove these acid gases is absorption process 6) both physical and chemical absorptions. The chemicals commercially used for acid gas removal

2 18 J. Jpn. Inst. Energy, Vol. 93, No. 11, 014 Table 1 Molecular structures of MEA and MDEA Name Molecular structure Monoethanolamine MEA Methyldiethanolamine MDEA are monoethanolamine (MEA) and methyldiethanolamine (MDEA) for chemical absorption and dimethylether of polyethylene glycol (DMPEG) or Selexol for physical absorption process 7). The molecular structures of MEA and MDEA are shown in Table 1. The advantages of these chemicals are that they can capture CO and H S at the same time with various conditions; then they can be used in single acid gas removal unit. Furthermore, they have been used in power generation and natural gas processing applications. Therefore, the data for solubility, physical properties and so on are available from literature 7). Much of the experimental work on acid gas removal has been studied. The simulation work of acid gas removal from BTL process is very interesting. Therefore, the objective of this work is to simulate the absorption process for acid-gas cleaning by using various solutions at different operating conditions in order to meet the requirement of product gas used as raw materials for liquid hydrocarbon production.. Simulation The simulation on acid gas cleaning process by means of absorption column with structured packing is investigated. The previous study used the ASPEN Plus software package to the process modelling based on aqueous MEA solution 8) for gas processing in power plants. In this work a variation in several parameters has been included, because the effects of several chemicals are expected to give a larger effect on the overall process performance. ASPEN plus 9), one of the simulators commercially used for chemical engineering work is chosen for this study. The detailed solubility or gas and liquid composition and temperature data is determined from the software for Selexol and aqueous MEA and MDEA solutions with CO and H S systems for gas processing application. The physical properties and column design are based on the data provided from Aspen simulator. Also, the effects of solution concentration, liquid flow rate, temperature and pressure for chemicals used are investigated for the optimal design in the future..1 Simulation of Physical Solvent The physical solvent used for this study is Selexol which is normally suited for the removal of CO from natural gas at high pressure. Because there is no chemical reaction, this solvent combined less strongly with CO and H S. The equilibrium concentration of the absorbing agent in the liquid phase is strongly dependent on the partial pressure of that component in the gas phase. The advantage of using such solvents is that these gases can be separated from them in the regenerator by reducing the pressure, resulting in much lower energy consumption losses 7).. Simulation of Chemical Solvent For chemical absorption, the chemical reactions illustrating the process are presented in Equation 1 to 7 which have been used in simulation. Tertiary amine solutions undergo reaction 1 to 4 but cannot react directly with CO to form a carbamate by reaction 5 because there is no hydrogen atom in the molecule attached to nitrogen atom. H O H + +OH - (1) H O+CO H + +HCO 3 - () HCO 3 - H + +CO 3 - RNH +H + RNHCOO - +H + (4) RNH +CO RNHCOO - +H + (5) H O+H S H 3O + +HS - (6) H S+HS - H 3O + +S - (7).3 Absorption Column Calculation The equation for column calculation is presented here as shown below 7) : p 1 p Bm dp L dc h = G m = k G a(p p) (P P e ) k L a(p p) (c e c) p Where: h = height of packed zone, m G'M = superficial molar mass velocity of inert gas, mol/(m -s) ρ L p 1, = partial pressure of solute in entering gas, Pa p = partial pressure of solute in leaving gas, Pa P = total pressure of system, Pa p = partial pressure of solute in main gas stream, Pa p e = partial pressure of solute in equilibrium with main body of solution, Pa p BM = log mean of inert gas pressures k'g = k G(p BM/p) = special gas-side mass-transfer coefficient which is independent of gas composition, mol/(m -s)(pa) k L = liquid-side mass-transfer coefficient, mol/(m -s) (mol/m 3 ) c 1 c (3) (8)

3 J. Jpn. Inst. Energy, Vol. 93, No. 11, a = interfacial area per unit volume of absorber, m / m 3 L = liquid flow rate, kg/(m -s) ρ L = liquid density, kg/m 3 (assumed constant) c 1 = solute concentration in liquid leaving bottom of column, mol/m 3 c = solute concentration in liquid fed to top of column, mol/m 3 c = solute concentration in main body of liquid, mol/ m 3 c e = concentration of solute in liquid phase in equilibrium with main body of gas, mol/m 3 The actual calculation was conducted in units of ft, lb, lb-mol, atm, and hr. These equations can be simplified by assuming the following conditions: 1. The equilibrium curve is linear over the range of concentrations encountered.. The partial pressure of the inert gas is essentially constant over the length of the column. 3. The solute contents of gaseous and liquid phases are sufficiently low that the partial pressure and liquid concentration values may be assumed proportional to the corresponding values when expressed in terms of moles of solute per mole of inert gas (or of solvent). y 1 x G 1 m dy L m dx h = = (9) K G ap y y y e ρ m K L a x x e x G m = superficial molar mass velocity, mol/(m-s) L m = molar values of liquid flow rate, mol/(m-s) K G = overall gas-phase mass transfer coefficient (mol/m -s- Pa) K L = overall liquid-phase mass transfer coefficient (m/s) ρ m = molar density, mol/m 3 y, x = mole fraction of solute in gas and in liquid, respectively Eq.9 can be arranged in terms of height of the transfer unit, HTU and number of transfer unit, N OG h = H OG N OG (10) G m where H OG = (11) K G ap y 1 dy and N OG = (1) y y y e Moreover, the conditions used in this simulation can be found in the literature 5). The gas contains 9.3 vol% CO, 1.9 vol% H, 11.0 vol% CO,.0 vol% CH 4, 35.5 vol% N, 9.1 ppm H S, and 5.3 ppm COS at 0.9 MPa and a flow rate of 5.5 Nm 3 /h. Furthermore, the details of the column used are 6 m packed height with 0 stages and 1 m in diameter. The packing type is Mellapak 50X dimension. 3. Results and Discussion The absorption column is practical for acid gas removal. It is used for simulation to study the effect of parameters such as types of absorbing agents, concentration of solutions and liquid flow rate on CO and H S removal from gasification. The simulation results are based on the data of experimental work from Hanaoka et al. 5) at 0.9 MPa absolute and 5 C and concentration of inlet acid gases. From experimental results, the composition of treated gas is 30.8 vol% CO, 5. vol% H, 0.9 vol% CO,.5 vol% CH 4, 40.6 vol% N, <5 ppb H S, and <5 ppb COS. 3.1 Effect of Chemicals Used The effects of the chemicals (MEA, MDEA and Selexol) on acid gas cleaning are shown in Figs. 1 and for CO and H S, respectively. The results are shown at conditions used of 0.9 MPa, 5 C, 150 kmol/h liquid flow rate. The simulation results show that MDEA and MEA are very good to remove CO and MDEA and Selexol are very outstanding in capturing H S. Moreover, the product gas consists of 3.64 vol% CO and 7.4 ppb H S for MEA, 0.74 % CO and 0.0 ppb H S for MDEA and 7.45 % CO Fig. 1 Effect of chemical and liquid flow rate on CO removal at 40 and 0.9 MPa Fig. Effect of chemical and liquid flow rate on H S removal at 40 and 0.9 MPa

4 130 J. Jpn. Inst. Energy, Vol. 93, No. 11, 014 and 1.1*10-6 ppb H S for Selexol. From the requirement, the product gas should have less than 5 % for CO and 10 ppb for H S. Therefore, MDEA is the suitable chemical for this application because it can capture both gases at the required level. 3. Effect of Liquid Flow Rate in the Packed Column The effects of the liquid flow rate on acid gas separation are shown in Figs. 1 and as well for all chemicals. It is evident that the liquid flow rate has a direct relationship with the efficiency of acid gas removal in that the efficiency increases as the liquid flow rate increases. This can be explained by the fact that applying a higher liquid flow rate to the system results in a larger driving force for acid gas removal, and consequently a higher rate of removing acid gas by the chemical per unit of time. This leads to a higher overall mass transfer coefficient that is directly proportional to the mass transfer values 10). From the viewpoint of liquid phase, at a higher liquid flow rate, the amount of free-amine molecules in the system becomes larger or the system has more capacity to absorb acid gas from the gas phase, thus enhancing the mass transfer 10). However, an increase in solution flow rate leads to a higher circulation rate and regeneration costs; thus, may not improve the overall system efficiency. Maximizing liquid flow rates may not lead to optimum operating conditions. Therefore, an optimum flow rate has to be determined. 3.3 Effect of Solution Concentration The effect of solution concentration is an interesting aspect since it affects the design of the process. The results of solution concentration of MEA at 3, 4 and 5 M are shown in Figs. 3 and 4 for both CO and H S removal. At flow rate of 150 kmol/h, the amount CO left in the product gas for 3M MEA is 5.03 % and that for H S is 39.6 ppb. For MDEA solution, the corresponding amount CO left in the product gas for 3M is found to be 4.00 % and that for H S is found to be 0.05 ppb although they are not shown in the figures. Also, the results of MDEA can meet the requirement for the criteria for the amount of acid gas in the product gas. From the results, it is clear that the efficiency of the process can increase as solution concentration increases because the amount of free molecules has increased per unit volume resulting in more acid gases removed. However, the higher concentration can cause more equipment corrosion and cost of chemical used as well as liquid viscosity 10). However, for the case of Selexol, there is no data for effect of concentration because normally Selexol is used as pure chemical. Fig. 3 Effect of solution concentration of MEA on CO removal at 40 and 0.9 MPa Fig. 4 Effect of solution concentration of MEA on H S removal at 40 and 0.9 MPa 3.4 Effect of Temperature The temperatures used in this study are 5 and 40 C and the effects of system temperature are presented in Figs. 5 and 6 for all chemicals. At flow rate of 150 kmol/ h, MDEA provides better performance among them. In addition, theoretically, at higher temperature gas has lower solubility because of higher kinetic energy to escape to gas phase. However, these two temperatures show little difference for solubility. Therefore, practical operating temperature can be at 40 C especially for tropical countries. 3.5 Effect of Pressure The effects of absolute pressure at 0.1, 0.9 and.5 MPa in the column are shown in Figs. 7 and 8 for both CO and H S removal with MEA. It is evident that solubility of gases in solution increases at higher pressure corresponding to Henry s law. Therefore, operation at.5 MPa offers the best solubility for MEA especially at flow rate of 150 kmol/h. However, increasing pressure means that extra equipment will be needed and leads to higher cost for acid gas removal.

5 J. Jpn. Inst. Energy, Vol. 93, No. 11, Fig. 5 Effect of temperature on CO removal at 0.9 MPa 4. Conclusion Among MEA, MDEA and Selexol, MDEA is the most effective chemical to capture both CO and H S. The effects of liquid flow rate, solution concentration, temperature and pressure on removal performance are also studied. From the results, it is evident that higher liquid flow rate can remove more gas because it has a higher driving force because of higher amount of free amine molecules per unit of time, and consequently a higher rate of removing acid gas by the chemical. For higher solution concentration, the amount of free-amine molecules in the system becomes larger or the system has more capacity to absorb acid gas. However, it is to be noted that increasing solution concentration or liquid flow rate can lead to a higher circulation rate, regeneration costs, more equipment corrosion, and cost of chemical used. In addition, high pressure and low temperature can provide higher solubility for acid gases but these conditions need some equipment to fulfill their requirement and the cost will be higher. Fig. 6 Effect of temperature on H S removal at 0.9 MPa Acknowledgements The authors would like to gratefully acknowledge the Research Support Grant from the Faculty of Engineering, Chulalongkorn University for financial support of this research. Also, partial support from Sliver Jubilee Fund of Chulalongkorn University (Climate Change Research Cluster) is also gratefully acknowledged. Fig. 7 Effect of pressure on CO removal in 4M MEA solution at 40 Fig. 8 Effect of pressure on H S removal in 4M MEA solution at 40 References 1)Leibold, H.; Hornung, A.; Seifert, H., Journal of Powder Technology, 180, (008) )Kumabe, K.; Sato, T.; Matsumoto, K.; Ishida, Y.; Hasegawa, T., Fuel, 89, (010) 3)Dunleavy, J., Platinum Metals Rev., 50, 110 (006) 4)Maneeintr K.; Idem, R. O.; Tontiwachwuthikul P.; Wee, A. G.H., Energy Procedia, 1, (009) 5)Hanaoka, T.; Liu Y.; Matsunaga, K.; Miyazawa, T.; Hirata, S.; Sakanishi, K., Fuel Processing Technology, 91, (010) 6)Gupta, M.; Coyle, I.; Thambimuthu, K., 1st Canadian CC7S Technology Roadmap Workshop, September 18-19, 003, Calgary, Alberta, Canada 7)Kohl, A. L; Nielsen, R. B., Gas Purification, 5ed., Gulf Publishing, )Abu-Zahra, M. R. M.; Schneiders, L. H. J.; Niederer, J. P. M.; Feron, P. H. M.; Versteeg, G. F., International Journal of Greenhouse Gas Control, 1, 7-46 (007) 9)Aspen Plus, Aspen Plus User Guide Version 1.1, Cambridge, )Maneeintr, K.; Idem, R. O.; Tontiwachwuthikul, P.; Wee, A. G. H., Ind. Eng. Chem. Res., 49, (010)