- Email: [email protected]
Liquid antimony anode direct carbon fuel cell fueled with mass-produced de-ash coal
Energy 75 (2014) 555e559
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Liquid antimony anode direct carbon fuel cell fueled with mass-produced de-ash coal Hongjian Wang a, Tianyu Cao a, Yixiang Shi a, *, Ningsheng Cai a, Wei Yuan b a
Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing 100084, China b Shenhua Ningxia Coal Industry Group Co. Ltd., Yinchuan, Ningxia Province 750011, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 January 2014 Received in revised form 17 July 2014 Accepted 4 August 2014 Available online 26 August 2014
A liquid antimony (Sb) anode DCFC (direct carbon fuel cell) is fabricated on a smooth single crystal YSZ (Yttria Stabilized Zirconia) electrolyte substrate with porous Pt cathode to reveal the intrinsic reaction kinetics of electrochemical oxidation of liquid Sb and the reduction reaction characteristics of Sb2O3 with the reaction mass-produced Taixi de-ash coal fuel. The reduction kinetics of Sb2O3 with the de-ash coal is obtained using a temperature programmed reaction testing system. The reaction kinetics of the Sb2O3 with the de-ash coal can be enhanced by decreasing the coal particle size, and by adding de-ash coal into the anode chamber. The Sb2O3 accumulation at the interface between anode and electrolyte lead to the increase of ohmic resistance. While effective reaction active sites increase when the mole content of oxygen ion conductor Sb2O3 increase at the earlier stage of the cell discharging processes which further decrease the electrode polarization. The Si and Fe in the ash possibly accumulate at the interface between anode and electrolyte. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Direct carbon fuel cell Liquid Sb anode Mass-produced de-ash coal Performance characteristics
1. Introduction Fuel cell is a clean, efficient and competitive power generation technology, which could convert fuel directly into electricity [1e3]. A hydrogen fueled PEMFC (Proton exchange membrane fuel cell) could exhibit satisfying performance on power generation and environment protection [4,5] under relatively mild working temperature (<100 C). But when the fuel becomes less electrochemically active, such as carbon fuel, one should raise the operation temperature of the fuel cell to activate the reaction. SOFCs (solid oxide fuel cell) mainly work between 600 and 1000 C is an ideal device for direct carbon power generation. An SOFC fueled by carbon fuel makes a SO-DCFC (solid oxide direct carbon fuel cell). The anode where the carbon is oxidized to produce electricity is a vital part of the whole power generating system, and there are mainly three types of anodes for SO-DCFC: Porous solid anode [6e8], molten carbonate anode [9,10] and liquid metal anode [11]. When compared to the porous solid anode, liquid metal offers better contacting condition between carbon fuel and electrode which promotes the SO-DCFC's performance. As the liquid metal could serve as a good electronic conductor, the liquid metal anode meets less problem than the molten carbonate anode when it comes to * Corresponding author. Tel./fax: þ86 10 62789955. E-mail address: [email protected] (Y. Shi). http://dx.doi.org/10.1016/j.energy.2014.08.017 0360-5442/© 2014 Elsevier Ltd. All rights reserved.
current collection. And what makes the liquid metal anode more fascinating is that a SO-DCFC utilizing a liquid metal anode can keep producing electricity for a couple of time without carbon supplement. This would decrease performance fluctuation brought by carbon fuel transportation. A number of liquid metals have been studied as the liquid metal anode, such as Sn [12,13], In Refs. [14], Bi [15], Ag [16], Sb [17e20]. Due to a non-oxygen-ion conducting layer formed at the interface between anode and electrolyte, the performance of liquid Sn [12,13] and In Ref. [14] anode decreased sharply at battery mode. Although Bi2O3 is an oxygen ion conductor, the OCV (open circuit voltage) of liquid Bi [15] is quite low. For liquid Ag, AgO decomposes when temperature is higher than 250 C which avoid the metal oxide layer formation. However, the impedance of liquid Ag anode was high, ~100 U cm2 [16]. But for Sb, both Sb and Sb2O3 are liquids at typical SOFC operating temperatures, which is beneficial for the transport of Sb towards the reaction active sites and the transport of Sb2O3 away from the reaction active sites, the liquid Sb anode showed excellent performance both under the “battery mode”, in which the metallic Sb is electrochemically oxidized by oxygen ions, and under the carbon fuel mode [15,16]. Raymond J. Gorte [17] used rice starch, carbon black and sugar char as fuel for liquid Sb anode DCFC, which suggested that the amount of ash in the carbon fuel, the carbon fuel density and the
556
H. Wang et al. / Energy 75 (2014) 555e559
initial reaction temperature were very important for the liquid Sb anode DCFC as the carbon fuel should contact the liquid Sb and the carbon oxidation should be thermo-dynamically favored. Recently, Chung-Hwan Jeon et al. [21] and Lee Injae [22] showed the feasibility of using de-ash coal in a porous-solid-anode SO-DCFC, since it has low ash content. In order to use de-ash coal as fuel in liquid Sb anode DCFC, the performance characteristics of liquid Sb anode SODCFC fuel with de-ash coal is necessary. Meanwhile, Shenhua Ningxia Coal Industry Group of China has successfully achieved mass production of Taixi de-ash coal (600,000 ton/year) using the skimping-floatation method. As shown in Fig. 1, the de-ash coal can be obtained by this novel coal washing system developed by Shenhua Ningxia Coal Industry Group of China. The de-ash coal can then be fed into the anode chamber of the liquid Sb anode DCFC (direct carbon fuel cell). The coal added in would reduce the Sb2O3 produced during fuel cell operation while power and heat can be generated cleanly and efficiently. For further consideration, syngas can also be produced at a certain operating conditions, which indicates the possibilities of developing an advanced power-gas cogeneration system based on the liquid Sb anode DCFC technology. In this study, the mass-produced de-ash coal from Shenhua Ningxia Coal Industry Group of China was directly used as the fuel of DCFC. The reduction kinetics of Sb2O3 with Taixi de-ash coal was obtained using a TPR (temperature programmed reaction) testing system. A liquid Sb anode direct carbon fuel cell was fabricated on a single crystal YSZ (Yttria Stabilized Zirconia) electrolyte substrate. IV (Current-voltage) characteristic curves and EIS (electrochemical impedance spectroscopy) were measured to evaluate the cell performance.
as current collector on the cathode. While on the anode side, a Ni20Cr80 ring was utilized as the current collector and also work as an antimony metal electrode holder above the electrolyte. The metallic antimony and the solid carbon fuel would be added into the anode chamber during the cell operation through the top of the alumina tube. In this study, 10 g antimony (99.5%, Sinopharm Chemical Reagent Limited Company, China) was added to the upper surface of the YSZ electrolyte. The thickness of the liquid metal thin layer should be around 7.5 mm according to the approximate calculation based on the density of the liquid Sb at 800 C (6.34 g cm3) [23]. The fuel cell preparation is described in more details in our previous work [24].
2.2. Characteristics of Taixi de-ash coal Taixi de-ash coal (Shenhua Ningxia Coal Industry Group Co. Ltd, China) was used as fuel for liquid Sb anode DCFC in this study. The proximate and ultimate analysis of the Taixi de-ash coal is shown in Table 1. The proximate analysis was carried out based on the standard of GB/T212 of China. The ultimate analysis was carried out based on the standard of GB/T476 of China. The sulfur in the coal was analyzed based on the standard of GB/T214 of China. The results shows that the weight percentage of the ash is lower than 3%, and fixed carbon in the de-ash coal is around 88.79%. Further, the specific surface area was 0.1836 m2 g1 tested by BET (BrunauerEm- mett-Teller method). The ash composition was shown in Table 2, which was analyzed based on the standard GB/T1574 of China.
2. Experimental procedures 2.3. Experimental testing setup 2.1. Liquid Sb anode fuel cell preparation A round single crystal smooth YSZ substrate with a diameter of 25 mm made of 13 mol%Y2O3 (Y0.13Zr0.87O1.935d, Crystal orientation <100>, Hefei Kejing Materials Technology Limited Company, China) was used as the electrolyte. The maximum roughness of the electrolyte surface is kept below 156 nm, and the thickness of the YSZ pellet is 500 mm. The platinum cathode of the cell was made from platinum paste (MC-Pt100, Grikin Advanced Materials Limited Company, China) by the method of screen printing. The as printed Pt paste layer was dried at 100 C in air for 15 min, and then a temperature of 800 C was set for cathode calcination. The post-calcination cathode is 16 mm in diameter and 15 mm in thickness. The cell was placed at the bottom end of a vertical alumina tube, fixed by an alumina plate with 3 springs. A platinum mesh was used
A test system was built for this liquid antimony anode fuel cell. The polarization curves were measured using the four-probe method with an electrochemical workstation (IM6ex, ZahnerElektrik GmbH, Germany). EIS (Electrochemical impedance spectroscopy) test was performed with amplitude of 20 mV, with frequency ranging from 100 kHz to 0.1 Hz. The ohmic resistance of the cell was estimated from the high frequency intercept of the EIS curve. A K-type thermocouple was placed next to the fuel cell as a temperature monitor. During the experiments, the button cell was heated to 800 C from room temperature, shielded in Ar. As soon as the fuel cell reaches the temperature steady state, the Ar gas flow rate in the anode chamber was increased, and the cathode gas was switched to air. And the Sb powder was added. The parameters of cell testing are presented elsewhere [24].
Fig. 1. Technology roadmap for liquid Sb anode DCFC using de-ash coal.
H. Wang et al. / Energy 75 (2014) 555e559 Table 1 Proximate and ultimate analysis of the Taixi de-ash coal. Proximate analysis (wt.%, air dry basis)
Ultimate analysis (wt.%, air dry basis)
Moisture Ash Volatiles Fixed C carbon Taixi de-ash coal 0.90
2.65 7.66
88.79
H
O
N
S
88.81 3.27 3.49 0.69 0.19
Table 2 Ash composition of the Taixi de-ash coal (wt.%). SiO2
Al2O3
Fe2O3
TiO2
CaO
MgO
K2O
Na2O
MnO2
SO3
P2O5
26.24
25.95
11.52
1.11
15.53
4.80
0.53
3.13
0.12
5.64
0.34
3. Results and discussion 3.1. Reaction characteristic of the de-ash coal with antimony oxide TPR measurements were carried out with different particle sizes of the de-ash coal reducing Sb2O3. The temperature was increased to 900 C at a rate of 10 C min1. And the carrier gas was Ar with 50 mL min1. The reactant mixture was 0.1 g coal power and 4.0 g Sb2O3. There are two main systematic errors: a) the tested
557
temperature is not exactly the temperature of liquid Sb2O3, as there is a Al2O3 tube between the thermocouple and the liquid Sb2O3; b) the flue gas molar concentration is not exact, as there is a tube from the reactor to the mass spectrometer. However, the variation trend for reaction characteristics of de-ash coal with different particle sizes is obvious, which meant that the error effects is not obvious. Fig. 2(a) shows that the particle sizes 50e100 mm had the better reaction kinetics performance, which effectively reduced the total reaction time due to an increase of the effective contact area between the de-ash coal and liquid Sb2O3. Fig. 2(b) further shows that CO is also generated, which was less than 10% of the CO/CO2 gas, while the balance gas component of the Boudouard reaction at 800 C should be 85% CO, 15% CO2 [25]. This indicated that the reaction between the carbon and Sb2O3 controlled the finial flue gas component other than the Boudouard reaction. In fact, the higher cell efficiency and higher fuel utilization efficiency can be achieved with higher content of the CO2 component in the gas product. It should be also noted that the CO content can be further reduced by tuning the electrode composition or by adding the flue gas recycling system. 3.2. Liquid Sb anode performance In order to investigate the performance of liquid Sb anode DCFC using Taixi de-ash coal, we discharged the cell at a constant voltage
Fig. 2. TPR test results for Taixi de-ash coal (a) different particle sizes, (b) gas component.
558
H. Wang et al. / Energy 75 (2014) 555e559
of 0.3 V for 4 h and then added 1.0 g de-ash coal in situ with the particle size 50e100 mm as shown in Fig. 3. It can be seen that in the battery mode, the current density decreased from 2425 A m2 to 890 A m2. It indicated that the products Sb2O3 accumulated at the interface between the electrolyte and the liquid Sb anode and the effective reaction boundaries decreased due to Sb consumption. Noteworthy, there were four times of obvious performance increasing, respectively at 15.6 min, 56.6 min, 1 h 42 min, 2 h 9 min. It is probably caused by the mass transportation near the anodeelectrolyte interface, especially the Sb2O3 transportation which is volatile with high vapor pressure (6432 Pa at 800 C [26]). When the product Sb2O3 is generated at the electrolyte surface, the reaction boundaries will increase and oxygen ion transport path will be extended. After adding the de-ash coal, the current density increased sharply from 890 A m2 to 2180 A m2 within 18 min and then the current density fluctuated between 2070 A m2 and 2220 A m2 for another 2 h. It indicated that the de-ash coal can greatly recover the cell performance up to 90% of the initial cell performance at battery mode. And the de-ash coal can keep the cell performance though the reduction of Sb2O3 by de-ash coal. It showed that the de-ash coal was a potential carbon fuel for liquid Sb anode DCFC. Fig. 4 shows the IV and EIS curves of the liquid Sb anode in the battery mode before discharging process and the carbon fuel mode after discharging process. There are also two main systematic errors in the test: a) voltage leakage; b) the ohm resistance of electrode lead. However, the systematic errors have no effects on the comparison of battery mode performance with carbon fuel cell mode performance. The OCVs (open circuit voltages) of the cell were almost kept the same, respectively 0.705 V and 0.704 V. In order to verify the electrochemical reaction at the interface, Nernst equation was applied to calculate the theory OCV of pure Sb and carbon.
Q vi aproducts RT Q ln E¼E i nF avreactants 0
(1)
where E is the OCV, V; E0 is the OCV with pure reactants and pure oxygen, V; R is universal gas constant, 8.314 J mol1 K1; T is the temperature, K; n is the number of transfer electron; F is Faraday's constant, 96,487 C mol1; vi is the stoichiometric coefficient of species i in electrochemical reaction; aproducts is the activity of product species and areactants is the activity of reactant species. The activity of solid carbon is equal to 1. And the activity of liquid Sb
Fig. 4. Performance characteristics of the liquid Sb anode DCFC before and after discharge (a) IV curves, (b) EIS.
and liquid Sb2O3 are both supposed to be 1. The OCV of carbon fuel with air at 800 C is 0.99 V, while the OCV of Sb with air at 800 C is 0.704 V. It indicated that the OCV was controlled by the liquid Sb other than the carbon fuel. And the IV curves were almost coincident and seemed to be straight lines. The maximum power density in the battery mode was 785 W m2 at 0.37 V, while the maximum power density after discharge was 721 W m2 at 0.39 V. The limit current density in the battery mode (4057 A m2) was a
Fig. 3. Performance of liquid Sb anode DCFC discharged at constant voltage 0.3 V.
H. Wang et al. / Energy 75 (2014) 555e559
559
between anode and electrolyte, and the effects of Si, Fe accumulation need further researches. Acknowledgments This work was supported by the National Natural Science Foundation of China (51106085, 20776078) and the Seed Funding of Low Carbon Energy University Alliance (300900002). References
Fig. 5. SEM image of the cell section after discharge.
little higher than that after discharge (3844 A m2). It further indicated that de-ash coal can recover the cell performance. In Fig. 4(b), it shows that the ohmic resistance increased from 0.71 U to 0.75 U. And the electrode polarization also decreased from 0.23 U to 0.18 U. Noteworthy, a Warburg resistance appeared in the carbon fuel mode. It indicated that Sb2O3 accumulated at the interface between anode and electrolyte during the discharging process, which possibly increased the path length of oxygen ion transport. However, liquid Sb2O3 as an oxygen ion conductor with a conductivity of approximately 0.0792 S cm1 at 828 C [27] can be well mixed with liquid Sb metal to form new reaction active sites in the interface between the ionic conductor phase and electronic conductor phase. The furnace was cooled down to room temperature with Ar shield gas after discharging process. Then, the cell cross section was checked by SEM (Scanning Electron Microscope). Fig. 5 shows that there was a Sb/Sb2O3 mixture layer on the top of the YSZ electrolyte. It indicated that the performance degradation was possibly due to Sb2O3 accumulation at the interface. And elementary Fe and Si were found in the layer, less than 2 wt.% by EDS analysis. The liquid Sb can accommodate the effects of the coal ash, and even not sensitive with the sulfur element in the coal. However, the Si and Fe in the ash possibly accumulated at the interface between anode and electrolyte, and effects of Si and Fe on the cell performance still need to be further considered. 4. Conclusions Liquid Sb anode DCFC fueled by Taixi de-ash coal was studied in this work which demonstrates the possibility of directly using the mass-produced de-ash coal in the liquid Sb anode DCFC. The results indicate that the Sb2O3 can be reduced by Taixi de-ash coal producing more than 90% CO2 and 10% CO in the flue gas. The smaller coal particle size accelerates the reduction rates of Sb2O3, as a consequence of increasing effective contact area between carbon fuel and liquid Sb2O3. The fuel cell performance after discharging process would recover after the de-ash coal was fed. The liquid Sb can accommodate the poisoning of the coal ash and sulfur. However, the Si and Fe in the ash possibly accumulated at the interface
[1] Wolk R, Lux S, Gelber S, Holcomb F. Direct carbon fuel cells converting waste to electricity. ERDC-CERL fuel cell program report; 2007. [2] Gür T. Mechanistic modes for solid carbon conversion in high temperature fuel cells. J Electrochem Soc 2010;157:B751e9. [3] Giddey S, Badwal S, Kulkarni A, Munnings C. A comprehensive review of direct carbon fuel cell technology. Prog Energy Comb Sci 2012;38:360e99. [4] Carton JG, Olabi AG. Design of experiment study of the parameters that affect performance of three flow plate configurations of a proton exchange membrane fuel cell. Energy 2010;35(7):2796e806. [5] Carton JG, Olabi AG. Wind/hydrogen hybrid systems: opportunity for Ireland's wind resource to provide consistent sustainable energy supply. Energy 2010;35(12):4536e44. [6] Nakagawa N, Ishida M. Performance of an internal direct-oxidation carbon fuel-cell and its evaluation by graphic exergy analysis. Industrial Eng Chem Res 1988;27(7):1181e5. [7] Gür TM, Huggins RA. Direct electrochemical conversion of carbon to electrical energy in a high temperature fuel cell. J Electrochem Soc 1992;139(10):L95e7. [8] Lee AC, Mitchell RE, Gür TM. Thermodynamic analysis of gasification-driven direct carbon fuel cells. J Power Sources 2009;194(2):774e85. [9] Balachov I, Dubois L, Hornbostel D, Lipilin A. Direct carbon fuel cells (DCFC): clean electricity from coal and carbon based fuels, direct carbon conversion workshop, U S Palm Springs CA; 2005. [10] Pointon K, Lakeman B, Irvine J, Bradley J, Jain B. The development of a carbonair semi fuel cell. J Power Sources 2006;162(2):750e6. [11] Toleuova A, Yufit V, Simons S, Maskell W, Brett D. A review of liquid metal anode solid oxide fuel cells. J Electrochem Sci Eng 2013;32:91e105. [12] Gemmen R, Abernathy H, Gerdes K, Koslowske M, McPhee W, Tao T. Fundamentals of liquid tin anode solid oxide fuel cell. J Am Cera Soc 2010;5:37e52. [13] Abernathy H, Gemmen R, Gerdes K, Koslowske M, Tao T. Basic properties of a liquid tin anode solid oxide fuel cell. J Power Sources 2011;196:4564e72. [14] Jayakumar A, Vohs J, Gorte R. Molten-metal electrodes for solid oxide fuel cells. Ind Eng Chem Res 2010;49:10237e41. s A, Vohs J, Gorte R. A comparison of molten Sn and [15] Jayakumar A, Lee S, Horne Bi for solid oxide fuel cell anodes. J Electroche Soc 2010;157:B365e9. [16] Javadekar A, Jayakumar A, Pujara R, Vohs JM, Gorte RJ. Molten silver as a direct carbon fuel cell anode. J Power Sources 2012;214:239e43. [17] Jayakumar A, Küngas R, Roy S, Javadekar A, Buttrey D, Vohs J, et al. A direct carbon fuel cell with a molten antimony anode. Energy Environ Sci 2011;4: 4133e7. [18] Javadekar A. Molten metal electrodes in solid oxide fuel cells. Doctor dissertation. University of Delaware; 2012. [19] Javadekar A, Jayakumar A, Gorte R, Vohs J, Buttrey D. Energy storage in electrochemical cells with molten Sb electrodes. J Electrochem Soc 2012;159: A386e9. [20] Jayakumar A, Javadekar A, Gissinger J, Vohs J, Huber G, Gorte R. The stability of direct carbon fuel cells with molten Sb and SbeBi alloy anodes. AIChE J 2012;9:3342e8. [21] Kim J, Choi H, Chang Y, Jeon C. Feasibility of using ash-free coal in a solidoxide-electrolyte direct carbon fuel cell. Int J Hydrogen Energy 2012;37: 11401e8. [22] Seong YA, Seong YE, Young HR, Yon MS, Cheor EM, Gyung MC, et al. Application of refuse fuels in a direct carbon fuel cell system. Energy 2013;51: 447e56. [23] Steinberg D. Simple relationship between temperature-dependence of density of liquid-metals and their boiling temperatures. Met Trans 1974;5:1341e3. [24] Wang H, Shi Y, Cai N. Polarization characteristics of liquid antimony anode with smooth single-crystal solid oxide electrolyte. J Power Sources 2014;245: 164e70. [25] Cao D, Sun Y, Wang G. Direct carbon fuel cell: fundamentals and recent developments. J Power Sources 2007;167:250e7. [26] Hicke W. The vapor pressure of Sb2O3. J Am Chem Soc 1930;52:10e2. [27] Arkel A, Flood E. The electrical conductivity of molten oxides. Can J Chem 1953;31:1009e19.
Contents lists available at ScienceDirect
Energy journal homepage: www.elsevier.com/locate/energy
Liquid antimony anode direct carbon fuel cell fueled with mass-produced de-ash coal Hongjian Wang a, Tianyu Cao a, Yixiang Shi a, *, Ningsheng Cai a, Wei Yuan b a
Key Laboratory for Thermal Science and Power Engineering of Ministry of Education, Department of Thermal Engineering, Tsinghua University, Beijing 100084, China b Shenhua Ningxia Coal Industry Group Co. Ltd., Yinchuan, Ningxia Province 750011, China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 January 2014 Received in revised form 17 July 2014 Accepted 4 August 2014 Available online 26 August 2014
A liquid antimony (Sb) anode DCFC (direct carbon fuel cell) is fabricated on a smooth single crystal YSZ (Yttria Stabilized Zirconia) electrolyte substrate with porous Pt cathode to reveal the intrinsic reaction kinetics of electrochemical oxidation of liquid Sb and the reduction reaction characteristics of Sb2O3 with the reaction mass-produced Taixi de-ash coal fuel. The reduction kinetics of Sb2O3 with the de-ash coal is obtained using a temperature programmed reaction testing system. The reaction kinetics of the Sb2O3 with the de-ash coal can be enhanced by decreasing the coal particle size, and by adding de-ash coal into the anode chamber. The Sb2O3 accumulation at the interface between anode and electrolyte lead to the increase of ohmic resistance. While effective reaction active sites increase when the mole content of oxygen ion conductor Sb2O3 increase at the earlier stage of the cell discharging processes which further decrease the electrode polarization. The Si and Fe in the ash possibly accumulate at the interface between anode and electrolyte. © 2014 Elsevier Ltd. All rights reserved.
Keywords: Direct carbon fuel cell Liquid Sb anode Mass-produced de-ash coal Performance characteristics
1. Introduction Fuel cell is a clean, efficient and competitive power generation technology, which could convert fuel directly into electricity [1e3]. A hydrogen fueled PEMFC (Proton exchange membrane fuel cell) could exhibit satisfying performance on power generation and environment protection [4,5] under relatively mild working temperature (<100 C). But when the fuel becomes less electrochemically active, such as carbon fuel, one should raise the operation temperature of the fuel cell to activate the reaction. SOFCs (solid oxide fuel cell) mainly work between 600 and 1000 C is an ideal device for direct carbon power generation. An SOFC fueled by carbon fuel makes a SO-DCFC (solid oxide direct carbon fuel cell). The anode where the carbon is oxidized to produce electricity is a vital part of the whole power generating system, and there are mainly three types of anodes for SO-DCFC: Porous solid anode [6e8], molten carbonate anode [9,10] and liquid metal anode [11]. When compared to the porous solid anode, liquid metal offers better contacting condition between carbon fuel and electrode which promotes the SO-DCFC's performance. As the liquid metal could serve as a good electronic conductor, the liquid metal anode meets less problem than the molten carbonate anode when it comes to * Corresponding author. Tel./fax: þ86 10 62789955. E-mail address: [email protected] (Y. Shi). http://dx.doi.org/10.1016/j.energy.2014.08.017 0360-5442/© 2014 Elsevier Ltd. All rights reserved.
current collection. And what makes the liquid metal anode more fascinating is that a SO-DCFC utilizing a liquid metal anode can keep producing electricity for a couple of time without carbon supplement. This would decrease performance fluctuation brought by carbon fuel transportation. A number of liquid metals have been studied as the liquid metal anode, such as Sn [12,13], In Refs. [14], Bi [15], Ag [16], Sb [17e20]. Due to a non-oxygen-ion conducting layer formed at the interface between anode and electrolyte, the performance of liquid Sn [12,13] and In Ref. [14] anode decreased sharply at battery mode. Although Bi2O3 is an oxygen ion conductor, the OCV (open circuit voltage) of liquid Bi [15] is quite low. For liquid Ag, AgO decomposes when temperature is higher than 250 C which avoid the metal oxide layer formation. However, the impedance of liquid Ag anode was high, ~100 U cm2 [16]. But for Sb, both Sb and Sb2O3 are liquids at typical SOFC operating temperatures, which is beneficial for the transport of Sb towards the reaction active sites and the transport of Sb2O3 away from the reaction active sites, the liquid Sb anode showed excellent performance both under the “battery mode”, in which the metallic Sb is electrochemically oxidized by oxygen ions, and under the carbon fuel mode [15,16]. Raymond J. Gorte [17] used rice starch, carbon black and sugar char as fuel for liquid Sb anode DCFC, which suggested that the amount of ash in the carbon fuel, the carbon fuel density and the
556
H. Wang et al. / Energy 75 (2014) 555e559
initial reaction temperature were very important for the liquid Sb anode DCFC as the carbon fuel should contact the liquid Sb and the carbon oxidation should be thermo-dynamically favored. Recently, Chung-Hwan Jeon et al. [21] and Lee Injae [22] showed the feasibility of using de-ash coal in a porous-solid-anode SO-DCFC, since it has low ash content. In order to use de-ash coal as fuel in liquid Sb anode DCFC, the performance characteristics of liquid Sb anode SODCFC fuel with de-ash coal is necessary. Meanwhile, Shenhua Ningxia Coal Industry Group of China has successfully achieved mass production of Taixi de-ash coal (600,000 ton/year) using the skimping-floatation method. As shown in Fig. 1, the de-ash coal can be obtained by this novel coal washing system developed by Shenhua Ningxia Coal Industry Group of China. The de-ash coal can then be fed into the anode chamber of the liquid Sb anode DCFC (direct carbon fuel cell). The coal added in would reduce the Sb2O3 produced during fuel cell operation while power and heat can be generated cleanly and efficiently. For further consideration, syngas can also be produced at a certain operating conditions, which indicates the possibilities of developing an advanced power-gas cogeneration system based on the liquid Sb anode DCFC technology. In this study, the mass-produced de-ash coal from Shenhua Ningxia Coal Industry Group of China was directly used as the fuel of DCFC. The reduction kinetics of Sb2O3 with Taixi de-ash coal was obtained using a TPR (temperature programmed reaction) testing system. A liquid Sb anode direct carbon fuel cell was fabricated on a single crystal YSZ (Yttria Stabilized Zirconia) electrolyte substrate. IV (Current-voltage) characteristic curves and EIS (electrochemical impedance spectroscopy) were measured to evaluate the cell performance.
as current collector on the cathode. While on the anode side, a Ni20Cr80 ring was utilized as the current collector and also work as an antimony metal electrode holder above the electrolyte. The metallic antimony and the solid carbon fuel would be added into the anode chamber during the cell operation through the top of the alumina tube. In this study, 10 g antimony (99.5%, Sinopharm Chemical Reagent Limited Company, China) was added to the upper surface of the YSZ electrolyte. The thickness of the liquid metal thin layer should be around 7.5 mm according to the approximate calculation based on the density of the liquid Sb at 800 C (6.34 g cm3) [23]. The fuel cell preparation is described in more details in our previous work [24].
2.2. Characteristics of Taixi de-ash coal Taixi de-ash coal (Shenhua Ningxia Coal Industry Group Co. Ltd, China) was used as fuel for liquid Sb anode DCFC in this study. The proximate and ultimate analysis of the Taixi de-ash coal is shown in Table 1. The proximate analysis was carried out based on the standard of GB/T212 of China. The ultimate analysis was carried out based on the standard of GB/T476 of China. The sulfur in the coal was analyzed based on the standard of GB/T214 of China. The results shows that the weight percentage of the ash is lower than 3%, and fixed carbon in the de-ash coal is around 88.79%. Further, the specific surface area was 0.1836 m2 g1 tested by BET (BrunauerEm- mett-Teller method). The ash composition was shown in Table 2, which was analyzed based on the standard GB/T1574 of China.
2. Experimental procedures 2.3. Experimental testing setup 2.1. Liquid Sb anode fuel cell preparation A round single crystal smooth YSZ substrate with a diameter of 25 mm made of 13 mol%Y2O3 (Y0.13Zr0.87O1.935d, Crystal orientation <100>, Hefei Kejing Materials Technology Limited Company, China) was used as the electrolyte. The maximum roughness of the electrolyte surface is kept below 156 nm, and the thickness of the YSZ pellet is 500 mm. The platinum cathode of the cell was made from platinum paste (MC-Pt100, Grikin Advanced Materials Limited Company, China) by the method of screen printing. The as printed Pt paste layer was dried at 100 C in air for 15 min, and then a temperature of 800 C was set for cathode calcination. The post-calcination cathode is 16 mm in diameter and 15 mm in thickness. The cell was placed at the bottom end of a vertical alumina tube, fixed by an alumina plate with 3 springs. A platinum mesh was used
A test system was built for this liquid antimony anode fuel cell. The polarization curves were measured using the four-probe method with an electrochemical workstation (IM6ex, ZahnerElektrik GmbH, Germany). EIS (Electrochemical impedance spectroscopy) test was performed with amplitude of 20 mV, with frequency ranging from 100 kHz to 0.1 Hz. The ohmic resistance of the cell was estimated from the high frequency intercept of the EIS curve. A K-type thermocouple was placed next to the fuel cell as a temperature monitor. During the experiments, the button cell was heated to 800 C from room temperature, shielded in Ar. As soon as the fuel cell reaches the temperature steady state, the Ar gas flow rate in the anode chamber was increased, and the cathode gas was switched to air. And the Sb powder was added. The parameters of cell testing are presented elsewhere [24].
Fig. 1. Technology roadmap for liquid Sb anode DCFC using de-ash coal.
H. Wang et al. / Energy 75 (2014) 555e559 Table 1 Proximate and ultimate analysis of the Taixi de-ash coal. Proximate analysis (wt.%, air dry basis)
Ultimate analysis (wt.%, air dry basis)
Moisture Ash Volatiles Fixed C carbon Taixi de-ash coal 0.90
2.65 7.66
88.79
H
O
N
S
88.81 3.27 3.49 0.69 0.19
Table 2 Ash composition of the Taixi de-ash coal (wt.%). SiO2
Al2O3
Fe2O3
TiO2
CaO
MgO
K2O
Na2O
MnO2
SO3
P2O5
26.24
25.95
11.52
1.11
15.53
4.80
0.53
3.13
0.12
5.64
0.34
3. Results and discussion 3.1. Reaction characteristic of the de-ash coal with antimony oxide TPR measurements were carried out with different particle sizes of the de-ash coal reducing Sb2O3. The temperature was increased to 900 C at a rate of 10 C min1. And the carrier gas was Ar with 50 mL min1. The reactant mixture was 0.1 g coal power and 4.0 g Sb2O3. There are two main systematic errors: a) the tested
557
temperature is not exactly the temperature of liquid Sb2O3, as there is a Al2O3 tube between the thermocouple and the liquid Sb2O3; b) the flue gas molar concentration is not exact, as there is a tube from the reactor to the mass spectrometer. However, the variation trend for reaction characteristics of de-ash coal with different particle sizes is obvious, which meant that the error effects is not obvious. Fig. 2(a) shows that the particle sizes 50e100 mm had the better reaction kinetics performance, which effectively reduced the total reaction time due to an increase of the effective contact area between the de-ash coal and liquid Sb2O3. Fig. 2(b) further shows that CO is also generated, which was less than 10% of the CO/CO2 gas, while the balance gas component of the Boudouard reaction at 800 C should be 85% CO, 15% CO2 [25]. This indicated that the reaction between the carbon and Sb2O3 controlled the finial flue gas component other than the Boudouard reaction. In fact, the higher cell efficiency and higher fuel utilization efficiency can be achieved with higher content of the CO2 component in the gas product. It should be also noted that the CO content can be further reduced by tuning the electrode composition or by adding the flue gas recycling system. 3.2. Liquid Sb anode performance In order to investigate the performance of liquid Sb anode DCFC using Taixi de-ash coal, we discharged the cell at a constant voltage
Fig. 2. TPR test results for Taixi de-ash coal (a) different particle sizes, (b) gas component.
558
H. Wang et al. / Energy 75 (2014) 555e559
of 0.3 V for 4 h and then added 1.0 g de-ash coal in situ with the particle size 50e100 mm as shown in Fig. 3. It can be seen that in the battery mode, the current density decreased from 2425 A m2 to 890 A m2. It indicated that the products Sb2O3 accumulated at the interface between the electrolyte and the liquid Sb anode and the effective reaction boundaries decreased due to Sb consumption. Noteworthy, there were four times of obvious performance increasing, respectively at 15.6 min, 56.6 min, 1 h 42 min, 2 h 9 min. It is probably caused by the mass transportation near the anodeelectrolyte interface, especially the Sb2O3 transportation which is volatile with high vapor pressure (6432 Pa at 800 C [26]). When the product Sb2O3 is generated at the electrolyte surface, the reaction boundaries will increase and oxygen ion transport path will be extended. After adding the de-ash coal, the current density increased sharply from 890 A m2 to 2180 A m2 within 18 min and then the current density fluctuated between 2070 A m2 and 2220 A m2 for another 2 h. It indicated that the de-ash coal can greatly recover the cell performance up to 90% of the initial cell performance at battery mode. And the de-ash coal can keep the cell performance though the reduction of Sb2O3 by de-ash coal. It showed that the de-ash coal was a potential carbon fuel for liquid Sb anode DCFC. Fig. 4 shows the IV and EIS curves of the liquid Sb anode in the battery mode before discharging process and the carbon fuel mode after discharging process. There are also two main systematic errors in the test: a) voltage leakage; b) the ohm resistance of electrode lead. However, the systematic errors have no effects on the comparison of battery mode performance with carbon fuel cell mode performance. The OCVs (open circuit voltages) of the cell were almost kept the same, respectively 0.705 V and 0.704 V. In order to verify the electrochemical reaction at the interface, Nernst equation was applied to calculate the theory OCV of pure Sb and carbon.
Q vi aproducts RT Q ln E¼E i nF avreactants 0
(1)
where E is the OCV, V; E0 is the OCV with pure reactants and pure oxygen, V; R is universal gas constant, 8.314 J mol1 K1; T is the temperature, K; n is the number of transfer electron; F is Faraday's constant, 96,487 C mol1; vi is the stoichiometric coefficient of species i in electrochemical reaction; aproducts is the activity of product species and areactants is the activity of reactant species. The activity of solid carbon is equal to 1. And the activity of liquid Sb
Fig. 4. Performance characteristics of the liquid Sb anode DCFC before and after discharge (a) IV curves, (b) EIS.
and liquid Sb2O3 are both supposed to be 1. The OCV of carbon fuel with air at 800 C is 0.99 V, while the OCV of Sb with air at 800 C is 0.704 V. It indicated that the OCV was controlled by the liquid Sb other than the carbon fuel. And the IV curves were almost coincident and seemed to be straight lines. The maximum power density in the battery mode was 785 W m2 at 0.37 V, while the maximum power density after discharge was 721 W m2 at 0.39 V. The limit current density in the battery mode (4057 A m2) was a
Fig. 3. Performance of liquid Sb anode DCFC discharged at constant voltage 0.3 V.
H. Wang et al. / Energy 75 (2014) 555e559
559
between anode and electrolyte, and the effects of Si, Fe accumulation need further researches. Acknowledgments This work was supported by the National Natural Science Foundation of China (51106085, 20776078) and the Seed Funding of Low Carbon Energy University Alliance (300900002). References
Fig. 5. SEM image of the cell section after discharge.
little higher than that after discharge (3844 A m2). It further indicated that de-ash coal can recover the cell performance. In Fig. 4(b), it shows that the ohmic resistance increased from 0.71 U to 0.75 U. And the electrode polarization also decreased from 0.23 U to 0.18 U. Noteworthy, a Warburg resistance appeared in the carbon fuel mode. It indicated that Sb2O3 accumulated at the interface between anode and electrolyte during the discharging process, which possibly increased the path length of oxygen ion transport. However, liquid Sb2O3 as an oxygen ion conductor with a conductivity of approximately 0.0792 S cm1 at 828 C [27] can be well mixed with liquid Sb metal to form new reaction active sites in the interface between the ionic conductor phase and electronic conductor phase. The furnace was cooled down to room temperature with Ar shield gas after discharging process. Then, the cell cross section was checked by SEM (Scanning Electron Microscope). Fig. 5 shows that there was a Sb/Sb2O3 mixture layer on the top of the YSZ electrolyte. It indicated that the performance degradation was possibly due to Sb2O3 accumulation at the interface. And elementary Fe and Si were found in the layer, less than 2 wt.% by EDS analysis. The liquid Sb can accommodate the effects of the coal ash, and even not sensitive with the sulfur element in the coal. However, the Si and Fe in the ash possibly accumulated at the interface between anode and electrolyte, and effects of Si and Fe on the cell performance still need to be further considered. 4. Conclusions Liquid Sb anode DCFC fueled by Taixi de-ash coal was studied in this work which demonstrates the possibility of directly using the mass-produced de-ash coal in the liquid Sb anode DCFC. The results indicate that the Sb2O3 can be reduced by Taixi de-ash coal producing more than 90% CO2 and 10% CO in the flue gas. The smaller coal particle size accelerates the reduction rates of Sb2O3, as a consequence of increasing effective contact area between carbon fuel and liquid Sb2O3. The fuel cell performance after discharging process would recover after the de-ash coal was fed. The liquid Sb can accommodate the poisoning of the coal ash and sulfur. However, the Si and Fe in the ash possibly accumulated at the interface
[1] Wolk R, Lux S, Gelber S, Holcomb F. Direct carbon fuel cells converting waste to electricity. ERDC-CERL fuel cell program report; 2007. [2] Gür T. Mechanistic modes for solid carbon conversion in high temperature fuel cells. J Electrochem Soc 2010;157:B751e9. [3] Giddey S, Badwal S, Kulkarni A, Munnings C. A comprehensive review of direct carbon fuel cell technology. Prog Energy Comb Sci 2012;38:360e99. [4] Carton JG, Olabi AG. Design of experiment study of the parameters that affect performance of three flow plate configurations of a proton exchange membrane fuel cell. Energy 2010;35(7):2796e806. [5] Carton JG, Olabi AG. Wind/hydrogen hybrid systems: opportunity for Ireland's wind resource to provide consistent sustainable energy supply. Energy 2010;35(12):4536e44. [6] Nakagawa N, Ishida M. Performance of an internal direct-oxidation carbon fuel-cell and its evaluation by graphic exergy analysis. Industrial Eng Chem Res 1988;27(7):1181e5. [7] Gür TM, Huggins RA. Direct electrochemical conversion of carbon to electrical energy in a high temperature fuel cell. J Electrochem Soc 1992;139(10):L95e7. [8] Lee AC, Mitchell RE, Gür TM. Thermodynamic analysis of gasification-driven direct carbon fuel cells. J Power Sources 2009;194(2):774e85. [9] Balachov I, Dubois L, Hornbostel D, Lipilin A. Direct carbon fuel cells (DCFC): clean electricity from coal and carbon based fuels, direct carbon conversion workshop, U S Palm Springs CA; 2005. [10] Pointon K, Lakeman B, Irvine J, Bradley J, Jain B. The development of a carbonair semi fuel cell. J Power Sources 2006;162(2):750e6. [11] Toleuova A, Yufit V, Simons S, Maskell W, Brett D. A review of liquid metal anode solid oxide fuel cells. J Electrochem Sci Eng 2013;32:91e105. [12] Gemmen R, Abernathy H, Gerdes K, Koslowske M, McPhee W, Tao T. Fundamentals of liquid tin anode solid oxide fuel cell. J Am Cera Soc 2010;5:37e52. [13] Abernathy H, Gemmen R, Gerdes K, Koslowske M, Tao T. Basic properties of a liquid tin anode solid oxide fuel cell. J Power Sources 2011;196:4564e72. [14] Jayakumar A, Vohs J, Gorte R. Molten-metal electrodes for solid oxide fuel cells. Ind Eng Chem Res 2010;49:10237e41. s A, Vohs J, Gorte R. A comparison of molten Sn and [15] Jayakumar A, Lee S, Horne Bi for solid oxide fuel cell anodes. J Electroche Soc 2010;157:B365e9. [16] Javadekar A, Jayakumar A, Pujara R, Vohs JM, Gorte RJ. Molten silver as a direct carbon fuel cell anode. J Power Sources 2012;214:239e43. [17] Jayakumar A, Küngas R, Roy S, Javadekar A, Buttrey D, Vohs J, et al. A direct carbon fuel cell with a molten antimony anode. Energy Environ Sci 2011;4: 4133e7. [18] Javadekar A. Molten metal electrodes in solid oxide fuel cells. Doctor dissertation. University of Delaware; 2012. [19] Javadekar A, Jayakumar A, Gorte R, Vohs J, Buttrey D. Energy storage in electrochemical cells with molten Sb electrodes. J Electrochem Soc 2012;159: A386e9. [20] Jayakumar A, Javadekar A, Gissinger J, Vohs J, Huber G, Gorte R. The stability of direct carbon fuel cells with molten Sb and SbeBi alloy anodes. AIChE J 2012;9:3342e8. [21] Kim J, Choi H, Chang Y, Jeon C. Feasibility of using ash-free coal in a solidoxide-electrolyte direct carbon fuel cell. Int J Hydrogen Energy 2012;37: 11401e8. [22] Seong YA, Seong YE, Young HR, Yon MS, Cheor EM, Gyung MC, et al. Application of refuse fuels in a direct carbon fuel cell system. Energy 2013;51: 447e56. [23] Steinberg D. Simple relationship between temperature-dependence of density of liquid-metals and their boiling temperatures. Met Trans 1974;5:1341e3. [24] Wang H, Shi Y, Cai N. Polarization characteristics of liquid antimony anode with smooth single-crystal solid oxide electrolyte. J Power Sources 2014;245: 164e70. [25] Cao D, Sun Y, Wang G. Direct carbon fuel cell: fundamentals and recent developments. J Power Sources 2007;167:250e7. [26] Hicke W. The vapor pressure of Sb2O3. J Am Chem Soc 1930;52:10e2. [27] Arkel A, Flood E. The electrical conductivity of molten oxides. Can J Chem 1953;31:1009e19.