- Email: [email protected]
Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas
Applied Thermal Engineering xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas Takaaki Somekawa ⇑, Kazuo Nakamura, Takuto Kushi, Takao Kume, Kenjiro Fujita, Hisataka Yakabe Fundamental Technology Dept., Tokyo Gas Co., Ltd., 1-7-7 Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa 230-0045, Japan
h i g h l i g h t s 2-stage SOFC system with fuel regenerating techniques was designed and fabricated. Total fuel utilization ratio was successfully enhanced to 92.0% in hot module tests. The maximum power generation efficiency reached 77.8% (DC, LHV). The requirement of heat loss was clarified by the heat balance analysis.
a r t i c l e
i n f o
Article history: Received 4 April 2016 Revised 11 October 2016 Accepted 14 October 2016 Available online xxxx Keywords: Solid oxide fuel cell Efficiency Multi-stage Regeneration Hot module
a b s t r a c t Enhancing the power generation efficiency, which is the main advantage of solid oxide fuel cells (SOFCs), can have valuable benefits to introduce SOFC systems into business and industrial markets of Japan, where power demands are higher than thermal demands. In this study, we examined a high-efficiency SOFC system with an off-gas regenerating technique. A two-stage SOFC stack configuration was employed. For off-gas regeneration, we used a CO2 absorber and a H2O condenser. A total of 92.0% of the fuel was successfully used with an electrical efficiency of 77.8% (DC, LHV). However, there existed the heat loss from the fuel cell system due to the thermal insulation performance. To compensate the heat loss, additional electric heaters were used to keep temperatures high, therefore heat sustainability remained an issue. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction The most important feature of a fuel cell is that it can directly convert chemical energy from fuels into electricity and achieve a high energy-conversion efficiency compared to existing thermal power generation systems. Among all fuel cell types, solid oxide fuel cells (SOFCs) are expected to have the highest power generation efficiency, which is consistent with the results of various research and development projects from industrial, government, and academic research groups. A micro combined heat and power SOFC system with rated power output of 700 W [1] and power generation efficiency of 45.0% LHV [1] (lower heating value) for residential use was introduced to the Japanese market in 2011, and its cumulated total number of installed systems has been increasing steadily. In contrast, in business and industrial areas, main activities of SOFC development still remains in the experimental or empirical research stage. In June 2014, the Ministry of Economy, ⇑ Corresponding author. E-mail address: [email protected] (T. Somekawa).
Trade and Industry of Japan formalized a ‘‘Strategic road map for hydrogen and fuel cells”, laying out objectives for the large-scale introduction of SOFCs with comparatively high electric power generation efficiency to the market in 2017 [2]. In the industrial regions of Japan, nearly six tenths of the market is comprised of the markets where power demands exceed thermal demands such as office buildings, convenience stores, and schools [3]. Therefore, to accelerate the introduction of SOFCs to this market, the development of high-efficiency power generation technology is very important. The power generation efficiency in a SOFC system, g, is generally given as follows:
g¼
nF V gFuel gaux gPC DH
ð1Þ
where n is the number of electrons that contribute to the reaction per mole of fuel, F is the Faraday constant, DH is the calorific power of the fuel, V is the operating voltage of the SOFC cell, gFuel is the fuel utilization ratio, gaux is the efficiency of the auxiliary machines, and gPC is the efficiency of the power conditioner. To improve the effi-
http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
2
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
ciency of a SOFC system, at least one these elements must be improved. It has been reported that fuel utilization ratio can be significantly increased compared to that of conventional SOFC systems shown in Fig. 1 by devising a system structure [4–7]. To create highefficiency power generation systems, developments not only for increasing efficiency but also for minimizing the heat radiation loss from the system are needed [8,9]. To study the feasibility of SOFC systems with higher power generation efficiencies, we analysed the heat balance of the SOFC system with regenerating devices of the anode off-gas. And we designed and manufactured the SOFC system. In this paper we report the investigation results of the manufactured system such as fuel utilization ratio, the electrical performance, and generation stability at continuous operation tests. Fig. 1. Composition of a conventional SOFC system.
2. SOFC systems with regenerating off-gas techniques 2.1. Conventional SOFC system At the anode of a SOFC, the following electrochemical reactions occur:
H2 þ O2 ! H2 O þ 2e
ð2Þ
CO þ O2 ! CO2 þ 2e
ð3Þ
These reactions produce H2O and CO2, the concentrations of which increase with the fuel utilization ratio during operation, which decrease the partial pressures of the fuels and prevent electrode reactions from progressing. When the fuel utilization ratio is high, the local partial pressure of oxygen at the anode can increase, which may cause a re-oxidation of the anode material, Ni, to NiO, which involves irreversible plastic deformation and causes a destruction of the cell. Therefore, in conventional SOFC systems, approximately 20–30% of the supplied fuel is generally not used for power generation but is burned for use as heat to maintain the cell temperature or the steam reforming reaction temperature. The remaining heat is emitted as exhaust gas. More specifically, when the CO2 and H2O molecules that disturb the electrode reactions are removed and the fuel concentrations increase, the fuels in the off-gas that are not used to supply heat can be reused as a fuel for power generation. This recycling enhances the total fuel utilization ratio and consequently power generation efficiency. To realize these concepts, fuel recycling SOFC systems and multi-stage SOFCs have been reported [4,5].
Heat Exchanger Fuel (CH4,13A,etc) Vaporizer
Reformer
Blower
Regenerator Anode
H 2O (gas)
H2O
Fuel + H 2O SOFC
Air Cathode Fig. 2. Composition of a fuel recycling SOFC system.
2.2. Fuel recycling SOFC system Fig. 2 shows a schematic of a fuel recycling SOFC system. In this figure, 1red lines represent fuel arrows, blue arrows represent water flows, green arrows represent air flows, purple arrows represent the flows of mixtures containing both fuels and water vapor, and black arrows represent off-gas flows. These colors are applicable in the same way for subsequent figures. The H2O and CO2 gases in the anode off-gas are removed by regenerators, such as a separation membrane and a CO2 absorption agent (the regenerator in Fig. 2), or H2 and CO gasses are selectively extracted. These regenerated gases are recirculated into the anode by a gas blower and are reused. This reuse of fuel can increase the total fuel utilization ratio. When the H2O needed for steam reformation is recirculated to the steam reformer, or a partial oxidation (POx) is employed to reform the fuel, as shown in Fig. 3, the system can be simplified by removing the water supply pumps. However, these fuel recycling systems need 1
For interpretation of color in Figs. 2 and 5, the reader is referred to the web version of this article.
Fig. 3. Composition of a fuel recycling SOFC system with a POx starter.
gas blowers to circulate the combustible gas that is ejected from the anode of the SOFC cells at high temperatures. Extensive peripheral equipment is needed for blower installations intended for hightemperature usage due to safety reasons. Therefore, the issues related to the use of fuel recycling systems are the complications imposed by the system structure and the durability of the blowers.
2.3. Multi-stage SOFC system Fig. 4 shows a two-stage stack configuration of a SOFC system. The SOFC stacks are arranged along the fuel flow, and the residual fuels that are not used at the 1st SOFC stack can be used at the 2nd
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
3
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Off-gas Heat Exchanger
Anode
Fuel (CH4,13A,, etc) Vaporizer H2O
Combustor
Reformer Anode R Regenerator t H2O (gas)
Fuel + H2O SOFC
SOFC
Cathode
Cathode
Air
Fig. 4. Composition of a multi-stage SOFC system.
Fig. 5. Composition of a multi-stage SOFC system with an off-gas regenerating techniques.
stack. In contrast, because the fuel that reaches the 2nd stack is enriched in H2O and CO2, which are produced by electrochemical reactions at the 1st stack, the partial pressure of oxygen at the 2nd stack is greater than at the 1st one. Thus, operations with lower fuel utilization ratios, lower current densities, or other appropriate conditions are necessary. However, when regenerators are installed for removing H2O and CO2, such as a separation membrane, absorbing agents or selective extraction devices for H2 and CO between the 1st and 2nd stacks (the regenerator in Fig. 4), the decrease in the partial pressure of oxygen at the 2nd stack can be achieved. As a result, the operating conditions of the 2nd stack can be close to those of the 1st one without increasing the risk of destroying the cell at the 2nd stack. For example, when the power ratio of the 1st and 2nd stacks is assumed 1:1 and the fuel utilization ratios 70.0% each, the 1st stack uses 70.0% of the fuel, while the 2nd stack uses 21.0% of the fuel. Thus, the total fuel utilization ratio becomes 91.0%, which is much higher than conventional SOFC systems. These combinations of the power ratio and fuel utilizations depend on each system. In this study, we focused on multi-stage SOFC systems because they do not require blowers for high-temperature use and their feasibility has already been demonstrated by Ceramic Fuel Cells Ltd [10–12]. To realize higher electrical power efficiency, we incorporate the fuel regenerating techniques with the multi-stage SOFC systems and examined a two-stage SOFC system with CO2 and H2O removal functions.
3. Experimental 3.1. SOFC stack performance test To demonstrate the feasibility of the multi-stage stack concept described in Section 2.2, i.e., ‘‘multi-stage SOFC systems” in a highefficiency system, it is important to know if the 2nd stack performs equivalently to the 1st stack. Therefore the performances of the 2nd stack under different conditions was examined. I-V performance tests were conducted under the following conditions. The electric furnace was set to 750 °C and the standard gas composition was set to its equilibrium composition (i.e., 20.0% H2, 65.0% H2O, and 15.0% CO2) at 750 °C after the 1st stack utilized 60.0% of the fuel. The CO2 removal device such as a CO2 absorber was assumed to pass a gas composed of 25.0% H2, 70.0% H2O, and 5.0% CO2; the H2O removal device such as a water condenser was assumed to pass a gas composed of 65.0% H2, 15.0% H2O, and 20.0% CO2. These three simulant gases were supplied to 10 planar SOFCs and their electrical performances were measured. 3.2. Hot module performance test Fig. 5 shows a schematic of the multi-stage SOFC system with the off-gas regeneration functions designed specifically in this experiment. In the system, a two-stage SOFC stack configuration was used. The CO2 absorber, which was rated to absorb CO2 for
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
4
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Efficiency䠖65% Power䠖10kW
Reformer 23.24 kW Heat loss 1.97 kW
Power 10.00 kW
Methane 15.38 kW
Cell Cell exotherm stack 10.92 kW
Air
Exhaust gas 3.41 kW
Exhaust 20.61 kW
7.37 kW
Off-gas 2.32 kW
Air prewarming 7.37 kW Water vaporizaon䞉water prewarming3.36 kW Methane prewarming䞉reforming 4.50 kW Fig. 6. Heat balance analysis of a hot module in the system.
6 h, was positioned between the 1st and 2nd stacks. The anodeside off-gas from the 1st stack passed through the heat exchanger with the cathode-side air before passing through the CO2 absorber. To remove H2O, the water condenser was positioned between the CO2 absorber and the 2nd stack; H2O was removed after removing CO2. Using these two processes, anode off-gas from the 1st stack was regenerated. Fig. 6 shows the heat balance of a hot module in the system for an output power of 10 kW assuming the power generation efficiency of 65.0% (LHV). The heat balance calculation was carried out by using VMGsim (Virtual Materials Group Inc.). In the hot module, heat must be reused for pre-heating air, vaporizing water, pre-heating water vapor and methane, and reforming the endotherm of the steam to use as much fuel as possible. For these purposes, 15.23 kW of the 20.61 kW of exhaust heat must be circulated in the hot module to maintain its temperature. Therefore it was necessary to keep the heat dissipation below 1.97 kW. The water vaporizer, integrated off-gas combustor and steam reformer, CO2 absorber, off-gas and air heat exchanger (Hex1 in Fig. 5), and heat exchanger for the off-gas ejected from the 1st stack and off-gas (Hex2 in Fig. 5) were designed and fabricated, and two SOFC stacks were incorporated into the system. A hot module was configured by surrounding hot elements by an insulator, shown by the orange dotted line in Fig. 5, and the system performance was measured. In the hot module, heat loss was compensated by supplementing the heat with electric heaters shown in Fig. 5 as yellow lines, allowing heat sustainability to be ignored. The heat loss was monitored by a heat flux meter (TR2C: Kyoto Electronics Manufacturing Co., Ltd.) on the surface of the insulator. In the performance test, the output power ratio of the 1st and 2nd stacks was set 2:1, and the fuel utilization ratio of the 1st stack was 60.0%, while that of 2nd stack was 75.0% to realize a total fuel utilization ratio exceeding 90.0%. CH4 was supplied as a fuel in this test with steam to carbon (S/C) ratio set 2.5. At the reformer, CH4 was reformed into H2 by the reactions as follow;
CH4 þ 2H2 O ! 4H2 þ CO2 ;
ð4Þ
CH4 þ H2 O ! 3H2 þ CO:
ð5Þ
After the steam reforming reactions, the reformed gas was supplied to the 1st stack and the fuel for the 1st stack includes a large amount of water vapor; in other words, the H2 concentration in the
Table 1 Overview of the multi-stage SOFC system with an off-gas regeneration techniques. Input
Fuel S/C
CH4 2.5
SOFC
Uf Uair Output ratio
(1st stack) 65% (2nd stack) 75% (1st stack) 50% (2nd stack) 50% 1st stack:2nd stack = 2:1
Off-gas regenerating
CO2 removal H2O removal
Li4SiO4 H2O condenser
fuel for the 1st stack was attenuated by the water vapor. In terms of the fuel for the 2nd stack, the H2O and CO2 were removed from the anode off-gas at the regenerator before the fuel was supplied to the 2nd stack. This means that H2 concentration in the fuel for the 2nd stack is high, which results in higher fuel utilization of 75%. Therefore, the fuel utilization of the 1st stack was designed as 60% which was lower than the fuel utilization of the 2nd stack. To check the gas composition, the sampling site was set after the H2O condenser as shown in Fig. 5. The collected gas was monitored by a gas analyzer (PG-344: HORIBA Ltd.) which use non-dispersive infrared absorption method. Table 1 shows the overview of the experimental setup pf the multi-stage SOFC system with an offgas regeneration devices. 4. Results and discussion 4.1. SOFC stack performance test Fig. 7 shows the I-V performance measurement results of stack using a simulant fuel. The open circuit voltage increased by 9.0 mV and 72.0 mV after CO2 and H2O removal, respectively. These improvements correspond to the calculated results at equilibrium. In the I-V performance tests, by comparing with the standard condition in Fig. 7, electrical efficiency improvements of 1.4 and 4.0 points were measured due to CO2 and H2O removal, respectively. These results indicate that a combination of a multi-stage SOFC stack and the removal of both CO2 and H2O can enhance the power generation efficiency. Some groups have reported that they operated their system for several thousand hours at fuel utilization of 70% and compared to the reported fuel utilization condition such as 70% [13,14], the stack was operated under low fuel utilization
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
5
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
10
9
Stack voltage /V
8
7
6
5
4
Fig. 9. The DC efficiency and total fuel utilization ratio dependencies of time. Fig. 7. Results of the I-V performance test of a SOFC stack using simulant fuel.
Fig. 8. The total DC efficiency dependence of the total fuel utilization ratio in the 1st and 2nd stacks.
condition to neglect concentration overpotentials in this test. The fuel utilization was set as less than 60%. In the experimental condition, the fuel concentration was thought to be high enough, and the effects of H2O and CO2 generated under high current density condition were insignificant. Therefore, the I-V curves showed a similar trend even at the higher current densities among different gas compositions.
ciency calculated from the stack-end voltage on the total fuel utilization ratio in the 1st and 2nd stacks is shown. The maximum efficiency was found to be 77.8% (DC, LHV) at fuel utilization ratio of 92.0%. When both of the efficiencies of the auxiliary machines and the power conditioner are supposed to be 95.0%, the AC power generation efficiency of the unit is calculated to be 70.2% (LHV). Table 2 shows the values of different parameters at the maximum efficiency point for the 1st and 2nd stack. These results successfully demonstrate that regenerating anode off-gas using CO2 absorption and H2O condensation can enhance the total fuel utilization ratio and increase the efficiency in a hot module. We also performed a continuous operation test to confirm the long-term power generation stability of the system. In the durability test, the total fuel utilization ratio was set to 87.0%, which is 5.0 points less than the maximum fuel utilization ratio confirmed in this study. Results of time dependencies of DC efficiency and total fuel utilization ratio are shown in Fig. 9. The measurement conditions were the same as for the performance test described above. In the figure, at approximately 0.5 h, CO2 absorption was started and the I-V performance was measured. Then, at approximately 1.0 h, the durability test was initiated and continued for approximately 40 min. During the test, the fuel utilization ratio remained nearly constant at 74.0% (LHV), which indicates that the system has its feasibility. In terms of the durability of the regenerators, the water condenser as a H2O remover was a matured technology and considered to have sustained durability. In the CO2 absorber, Li4SiO4 was used as a CO2 absorption agent. In terms of the cyclic durability of the Li4SiO4, it has been reported [15] that Li4SiO4 could be repeated about 175 times of absorption and desorption of CO2. This indicates that CO2 absorber has a potential for the practical use.
4.2. Hot module performance test 5. Conclusion Fig. 8 shows the power generation efficiency performance test results of the hot module. The total oxygen utilization was 50.0% in both-stage stacks. In Fig. 8, the dependence of the total DC effi-
We examined a high-efficiency SOFC system that uses off-gas regenerating techniques. In this study, CO2 and H2O were removed
Table 2 Values of different parameters at the maximum efficiency point for the 1st and 2nd stack.
1st stack 2nd stack
Cell voltage (mV)
Current density (A cm2)
Output density (mW cm2)
Output (W)
Fuel utilization ratio (%)
889 868
0.286 0.229
254 199
1026 402
65.7 76.9
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
6
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
using a CO2 absorber and a H2O condenser, respectively. The heat balance analysis of a hot module inside the system shows that suppressing heat losses of less than 1.97 kW is required to maintain the necessary temperature when the electrical power output and efficiency are supposed to be 10 kW and 65% (LHV), respectively. In the hot module test, we showed that the total fuel utilization ratio was successfully enhanced to 92.0%, and the maximum power generation efficiency reached 77.8% (DC, LHV). However, in this hot module test, the lost heat was compensated by electric heaters, and the issue of the system heat sustainability remains a problem. The continuous operation test showed that the system was stable for 40 min. These results indicate that a multi-stage SOFC system with regenerating anode off-gas techniques can enhance total fuel utilization ration and power generation efficiency and has its feasibility.
[4] [5] [6]
[7]
[8]
[9]
[10] [11]
[12]
References [1] K. Horiuchi, Development and application of SOFC technology in Japan, in: 11th European SOFC & SOE Forum, Lucerne/Switzerland, 2014, pp. 5–13. [2] Agency for Natural Resources and Energy, Strategic Road Map for Hydrogen and Fuel Cell,, 28 May 2015. [3] K. Fujita, T. Kushi, Y. Ikeda, T. Kume, K. Nakamura, H. Yakabe, Study of highefficiency power generation SOFC system with regenerative technology from
[13] [14] [15]
exhaust gas, in: Proceedings of 23rd Symposium on Solid Oxide Fuel Cells in Japan, Tokyo/Japan, 2014, pp. 230–235. L.J. Frost, R.M. Privette, A.C. Khandkar, Progress in the planar CPn SOFC system design, J. Power Sources 61 (1996) 135–139. S. Elangovan, J. Hartvigsen, A. Khandkar, Efficiency enhancement in fuel cells: multistage oxidation concept, J. Electrochem. Soc. 144 (1997) 3337–3342. M. Powerll, K. Meinhardt, V. Sprenkle, L. Chick, G. Mcvay, Demonstration of a highly efficient solid oxide fuel cell power system using adiabatic steam reforming and anode gas recirculation, J. Power Sources 205 (2012) 377–384. R. Peters, R. Deja, L. Blum, J. Pennanen, J. Kiviaho, T. Hakala, Analysis of solid oxide fuel cell system concepts with anode recycling, Int. J. Hydrogen Energy 38 (2013) 6809–6820. T.H. Lim, J.L. Park, S.B. Lee, S.J. Park, R.H. Song, D.R. Shin, Fabrication and operation of a 1 kW class anode-supported flat tubular SOFC stack, Int. J. Hydrogen Energy 35 (2010) 9687–9692. N. Watanabe, T. Ooe, T. Ishihara, Design of thermal self supported 700 W class, solid oxide fuel cell module using, LSGM thin film micro tubular cells, J. Power Sources 199 (2012) 117–123. R. Payne, CFCL’s BlueGen Product, Abstracts of 219th ECS Meeting, #651, 2011. J. Love, S. Amarasinghe, D. Selvey, X. Zheng, L. Christiansen, Development of SOFC Stacks at Ceramic Fuel Cells Limited, Solid Oxide Fuel Cells 11 (Sofc-Xi) 25 (2) (2009) p. 115. R. Payne, J. Love, M. Kah, Generating Electricity at 60% Electrical Efficiency from 1–2 kWe SOFC Products, Solid Oxide Fuel Cells 11 (Sofc-Xi) 25 (2) (2009) p. 231. M. Suzuki, T. Sogi, K. Higaki, T. Ono, N. Takahashi, K. Shimazu, T. Shigehisa, ECS Trans. 7 (1) (2007) 27–30. R. Peters, L. Blum, R. Deja, I. Hoven, W. Tiedemann, S. Küpper, D. Stolten, Fuel Cells 14 (3) (2014) 489–499. Y. Ikeda, T. Iseki, Highly efficient solid oxide fuel cell system using a CO2 absorbent, Abstracts of Fuel Cell Seminar 2014, Los Angeles, CA, November 10– 13, 2014.
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
Contents lists available at ScienceDirect
Applied Thermal Engineering journal homepage: www.elsevier.com/locate/apthermeng
Research Paper
Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas Takaaki Somekawa ⇑, Kazuo Nakamura, Takuto Kushi, Takao Kume, Kenjiro Fujita, Hisataka Yakabe Fundamental Technology Dept., Tokyo Gas Co., Ltd., 1-7-7 Suehiro-cho, Tsurumi-ku, Yokohama City, Kanagawa 230-0045, Japan
h i g h l i g h t s 2-stage SOFC system with fuel regenerating techniques was designed and fabricated. Total fuel utilization ratio was successfully enhanced to 92.0% in hot module tests. The maximum power generation efficiency reached 77.8% (DC, LHV). The requirement of heat loss was clarified by the heat balance analysis.
a r t i c l e
i n f o
Article history: Received 4 April 2016 Revised 11 October 2016 Accepted 14 October 2016 Available online xxxx Keywords: Solid oxide fuel cell Efficiency Multi-stage Regeneration Hot module
a b s t r a c t Enhancing the power generation efficiency, which is the main advantage of solid oxide fuel cells (SOFCs), can have valuable benefits to introduce SOFC systems into business and industrial markets of Japan, where power demands are higher than thermal demands. In this study, we examined a high-efficiency SOFC system with an off-gas regenerating technique. A two-stage SOFC stack configuration was employed. For off-gas regeneration, we used a CO2 absorber and a H2O condenser. A total of 92.0% of the fuel was successfully used with an electrical efficiency of 77.8% (DC, LHV). However, there existed the heat loss from the fuel cell system due to the thermal insulation performance. To compensate the heat loss, additional electric heaters were used to keep temperatures high, therefore heat sustainability remained an issue. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction The most important feature of a fuel cell is that it can directly convert chemical energy from fuels into electricity and achieve a high energy-conversion efficiency compared to existing thermal power generation systems. Among all fuel cell types, solid oxide fuel cells (SOFCs) are expected to have the highest power generation efficiency, which is consistent with the results of various research and development projects from industrial, government, and academic research groups. A micro combined heat and power SOFC system with rated power output of 700 W [1] and power generation efficiency of 45.0% LHV [1] (lower heating value) for residential use was introduced to the Japanese market in 2011, and its cumulated total number of installed systems has been increasing steadily. In contrast, in business and industrial areas, main activities of SOFC development still remains in the experimental or empirical research stage. In June 2014, the Ministry of Economy, ⇑ Corresponding author. E-mail address: [email protected] (T. Somekawa).
Trade and Industry of Japan formalized a ‘‘Strategic road map for hydrogen and fuel cells”, laying out objectives for the large-scale introduction of SOFCs with comparatively high electric power generation efficiency to the market in 2017 [2]. In the industrial regions of Japan, nearly six tenths of the market is comprised of the markets where power demands exceed thermal demands such as office buildings, convenience stores, and schools [3]. Therefore, to accelerate the introduction of SOFCs to this market, the development of high-efficiency power generation technology is very important. The power generation efficiency in a SOFC system, g, is generally given as follows:
g¼
nF V gFuel gaux gPC DH
ð1Þ
where n is the number of electrons that contribute to the reaction per mole of fuel, F is the Faraday constant, DH is the calorific power of the fuel, V is the operating voltage of the SOFC cell, gFuel is the fuel utilization ratio, gaux is the efficiency of the auxiliary machines, and gPC is the efficiency of the power conditioner. To improve the effi-
http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096 1359-4311/Ó 2016 Elsevier Ltd. All rights reserved.
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
2
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
ciency of a SOFC system, at least one these elements must be improved. It has been reported that fuel utilization ratio can be significantly increased compared to that of conventional SOFC systems shown in Fig. 1 by devising a system structure [4–7]. To create highefficiency power generation systems, developments not only for increasing efficiency but also for minimizing the heat radiation loss from the system are needed [8,9]. To study the feasibility of SOFC systems with higher power generation efficiencies, we analysed the heat balance of the SOFC system with regenerating devices of the anode off-gas. And we designed and manufactured the SOFC system. In this paper we report the investigation results of the manufactured system such as fuel utilization ratio, the electrical performance, and generation stability at continuous operation tests. Fig. 1. Composition of a conventional SOFC system.
2. SOFC systems with regenerating off-gas techniques 2.1. Conventional SOFC system At the anode of a SOFC, the following electrochemical reactions occur:
H2 þ O2 ! H2 O þ 2e
ð2Þ
CO þ O2 ! CO2 þ 2e
ð3Þ
These reactions produce H2O and CO2, the concentrations of which increase with the fuel utilization ratio during operation, which decrease the partial pressures of the fuels and prevent electrode reactions from progressing. When the fuel utilization ratio is high, the local partial pressure of oxygen at the anode can increase, which may cause a re-oxidation of the anode material, Ni, to NiO, which involves irreversible plastic deformation and causes a destruction of the cell. Therefore, in conventional SOFC systems, approximately 20–30% of the supplied fuel is generally not used for power generation but is burned for use as heat to maintain the cell temperature or the steam reforming reaction temperature. The remaining heat is emitted as exhaust gas. More specifically, when the CO2 and H2O molecules that disturb the electrode reactions are removed and the fuel concentrations increase, the fuels in the off-gas that are not used to supply heat can be reused as a fuel for power generation. This recycling enhances the total fuel utilization ratio and consequently power generation efficiency. To realize these concepts, fuel recycling SOFC systems and multi-stage SOFCs have been reported [4,5].
Heat Exchanger Fuel (CH4,13A,etc) Vaporizer
Reformer
Blower
Regenerator Anode
H 2O (gas)
H2O
Fuel + H 2O SOFC
Air Cathode Fig. 2. Composition of a fuel recycling SOFC system.
2.2. Fuel recycling SOFC system Fig. 2 shows a schematic of a fuel recycling SOFC system. In this figure, 1red lines represent fuel arrows, blue arrows represent water flows, green arrows represent air flows, purple arrows represent the flows of mixtures containing both fuels and water vapor, and black arrows represent off-gas flows. These colors are applicable in the same way for subsequent figures. The H2O and CO2 gases in the anode off-gas are removed by regenerators, such as a separation membrane and a CO2 absorption agent (the regenerator in Fig. 2), or H2 and CO gasses are selectively extracted. These regenerated gases are recirculated into the anode by a gas blower and are reused. This reuse of fuel can increase the total fuel utilization ratio. When the H2O needed for steam reformation is recirculated to the steam reformer, or a partial oxidation (POx) is employed to reform the fuel, as shown in Fig. 3, the system can be simplified by removing the water supply pumps. However, these fuel recycling systems need 1
For interpretation of color in Figs. 2 and 5, the reader is referred to the web version of this article.
Fig. 3. Composition of a fuel recycling SOFC system with a POx starter.
gas blowers to circulate the combustible gas that is ejected from the anode of the SOFC cells at high temperatures. Extensive peripheral equipment is needed for blower installations intended for hightemperature usage due to safety reasons. Therefore, the issues related to the use of fuel recycling systems are the complications imposed by the system structure and the durability of the blowers.
2.3. Multi-stage SOFC system Fig. 4 shows a two-stage stack configuration of a SOFC system. The SOFC stacks are arranged along the fuel flow, and the residual fuels that are not used at the 1st SOFC stack can be used at the 2nd
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
3
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Off-gas Heat Exchanger
Anode
Fuel (CH4,13A,, etc) Vaporizer H2O
Combustor
Reformer Anode R Regenerator t H2O (gas)
Fuel + H2O SOFC
SOFC
Cathode
Cathode
Air
Fig. 4. Composition of a multi-stage SOFC system.
Fig. 5. Composition of a multi-stage SOFC system with an off-gas regenerating techniques.
stack. In contrast, because the fuel that reaches the 2nd stack is enriched in H2O and CO2, which are produced by electrochemical reactions at the 1st stack, the partial pressure of oxygen at the 2nd stack is greater than at the 1st one. Thus, operations with lower fuel utilization ratios, lower current densities, or other appropriate conditions are necessary. However, when regenerators are installed for removing H2O and CO2, such as a separation membrane, absorbing agents or selective extraction devices for H2 and CO between the 1st and 2nd stacks (the regenerator in Fig. 4), the decrease in the partial pressure of oxygen at the 2nd stack can be achieved. As a result, the operating conditions of the 2nd stack can be close to those of the 1st one without increasing the risk of destroying the cell at the 2nd stack. For example, when the power ratio of the 1st and 2nd stacks is assumed 1:1 and the fuel utilization ratios 70.0% each, the 1st stack uses 70.0% of the fuel, while the 2nd stack uses 21.0% of the fuel. Thus, the total fuel utilization ratio becomes 91.0%, which is much higher than conventional SOFC systems. These combinations of the power ratio and fuel utilizations depend on each system. In this study, we focused on multi-stage SOFC systems because they do not require blowers for high-temperature use and their feasibility has already been demonstrated by Ceramic Fuel Cells Ltd [10–12]. To realize higher electrical power efficiency, we incorporate the fuel regenerating techniques with the multi-stage SOFC systems and examined a two-stage SOFC system with CO2 and H2O removal functions.
3. Experimental 3.1. SOFC stack performance test To demonstrate the feasibility of the multi-stage stack concept described in Section 2.2, i.e., ‘‘multi-stage SOFC systems” in a highefficiency system, it is important to know if the 2nd stack performs equivalently to the 1st stack. Therefore the performances of the 2nd stack under different conditions was examined. I-V performance tests were conducted under the following conditions. The electric furnace was set to 750 °C and the standard gas composition was set to its equilibrium composition (i.e., 20.0% H2, 65.0% H2O, and 15.0% CO2) at 750 °C after the 1st stack utilized 60.0% of the fuel. The CO2 removal device such as a CO2 absorber was assumed to pass a gas composed of 25.0% H2, 70.0% H2O, and 5.0% CO2; the H2O removal device such as a water condenser was assumed to pass a gas composed of 65.0% H2, 15.0% H2O, and 20.0% CO2. These three simulant gases were supplied to 10 planar SOFCs and their electrical performances were measured. 3.2. Hot module performance test Fig. 5 shows a schematic of the multi-stage SOFC system with the off-gas regeneration functions designed specifically in this experiment. In the system, a two-stage SOFC stack configuration was used. The CO2 absorber, which was rated to absorb CO2 for
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
4
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
Efficiency䠖65% Power䠖10kW
Reformer 23.24 kW Heat loss 1.97 kW
Power 10.00 kW
Methane 15.38 kW
Cell Cell exotherm stack 10.92 kW
Air
Exhaust gas 3.41 kW
Exhaust 20.61 kW
7.37 kW
Off-gas 2.32 kW
Air prewarming 7.37 kW Water vaporizaon䞉water prewarming3.36 kW Methane prewarming䞉reforming 4.50 kW Fig. 6. Heat balance analysis of a hot module in the system.
6 h, was positioned between the 1st and 2nd stacks. The anodeside off-gas from the 1st stack passed through the heat exchanger with the cathode-side air before passing through the CO2 absorber. To remove H2O, the water condenser was positioned between the CO2 absorber and the 2nd stack; H2O was removed after removing CO2. Using these two processes, anode off-gas from the 1st stack was regenerated. Fig. 6 shows the heat balance of a hot module in the system for an output power of 10 kW assuming the power generation efficiency of 65.0% (LHV). The heat balance calculation was carried out by using VMGsim (Virtual Materials Group Inc.). In the hot module, heat must be reused for pre-heating air, vaporizing water, pre-heating water vapor and methane, and reforming the endotherm of the steam to use as much fuel as possible. For these purposes, 15.23 kW of the 20.61 kW of exhaust heat must be circulated in the hot module to maintain its temperature. Therefore it was necessary to keep the heat dissipation below 1.97 kW. The water vaporizer, integrated off-gas combustor and steam reformer, CO2 absorber, off-gas and air heat exchanger (Hex1 in Fig. 5), and heat exchanger for the off-gas ejected from the 1st stack and off-gas (Hex2 in Fig. 5) were designed and fabricated, and two SOFC stacks were incorporated into the system. A hot module was configured by surrounding hot elements by an insulator, shown by the orange dotted line in Fig. 5, and the system performance was measured. In the hot module, heat loss was compensated by supplementing the heat with electric heaters shown in Fig. 5 as yellow lines, allowing heat sustainability to be ignored. The heat loss was monitored by a heat flux meter (TR2C: Kyoto Electronics Manufacturing Co., Ltd.) on the surface of the insulator. In the performance test, the output power ratio of the 1st and 2nd stacks was set 2:1, and the fuel utilization ratio of the 1st stack was 60.0%, while that of 2nd stack was 75.0% to realize a total fuel utilization ratio exceeding 90.0%. CH4 was supplied as a fuel in this test with steam to carbon (S/C) ratio set 2.5. At the reformer, CH4 was reformed into H2 by the reactions as follow;
CH4 þ 2H2 O ! 4H2 þ CO2 ;
ð4Þ
CH4 þ H2 O ! 3H2 þ CO:
ð5Þ
After the steam reforming reactions, the reformed gas was supplied to the 1st stack and the fuel for the 1st stack includes a large amount of water vapor; in other words, the H2 concentration in the
Table 1 Overview of the multi-stage SOFC system with an off-gas regeneration techniques. Input
Fuel S/C
CH4 2.5
SOFC
Uf Uair Output ratio
(1st stack) 65% (2nd stack) 75% (1st stack) 50% (2nd stack) 50% 1st stack:2nd stack = 2:1
Off-gas regenerating
CO2 removal H2O removal
Li4SiO4 H2O condenser
fuel for the 1st stack was attenuated by the water vapor. In terms of the fuel for the 2nd stack, the H2O and CO2 were removed from the anode off-gas at the regenerator before the fuel was supplied to the 2nd stack. This means that H2 concentration in the fuel for the 2nd stack is high, which results in higher fuel utilization of 75%. Therefore, the fuel utilization of the 1st stack was designed as 60% which was lower than the fuel utilization of the 2nd stack. To check the gas composition, the sampling site was set after the H2O condenser as shown in Fig. 5. The collected gas was monitored by a gas analyzer (PG-344: HORIBA Ltd.) which use non-dispersive infrared absorption method. Table 1 shows the overview of the experimental setup pf the multi-stage SOFC system with an offgas regeneration devices. 4. Results and discussion 4.1. SOFC stack performance test Fig. 7 shows the I-V performance measurement results of stack using a simulant fuel. The open circuit voltage increased by 9.0 mV and 72.0 mV after CO2 and H2O removal, respectively. These improvements correspond to the calculated results at equilibrium. In the I-V performance tests, by comparing with the standard condition in Fig. 7, electrical efficiency improvements of 1.4 and 4.0 points were measured due to CO2 and H2O removal, respectively. These results indicate that a combination of a multi-stage SOFC stack and the removal of both CO2 and H2O can enhance the power generation efficiency. Some groups have reported that they operated their system for several thousand hours at fuel utilization of 70% and compared to the reported fuel utilization condition such as 70% [13,14], the stack was operated under low fuel utilization
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
5
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
10
9
Stack voltage /V
8
7
6
5
4
Fig. 9. The DC efficiency and total fuel utilization ratio dependencies of time. Fig. 7. Results of the I-V performance test of a SOFC stack using simulant fuel.
Fig. 8. The total DC efficiency dependence of the total fuel utilization ratio in the 1st and 2nd stacks.
condition to neglect concentration overpotentials in this test. The fuel utilization was set as less than 60%. In the experimental condition, the fuel concentration was thought to be high enough, and the effects of H2O and CO2 generated under high current density condition were insignificant. Therefore, the I-V curves showed a similar trend even at the higher current densities among different gas compositions.
ciency calculated from the stack-end voltage on the total fuel utilization ratio in the 1st and 2nd stacks is shown. The maximum efficiency was found to be 77.8% (DC, LHV) at fuel utilization ratio of 92.0%. When both of the efficiencies of the auxiliary machines and the power conditioner are supposed to be 95.0%, the AC power generation efficiency of the unit is calculated to be 70.2% (LHV). Table 2 shows the values of different parameters at the maximum efficiency point for the 1st and 2nd stack. These results successfully demonstrate that regenerating anode off-gas using CO2 absorption and H2O condensation can enhance the total fuel utilization ratio and increase the efficiency in a hot module. We also performed a continuous operation test to confirm the long-term power generation stability of the system. In the durability test, the total fuel utilization ratio was set to 87.0%, which is 5.0 points less than the maximum fuel utilization ratio confirmed in this study. Results of time dependencies of DC efficiency and total fuel utilization ratio are shown in Fig. 9. The measurement conditions were the same as for the performance test described above. In the figure, at approximately 0.5 h, CO2 absorption was started and the I-V performance was measured. Then, at approximately 1.0 h, the durability test was initiated and continued for approximately 40 min. During the test, the fuel utilization ratio remained nearly constant at 74.0% (LHV), which indicates that the system has its feasibility. In terms of the durability of the regenerators, the water condenser as a H2O remover was a matured technology and considered to have sustained durability. In the CO2 absorber, Li4SiO4 was used as a CO2 absorption agent. In terms of the cyclic durability of the Li4SiO4, it has been reported [15] that Li4SiO4 could be repeated about 175 times of absorption and desorption of CO2. This indicates that CO2 absorber has a potential for the practical use.
4.2. Hot module performance test 5. Conclusion Fig. 8 shows the power generation efficiency performance test results of the hot module. The total oxygen utilization was 50.0% in both-stage stacks. In Fig. 8, the dependence of the total DC effi-
We examined a high-efficiency SOFC system that uses off-gas regenerating techniques. In this study, CO2 and H2O were removed
Table 2 Values of different parameters at the maximum efficiency point for the 1st and 2nd stack.
1st stack 2nd stack
Cell voltage (mV)
Current density (A cm2)
Output density (mW cm2)
Output (W)
Fuel utilization ratio (%)
889 868
0.286 0.229
254 199
1026 402
65.7 76.9
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096
6
T. Somekawa et al. / Applied Thermal Engineering xxx (2016) xxx–xxx
using a CO2 absorber and a H2O condenser, respectively. The heat balance analysis of a hot module inside the system shows that suppressing heat losses of less than 1.97 kW is required to maintain the necessary temperature when the electrical power output and efficiency are supposed to be 10 kW and 65% (LHV), respectively. In the hot module test, we showed that the total fuel utilization ratio was successfully enhanced to 92.0%, and the maximum power generation efficiency reached 77.8% (DC, LHV). However, in this hot module test, the lost heat was compensated by electric heaters, and the issue of the system heat sustainability remains a problem. The continuous operation test showed that the system was stable for 40 min. These results indicate that a multi-stage SOFC system with regenerating anode off-gas techniques can enhance total fuel utilization ration and power generation efficiency and has its feasibility.
[4] [5] [6]
[7]
[8]
[9]
[10] [11]
[12]
References [1] K. Horiuchi, Development and application of SOFC technology in Japan, in: 11th European SOFC & SOE Forum, Lucerne/Switzerland, 2014, pp. 5–13. [2] Agency for Natural Resources and Energy, Strategic Road Map for Hydrogen and Fuel Cell,
[13] [14] [15]
exhaust gas, in: Proceedings of 23rd Symposium on Solid Oxide Fuel Cells in Japan, Tokyo/Japan, 2014, pp. 230–235. L.J. Frost, R.M. Privette, A.C. Khandkar, Progress in the planar CPn SOFC system design, J. Power Sources 61 (1996) 135–139. S. Elangovan, J. Hartvigsen, A. Khandkar, Efficiency enhancement in fuel cells: multistage oxidation concept, J. Electrochem. Soc. 144 (1997) 3337–3342. M. Powerll, K. Meinhardt, V. Sprenkle, L. Chick, G. Mcvay, Demonstration of a highly efficient solid oxide fuel cell power system using adiabatic steam reforming and anode gas recirculation, J. Power Sources 205 (2012) 377–384. R. Peters, R. Deja, L. Blum, J. Pennanen, J. Kiviaho, T. Hakala, Analysis of solid oxide fuel cell system concepts with anode recycling, Int. J. Hydrogen Energy 38 (2013) 6809–6820. T.H. Lim, J.L. Park, S.B. Lee, S.J. Park, R.H. Song, D.R. Shin, Fabrication and operation of a 1 kW class anode-supported flat tubular SOFC stack, Int. J. Hydrogen Energy 35 (2010) 9687–9692. N. Watanabe, T. Ooe, T. Ishihara, Design of thermal self supported 700 W class, solid oxide fuel cell module using, LSGM thin film micro tubular cells, J. Power Sources 199 (2012) 117–123. R. Payne, CFCL’s BlueGen Product, Abstracts of 219th ECS Meeting, #651, 2011. J. Love, S. Amarasinghe, D. Selvey, X. Zheng, L. Christiansen, Development of SOFC Stacks at Ceramic Fuel Cells Limited, Solid Oxide Fuel Cells 11 (Sofc-Xi) 25 (2) (2009) p. 115. R. Payne, J. Love, M. Kah, Generating Electricity at 60% Electrical Efficiency from 1–2 kWe SOFC Products, Solid Oxide Fuel Cells 11 (Sofc-Xi) 25 (2) (2009) p. 231. M. Suzuki, T. Sogi, K. Higaki, T. Ono, N. Takahashi, K. Shimazu, T. Shigehisa, ECS Trans. 7 (1) (2007) 27–30. R. Peters, L. Blum, R. Deja, I. Hoven, W. Tiedemann, S. Küpper, D. Stolten, Fuel Cells 14 (3) (2014) 489–499. Y. Ikeda, T. Iseki, Highly efficient solid oxide fuel cell system using a CO2 absorbent, Abstracts of Fuel Cell Seminar 2014, Los Angeles, CA, November 10– 13, 2014.
Please cite this article in press as: T. Somekawa et al., Examination of a high-efficiency solid oxide fuel cell system that reuses exhaust gas, Appl. Therm. Eng. (2016), http://dx.doi.org/10.1016/j.applthermaleng.2016.10.096