Performance test on a 5kW SOFC system under high fuel utility with practical syngas feeding

With increasing demand of green energy supply with high eciency and low CO 2 emission, Solid oxide fuel cell (SOFC) has been intensively developed in recent years. And the integration of gasication with fuel cell (IGFC) shows potential in large scale power generation to further increase the system eciency. Reliable design of multi-stacks for large system and long term stability of stacks with practical fuel gas from industrial equipment are the key for commercial application of IGFC. In this work, a test rig of 5kW SOFC system was fabricated using practical syngas from industrial gasiers as fuel and long term test under high fuel utility was conducted to investigate the system performance. The results show that the maximum steady output power of system is 5700W for hydrogen case and 5660W for syngas case, and the maximum steady electrical eciency is 61.24% while the fuel utility eciency is 89.25%. The test lasted for more than 500h as the fuel utility eciency was larger than 83%. The performances of each stack tower are almost identical at both initial stage and after long term operation. After 500h operation, the performances of stack towers just slight decrease under lower current and almost not change under higher current. Therefore, the results illustrate that the reliability of multi-stacks design and the prospect of SOFC power generation system for further enlarging its application in a MW th demonstration.


Introduction
With increasing demand of green energy supply with high e ciency and low CO 2 emission, solid oxide fuel cell (SOFC) is an attractive choice, thus get more attention of researchers and has been intensively developed in recent years [1]. Its advantages include variety of fuels, high electrical e ciency and full utility of heat, quiet operation and versatility in the electrolyte material [2]. It provides an high electrical e ciency of about 60% in normal operation and up to 90% in combined heat and power operation [3]. It also has the flexibility to be integrated with another power generation source, water heating device and cooling device used in residential homes [4]. To further its application for large scale, the integration of SOFC with gasi cation using coal or biomass as feedstock shows potential in power generation [5][6][7].
And it seems a feasible and prospective way for integration gas cation with fuel cell (IGFC).
Most of researches on SOFC focus on materials for the development of anode, electrolyte and cathode as well as performance test of cell [8,9]. And different cells have been developed for better chemical stability and electrochemical performance. As the performance of cells improving, SOFC stacks consisting of multi-layer cells should obtain more attention to widen the applications, while not much research work on them. Lim [11]. They also tested two stacks of different design under high utilization of up to 90% with humidi ed hydrogen or 10% pre-reformed LNG, and the results found that high fuel utilization could introduce polarization and a high risk of fuel starvation [12].The fuel utilization ratio is between 64%~80% and mostly about 70% while the feeding fuels of the anode side are H 2 or simulated reformate gas, and the stability test period is more than 5000h under 700℃. Edison R&D center built a 5kW SOFC system and conducted life test for 1500h at xed power output (1500W) over four start-up/shutdown cycles, and the system power was upt to 3000W during the rst test sessions. [13].
Researches also have been explored on combine SOFC with gas cation to widen the application in more elds for heat utilization besides electrical supply. Lim et al. constructed a pressurized 5kW class anodesupported planar SOFC power generation system with a pre-former for a fuel cell/gas turbine hybrid system [14]. The results show that the output of the SOFC stack was 4.7kW for the pre-performed gas while the output increased to 5.1kW at 3.5atm (abs.). Modelling of common hybrid con guration of the SOFC-Gas turbine system illustrated a signi cant e ciency upgrade. The combined SOFC-Gas turbine system produces an electric e ciency up to 50% and better syngas utilization compared to the implementation of each single technology [15]. Subotic et al. discussed the applicability of the SOFC technology for coupling with biomass-gasi er systems, using commercial SOFC single cells of industrial size fueled with different representative producer gas compositions of industrial relevance at two relevant operating temperatures [7]. The results show that feeding SOFC with a producer gas from a downdraft gasi er, with hot gas cleaning operating temperature of 750 °C represents the most favorable setting, considering system integration and the highest fuel utilization.
There are still few commercial SOFC stacks available in market. To build a demonstration of SOFC power generation system in MW class scale, hundreds of stacks should be well assembled and it is important to distribute the fuel gas for each stack in order to prevent fuel starvation. When coupled with gas cation to use practical produce gas as fuel, long term durability of stacks should be tested as most research using simulated produce gas. As there still lack of performance results of multi-stacks SOFC system using syngas as fuel from industrial gas ers fed with coal, we fabricated a test rig of 5kW SOFC system using practical syngas from industrial gasi er as fuel to explore the feasibility of a MW class IGFC demonstration equipment. As syngas used as fuel for the anode side, it has been reported that it might lead to fast degradation phenomena due to the impurities such as hydrogen sulphide or tars as well as coke deposition [16][17][18][19]. Thus an long term test using practical industrial syngas under high fuel utility was conducted to investigate the in uence of syngas on the performance of stacks in this work.

Test rig
As shown in Fig.1, a test rig of 5kW class was designed and implemented, which mainly composed of two parts, a fuel gas supply system and a hotbox shown in Fig.2. The fuel gas supply system could provide H 2 , syngas, N 2 and stream for the anode of stacks and air for the cathode. The hotbox consists of a heating furnace and 4 SOFC stacks assembled in parallel connection in the heating furnace. The nominal power of the commercial SOFC stack is designed about 1kW for syngas and H 2 as fuel feeding while it is 1.5kW for CH 4 case. The test rig was operated in the Synthesis Oil Plant in Ningmei Coal to Liquid (CTL) Company in Ningxia province, and all the fuel gases were provided by industrial equipment.
For the power generation control of test rig, the current of stacks were adjusted with two ITEC IT8904 electronic loads.
In the hotbox, four SOFC stacks were divided into two groups, i.e. two stack towers, and two stacks of each stack tower shared one gas distributor. For both anode and cathode side, two branches of fuel feeding gas connected with the inlet of each distributor. And the tail gases were also gathered into the outlet of the distributor and ew out of the heating furnace. The heating furnace had four heating walls and the temperature could be controlled by the heating power with the precision of ±1℃. The current and power of each stack tower was controlled with a electronic load.

Test methods
The schematic owchart of the experimental setup is shown in Fig.3. Hydrogen and syngas from the 4Mt/a CTL industrial equipment are used as fuel gas supplied to the anode side. Practical stream is added for syngas case and its molar ow rate is equal to CO in the syngas to prevent coke deposition. Nitrogen is used as an inert gas for the anode side. The air used as an oxidant for the cathode is also provided by the CTL plant. The ow rates of gases are controlled by mass ow controllers. The purity of hydrogen is 99.9% and the components of syngas are listed as Table 1. For syngas case, syngas has been desulphurized before it enters into stacks. As the stacks have been reduced before experimental, a mixture of nitrogen and hydrogen is used as shielding gas for the anode side while air is also used for the cathode side at the start up stage after the stacks were assembled. The inlet pressure of the shielding gas and air are regulated with pressure control valves, and the inlet temperature is controlled by pre-heaters. To save energy, the tail gas owing out of the stacks transfer heat to the feeding gas owing into the stacks by high temperature heat exchanger. Then the heating furnace is also controlled the heating power to keep the temperature increasing rate of stacks within 30℃/h. As the system temperature increase, the ow rate of cooling water of secondary heat exchanger is regulated to maintain the temperature of tail gas out of the system, and fans are applied to maintain the system pressure. When the temperature of stacks reaches the working temperature, it maintains 1h and the shielding gas shift to fuel gas. As the test goes, the current is increased and more heat would be produced. Thus the ow rate of air is also adapted to keep the system temperature stable. For the stack, most data is for methane case and no data for syngas case. Therefore, test of hydrogen was performed initially and then hydrogen was changed to syngas. In both cases, the furnace was operated at 770℃.

Performance veri cation of stack towers design
As the power generation system consists of two stack towers and each tower consist of two stacks, the feasibility of stack towers design was rstly identi ed. As shown in Fig.4, the results of I-V and I-P of multi-stacks illustrate that the average open circuit voltage (OCV) is 62.35V and the voltage decrease to 41.89V when the current increase to 36A. The system power reaches the design value i.e. 5kW when the current is 29A. The consistence between the two stack towers also could be characterized by the data of electronic load data of each stack tower, as shown in Fig.5. The voltage and power of each stack towers under the same current is almost the same all the test range. As shown in Fig.6, area speci c resistance (ASR) of the multi-stacks power generation system stabilize at about 0.35Ω·cm 2 as the current density increase up to 200mA/cm 2 . The maximum current density is about 350mA/cm 2 and the fuel utility is larger than 90%. Thus, the performance test results of stack towers identify the reliable design of stack towers.

Performance results of long term test
A long term performance test using hydrogen and syngas as fuel feeding was conducted. The experimental result of hydrogen as fuel feeding is shown in Fig.7. As the fuel owrate increases, and the output power increases from 3700W to 5700W. The operation data is very stable and few uctuations appear in the operation. The detailed operation conditions and results of cases would be discussed in following.
The experimental results of syngas as fuel feeding are shown in Fig.8. As the fuel owrate increases, the output power increases from 4500W to 5600W. The currents of different cases are very stable while there are uctuations appeared in the output power. The uctuations in the output power are caused by uctuations in the feeding stream owrate as the pressure of stream from the plant is not very stable. The detailed operation conditions and results of cases would be discussed in following.
The long term performance test lasted about 600h and the overview of the results is illustrated in Fig.9. Totally ten cases of different feeding have been conducted, and the operation conditions as well as the results of electrical e ciency E f and fuel utility e ciency U f are listed in Table 2. Electrical e ciency E f is calculated with input heat of fuels Q in and output electric power P out by E f = P out /Q in. And fuel utility e ciency U f is calculated by equation as following: , in which I is the current and F is Faraday constant as well as x is the fraction of fuel.
At rst, hydrogen is used as fuel feeding to start up the generation system at low fuel utility level. When the operation becomes stable, the fuel ow rate is increased and the current accordingly is increased to keep the fuel utility e ciency above 80%. Then the operation at high fuel utility level lasted for 500h. As for the syngas cases, the ratio of H 2 /CO was adjusted to investigate its in uence on the system performance. As shown in Table 2, the maximum electrical e ciency is up to 61.24% and the according fuel utility e ciency is 89.25% in steady operation with hydrogen feeding. For syngas case, the maximum electrical e ciency is up to 56.15% and the according fuel utility e ciency is 88.22%. The maximum output power in steady operation is 5700W for hydrogen feeding cases and it is 5660 W for syngas cases with the approximate feeding owrate, while the electrical e ciency of hydrogen case was about 2% larger than that of syngas case. No obvious in uence of H 2 /CO ratio has been found in the experiment as the operation time is not long for each case and should be further investigated. Performance comparison of stack tower1 of initial with after 500h operation is shown in Fig 10. The voltage and power of stack tower1 is slightly decreased at low current after 500h operation compared with that of initial stage, while the performance is almost the same as initial when the current is increased above 50A. The result of stack tower2 shown in Fig.11 illustrates the similar tendency. As shown in Fig.12, comparison of stack tower1and tower2 after 500h operation shows almost same performance, just slight different when the current is larger than 60A. The results also illustrate the long term stability of stack towers and the whole system.

Conclusions
In this work, a test rig of 5kW SOFC system was fabricated using practical syngas from industrial gasi ers as fuel and long term test under high fuel utility was conducted to investigate the system performance. The results show that the maximum steady output power of system is 5700W for hydrogen case and 5660W for syngas case, and the maximum steady electrical e ciency is 61.24% while the fuel utility e ciency is 89.25%. The test lasted for more than 500h as the fuel utility e ciency larger than 83%. The mean performance of stack towers is better than that of single stack, and the performances of each stack tower are almost identical. After 500h operation, the performances of stack towers just slight Photo of the test rig Photos of the fuel gas supply system and the hotbox.  System performance using H2 as fuel feeding Figure 8 System performance using syngas as fuel feeding Overview of long term performance test results

Figure 10
Performance comparison of stack tower1 of initial with after 500h operation Page 15/16

Figure 11
Performance comparison of stack tower2 of initial with after 500h operation Figure 12 Performance comparison of stack tower1and tower2 after 500h operation