sensors Letter
In-Situ Measurement of High-Temperature Proton Exchange Membrane Fuel Cell Stack Using Flexible Five-in-One Micro-Sensor Chi-Yuan Lee *, Fang-Bor Weng, Yzu-Wei Kuo, Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin Department of Mechanical Engineering, Yuan Ze Fuel Cell Center, Yuan Ze University, Taoyuan 320, Taiwan; [email protected] (F.-B.W.); [email protected] (Y.-W.K.); [email protected] (C.-H.T.); [email protected] (Y.-T.C.); [email protected] (C.-K.C.); [email protected] (J.-T.L.) * Correspondence: [email protected]; Tel: +886-3-463-8800 (ext. 2478); Fax: +886-3-455-8013 Academic Editor: Stefano Mariani Received: 10 June 2016; Accepted: 13 October 2016; Published: 18 October 2016
Abstract: In the chemical reaction that proceeds in a high-temperature proton exchange membrane fuel cell stack (HT-PEMFC stack), the internal local temperature, voltage, pressure, flow and current nonuniformity may cause poor membrane material durability and nonuniform fuel distribution, thus influencing the performance and lifetime of the fuel cell stack. In this paper micro-electro-mechanical systems (MEMS) are utilized to develop a high-temperature electrochemical environment-resistant five-in-one micro-sensor embedded in the cathode channel plate of an HT-PEMFC stack, and materials and process parameters are appropriately selected to protect the micro-sensor against failure or destruction during long-term operation. In-situ measurement of the local temperature, voltage, pressure, flow and current distributions in the HT-PEMFC stack is carried out. This integrated micro-sensor has five functions, and is favorably characterized by small size, good acid resistance and temperature resistance, quick response, real-time measurement, and the goal is being able to be put in any place for measurement without affecting the performance of the battery. Keywords: five-in-one micro-sensor; HT-PEMFC stack; in-situ measurement
1. Introduction Facing the challenges of global climate change and the need to reduce greenhouse gas emissions, people are becoming increasingly aware of the importance of sustainable development and green technology. Therefore, various countries are promoting the green energy industry. In Taiwan, the Executive Yuan launched the “Dawning Green Energy Industry Program” to develop the green energy sector. This program is expected to facilitate research and development into hydrogen energy and fuel cell technology and product development [1]. Tan [2] suggested that the unreasonable consumption of energy may inhibit the sustainable development of the national economy. Newly developed energy generation technology must be efficient, clean and safe. Therefore, the fuel cell has become an important new source of energy. The fuel cell stack is essential to the commercialization of fuel cells. A good fuel cell stack must exhibit close to ideal flow, reacting gas, temperature distribution and uniform stacking pressure [3]. The HT-PEMFC stack is characterized by its portability, high energy conversion efficiency, lack of electrolyte loss, ease of assembly and production, and long operating lifetime [4]. In the last decade, research teams around the world have paid close attention to the high-temperature proton exchange membrane fuel cell stack (HT-PEMFC) stack (120~200 ◦ C) [5–7]. Harms [8] showed that the voltage of the fuel cell stack is governed by the cathode, particularly the voltage drop at low relative humidity. A high current load can dry the proton exchange membrane and Sensors 2016, 16, 1731; doi:10.3390/s16101731
www.mdpi.com/journal/sensors
Sensors 2016, 16, 1731
2 of 10
Harms [8] showed that the voltage of the fuel cell stack is governed by the cathode, particularly the voltage drop at low relative humidity. A high current load can dry the proton exchange Sensors 2016, 16, 1731 2 of 10 membrane and degrade the performance. Therefore, the measurement of both current and voltage is very important. Our research team has successfully embedded micro temperature and voltage sensors inside the HT-PEMFC stack [9–11]. degrade the performance. Therefore, the measurement of both current and voltage is very important. The shortcomings of the low-temperature fuel cell stack include following; (1) inside the anode Our research team has successfully embedded micro temperature andthe voltage sensors the catalyst has poor resistance to CO poisoning at low temperature; (2) the perfluorosulfonic acid HT-PEMFC stack [9–11]. membrane has good of ionic conductivity only atcell high humidity; (3) following; the cathodic reduction The shortcomings the low-temperature fuel stack include the (1) the anode overpotential is high; (4) liquid water and heatatremoval management bulk production. catalyst has poor resistance to CO poisoning low temperature; (2)[12] the delays perfluorosulfonic acid However, problems of the HT-PEMFC stack, such as durability of the membrane material, catalyst membrane has good ionic conductivity only at high humidity; (3) the cathodic reduction overpotential flow, pressure, temperature, and voltage and inside the fuel iscorrosion, high; (4)local liquid water and heat removal management [12]current delaysnonuniformity bulk production. However, cell stack, must be solved before such a fuel cell can be commercialized. This investigation concerns problems of the HT-PEMFC stack, such as durability of the membrane material, catalyst corrosion, the in-situ monitoring of local flow, temperature, voltage and current in the local flow, pressure, temperature, andpressure, voltage and current nonuniformity inside the fuelHT-PEMFC cell stack, stack.be solved before such a fuel cell can be commercialized. This investigation concerns the in-situ must monitoring of local flow, pressure, temperature, voltage and current in the HT-PEMFC stack. 2. Sensing Principle and Design of Five-in-One Micro-Sensor 2. Sensing Principle and Design of Five-in-One Micro-Sensor 2.1. Sensing Principle of Five-in-One Micro-Sensor 2.1. Sensing Principle of Five-in-One Micro-Sensor 2.1.1. Micro Temperature Sensor 2.1.1. Micro Temperature Sensor Figure 1 displays the design of a micro temperature sensor. The sensing principle is that, since displaystemperature the design ofcoefficient a micro temperature The sensing principle is that, gold goldFigure has a 1positive (PTC), as sensor. the ambient temperature rises, its since resistivity has a positive temperature coefficient (PTC), as the ambient temperature rises, its resistivity increases, increases, as according to Equation (1): as according to Equation (1): d 11 dρ α= (1) (1) ρ00dT dT When temperature detector is used the range linear of variation resistance Whena aresistance-based resistance-based temperature detector is in used in theofrange linear of variation of with temperature, Equation (1) reduces to Equation (2). resistance with temperature, Equation (1) reduces to Equation (2).
(1+ T )) Rt R =t RR α11∆T r (r1
(2) (2)
Equation (2) can to Equation (3). (3). Equation (2) be canchanged be changed to Equation
R R
1 Rt t− Rrr α1 = T)) RRrr((∆T
(3) (3)
Figure1.1. Design Design of ofmicro microtemperature temperaturesensor. sensor. Figure
2.1.2.Micro MicroVoltage VoltageSensor Sensor 2.1.2. Figure22displays displaysthe thedesign designof ofthe themicro microvoltage voltagesensor. sensor.AA200 200µm μm×× 200 200 µm μmsensing sensingarea areaof ofthe the Figure micro voltage sensor is exposed. The conductors are insulated by an insulating layer, ensuring that micro voltage sensor is exposed. The conductors are insulated by an insulating layer, ensuring that thevoltage voltagethat thatisisdetected detectedherein hereinby bythe theflaky flakyprobe probedeep deepin inthe thehigh-temperature high-temperaturefuel fuelcell cellstack stackisis the detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and the anode plate.
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731 Sensors 2016, 2016, 16, 16, 1731 1731 Sensors
3 of 10 3 of 10 of 10 10 33 of
detected locally. The voltage voltage is is measured measured between between the the probe probe (inserted (inserted in in the the cathode cathode flow-field) flow-field)3 and and of 10 the anode plate. detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and the anode plate. the anode anode plate. plate. the
detected Sensors 2016,locally. 16, 1731 The
Figure 2. Design sensor. Figure Design of of micro micro voltage voltage Figure 2. 2. Design of micro voltage sensor. sensor. Figure 2. Design of micro voltage sensor. Figure 2. Design of micro voltage sensor.
2.1.3. Sensor 2.1.3. Micro Pressure Sensor Micro Pressure 2.1.3. Micro Pressure Pressure Sensor 2.1.3. Micro Sensor The sensing area is designed as 800 μm ×× 800 The capacitive capacitive sensing area 800 μm, µm, as as presented presented in area is is designed designed as as 800 800 µm μm × 800 μm, as presented in Figure Figure 3. 3. The capacitive sensing area is designed as 800 μm × 800 μm, as presented in Figure 3. 3. The capacitive sensing area is designed as 800 μm × 800 μm, as presented in Figure
Figure 3. Design of capacitive micro pressure sensor. Figure 3. Design of capacitive micro pressure sensor. Figure 3. Design of capacitive micro pressure sensor. Figure 3. 3. Design Design of of capacitive capacitive micro micro pressure pressure sensor. sensor. Figure
2.1.4. 2.1.4. Micro Micro Flow Flow Sensor Sensor 2.1.4. Micro Flow Sensor 2.1.4. Micro Flow Sensor 2.1.4. Micro Flow Sensor Figure Figure 44 shows shows the the sensing sensing principle principle of of aa hot-wire hot-wire micro micro flow flow sensor. sensor. Figure Figure 11 also also displays displays the the Figure 4 4shows the sensing principle ofofa ahot-wire micro flow sensor. Figure 1 1also displays design of a micro flow sensor. Figure shows the sensing principle hot-wire micro flow sensor. Figure also displaysthe the 4 shows sensing principle of a hot-wire micro flow sensor. Figure 1 also displays the designFigure of a micro flowthe sensor. design of a micro flow sensor. design of of aa micro micro flow flow sensor. sensor. design
Figure 4. Sensing principle of micro flow sensor. Figure 4. Sensing principle of micro flow sensor. Figure Figure4.4. 4.Sensing Sensingprinciple principleofof ofmicro microflow flowsensor. sensor. Figure Sensing principle micro flow sensor.
2.1.5. 2.1.5. Micro Micro Current Current Sensor Sensor 2.1.5. Micro Current Sensor designed herein with sizes 350 μm × 350 μm and 400 μm × 350 μm 2.1.5. Micro Current Sensor 2.1.5. Micro Current Sensor Micro Micro current current sensors sensors are are designed herein with sizes 350 μm × 350 μm and 400 μm × 350 μm integrated in the of aa 40 μm steel sheet (stainless steel foil) of the flexible substrate, Micro current sensors are designed herein with sizes 350 μm 350 μm and400 400 μm 350µm μm Micro current sensors are designed herein with sizes 350 µm × µm µm × Micro current sensors are designed herein with sizes 350 μm ××350 350 μm and 400 μm ××350 350 μm integrated in the thickness thickness of 40 μm stainless stainless steel sheet (stainless steel foil) ofand the flexible substrate, as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while the other integrated in the thickness ofa a4040 40 μmstainless stainlesssensor, steelsheet sheet (stainlesssteel steelfoil) foil) ofthe theflexible flexible substrate, integrated inin the thickness µm steel ofof the thickness μm stainless steel sheet (stainless steel foil) the flexible substrate, asintegrated displayed in Figure 5. Asofof in micro voltage the (stainless front-end probe is exposed whilesubstrate, the other parts are covered with an insulating layer, ensuring that the probe penetrating into the fuel cell stack as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while theother other as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while the as displayed in Figure 5. insulating As in a micro voltage sensor, probe is exposed the other parts are covered with an layer, ensuring thatthe thefront-end probe penetrating into the while fuel cell stack can detect the current of the local specific area, and the external measuring instrument measures the parts are covered with an insulating layer, ensuring that the probe penetrating into the fuel cell stack parts are with anan insulating layer, ensuring that the penetrating into parts arecovered covered with insulating layer, ensuring that theprobe probe penetrating intothe thefuel fuelcell cellstack stack can detect the current of the local specific area, and the external measuring instrument measures the voltage difference and the resistivity. can detect the current of the local specific area, and the external measuring instrument measures the can detect the current of the local specific area, and the external measuring instrument measures the can detect the current of the local specific area, and the external measuring instrument measures the voltage difference and the resistivity. voltage difference and the resistivity. voltage difference and the resistivity. voltage difference and the resistivity.
Figure 5. Design of micro current sensor. Figure 5. Design of micro current sensor. Figure 5. 5. Design Design of of micro micro current current sensor. sensor. Figure Figure 5. Design of micro current sensor.
Sensors 2016, 16, 1731
4 of 10
Sensors 2016, 16, 1731 Sensors 2016, 16,Product 1731 2.1.6. Finished of Five-in-One Micro-Sensor
2.1.6. Finished Product of Five-in-One Micro-Sensor
4 of 10 4 of 10
Figure 6 shows thethe real product optical ofthe thefive-in-one five-in-onemicro-sensor. micro-sensor. 2.1.6. Finished Product ofreal Five-in-One Micro-Sensor Figure 6 shows productand and opticalmicrograph micrograph of A 4Ain4 in stainless steel substrate can be made into seven five-in-one micro-sensors. The five-in-one micro-sensor stainless steel substrate can product be madeand intooptical sevenmicrograph five-in-one of micro-sensors. The five-in-one A microFigure 6 shows the real the five-in-one micro-sensor. 4 in has stainless many characteristics—compactness, resistance against acid corrosion, favorable thermal sensorimportant has many important characteristics—compactness, resistance against acid corrosion, favorable steel substrate can be made into seven five-in-one micro-sensors. The five-in-one microtolerance, a tolerance, short response and the to ability make measurements in real thermal a shorttime, response time,ability and the toresistance make measurements in real time. sensor has many important characteristics—compactness, against acid time. corrosion, favorable thermal tolerance, a short response time, and the ability to make measurements in real time.
Figure 6. Real product of five-in-one five-in-onemicro-sensor. micro-sensor. Figure 6. Real productand andoptical opticalmicrograph micrograph of Figure 6. Real product and optical micrograph of five-in-one micro-sensor.
2.1.7.2.1.7. Sensor Calibration Sensor Calibration When the five-in-one micro-sensor is inserted into the HT-PEMFC stack to make local 2.1.7. Sensor When the Calibration five-in-one micro-sensor is inserted into the HT-PEMFC stack to make local measurements in real time, micro-sensor corrections must be made for the its temperature, flow andlocal the When the five-in-one is inserted HT-PEMFC the stack to rate make measurements in real time, corrections must be madeinto for its temperature, the flow rate and the pressure, to yield that correspond physical quantities. has measurements in electrical real time,measurements corrections must be made fortoits temperature, theAfter flow the ratedevice and the pressure, to yield electrical measurements that correspond to physical quantities. After the device has been calibrated times,measurements the mean value is correspond obtained and reliability of its signals is tested. pressure, to yieldthree electrical that to the physical quantities. After the device has been calibrated three times, the mean value is obtained and theHT-PEMFC reliability of its signals isdirectly tested. In this work, the external temperature controller of the stack is used been calibrated three times, the mean value is obtained and the reliability of its signals is tested. to In this work, the external temperature controller of the HT-PEMFC stack is used directly provide the work, benchmark to calibrate the temperature of the five-in-one micro-sensors, as presented in to In this the external temperature controller of the HT-PEMFC stack is used directly to provide the benchmark totocalibrate the micro temperature five-in-one as the presented Figure 7.the The calibration curves of the temperature sensors thatmicro-sensors, aremicro-sensors, embeddedasinside HTprovide benchmark calibrate temperature ofof thethe five-in-one presented in in Figure 7. The calibration curves of the micro temperature sensors that are embedded inside PEMFC are obtained fromof three calibrations at temperatures from C to 190inside C. The Figure 7.stack The calibration curves the micro temperature sensors that are140 embedded themicro HT- the ◦ to 190 ◦ C. The micro HT-PEMFC stacksensor areobtained obtained fromthree three calibrations at temperatures from temperature has an from accuracy better than 0.5atC. PEMFC stack are calibrations temperatures from 140140 C toC190 C. The micro ◦ temperature sensor has anan accuracy C. temperature sensor has accuracybetter betterthan than 0.5 0.5 C.
Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of the five-in-one micro-sensors. Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of the five-in-one micro-sensors.
the five-in-one Regarding micro-sensors. the calibration based on pressure, a Wayne Kerr Electronics (London, UK) 4230 LCR
(Inductance, Capacitance, Resistance) was used to capture the relevant data. The range the Regarding the calibration based onmeter pressure, a Wayne Kerr Electronics (London, UK) 4230of LCR Regarding Capacitance, the basedusing onmeter pressure, a Wayne Kerr Electronics UK) 4230 capacitances thatcalibration can be measured the was instrument 0.01 pF~1 F,relevant and its (London, precision ±0.1%. (Inductance, Resistance) used toiscapture the data. The is range of The theLCR (Inductance, Capacitance, Resistance) meter relevant data. The rangeThe of the micro pressure an accuracy better thanused 0.01 to kgf/cm . the capacitances thatsensor can behas measured using the was instrument is capture 0.01 2pF~1 F, and its precision is ±0.1%. 2 The micro flow rate sensors can be calibrated in a stable temperature field in air which the power capacitances that can be measured using the instrument is 0.01 pF~1 F, and its precision is ± 0.1%. micro pressure sensor has an accuracy better than 0.01 kgf/cm . 2 supply generates at the given voltage. The range of flow rates in the calibration is 0~30 L/min. The The micro sensor has an accuracy better than 0.01 kgf/cm . field in air which the power Thepressure micro flow rate sensors can be calibrated in a stable temperature micro flow rate sensor an accuracy ofcalibrated better 0.1aL/min. supply generates atrate thehas given voltage. The rangethan of flow rates in temperature the calibrationfield is 0~30 The the The micro flow sensors can be in stable in L/min. air which micro flow rate sensor at has angiven accuracy of better 0.1of L/min. power supply generates the voltage. Thethan range flow rates in the calibration is 0~30 L/min.
The micro flow rate sensor has an accuracy of better than 0.1 L/min.
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
5 of 10 5 of 10
3. Test TestResult Resultof ofConstant ConstantCurrent Current(5, (5,13, 13,20 20 A) A) Output Output An experiment onon thethe HT-PEMFC stack at a stable temperature and anand open experimentisiscarried carriedout out HT-PEMFC stack at a stable temperature ancircuit open voltage (OCV).(OCV). In thisInexperiment, the operating temperature of theoffuel cell cell stackstack is 160 C, ◦as circuit voltage this experiment, the operating temperature the fuel is 160 C, 2 2 2 2 indicated in in Table 1; 1; the size as indicated Table the sizeofoffuel fuelcell cellstack stackisis33 33cm cm and andthe thereaction reaction area area is is 31.4 cm . The The layers are made made of ofgraphite graphitebecause becauseit itresists resists corrosion. Five-in-one micro-sensors installed upstream corrosion. Five-in-one micro-sensors are are installed upstream and and downstream in Cells 1, 510, and as displayed in Figure 8. The machine gives theand anode and downstream in Cells 1, 5 and as 10, displayed in Figure 8. The machine gives the anode cathode cathode fixed, not-humidified gases H (anode H2: 5cathode slpm; cathode 30 slpm), and different loads are fixed, not-humidified gases (anode Air: 30Air: slpm), and different loads are given 2 : 5 slpm; given a constant (5,A); 13,the 20 output A); the time output time 60 min. local temperature, voltage, a constant currentcurrent (5, 13, 20 is 60 min.is The localThe temperature, voltage, pressure, pressure, flow rate and in the HT-PEMFC stack are obtained by real-time continuous flow rate and current in thecurrent HT-PEMFC stack are obtained by real-time continuous measurement using measurement an NI PXI 2575 acquisitionfive-in-one unit. The embedded five-in-one micro-sensors an NI PXI 2575using data acquisition unit. data The embedded micro-sensors inside the HT-PEMFC inside the HT-PEMFC stack reaction area of Electrode the MEA (Membrane Electrode to stack cover the reaction areacover of thethe MEA (Membrane Assembly) to prevent Assembly) the gas from prevent the gas from reacting with the covered (the HT-MEA used inavailable this study is a reacting with the covered MEA (the HT-MEA used inMEA this study is a commercially Advent commercially available Advent Energy Company’s (Patras, Greece) specifications are Energy Company’s (Patras, Greece) HT-MEA; the specifications are HT-MEA; in Table 2).the However, since the in Table 2). micro-sensors However, since five-in-one micro-sensors are very small, their total areareaction equals only five-in-one arethe very small, their total area equals only 1.34% of the MEA area. 1.34% of the MEA reactionmicro-sensors area. The embedded five-in-one micro-sensors in the HT-PEMFC stack The embedded five-in-one in the HT-PEMFC stack alter the performance of the stack by alter the1.3%. performance of the stack by around 1.3%. around Table 1. 1. Testing Testing conditions Table conditions for for the the HT-PEMFC HT-PEMFC stack. stack.
Subject Subject Temperature stack(◦(C) Temperature of of the the stack C) Cathode flow rate (Air) (slpm) Cathode flow rate (Air) (slpm) Anodeflow flow rate rate (H Anode (H22)) (slpm) (slpm) Constant current Constant current (A) (A) Gas temperature Gas temperature Reaction area (cm2 ) Reaction area (cm2)
Condition Condition 160 160 3030 55 5,5,13, 13,2020 Room temperature Room temperature 31.4 31.4
Table 2. Specifications of the HT-MEA. Table 2. Specifications of the HT-MEA. Subject Subject HT-MEA HT-MEA Membrane thickness Membrane thickness Conductivity Conductivity
Specificationsis Specificationsis Advent energy Advent energy 60~65 µm 60~65 μm −2 S/cm 88××10 −2 10 S/cm
Figure channel plates. plates. Figure 8. 8. Position Position of of five-in-one five-in-one micro-sensors micro-sensors in in cathode cathode channel
3.1. Local Temperature and Voltage Distributions in Different Cells 3.1. Local Temperature and Voltage Distributions in Different Cells The external voltages of different cells and the local temperature and voltage inside the HTThe external voltages of different cells and the local temperature and voltage inside the HT-PEMFC PEMFC stack, measured between the probe (inserted in the cathode flow-field) and the anode plate stack, measured between the probe (inserted in the cathode flow-field) and the anode plate at at an operating temperature of 160 C, are discussed and analyzed as follows. an operating temperature of 160 ◦ C, are discussed and analyzed as follows. 1. Figure 9 compares the internal and external voltages. The voltage curve is smooth at a constant current (5, 13 A), because the electrochemical reaction is uniform. At a high current (20 A), the internal electrochemical reaction is violent, and the voltage distribution is nonuniform. The
be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction is vigorous at a high current (20 A), and heat is concentrated, causing Cell 5 to be the hottest. Sensors 2016, 16, 1731
1
6 of 10
Figure 10 compares the internal and external voltages. The voltage curve is smooth at a constant current (5, 13 A), because the electrochemical reaction is uniform. At a high current (20 A), the internal electrochemical reaction is violent, and the voltage distribution is nonuniform. The internal local voltage is measured using a micro voltage sensor. The curve of the variation of the internal voltage is consistent with that of the external voltage. Figure 9 plots the local temperature and voltage in different cells. The temperature increases gradually with the current (5, 13, 20 A), because a higher operating current causes more heat to be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction is vigorous at a high current (20 A), and heat is concentrated, causing Cell 5 to be the hottest.
2
Figure 9. Comparison of internal and external voltages.
Sensors 2016, 16, 1731
2.
6 of 10
internal local voltage is measured using a micro voltage sensor. The curve of the variation of the internal voltage is consistent with that of the external voltage. Figure 10 plots the local temperature and voltage in different cells. The temperature increases gradually with the current (5, 13, 20 A), because a higher operating current causes more heat to be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction Figure 10. Comparison of local temperatures and voltages in different cells. 9. Comparison local and voltages in different is vigorousFigure at a high current (20ofA), andtemperatures heat is concentrated, causing Cell 5cells. to be the hottest.
3.2. Local Pressure Distribution in Different Cells Figure 11 plots the local pressure in different cells, and the pressure control in the experimental work is 1.3 kgf/cm2. The local pressure in the HT-PEMFC stack does not vary as the current (5, 13, 20 A) increases. The upstream pressure slightly exceeds the downstream pressure. The mean pressure in the HT-PEMFC stack is 1.291 kgf/cm2. The upstream end pressure of Cell 1 and Cell 10 is unstable, but is stabilized at the downstream end. The difference between the end pressure upstream and that downstream in Cell 5 is small, so the pressure in the HT-PEMFC stack is stable. This can be used to assess whether the design of the flow field inside the cell stack is good or not.
Figure 9. Comparison of internal and external voltages. Figure 10. Comparison of internal and external voltages.
3.2. Local Pressure Distribution in Different Cells Figure 11 plots the local pressure in different cells, and the pressure control in the experimental work is 1.3 kgf/cm2 . The local pressure in the HT-PEMFC stack does not vary as the current (5, 13, 20 A) increases. The upstream pressure slightly exceeds the downstream pressure. The mean pressure in the HT-PEMFC stack is 1.291 kgf/cm2 . The upstream end pressure of Cell 1 and Cell 10 is unstable,
Sensors 2016, 16, 1731
7 of 10
but is stabilized at the downstream end. The difference between the end pressure upstream and that downstream in Cell 5 is small, so the pressure in the HT-PEMFC stack is stable. This can be used to Sensors 2016, 16, 1731 7 of 10 assess whether the design of the flow field inside the cell stack is good or not.
Figure 11.11. Comparison of of local pressures in in different cells. Figure Comparison local pressures different cells.
3.3. Local Local Flow Flow Distribution Distribution in in Different Different Cells Cells 3.3. Figure1212displays displays flow in different cells. The upstream flow rate exceeds the Figure the the locallocal flow in different cells. The upstream flow rate exceeds the downstream downstream flow rate, and the flow rates in Cell 1 and Cell 10 exceed those in Cell 5. Since the flow rate, and the flow rates in Cell 1 and Cell 10 exceed those in Cell 5. Since the upstream flowing upstream flowing in Cell 1 and Cell is close inlet, the gas is sufficient. Theflow downstream flow in Cell 1 and Cell 10 is close the air10 inlet, the the gasair is sufficient. The downstream rate is lower, rate is lower, because multiple snakelike runners (commonly called as “serpentine flow channels”) because multiple snakelike runners (commonly called as “serpentine flow channels”) are used to are used the to improve the performance of the stack, HT-PEMFC stack, and distribution the internal of distribution ofgas the improve performance of the HT-PEMFC and the internal the internal internal gas is nonuniform. is nonuniform.
Figure 12. 12. Comparison Comparison of of local local flow flow rates rates in in different differentcells. cells. Figure
3.4. Local Current Distribution in Different Cells 3.4. Local Current Distribution in Different Cells Figures 13–16 plot the local current density in different cells and this in-plane current does not Figures 13–16 plot the local current density in different cells and this in-plane current directly the through-plane current flow through the catalyst layers and PEM (Proton does not represent directly represent the through-plane current flow through the catalyst layers and PEM Exchange Membrane). The current density in the HT-PEMFC stack varies with the current (5, 13, 20 (Proton Exchange Membrane). The current density in the HT-PEMFC stack varies with the current A). The current density upstream exceeds that downstream, because the gas is present at the (5, 13, 20 A). The current density upstream exceeds that downstream, because the gas is present at upstream inletinlet to initiate a electrochemical reaction, but the serpentine flow channels cause the upstream to initiate a electrochemical reaction, butmultiple the multiple serpentine flow channels the internal gas distribution to be nonuniform. The Nernst loss due to gas (especially oxygen) cause the internal gas distribution to be nonuniform. The Nernst loss due to gas (especially oxygen) depletion increases increases toward towardthe thedownstream downstreampart. part. depletion
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
8 of 10
8 of 10
8 of 10
Figure 13. Comparison of local in different cells. cells. Figure13. 13.Comparison Comparison of of local local current current densities densities Figure current densitiesinindifferent different cells.
Figure 14. Comparison of local current densities in different cells (part of 5 A). Figure14. 14.Comparison Comparisonof oflocal local current current densities ofof 5 A). Figure densitiesin indifferent differentcells cells(part (part 5 A).
Figure 15. Comparison of local current densities in different cells (part of 13 A). Figure15. 15.Comparison Comparisonof oflocal local current current densities ofof 1313 A).A). Figure densitiesin indifferent differentcells cells(part (part
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
9 of 10
9 of 10
Figure 16.Comparison Comparisonof oflocal localcurrent current densities densities in 20A). Figure 16. indifferent differentcells cells(part (partofof 20A).
4. Conclusions 4. Conclusions thiswork, work,MEMS MEMS technology to to develop a five-in-one micro-sensor on a 40-µm-thick InIn this technologyisisutilized utilized develop a five-in-one micro-sensor on a 40-μmstainless steel substrate. The protective layer is made of PI (Polyimide) with a favorable temperature thick stainless steel substrate. The protective layer is made of PI (Polyimide) with a favorable tolerance. This five-in-one micro-sensor has five functions. It is characterized by its compactness, temperature tolerance. This five-in-one micro-sensor has five functions. It is characterized by its resistance to acid corrosion, favorable temperature tolerance, rapid response and usefulness in compactness, resistance to acid corrosion, favorable temperature tolerance, rapid response and real-time measurement. usefulness in real-time measurement. The local temperature, flow and pressure information inside the HT-PEMFC stack are extracted The local temperature, flow and pressure information inside the HT-PEMFC stack are extracted successfully. The test result of the operating temperature of 160 ◦ C and the constant current (5, 13, successfully. test resulttemperature of the operating temperature of 160 C HT-PEMFC and the constant 20 A) showsThe nonuniform and flow distributions in the stack. current (5, 13, 20 A) shows andthe flow distributions in the HT-PEMFC Thenonuniform experimentstemperature indicated that nonuniform temperature and flowstack. distributions in the The experiments indicated that the nonuniform temperature and flow distributions the HTHT-PEMFC stack were responsible for a large internal temperature difference, nonuniformin voltage PEMFC stack distributions, were responsible largeThe internal temperature difference, stack nonuniform voltage and and current and afor hotastack. pressure inside the HT-PEMFC was stable, but the current distributions, and a hot stack. The pressure inside the HT-PEMFC stack was stable, but the air inlet upstream pressure was unstable at both the cathode and the anode.
air inlet upstream pressure was unstable at both the cathode and the anode. Acknowledgments: This work was accomplished with much needed support and the authors would
Acknowledgments: work of was accomplished with much and thesupport authorsthrough would like like to thank the This Ministry Science and Technology of needed R. O. C.support for financial the to grant MOST 102-2221-E-155-033-MY3, 103-2622-E-155-006-CC2, 104-2623-E-155-004-ET, 103-2622-E-155-018-CC2, thank the Ministry of Science and Technology of R. O. C. for financial support through the grant MOST 102104-2622-E-155-004, 104-2622-E-155-007-CC2, 105-ET-E-155-002-ET, 105-2622-8-155-003-TE3, 105-2622-E-155-013-CC2 2221-E-155-033-MY3, 103-2622-E-155-006-CC2, 104-2623-E-155-004-ET, 103-2622-E-155-018-CC2, 104-2622-Eand 105-2221-E-155-005. The authors also like to thank Shih Hung Chan, Ay Su and Guo-Bin Jung of Yuan Ze 155-004, 104-2622-E-155-007-CC2, 105-ET-E-155-002-ET, 105-2622-8-155-003-TE3, 105-2622-E-155-013-CC2 and University for their valuable advice and assistance in the experiments. In addition, we would like to thank the 105-2221-E-155-005. The authors also like to thank Shih Hung Chan, Ay Su and Guo-Bin Jung of Yuan Ze YZU Fuel Cell Center and NENS Common Lab for providing access to their research facilities. University for their valuable advice assistance in was the experiments. In addition, we would likealltoauthors. thank the Author Contributions: The workand presented here carried out in collaboration between YZU Fuel Cell and NENS Lab for providing access this to their research facilities. Chi-Yuan LeeCenter and Fang-Bor WengCommon conceived, designed and discussed study. Yzu-Wei Kuo, Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin performed the experiments and analyzed the data. All authors
Author Contributions: The and work presented here was carried out in collaboration between all authors. have contributed, reviewed improved the manuscript. Chi-Yuan Lee and Fang-Bor Weng conceived, designed and discussed this study. Yzu-Wei Kuo, Conflicts of Interest: The authors declare no conflict of interest. Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin performed the experiments and analyzed the data. All authors have contributed, reviewed and improved the manuscript.
References
Conflicts of Interest: The authors declare no conflict of interest. 1.
Bureau of Energy. Ministry of Economic Affairs. Available online: http://web3.moeaboe.gov.tw/ecw/ populace/content/SubMenu.aspx?menu_id=2324 (accessed on 4 May 2014). References 2. Tan, Z.; Zhang, C.; Liu, P.; Reed, B.; Zhao, J. Focus on fuel cell systems in China. Renew. Sustain. Energy Rev. 2015, 47, 1. Bureau of 912–923. Energy. Ministry of Economic Affairs. Available online: http://web3.moeaboe.gov.tw/ecw/ 3. populace/content/SubMenu.aspx?menu_id=2324 Weng, F.B.; Jou, B.S.; Su, A.; Chan, S.H.; Chi, P.H. (accessed Design, fabrication performance analysis of a 200 W on 4 Mayand 2014). PEM fuel cell short stack. J. Power Sources 2007, 171, 179–185. [CrossRef] 2. Tan, Z.; Zhang, C.; Liu, P.; Reed, B.; Zhao, J. Focus on fuel cell systems in China. Renew. Sustain. Energy Rev. 4. Janssen, H.; Supra, J.; Lueke, L.; Lehnert, W.; Stolten, D. Development of HT-PEFC stacks in the kW range. 2015, 47, 912–923. Int. J. Hydrogen Energy 2013, 38, 4705–4713. [CrossRef]
3. 4.
Weng, F.B.; Jou, B.S.; Su, A.; Chan, S.H.; Chi, P.H. Design, fabrication and performance analysis of a 200 W PEM fuel cell short stack. J. Power Sources 2007, 171, 179–185. Janssen, H.; Supra, J.; Lueke, L.; Lehnert, W.; Stolten, D. Development of HT-PEFC stacks in the kW range. Int. J. Hydrogen Energy 2013, 38, 4705–4713.
Sensors 2016, 16, 1731
5. 6. 7.
8. 9.
10.
11. 12.
10 of 10
Zhang, J.; Xie, Z.; Zhang, J.; Tang, Y.; Song, C.; Navessin, T.; Shi, Z.; Song, D.; Wang, H.; Wilkinson, D.P.; et al. High temperature PEM fuel cells. J. Power Sources 2006, 160, 872–891. [CrossRef] Li, Q.; Jensen, J.O.; Savinell, R.F.; Bjerruma, N.J. High temperature proton exchange membranes based on polybenzimidazoles for fuel cells. Prog. Polym. Sci. 2009, 34, 449–477. [CrossRef] Chandan, A.; Hattenberger, M.; El-Kharouf, A.; Du, S.F.; Dhir, A.; Self, V.; Pollet, B.G.; Ingram, A.; Bujalski, W. High temperature (HT) polymer electrolyte membrane fuel cells (PEMFC)—A review. J. Power Sources 2013, 231, 264–278. [CrossRef] Harms, C.; Köhrmann, F.; Dyck, A. Study of the influence of key test parameters on the performance of a PEMFC stack. Solid State Ionics 2015, 275, 75–79. [CrossRef] Lee, C.Y.; Weng, F.B.; Cheng, C.H.; Shiu, H.R.; Jung, S.P.; Chang, W.C.; Chan, P.C.; Chen, W.T.; Lee, C.J. Use of flexible micro-temperature sensor to determine temperature in situ and to simulate a proton exchange membrane fuel cell. J. Power Sources 2011, 196, 228–234. [CrossRef] Weng, F.B.; Cheng, C.K.; Lee, C.Y.; Chang, C.P. Analysis of thermal balance in high-temperature proton exchange membrane fuel cells with short stacks via in situ monitoring with a flexible micro sensor. Int. J. Hydrogen Energy 2014, 39, 13681–13686. [CrossRef] Lee, C.Y.; Weng, F.B.; Huang, Y.P.; Chang, C.P.; Cheng, C.K. Real-time monitoring of internal temperature and voltage of high-temperature fuel cell stack. Electrochim. Acta 2015, 161, 413–419. [CrossRef] Bose, S.; Kuila, T.; Nguyen, T.X. H.; Kim, N.H.; Lau, K.T.; Lee, J.H. Polymer membranes for high temperature proton exchange membrane fuel cell: Recent advances and challenges. Prog. Polym. Sci. 2011, 36, 813–843. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
In-Situ Measurement of High-Temperature Proton Exchange Membrane Fuel Cell Stack Using Flexible Five-in-One Micro-Sensor Chi-Yuan Lee *, Fang-Bor Weng, Yzu-Wei Kuo, Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin Department of Mechanical Engineering, Yuan Ze Fuel Cell Center, Yuan Ze University, Taoyuan 320, Taiwan; [email protected] (F.-B.W.); [email protected] (Y.-W.K.); [email protected] (C.-H.T.); [email protected] (Y.-T.C.); [email protected] (C.-K.C.); [email protected] (J.-T.L.) * Correspondence: [email protected]; Tel: +886-3-463-8800 (ext. 2478); Fax: +886-3-455-8013 Academic Editor: Stefano Mariani Received: 10 June 2016; Accepted: 13 October 2016; Published: 18 October 2016
Abstract: In the chemical reaction that proceeds in a high-temperature proton exchange membrane fuel cell stack (HT-PEMFC stack), the internal local temperature, voltage, pressure, flow and current nonuniformity may cause poor membrane material durability and nonuniform fuel distribution, thus influencing the performance and lifetime of the fuel cell stack. In this paper micro-electro-mechanical systems (MEMS) are utilized to develop a high-temperature electrochemical environment-resistant five-in-one micro-sensor embedded in the cathode channel plate of an HT-PEMFC stack, and materials and process parameters are appropriately selected to protect the micro-sensor against failure or destruction during long-term operation. In-situ measurement of the local temperature, voltage, pressure, flow and current distributions in the HT-PEMFC stack is carried out. This integrated micro-sensor has five functions, and is favorably characterized by small size, good acid resistance and temperature resistance, quick response, real-time measurement, and the goal is being able to be put in any place for measurement without affecting the performance of the battery. Keywords: five-in-one micro-sensor; HT-PEMFC stack; in-situ measurement
1. Introduction Facing the challenges of global climate change and the need to reduce greenhouse gas emissions, people are becoming increasingly aware of the importance of sustainable development and green technology. Therefore, various countries are promoting the green energy industry. In Taiwan, the Executive Yuan launched the “Dawning Green Energy Industry Program” to develop the green energy sector. This program is expected to facilitate research and development into hydrogen energy and fuel cell technology and product development [1]. Tan [2] suggested that the unreasonable consumption of energy may inhibit the sustainable development of the national economy. Newly developed energy generation technology must be efficient, clean and safe. Therefore, the fuel cell has become an important new source of energy. The fuel cell stack is essential to the commercialization of fuel cells. A good fuel cell stack must exhibit close to ideal flow, reacting gas, temperature distribution and uniform stacking pressure [3]. The HT-PEMFC stack is characterized by its portability, high energy conversion efficiency, lack of electrolyte loss, ease of assembly and production, and long operating lifetime [4]. In the last decade, research teams around the world have paid close attention to the high-temperature proton exchange membrane fuel cell stack (HT-PEMFC) stack (120~200 ◦ C) [5–7]. Harms [8] showed that the voltage of the fuel cell stack is governed by the cathode, particularly the voltage drop at low relative humidity. A high current load can dry the proton exchange membrane and Sensors 2016, 16, 1731; doi:10.3390/s16101731
www.mdpi.com/journal/sensors
Sensors 2016, 16, 1731
2 of 10
Harms [8] showed that the voltage of the fuel cell stack is governed by the cathode, particularly the voltage drop at low relative humidity. A high current load can dry the proton exchange Sensors 2016, 16, 1731 2 of 10 membrane and degrade the performance. Therefore, the measurement of both current and voltage is very important. Our research team has successfully embedded micro temperature and voltage sensors inside the HT-PEMFC stack [9–11]. degrade the performance. Therefore, the measurement of both current and voltage is very important. The shortcomings of the low-temperature fuel cell stack include following; (1) inside the anode Our research team has successfully embedded micro temperature andthe voltage sensors the catalyst has poor resistance to CO poisoning at low temperature; (2) the perfluorosulfonic acid HT-PEMFC stack [9–11]. membrane has good of ionic conductivity only atcell high humidity; (3) following; the cathodic reduction The shortcomings the low-temperature fuel stack include the (1) the anode overpotential is high; (4) liquid water and heatatremoval management bulk production. catalyst has poor resistance to CO poisoning low temperature; (2)[12] the delays perfluorosulfonic acid However, problems of the HT-PEMFC stack, such as durability of the membrane material, catalyst membrane has good ionic conductivity only at high humidity; (3) the cathodic reduction overpotential flow, pressure, temperature, and voltage and inside the fuel iscorrosion, high; (4)local liquid water and heat removal management [12]current delaysnonuniformity bulk production. However, cell stack, must be solved before such a fuel cell can be commercialized. This investigation concerns problems of the HT-PEMFC stack, such as durability of the membrane material, catalyst corrosion, the in-situ monitoring of local flow, temperature, voltage and current in the local flow, pressure, temperature, andpressure, voltage and current nonuniformity inside the fuelHT-PEMFC cell stack, stack.be solved before such a fuel cell can be commercialized. This investigation concerns the in-situ must monitoring of local flow, pressure, temperature, voltage and current in the HT-PEMFC stack. 2. Sensing Principle and Design of Five-in-One Micro-Sensor 2. Sensing Principle and Design of Five-in-One Micro-Sensor 2.1. Sensing Principle of Five-in-One Micro-Sensor 2.1. Sensing Principle of Five-in-One Micro-Sensor 2.1.1. Micro Temperature Sensor 2.1.1. Micro Temperature Sensor Figure 1 displays the design of a micro temperature sensor. The sensing principle is that, since displaystemperature the design ofcoefficient a micro temperature The sensing principle is that, gold goldFigure has a 1positive (PTC), as sensor. the ambient temperature rises, its since resistivity has a positive temperature coefficient (PTC), as the ambient temperature rises, its resistivity increases, increases, as according to Equation (1): as according to Equation (1): d 11 dρ α= (1) (1) ρ00dT dT When temperature detector is used the range linear of variation resistance Whena aresistance-based resistance-based temperature detector is in used in theofrange linear of variation of with temperature, Equation (1) reduces to Equation (2). resistance with temperature, Equation (1) reduces to Equation (2).
(1+ T )) Rt R =t RR α11∆T r (r1
(2) (2)
Equation (2) can to Equation (3). (3). Equation (2) be canchanged be changed to Equation
R R
1 Rt t− Rrr α1 = T)) RRrr((∆T
(3) (3)
Figure1.1. Design Design of ofmicro microtemperature temperaturesensor. sensor. Figure
2.1.2.Micro MicroVoltage VoltageSensor Sensor 2.1.2. Figure22displays displaysthe thedesign designof ofthe themicro microvoltage voltagesensor. sensor.AA200 200µm μm×× 200 200 µm μmsensing sensingarea areaof ofthe the Figure micro voltage sensor is exposed. The conductors are insulated by an insulating layer, ensuring that micro voltage sensor is exposed. The conductors are insulated by an insulating layer, ensuring that thevoltage voltagethat thatisisdetected detectedherein hereinby bythe theflaky flakyprobe probedeep deepin inthe thehigh-temperature high-temperaturefuel fuelcell cellstack stackisis the detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and the anode plate.
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731 Sensors 2016, 2016, 16, 16, 1731 1731 Sensors
3 of 10 3 of 10 of 10 10 33 of
detected locally. The voltage voltage is is measured measured between between the the probe probe (inserted (inserted in in the the cathode cathode flow-field) flow-field)3 and and of 10 the anode plate. detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and detected locally. The voltage is measured between the probe (inserted in the cathode flow-field) and the anode plate. the anode anode plate. plate. the
detected Sensors 2016,locally. 16, 1731 The
Figure 2. Design sensor. Figure Design of of micro micro voltage voltage Figure 2. 2. Design of micro voltage sensor. sensor. Figure 2. Design of micro voltage sensor. Figure 2. Design of micro voltage sensor.
2.1.3. Sensor 2.1.3. Micro Pressure Sensor Micro Pressure 2.1.3. Micro Pressure Pressure Sensor 2.1.3. Micro Sensor The sensing area is designed as 800 μm ×× 800 The capacitive capacitive sensing area 800 μm, µm, as as presented presented in area is is designed designed as as 800 800 µm μm × 800 μm, as presented in Figure Figure 3. 3. The capacitive sensing area is designed as 800 μm × 800 μm, as presented in Figure 3. 3. The capacitive sensing area is designed as 800 μm × 800 μm, as presented in Figure
Figure 3. Design of capacitive micro pressure sensor. Figure 3. Design of capacitive micro pressure sensor. Figure 3. Design of capacitive micro pressure sensor. Figure 3. 3. Design Design of of capacitive capacitive micro micro pressure pressure sensor. sensor. Figure
2.1.4. 2.1.4. Micro Micro Flow Flow Sensor Sensor 2.1.4. Micro Flow Sensor 2.1.4. Micro Flow Sensor 2.1.4. Micro Flow Sensor Figure Figure 44 shows shows the the sensing sensing principle principle of of aa hot-wire hot-wire micro micro flow flow sensor. sensor. Figure Figure 11 also also displays displays the the Figure 4 4shows the sensing principle ofofa ahot-wire micro flow sensor. Figure 1 1also displays design of a micro flow sensor. Figure shows the sensing principle hot-wire micro flow sensor. Figure also displaysthe the 4 shows sensing principle of a hot-wire micro flow sensor. Figure 1 also displays the designFigure of a micro flowthe sensor. design of a micro flow sensor. design of of aa micro micro flow flow sensor. sensor. design
Figure 4. Sensing principle of micro flow sensor. Figure 4. Sensing principle of micro flow sensor. Figure Figure4.4. 4.Sensing Sensingprinciple principleofof ofmicro microflow flowsensor. sensor. Figure Sensing principle micro flow sensor.
2.1.5. 2.1.5. Micro Micro Current Current Sensor Sensor 2.1.5. Micro Current Sensor designed herein with sizes 350 μm × 350 μm and 400 μm × 350 μm 2.1.5. Micro Current Sensor 2.1.5. Micro Current Sensor Micro Micro current current sensors sensors are are designed herein with sizes 350 μm × 350 μm and 400 μm × 350 μm integrated in the of aa 40 μm steel sheet (stainless steel foil) of the flexible substrate, Micro current sensors are designed herein with sizes 350 μm 350 μm and400 400 μm 350µm μm Micro current sensors are designed herein with sizes 350 µm × µm µm × Micro current sensors are designed herein with sizes 350 μm ××350 350 μm and 400 μm ××350 350 μm integrated in the thickness thickness of 40 μm stainless stainless steel sheet (stainless steel foil) ofand the flexible substrate, as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while the other integrated in the thickness ofa a4040 40 μmstainless stainlesssensor, steelsheet sheet (stainlesssteel steelfoil) foil) ofthe theflexible flexible substrate, integrated inin the thickness µm steel ofof the thickness μm stainless steel sheet (stainless steel foil) the flexible substrate, asintegrated displayed in Figure 5. Asofof in micro voltage the (stainless front-end probe is exposed whilesubstrate, the other parts are covered with an insulating layer, ensuring that the probe penetrating into the fuel cell stack as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while theother other as displayed in Figure 5. As in a micro voltage sensor, the front-end probe is exposed while the as displayed in Figure 5. insulating As in a micro voltage sensor, probe is exposed the other parts are covered with an layer, ensuring thatthe thefront-end probe penetrating into the while fuel cell stack can detect the current of the local specific area, and the external measuring instrument measures the parts are covered with an insulating layer, ensuring that the probe penetrating into the fuel cell stack parts are with anan insulating layer, ensuring that the penetrating into parts arecovered covered with insulating layer, ensuring that theprobe probe penetrating intothe thefuel fuelcell cellstack stack can detect the current of the local specific area, and the external measuring instrument measures the voltage difference and the resistivity. can detect the current of the local specific area, and the external measuring instrument measures the can detect the current of the local specific area, and the external measuring instrument measures the can detect the current of the local specific area, and the external measuring instrument measures the voltage difference and the resistivity. voltage difference and the resistivity. voltage difference and the resistivity. voltage difference and the resistivity.
Figure 5. Design of micro current sensor. Figure 5. Design of micro current sensor. Figure 5. 5. Design Design of of micro micro current current sensor. sensor. Figure Figure 5. Design of micro current sensor.
Sensors 2016, 16, 1731
4 of 10
Sensors 2016, 16, 1731 Sensors 2016, 16,Product 1731 2.1.6. Finished of Five-in-One Micro-Sensor
2.1.6. Finished Product of Five-in-One Micro-Sensor
4 of 10 4 of 10
Figure 6 shows thethe real product optical ofthe thefive-in-one five-in-onemicro-sensor. micro-sensor. 2.1.6. Finished Product ofreal Five-in-One Micro-Sensor Figure 6 shows productand and opticalmicrograph micrograph of A 4Ain4 in stainless steel substrate can be made into seven five-in-one micro-sensors. The five-in-one micro-sensor stainless steel substrate can product be madeand intooptical sevenmicrograph five-in-one of micro-sensors. The five-in-one A microFigure 6 shows the real the five-in-one micro-sensor. 4 in has stainless many characteristics—compactness, resistance against acid corrosion, favorable thermal sensorimportant has many important characteristics—compactness, resistance against acid corrosion, favorable steel substrate can be made into seven five-in-one micro-sensors. The five-in-one microtolerance, a tolerance, short response and the to ability make measurements in real thermal a shorttime, response time,ability and the toresistance make measurements in real time. sensor has many important characteristics—compactness, against acid time. corrosion, favorable thermal tolerance, a short response time, and the ability to make measurements in real time.
Figure 6. Real product of five-in-one five-in-onemicro-sensor. micro-sensor. Figure 6. Real productand andoptical opticalmicrograph micrograph of Figure 6. Real product and optical micrograph of five-in-one micro-sensor.
2.1.7.2.1.7. Sensor Calibration Sensor Calibration When the five-in-one micro-sensor is inserted into the HT-PEMFC stack to make local 2.1.7. Sensor When the Calibration five-in-one micro-sensor is inserted into the HT-PEMFC stack to make local measurements in real time, micro-sensor corrections must be made for the its temperature, flow andlocal the When the five-in-one is inserted HT-PEMFC the stack to rate make measurements in real time, corrections must be madeinto for its temperature, the flow rate and the pressure, to yield that correspond physical quantities. has measurements in electrical real time,measurements corrections must be made fortoits temperature, theAfter flow the ratedevice and the pressure, to yield electrical measurements that correspond to physical quantities. After the device has been calibrated times,measurements the mean value is correspond obtained and reliability of its signals is tested. pressure, to yieldthree electrical that to the physical quantities. After the device has been calibrated three times, the mean value is obtained and theHT-PEMFC reliability of its signals isdirectly tested. In this work, the external temperature controller of the stack is used been calibrated three times, the mean value is obtained and the reliability of its signals is tested. to In this work, the external temperature controller of the HT-PEMFC stack is used directly provide the work, benchmark to calibrate the temperature of the five-in-one micro-sensors, as presented in to In this the external temperature controller of the HT-PEMFC stack is used directly to provide the benchmark totocalibrate the micro temperature five-in-one as the presented Figure 7.the The calibration curves of the temperature sensors thatmicro-sensors, aremicro-sensors, embeddedasinside HTprovide benchmark calibrate temperature ofof thethe five-in-one presented in in Figure 7. The calibration curves of the micro temperature sensors that are embedded inside PEMFC are obtained fromof three calibrations at temperatures from C to 190inside C. The Figure 7.stack The calibration curves the micro temperature sensors that are140 embedded themicro HT- the ◦ to 190 ◦ C. The micro HT-PEMFC stacksensor areobtained obtained fromthree three calibrations at temperatures from temperature has an from accuracy better than 0.5atC. PEMFC stack are calibrations temperatures from 140140 C toC190 C. The micro ◦ temperature sensor has anan accuracy C. temperature sensor has accuracybetter betterthan than 0.5 0.5 C.
Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of the five-in-one micro-sensors. Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of Figure 7. HT-PEMFC stack as the temperature calibration benchmark to calibrate the temperature of the five-in-one micro-sensors.
the five-in-one Regarding micro-sensors. the calibration based on pressure, a Wayne Kerr Electronics (London, UK) 4230 LCR
(Inductance, Capacitance, Resistance) was used to capture the relevant data. The range the Regarding the calibration based onmeter pressure, a Wayne Kerr Electronics (London, UK) 4230of LCR Regarding Capacitance, the basedusing onmeter pressure, a Wayne Kerr Electronics UK) 4230 capacitances thatcalibration can be measured the was instrument 0.01 pF~1 F,relevant and its (London, precision ±0.1%. (Inductance, Resistance) used toiscapture the data. The is range of The theLCR (Inductance, Capacitance, Resistance) meter relevant data. The rangeThe of the micro pressure an accuracy better thanused 0.01 to kgf/cm . the capacitances thatsensor can behas measured using the was instrument is capture 0.01 2pF~1 F, and its precision is ±0.1%. 2 The micro flow rate sensors can be calibrated in a stable temperature field in air which the power capacitances that can be measured using the instrument is 0.01 pF~1 F, and its precision is ± 0.1%. micro pressure sensor has an accuracy better than 0.01 kgf/cm . 2 supply generates at the given voltage. The range of flow rates in the calibration is 0~30 L/min. The The micro sensor has an accuracy better than 0.01 kgf/cm . field in air which the power Thepressure micro flow rate sensors can be calibrated in a stable temperature micro flow rate sensor an accuracy ofcalibrated better 0.1aL/min. supply generates atrate thehas given voltage. The rangethan of flow rates in temperature the calibrationfield is 0~30 The the The micro flow sensors can be in stable in L/min. air which micro flow rate sensor at has angiven accuracy of better 0.1of L/min. power supply generates the voltage. Thethan range flow rates in the calibration is 0~30 L/min.
The micro flow rate sensor has an accuracy of better than 0.1 L/min.
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
5 of 10 5 of 10
3. Test TestResult Resultof ofConstant ConstantCurrent Current(5, (5,13, 13,20 20 A) A) Output Output An experiment onon thethe HT-PEMFC stack at a stable temperature and anand open experimentisiscarried carriedout out HT-PEMFC stack at a stable temperature ancircuit open voltage (OCV).(OCV). In thisInexperiment, the operating temperature of theoffuel cell cell stackstack is 160 C, ◦as circuit voltage this experiment, the operating temperature the fuel is 160 C, 2 2 2 2 indicated in in Table 1; 1; the size as indicated Table the sizeofoffuel fuelcell cellstack stackisis33 33cm cm and andthe thereaction reaction area area is is 31.4 cm . The The layers are made made of ofgraphite graphitebecause becauseit itresists resists corrosion. Five-in-one micro-sensors installed upstream corrosion. Five-in-one micro-sensors are are installed upstream and and downstream in Cells 1, 510, and as displayed in Figure 8. The machine gives theand anode and downstream in Cells 1, 5 and as 10, displayed in Figure 8. The machine gives the anode cathode cathode fixed, not-humidified gases H (anode H2: 5cathode slpm; cathode 30 slpm), and different loads are fixed, not-humidified gases (anode Air: 30Air: slpm), and different loads are given 2 : 5 slpm; given a constant (5,A); 13,the 20 output A); the time output time 60 min. local temperature, voltage, a constant currentcurrent (5, 13, 20 is 60 min.is The localThe temperature, voltage, pressure, pressure, flow rate and in the HT-PEMFC stack are obtained by real-time continuous flow rate and current in thecurrent HT-PEMFC stack are obtained by real-time continuous measurement using measurement an NI PXI 2575 acquisitionfive-in-one unit. The embedded five-in-one micro-sensors an NI PXI 2575using data acquisition unit. data The embedded micro-sensors inside the HT-PEMFC inside the HT-PEMFC stack reaction area of Electrode the MEA (Membrane Electrode to stack cover the reaction areacover of thethe MEA (Membrane Assembly) to prevent Assembly) the gas from prevent the gas from reacting with the covered (the HT-MEA used inavailable this study is a reacting with the covered MEA (the HT-MEA used inMEA this study is a commercially Advent commercially available Advent Energy Company’s (Patras, Greece) specifications are Energy Company’s (Patras, Greece) HT-MEA; the specifications are HT-MEA; in Table 2).the However, since the in Table 2). micro-sensors However, since five-in-one micro-sensors are very small, their total areareaction equals only five-in-one arethe very small, their total area equals only 1.34% of the MEA area. 1.34% of the MEA reactionmicro-sensors area. The embedded five-in-one micro-sensors in the HT-PEMFC stack The embedded five-in-one in the HT-PEMFC stack alter the performance of the stack by alter the1.3%. performance of the stack by around 1.3%. around Table 1. 1. Testing Testing conditions Table conditions for for the the HT-PEMFC HT-PEMFC stack. stack.
Subject Subject Temperature stack(◦(C) Temperature of of the the stack C) Cathode flow rate (Air) (slpm) Cathode flow rate (Air) (slpm) Anodeflow flow rate rate (H Anode (H22)) (slpm) (slpm) Constant current Constant current (A) (A) Gas temperature Gas temperature Reaction area (cm2 ) Reaction area (cm2)
Condition Condition 160 160 3030 55 5,5,13, 13,2020 Room temperature Room temperature 31.4 31.4
Table 2. Specifications of the HT-MEA. Table 2. Specifications of the HT-MEA. Subject Subject HT-MEA HT-MEA Membrane thickness Membrane thickness Conductivity Conductivity
Specificationsis Specificationsis Advent energy Advent energy 60~65 µm 60~65 μm −2 S/cm 88××10 −2 10 S/cm
Figure channel plates. plates. Figure 8. 8. Position Position of of five-in-one five-in-one micro-sensors micro-sensors in in cathode cathode channel
3.1. Local Temperature and Voltage Distributions in Different Cells 3.1. Local Temperature and Voltage Distributions in Different Cells The external voltages of different cells and the local temperature and voltage inside the HTThe external voltages of different cells and the local temperature and voltage inside the HT-PEMFC PEMFC stack, measured between the probe (inserted in the cathode flow-field) and the anode plate stack, measured between the probe (inserted in the cathode flow-field) and the anode plate at at an operating temperature of 160 C, are discussed and analyzed as follows. an operating temperature of 160 ◦ C, are discussed and analyzed as follows. 1. Figure 9 compares the internal and external voltages. The voltage curve is smooth at a constant current (5, 13 A), because the electrochemical reaction is uniform. At a high current (20 A), the internal electrochemical reaction is violent, and the voltage distribution is nonuniform. The
be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction is vigorous at a high current (20 A), and heat is concentrated, causing Cell 5 to be the hottest. Sensors 2016, 16, 1731
1
6 of 10
Figure 10 compares the internal and external voltages. The voltage curve is smooth at a constant current (5, 13 A), because the electrochemical reaction is uniform. At a high current (20 A), the internal electrochemical reaction is violent, and the voltage distribution is nonuniform. The internal local voltage is measured using a micro voltage sensor. The curve of the variation of the internal voltage is consistent with that of the external voltage. Figure 9 plots the local temperature and voltage in different cells. The temperature increases gradually with the current (5, 13, 20 A), because a higher operating current causes more heat to be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction is vigorous at a high current (20 A), and heat is concentrated, causing Cell 5 to be the hottest.
2
Figure 9. Comparison of internal and external voltages.
Sensors 2016, 16, 1731
2.
6 of 10
internal local voltage is measured using a micro voltage sensor. The curve of the variation of the internal voltage is consistent with that of the external voltage. Figure 10 plots the local temperature and voltage in different cells. The temperature increases gradually with the current (5, 13, 20 A), because a higher operating current causes more heat to be released from the electrochemical reaction, and as the current increases to a high value (20 A), the thermal nonuniformity becomes gradually worse. Cell 5 has the highest temperature, perhaps because Cell 5 is located in the center of the HT-PEMFC stack. The gases on both ends transfer heat from the front end of the stack to the tail end. The internal electrochemical reaction Figure 10. Comparison of local temperatures and voltages in different cells. 9. Comparison local and voltages in different is vigorousFigure at a high current (20ofA), andtemperatures heat is concentrated, causing Cell 5cells. to be the hottest.
3.2. Local Pressure Distribution in Different Cells Figure 11 plots the local pressure in different cells, and the pressure control in the experimental work is 1.3 kgf/cm2. The local pressure in the HT-PEMFC stack does not vary as the current (5, 13, 20 A) increases. The upstream pressure slightly exceeds the downstream pressure. The mean pressure in the HT-PEMFC stack is 1.291 kgf/cm2. The upstream end pressure of Cell 1 and Cell 10 is unstable, but is stabilized at the downstream end. The difference between the end pressure upstream and that downstream in Cell 5 is small, so the pressure in the HT-PEMFC stack is stable. This can be used to assess whether the design of the flow field inside the cell stack is good or not.
Figure 9. Comparison of internal and external voltages. Figure 10. Comparison of internal and external voltages.
3.2. Local Pressure Distribution in Different Cells Figure 11 plots the local pressure in different cells, and the pressure control in the experimental work is 1.3 kgf/cm2 . The local pressure in the HT-PEMFC stack does not vary as the current (5, 13, 20 A) increases. The upstream pressure slightly exceeds the downstream pressure. The mean pressure in the HT-PEMFC stack is 1.291 kgf/cm2 . The upstream end pressure of Cell 1 and Cell 10 is unstable,
Sensors 2016, 16, 1731
7 of 10
but is stabilized at the downstream end. The difference between the end pressure upstream and that downstream in Cell 5 is small, so the pressure in the HT-PEMFC stack is stable. This can be used to Sensors 2016, 16, 1731 7 of 10 assess whether the design of the flow field inside the cell stack is good or not.
Figure 11.11. Comparison of of local pressures in in different cells. Figure Comparison local pressures different cells.
3.3. Local Local Flow Flow Distribution Distribution in in Different Different Cells Cells 3.3. Figure1212displays displays flow in different cells. The upstream flow rate exceeds the Figure the the locallocal flow in different cells. The upstream flow rate exceeds the downstream downstream flow rate, and the flow rates in Cell 1 and Cell 10 exceed those in Cell 5. Since the flow rate, and the flow rates in Cell 1 and Cell 10 exceed those in Cell 5. Since the upstream flowing upstream flowing in Cell 1 and Cell is close inlet, the gas is sufficient. Theflow downstream flow in Cell 1 and Cell 10 is close the air10 inlet, the the gasair is sufficient. The downstream rate is lower, rate is lower, because multiple snakelike runners (commonly called as “serpentine flow channels”) because multiple snakelike runners (commonly called as “serpentine flow channels”) are used to are used the to improve the performance of the stack, HT-PEMFC stack, and distribution the internal of distribution ofgas the improve performance of the HT-PEMFC and the internal the internal internal gas is nonuniform. is nonuniform.
Figure 12. 12. Comparison Comparison of of local local flow flow rates rates in in different differentcells. cells. Figure
3.4. Local Current Distribution in Different Cells 3.4. Local Current Distribution in Different Cells Figures 13–16 plot the local current density in different cells and this in-plane current does not Figures 13–16 plot the local current density in different cells and this in-plane current directly the through-plane current flow through the catalyst layers and PEM (Proton does not represent directly represent the through-plane current flow through the catalyst layers and PEM Exchange Membrane). The current density in the HT-PEMFC stack varies with the current (5, 13, 20 (Proton Exchange Membrane). The current density in the HT-PEMFC stack varies with the current A). The current density upstream exceeds that downstream, because the gas is present at the (5, 13, 20 A). The current density upstream exceeds that downstream, because the gas is present at upstream inletinlet to initiate a electrochemical reaction, but the serpentine flow channels cause the upstream to initiate a electrochemical reaction, butmultiple the multiple serpentine flow channels the internal gas distribution to be nonuniform. The Nernst loss due to gas (especially oxygen) cause the internal gas distribution to be nonuniform. The Nernst loss due to gas (especially oxygen) depletion increases increases toward towardthe thedownstream downstreampart. part. depletion
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
8 of 10
8 of 10
8 of 10
Figure 13. Comparison of local in different cells. cells. Figure13. 13.Comparison Comparison of of local local current current densities densities Figure current densitiesinindifferent different cells.
Figure 14. Comparison of local current densities in different cells (part of 5 A). Figure14. 14.Comparison Comparisonof oflocal local current current densities ofof 5 A). Figure densitiesin indifferent differentcells cells(part (part 5 A).
Figure 15. Comparison of local current densities in different cells (part of 13 A). Figure15. 15.Comparison Comparisonof oflocal local current current densities ofof 1313 A).A). Figure densitiesin indifferent differentcells cells(part (part
Sensors 2016, 16, 1731 Sensors 2016, 16, 1731
9 of 10
9 of 10
Figure 16.Comparison Comparisonof oflocal localcurrent current densities densities in 20A). Figure 16. indifferent differentcells cells(part (partofof 20A).
4. Conclusions 4. Conclusions thiswork, work,MEMS MEMS technology to to develop a five-in-one micro-sensor on a 40-µm-thick InIn this technologyisisutilized utilized develop a five-in-one micro-sensor on a 40-μmstainless steel substrate. The protective layer is made of PI (Polyimide) with a favorable temperature thick stainless steel substrate. The protective layer is made of PI (Polyimide) with a favorable tolerance. This five-in-one micro-sensor has five functions. It is characterized by its compactness, temperature tolerance. This five-in-one micro-sensor has five functions. It is characterized by its resistance to acid corrosion, favorable temperature tolerance, rapid response and usefulness in compactness, resistance to acid corrosion, favorable temperature tolerance, rapid response and real-time measurement. usefulness in real-time measurement. The local temperature, flow and pressure information inside the HT-PEMFC stack are extracted The local temperature, flow and pressure information inside the HT-PEMFC stack are extracted successfully. The test result of the operating temperature of 160 ◦ C and the constant current (5, 13, successfully. test resulttemperature of the operating temperature of 160 C HT-PEMFC and the constant 20 A) showsThe nonuniform and flow distributions in the stack. current (5, 13, 20 A) shows andthe flow distributions in the HT-PEMFC Thenonuniform experimentstemperature indicated that nonuniform temperature and flowstack. distributions in the The experiments indicated that the nonuniform temperature and flow distributions the HTHT-PEMFC stack were responsible for a large internal temperature difference, nonuniformin voltage PEMFC stack distributions, were responsible largeThe internal temperature difference, stack nonuniform voltage and and current and afor hotastack. pressure inside the HT-PEMFC was stable, but the current distributions, and a hot stack. The pressure inside the HT-PEMFC stack was stable, but the air inlet upstream pressure was unstable at both the cathode and the anode.
air inlet upstream pressure was unstable at both the cathode and the anode. Acknowledgments: This work was accomplished with much needed support and the authors would
Acknowledgments: work of was accomplished with much and thesupport authorsthrough would like like to thank the This Ministry Science and Technology of needed R. O. C.support for financial the to grant MOST 102-2221-E-155-033-MY3, 103-2622-E-155-006-CC2, 104-2623-E-155-004-ET, 103-2622-E-155-018-CC2, thank the Ministry of Science and Technology of R. O. C. for financial support through the grant MOST 102104-2622-E-155-004, 104-2622-E-155-007-CC2, 105-ET-E-155-002-ET, 105-2622-8-155-003-TE3, 105-2622-E-155-013-CC2 2221-E-155-033-MY3, 103-2622-E-155-006-CC2, 104-2623-E-155-004-ET, 103-2622-E-155-018-CC2, 104-2622-Eand 105-2221-E-155-005. The authors also like to thank Shih Hung Chan, Ay Su and Guo-Bin Jung of Yuan Ze 155-004, 104-2622-E-155-007-CC2, 105-ET-E-155-002-ET, 105-2622-8-155-003-TE3, 105-2622-E-155-013-CC2 and University for their valuable advice and assistance in the experiments. In addition, we would like to thank the 105-2221-E-155-005. The authors also like to thank Shih Hung Chan, Ay Su and Guo-Bin Jung of Yuan Ze YZU Fuel Cell Center and NENS Common Lab for providing access to their research facilities. University for their valuable advice assistance in was the experiments. In addition, we would likealltoauthors. thank the Author Contributions: The workand presented here carried out in collaboration between YZU Fuel Cell and NENS Lab for providing access this to their research facilities. Chi-Yuan LeeCenter and Fang-Bor WengCommon conceived, designed and discussed study. Yzu-Wei Kuo, Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin performed the experiments and analyzed the data. All authors
Author Contributions: The and work presented here was carried out in collaboration between all authors. have contributed, reviewed improved the manuscript. Chi-Yuan Lee and Fang-Bor Weng conceived, designed and discussed this study. Yzu-Wei Kuo, Conflicts of Interest: The authors declare no conflict of interest. Chao-Hsuan Tsai, Yen-Ting Cheng, Chih-Kai Cheng and Jyun-Ting Lin performed the experiments and analyzed the data. All authors have contributed, reviewed and improved the manuscript.
References
Conflicts of Interest: The authors declare no conflict of interest. 1.
Bureau of Energy. Ministry of Economic Affairs. Available online: http://web3.moeaboe.gov.tw/ecw/ populace/content/SubMenu.aspx?menu_id=2324 (accessed on 4 May 2014). References 2. Tan, Z.; Zhang, C.; Liu, P.; Reed, B.; Zhao, J. Focus on fuel cell systems in China. Renew. Sustain. Energy Rev. 2015, 47, 1. Bureau of 912–923. Energy. Ministry of Economic Affairs. Available online: http://web3.moeaboe.gov.tw/ecw/ 3. populace/content/SubMenu.aspx?menu_id=2324 Weng, F.B.; Jou, B.S.; Su, A.; Chan, S.H.; Chi, P.H. (accessed Design, fabrication performance analysis of a 200 W on 4 Mayand 2014). PEM fuel cell short stack. J. Power Sources 2007, 171, 179–185. [CrossRef] 2. Tan, Z.; Zhang, C.; Liu, P.; Reed, B.; Zhao, J. Focus on fuel cell systems in China. Renew. Sustain. Energy Rev. 4. Janssen, H.; Supra, J.; Lueke, L.; Lehnert, W.; Stolten, D. Development of HT-PEFC stacks in the kW range. 2015, 47, 912–923. Int. J. Hydrogen Energy 2013, 38, 4705–4713. [CrossRef]
3. 4.
Weng, F.B.; Jou, B.S.; Su, A.; Chan, S.H.; Chi, P.H. Design, fabrication and performance analysis of a 200 W PEM fuel cell short stack. J. Power Sources 2007, 171, 179–185. Janssen, H.; Supra, J.; Lueke, L.; Lehnert, W.; Stolten, D. Development of HT-PEFC stacks in the kW range. Int. J. Hydrogen Energy 2013, 38, 4705–4713.
Sensors 2016, 16, 1731
5. 6. 7.
8. 9.
10.
11. 12.
10 of 10
Zhang, J.; Xie, Z.; Zhang, J.; Tang, Y.; Song, C.; Navessin, T.; Shi, Z.; Song, D.; Wang, H.; Wilkinson, D.P.; et al. High temperature PEM fuel cells. J. Power Sources 2006, 160, 872–891. [CrossRef] Li, Q.; Jensen, J.O.; Savinell, R.F.; Bjerruma, N.J. High temperature proton exchange membranes based on polybenzimidazoles for fuel cells. Prog. Polym. Sci. 2009, 34, 449–477. [CrossRef] Chandan, A.; Hattenberger, M.; El-Kharouf, A.; Du, S.F.; Dhir, A.; Self, V.; Pollet, B.G.; Ingram, A.; Bujalski, W. High temperature (HT) polymer electrolyte membrane fuel cells (PEMFC)—A review. J. Power Sources 2013, 231, 264–278. [CrossRef] Harms, C.; Köhrmann, F.; Dyck, A. Study of the influence of key test parameters on the performance of a PEMFC stack. Solid State Ionics 2015, 275, 75–79. [CrossRef] Lee, C.Y.; Weng, F.B.; Cheng, C.H.; Shiu, H.R.; Jung, S.P.; Chang, W.C.; Chan, P.C.; Chen, W.T.; Lee, C.J. Use of flexible micro-temperature sensor to determine temperature in situ and to simulate a proton exchange membrane fuel cell. J. Power Sources 2011, 196, 228–234. [CrossRef] Weng, F.B.; Cheng, C.K.; Lee, C.Y.; Chang, C.P. Analysis of thermal balance in high-temperature proton exchange membrane fuel cells with short stacks via in situ monitoring with a flexible micro sensor. Int. J. Hydrogen Energy 2014, 39, 13681–13686. [CrossRef] Lee, C.Y.; Weng, F.B.; Huang, Y.P.; Chang, C.P.; Cheng, C.K. Real-time monitoring of internal temperature and voltage of high-temperature fuel cell stack. Electrochim. Acta 2015, 161, 413–419. [CrossRef] Bose, S.; Kuila, T.; Nguyen, T.X. H.; Kim, N.H.; Lau, K.T.; Lee, J.H. Polymer membranes for high temperature proton exchange membrane fuel cell: Recent advances and challenges. Prog. Polym. Sci. 2011, 36, 813–843. [CrossRef] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).
