Open Access
Add to BibSonomy
Abstract
The dehydration of electrolyte membranes in polymer electrolyte fuel cells (PEFCs) operating under low-humidity conditions is a critical issue for achieving their high efficiency and high power density. To reduce the membrane dryout, it’s necessary to investigate and control the water transport within working fuel cells. This study developed a single-ended fiber-optic sensor based on tunable diode laser absorption spectroscopy (TDLAS) and applied it to the real-time monitoring of the water vapor concentration in the narrow flow channel of a PEFC. The newly proposed wavelength modulation spectroscopy (WMS) technique enabled to quantify the mole fraction of water in the channel over the wide concentration range with high accuracy. The in-situ TDLAS measurement in the PEFC during a low-humidity and load-change operation revealed that the dynamic change of cell voltage is strongly correlated to the dry-wet transition in the anode channel.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Polymer electrolyte fuel cells (PEFCs) have been attracting attention as the next-generation power source for transportation and stationary applications. However, there are still technical issues to be solved for achieving their high efficiency, long-term durability and cost reduction. Especially, water management in PEFCs is recognized as a critical factor influencing their performance [1–9]. For reduced system complexity and compactness, it is desirable to operate PEFCs under low-humidity conditions without the use of external humidifiers. During low-humidity operations, the electro-osmotic transport of water from the anode to cathode through the electrolyte membrane induces the membrane dehydration on the anode side, resulting in the deterioration of proton conductivity [4,7,8]. To alleviate this issue called “dryout”, it is essential to deeply understand and control the water transport inside operating fuel cells.
In the previous studies, the water vapor distribution in flow channels of PEFCs has been widely investigated using a micro gas chromatograph (µGC) [10–15]. Mench et al. [10] first used a micro GC system to directly map the in-situ distributions of water and other species within the anode and cathode channel of an operating fuel cell with a time resolution of approximately 2 minutes. Yang et al. [11] and Dong et al. [12] attempted the simultaneous measurements of water, species and current distributions in a low-humidity PEFC using a segmented cell. They disclosed that the anode-side water profile has strong effect on the current distribution and the overall performance under low-humidity conditions. GC technique enables to provide more accurate species concentrations, but on the other hand it takes a long time (at least two minutes) to collect a gas sample and identify a species. The special resolution is also limited by the proximity of sample extraction ports located in gas channels. As an alternative approach, several researchers have attempted to apply tunable diode laser absorption spectroscopy (TDLAS) to the in-situ measurement of the water transport in working fuel cells [16–22]. TDLAS is a high-sensitive laser absorption spectroscopy which detects the absorbance of a chemical species and identifies its concentration at a sub-second time scale using infrared diode laser. Basu et al. [16] performed the simultaneous measurements of water partial pressure and temperature in a gas channel of a PEFC under dynamic operating conditions by monitoring the laser transmission and water absorption through the flow passage. They found that the water partial pressure rapidly responds to the load changes and follows the current profile. Fujii et al. [21] developed a single-ended back-reflection TDLAS sensor for measuring the water vapor and oxygen concentration in a PEFC channel. To allow the laser access to the flow channel, a part of the experimental fuel cell was replaced with an anti-reflective IR transparent window. Furthermore, to secure the sufficient optical path and reduce the multiple reflection, the channel depth and width were set to 5.0 mm and 3.0 mm that are much wider than those of standard cells. Other than fuel cell diagnostics, TDLAS technique has been widely used in various applications such as combustion diagnostics [23–28], industrial sensors [29–32], environmental monitoring [33–36] and biomedical diagnostics [37,38].
Compared to GC technique, TDLAS has a feature of quantifying species concentration with a high temporal resolution (< 1 s) and enables to grasp the transient water transport in PEFCs under load-change operations. However, since millimeter-scale gas channels inside fuel cells which are the measurement region are extremely narrow, the multiple reflection of the injected laser beam tends to generate the interference fringe noise. In addition, it is difficult to keep a sufficient optical path length. To realize a high-accuracy TDLAS measurement in narrow channels of fuel cells, it is necessary to improve the signal-to-noise (S/N) ratio of output signals. This study first developed a single-ended fiber-optic gas sensor based on TDLAS for monitoring the water vapor concentration in gas flow channels of operating PEFCs. The reflection-type fiber-bundle probe developed by the authors is compactly composed of one single-mode pitch fiber and six multi-mode catch fibers which are arranged coaxially, and enables the in-situ local laser spectroscopy within millimeter-scale narrow channels. The incident beam emitted from a distributed feedback (DFB) laser diode was directly injected into a fuel cell channel through the pitch fiber and the diffuse-reflected lights from the porous electrode were received by the six catch fibers. To improve the signal-to-noise (S/N) ratio of the output, the multiple signals detected by photodiodes were synthesized and the optical fringe noises were effectively canceled. The water vapor concentration in the channel was successfully quantified using the newly proposed wavelength modulation spectroscopy (WMS-2f/4f) technique with a wide concentration range. In the study, this gas sensor was applied to the in-situ measurement of water distribution in the anode channel of an operating PEFC and the correlation between the dynamic response of cell voltage and the dry-wet transition was evaluated under a low-humidity and load-change condition.
2. Measurement principle
2.1 WMS-based TDLAS technique
Several molecules such as hydrocarbons, water and carbon dioxide have the property of absorbing infrared lights at specific wavelengths. TDLAS is one of the laser absorption spectroscopy techniques which measure the absorbance of a chemical species and identifies its concentration using a diode laser. The infrared absorption due to a species is based on Lambert-Beer's law:
where ${I_0}(\lambda )$ is the intensity of incident light, $I(\lambda )$ is the intensity of transmitted light, $\alpha (\lambda )$ is the absorption coefficient, and L is the optical path length. This study applied a high-speed and high-sensitive TDLAS with wavelength modulation spectroscopy (WMS), which adds the high-frequency current modulation to the fundamental current sweep. The basic principle of WMS-based TDLAS approach is shown in Fig. 1. To detect the absorption spectrum of water vapor at 1392 nm in this study, the laser wavelength was swept across the selected absorption line by tuning the laser drive current with a 10 Hz triangular waveform. Additionally, the high-frequency wavelength modulation of f = 10 kHz was superimposed on the triangular wave of incident light. After the detected signal was transported to a lock-in amplifier, the harmonic signals can be extracted from the output by phase-sensitive detection. The n-th harmonic component, $S({{\lambda_0}} )$, is expressed as follows [39]:
Fig. 1. Basic principle of WMS-based TDLAS technique: (a) triangular scanning modulation of incident laser beam, (b) detected signal.
2.2 Quantification of water concentration
The amplitudes of even-order harmonic spectra closely correlate with the concentration of a target species. In this study, to quantitatively estimate the mole fraction of water vapor over a wide concentration range, both the second-harmonic (2f) and the fourth-harmonic (4f) spectra were detected using a lock-in amplifier. Figures 2(a)–2(d) present the 2f and 4f spectra of water in atmospheric air at the wavelength of 1392 nm simulated using HITRAN2012 database [40]. The absorption line at 1392 nm selected in this study has been often used in the fields of the water detection based on near-infrared TDLAS [18,19,21,24–26,36]. Furthermore, a 1.39 µm near-infrared fiber-coupled DFB diode laser manufactured by NTT Electronics is available to the fiber-optic moisture sensor developed in this study. The vertical axis shows the intensity of 2f or 4f signal and the horizontal axis shows the wavelength. All spectral analyses were performed at 70 °C that is the typical operating temperature of PEFCs. The partial pressure of water vapor was changed from 0.003 to 0.257 atm. At the low pressure of 0.003-0.089 atm, the difference in height between the peak and valley value of 2f spectrum increases with an increase in pressure. On the other hand, when the partial pressure becomes higher than 0.089 atm, the peak-valley height of 2f gradually decreases with increasing the pressure due to the pressure broadening of absorption spectra. As shown in Figs. 2(c) and 2(d), the dependence of pressure on the peak-valley height of 4f spectrum also has the same tendency.
Fig. 2. The n-th harmonic (nf) spectra of water at the wavelength of 1392 nm simulated using HITRAN database [40]: (a) 2f spectra at low pressure, (b) 2f spectra at high pressure, (c) 4f spectra at low pressure, and (d) 4f spectra at high pressure. The partial pressure of water was varied from 0.003 to 0.257 atm at the temperature of 70 °C.
Figures 3(a) and 3(b) summarize the variations of the simulated peak-valley heights of 2f and 4f spectrum with the partial pressure of water vapor. Since the both peak-valley heights of 2f and 4f signal doesn’t change linearly for variation of pressure, the water concentration cannot be quantitatively estimated only by the 2f and 4f spectrum data. To quantify the water concentration, this study newly proposed the WMS-2f/4f technique with the use of both 2f and 4f data. Figure 3(c) shows the relationship between the ratio of 2f and 4f peak-valley height (2fpv/4fpv) and the partial pressure of water. It was noted that the peak-valley ratio of 2f and 4f monotonically increases with an increase in pressure. This relationship can be approximately expressed by a cubic function over the wide pressure range and applied to the quantification of water vapor concentration. In addition, the normalization of 2fpv/4fpv eliminates the non-absorption laser transmission losses such as beam steering and scattering.
Fig. 3. (a) 2f peak-valley height vs. partial pressure of water vapor at the wavelength of 1392 nm and the temperature of 70 °C. (b) 4f peak-valley height vs. water vapor pressure. (c) The ratio of 2f and 4f peak-valley height (2fpv/4fpv) vs. water vapor pressure.
3. Experimental setup
Figure 4 shows the schematic diagram of the TDLAS-based single-ended fiber-optic gas sensor developed in this study. The fiber-optic sensing technology enables in-situ laser diagnostics within millimeter-scale narrow channels of fuel cells. A single-ended fiber bundle probe was newly developed to accurately detect the absorption spectra of water and estimate its concentration in flow channels of PEFCs. This probe compactly composed of one single-mode pitch fiber (core material: silica, type: SM10/125, numerical aperture (NA): 0.1) and six multi-mode catch fibers (core material: silica, type: GI62.5/125, NA: 0.2) was inserted into the metal separator of an operating PEFC. At the end of probe, the gap between the outer housing and optical fibers was filled with an adhesive to prevent the inflow of gas inside the probe from the fuel cell channel. Furthermore, the gap between the insertion hole of the separator and the probe was sealed with a sealing tape to prevent the gas leak from the channel. The triangular modulated beam (triangular sweep frequency: 10 Hz, modulation frequency: 10 kHz) emitted from a DFB laser diode (NTT Electronics Inc., Wavelength: 1392.5 nm) was injected into the gas channel through the pitch fiber. The back-reflection light from the electrode surface was captured by six catch fibers and converted to electric signals by six photodiodes (PDs). After six signals were synthesized, the composite output signal was transmitted to a lock-in amplifier and the high-sensitive harmonic spectra of 2f and 4f were obtained by phase-sensitive detection.
Fig. 4. Schematic diagram of the TDLAS-based single-ended fiber-optic gas sensor for monitoring the water vapor concentration in fuel cell channels.
The laser spectroscopy performed in narrow channels of fuel cells is easily affected by the interference fringe noise due to the multiple reflection of the laser beam at the channel walls and the electrode surface. To reduce the interference fringes included in the output signal, this study adopted the noise cancellation scheme as shown in Fig. 5 after the PD detection. In the measurement system, six electric signals converted at six PDs were separately amplified using amplifiers and their gains were adjusted to become the same level. After that, the six gain-adjusted signals were summed by using the adding circuit. The signal synthesis process enhances the intensity of the absorption spectra of 2f or 4f at the specific wavelength, on the other hand, it doesn’t increase the amplitude of fringe noise because the frequencies of fringes in six signals are different. Thus, this scheme enables to improve the S/N ratio of the output signal. It may be preferable to increase the storage number of catch fibers in the probe to enhance the noise cancellation. However, due to the space limitation inside the outer housing of the probe, the number of catch fibers was set to six.
Fig. 5. Noise cancellation scheme of the output signal by summing six electric signals including different frequency noise.
4. Calibration of the TDLAS sensor
The authors first conducted the calibration of the TDLAS-based fiber-optic gas sensor shown in Fig. 4 using a simulated flow cell under non-operating conditions. Figures 6(a) and 6(b) show the photograph and outline drawing of the flow cell fabricated in this study, respectively. The carbon paper (Toray, TGP-H-060) for a PEFC electrode was sandwiched between two stainless steel plates. A straight narrow channel (30 mm length, 1.5 mm width, 1.5 mm depth) was formed in one of the two plates and a penetration hole was installed toward the channel to insert the fiber-optic probe. The laser beam emitted from the laser diode was directly injected into the flow channel in the through-plane direction. The incident beam was reflected at the surface of gas diffusion electrode (GDE) and returned to the probe. The total length of optical path is 3 mm. The humidified oxygen was fed to the gas channel at 1 atm, 70 °C and 200 mL/min. In this experiment, the gas temperature inside the flow cell was strictly managed. Two stainless steel plates were held at a constant temperature by using a ribbon heater and temperature controller. Furthermore, it was confirmed that the gas flows at the cell inlet and outlet are kept at the target temperature of 70 °C using a thermocouple. The variation of gas temperature along the flow direction in the cell is negligibly small because the length of the straight channel is short. The inlet mole fraction of water vapor was controlled in the range of 0.003 to 0.257 using a humidifier. To calibrate the TDLAS sensor, the humidity and temperature of the supplied gas were simultaneously monitored by using a capacitive humidity sensor (Vaisala, HMI41).
Fig. 6. (a) Photograph and (b) outline drawing of the flow cell used for the calibration of the TDLAS sensor.
Figures 7(a)–7(d) present the 2f and 4f spectra of water vapor obtained from the TDLAS measurements in the simulated flow cell working at 1 atm and 70 °C. The mole fraction of water was changed from 0.003 to 0.257. At the low concentration of 0.003-0.089, the peak-valley height of 2f spectrum increases with an increase in water mole fraction. When the water concentration is higher than 0.143, the peak-valley height of 2f gradually decreases with increasing its mole fraction. As shown in Figs. 7(c) and 7(d), a similar tendency can also be seen with respect to the dependency of water concentration on the 4f peak-valley height. These agree very well with the results of spectra simulation using HITRAN database shown in Fig. 2(a)–2(d). To quantify the water concentration in the flow cell, the authors investigated the relationship between the ratio of 2f and 4f peak-valley height (2fpv/4fpv) and the water mole fraction (Fig. 8). It was confirmed that the value of 2fpv/4fpv monotonically increases with the mole fraction of water. The WMS-2f/4f technique enables to quantitatively estimate the mole fraction of water in narrow flow channels over the concentration range of 0.003-0.257 under the condition of 1 atm and 70 °C. The measurement accuracy was less than 0.005.
Fig. 7. The nf spectra of water at the wavelength of 1392 nm obtained from TDLAS measurement in the simulated flow cell working at 1 atm and 70 °C: (a) 2f spectra at low concentration, (b) 2f spectra at high concentration, (c) 4f spectra at low concentration, and (d) 4f spectra at high concentration. The mole fraction of water was varied from Xw = 0.003 to 0.257.
5. Measurement of water distribution in operating PEFC
During the operation of PEFCs under dry conditions, the insufficient supply of water unfortunately induces the dehydration of electrolyte membrane and deteriorates the proton conductivity through the membrane. Especially, when the output current rapidly rises in steps, the remarkable decrease of the water concentration on the anode side temporarily occurs due to the electro-osmotic drag and it causes the sudden drop of cell voltage called the “undershoot”. To alleviate this issue, it is necessary to investigate the dry-wet transient phenomena in PEFCs operated under low-humidity and load-change conditions. This study applied the TDLAS-based fiber-optic gas sensor to monitor the water concentration in the anode channel of a working PEFC in real time.
Figures 9(a) and 9(b) show the photograph of the experimental fuel cell used in this study and the schematic of the anode flow field, respectively. The five-layer membrane electrode assembly (MEA) constructed of an electrolyte membrane (Nafion NRE-212), two catalyst layers (CLs) and two gas diffusion layers (GDLs, TGP-H-060) was sandwiched between two current-collecting plates with a single-pass serpentine flow channel (1.0 mm width, 1.0 mm depth and 241 mm length). Two stainless steel separators placed outside the current collectors were held together by eight bolts and tightened with a controlled torque. The effective electrode area is 5.3 cm2. To insert the fiber-optic probe into the cell, thirteen penetration holes were installed in the anode separator at the fractional distance (x/L) of 0.03, 0.07, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 0.93 and 0.97. The TDLAS measurement of water vapor distribution along the anode channel can be performed by changing the probe position. The experimental cell was operated at 70 °C. Pure hydrogen and oxygen as the fuel and oxidant were supplied to the anode and cathode channel in the co-flow arrangement, respectively. The stoichiometric ratios of H2 and O2 were both 5.0 at 0.3 A/cm2. The stoichiometric ratio is defined as the ratio between the actual volumetric flow rate of a reactant at the fuel cell inlet and the volumetric consumption rate of the same reactant by the electrochemical reaction. The temperature management of gases in the fuel cell was strictly conducted by the same method as that in the flow cell shown in Fig. 6. Since the amount of H2 and O2 supplied to the cell inlet is much more than the amount consumed by the electrochemical reactions, the gas temperature along the flow direction is hardly changed by the reactions. The inlet gas humidifications of the anode and cathode were set to be 45 and 0%RH. In this experiment, to evaluate the dry-wet transient phenomena during a load-change operation, the repetitive cycling test of current density was performed between 0.2 and 0.6 A/cm2.
Fig. 9. (a) Photograph and (b) schematic of the anode flow field of the experimental fuel cell used in this study. The effective electrode area is 5.3 cm2. Thirteen holes were installed in the anode separator to insert the fiber-optic probe. Reprinted from [41] with permission from IOP Publishing.
Figures 10(a), 10(b) and 10(c) present the time-series variation of current density, the dynamic response of cell voltage and the transitional water concentrations along the anode channel (x/L = 0.03, 0.1, 0.2, 0.5, 0.8, 0.97) during a low-humidity and current-cycling operation. The operation test and TDLAS measurement were repeated six times by changing the insertion position of probe. As shown in Fig. 10(a), the current density was repeatedly changed in steps between 0.2 and 0.6 A/cm2 every 23 s. It can be seen from Fig. 10(c) that the mole fraction of water in the upstream region of anode (x/L = 0.03, 0.1, 0.2) hardly changes during the current cycling. On the other hand, the water concentration in the downstream (x/L = 0.5, 0.8, 0.97) is remarkably affected by the variation of current density. When the current density is steeply increased at t = 23 s, the large undershoot of cell voltage is observed (see Fig. 10(b)) because the water content at the anode side of the electrolyte membrane is suddenly decreased by the strong electro-osmotic effect and the proton conductivity through the membrane is deteriorated.
Fig. 10. (a) Time-series variation of current density, (b) dynamic response of cell voltage, and (c) transitional water concentration in the anode channel during the low-humidity and load-change operation of a PEFC.
It should also be noted that the water concentration in the anode drops temporarily at the downstream section (x/L = 0.5, 0.8, 0.97). After the undershoot, the cell voltage is gradually recovered because the back-diffusion of product water hydrates the membrane from the cathode to anode and reduces its ohmic resistance. The anode water concentration is also gradually recuperated by the back-diffusion effect. It was found that the output cell voltage has a strong correlation with the water vapor concentration in the anode channel under low-humidity and load-change operations. The high-speed TDLAS technique used in this study enables to investigate the instantaneous fluctuations of water concentration in narrow flow channels of operating fuel cells with high accuracy.
6. Conclusions
To alleviate the dryout in PEFCs operated under dry conditions, it is necessary to understand and control the water transport inside fuel cells. This study developed a single-ended fiber-optic gas sensor based on TDLAS technique to quantitatively measure the water vapor concentration in narrow flow channels of working PEFCs at high speed and with high accuracy. The WMS-2f/4f method with the use of both 2f and 4f spectra enabled to estimate the mole fraction of water in flow channels over the wide concentration range. Furthermore, the TDLAS-based fiber-optic sensor was applied to investigate the dry-wet transient phenomena in the anode channel of a PEFC under a low-humidity and load-change operation. The results reveled that the dynamic response of cell voltage during the current cycling is strongly correlated with the transient change of water vapor concentration in the anode. When the current density increased stepwise, the voltage undershoot and the temporal drop of water concentration in the anode simultaneously occurred because of the strong electro-osmotic transport of water from the anode to cathode through the electrolyte membrane and the deterioration of proton conductivity. After the undershoot, both the cell voltage and the water concentration in the anode were gradually recuperated by the back-diffusion of product water from the cathode and the membrane hydration.
Funding
JKA and its promotion funds from KEIRIN RACE (2022M-253).
Acknowledgments
This work was supported by JKA and its promotion funds from KEIRIN RACE.
Disclosures
The authors declare that there are no conflicts of interest related to this article.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. T. F. Fuller and J. Newman, “Water and Thermal Management in Solid-Polymer-Electrolyte Fuel Cells,” J. Electrochem. Soc. 140(5), 1218–1225 (1993). [CrossRef]
2. T. V. Nguyen and R. E. White, “A Water and Heat Management Model for Proton-Exchange-Membrane Fuel Cells,” J. Electrochem. Soc. 140(8), 2178–2186 (1993). [CrossRef]
3. P. Berg, K. Promislow, J. St Pierre, J. Stumper, and B. Wetton, “Water management in PEM fuel cells,” J. Electrochem. Soc. 151(3), A341–A353 (2004). [CrossRef]
4. W. Dai, H. J. Wang, X. Z. Yuan, J. J. Martin, D. J. Yang, J. L. Qiao, and J. X. Ma, “A review on water balance in the membrane electrode assembly of proton exchange membrane fuel cells,” Int. J. Hydrogen Energy 34(23), 9461–9478 (2009). [CrossRef]
5. J. P. Owejan, J. J. Gagliardo, J. M. Sergi, S. G. Kandlikar, and T. A. Trabold, “Water management studies in PEM fuel cells, Part I: Fuel cell design and in situ water distributions,” Int. J. Hydrogen Energy 34(8), 3436–3444 (2009). [CrossRef]
6. S. G. Kim and S. J. Lee, “A review on experimental evaluation of water management in a polymer electrolyte fuel cell using X-ray imaging technique,” J. Power Sources 230, 101–108 (2013). [CrossRef]
7. K. Nishida, “Optical Visualization and Spectroscopic Techniques for Probing Water Transport in a Polymer Electrolyte Fuel Cell,” ChemElectroChem 2(10), 1458–1470 (2015). [CrossRef]
8. P. C. Okonkwo and C. Otor, “A review of gas diffusion layer properties and water management in proton exchange membrane fuel cell system,” Int. J. Energy Res. 45(3), 3780–3800 (2021). [CrossRef]
9. K. Nishida, Y. Kono, R. Funaoka, and T. Furukawa, “Acceleration of through-plane water removal in polymer electrolyte fuel cell by channel hydrophilization and electrode perforation,” J. Power Sources Adv. 16, 100102 (2022). [CrossRef]
10. M. M. Mench, Q. L. Dong, and C. Y. Wang, “In situ water distribution measurements in a polymer electrolyte fuel cell,” J. Power Sources 124(1), 90–98 (2003). [CrossRef]
11. X. G. Yang, N. Burke, C. Y. Wang, K. Tajiri, and K. Shinohara, “Simultaneous measurements of species and current distributions in a PEFC under low-humidity operation,” J. Electrochem. Soc. 152(4), A759–A766 (2005). [CrossRef]
12. Q. Dong, M. M. Mench, S. Cleghorn, and U. Beuscher, “Distributed performance of polymer electrolyte fuel cells under low-humidity conditions,” J. Electrochem. Soc. 152(11), A2114–A2122 (2005). [CrossRef]
13. Q. Dong, J. Kull, and M. M. Mench, “Real-time water distribution in a polymer electrolyte fuel cell,” J. Power Sources 139(1-2), 106–114 (2005). [CrossRef]
14. F. Q. Liu, G. Q. Lu, and C. Y. Wang, “Water transport coefficient distribution through the membrane in a polymer electrolyte fuel cell,” J. Membr. Sci. 287(1), 126–131 (2007). [CrossRef]
15. G. Q. Lu, F. Q. Liu, and C. Y. Wang, “An approach to measuring spatially resolved water crossover coefficient in a polymer electrolyte fuel cell,” J. Power Sources 164(1), 134–140 (2007). [CrossRef]
16. S. Basu, M. W. Renfro, H. Gorgun, and B. M. Cetegen, “In situ simultaneous measurements of temperature and water partial pressure in a PEM fuel cell under steady state and dynamic cycling,” J. Power Sources 159(2), 987–994 (2006). [CrossRef]
17. S. Basu, M. W. Renfro, and B. M. Cetegen, “Spatially resolved optical measurements of water partial pressure and temperature in a PEM fuel cell under dynamic operating conditions,” J. Power Sources 162(1), 286–293 (2006). [CrossRef]
18. M. H. Bayer, A. Wokaun, G. G. Scherer, and I. A. Schneider, “Dynamic Measurement and Modeling of the Water Vapor Concentration during AC Impedance Measurements in Polymer Electrolyte Fuel Cells (PEFCs),” ECS Trans. 25(1), 949–960 (2009). [CrossRef]
19. M. H. Bayer and I. A. Schneider, “Dynamic response of a subsaturated polymer electrolyte fuel cell in counterflow mode,” Fuel Cells 12(3), 320–325 (2012). [CrossRef]
20. R. Sur, T. J. Boucher, M. W. Renfro, and B. M. Cetegen, “In Situ Measurements of Water Vapor Partial Pressure and Temperature Dynamics in a PEM Fuel Cell,” J. Electrochem. Soc. 157(1), B45–B53 (2010). [CrossRef]
21. Y. Fujii, S. Tsushima, and S. Hirai, “In-situ Measurement of Water Vapor and Oxygen Concentration in a PEMFC Channel by Tunable Diode Laser Absorption Spectroscopy Techniques,” ECS Trans. 11(1), 451–459 (2007). [CrossRef]
22. Y. Fujii, S. Tsushima, K. Fukuzato, and S. Hirai, “Effect of the Micro Porous Layer on Water Transport under Low Humidity Condition,” ECS Trans. 16(2), 1635–1642 (2008). [CrossRef]
23. R. K. Hanson, “Applications of quantitative laser sensors to kinetics, propulsion and practical energy systems,” Proc. Combust. Inst. 33(1), 1–40 (2011). [CrossRef]
24. J. T. C. Liu, G. B. Rieker, J. B. Jeffries, M. R. Gruber, C. D. Carter, T. Mathur, and R. K. Hanson, “Near-infrared diode laser absorption diagnostic for temperature and water vapor in a scramjet combustor,” Appl. Opt. 44(31), 6701–6711 (2005). [CrossRef]
25. M. A. Bolshov, Yu. A. Kuritsyn, and Yu. V. Romanovskii, “Tunable diode laser spectroscopy as a technique for combustion diagnostics,” Spectrochim. Acta, Part B 106(1), 45–66 (2015). [CrossRef]
26. C. S. Goldenstein, R. M. Spearrin, and R. K. Hanson, “Fiber-coupled diode-laser sensors for calibration-free stand-off measurements of gas temperature, pressure, and composition,” Appl. Opt. 55(3), 479–484 (2016). [CrossRef]
27. W. Y. Peng, C. S. Goldenstein, R. M. Spearrin, J. B. Jeffries, and R. K. Hanson, “Single-ended mid-infrared laser-absorption sensor for simultaneous in situ measurements of H2O, CO2, CO, and temperature in combustion flows,” Appl. Opt. 55(33), 9347–9359 (2016). [CrossRef]
28. W. Y. Peng, S. J. Cassady, C. L. Strand, C. S. Goldenstein, R. M. Spearrin, C. M. Brophy, J. B. Jeffries, and R. K. Hanson, “Single-ended mid-infrared laser-absorption sensor for time-resolved measurements of water concentration and temperature within the annulus of a rotating detonation engine,” Proc. Combust. Inst. 37(2), 1435–1443 (2019). [CrossRef]
29. M. B. Frish, R. T. Wainner, B. D. Green, J. Stafford-Evans, M. C. Laderer, and M. G. Allen, “Progress in reducing size and cost of trace gas analyzers based on Tunable Diode Laser Absorption Spectroscopy,” Proc. SPIE 5586, 76–82 (2004). [CrossRef]
30. M. B. Frish, R. T. Wainner, B. D. Green, M. C. Laderer, and M. G. Allen, “Standoff gas leak detectors based on tunable diode laser absorption spectroscopy,” Proc. SPIE 6010, 60100D (2005). [CrossRef]
31. M. B. Frish, R. T. Wainner, M. C. Laderer, K. R. Parameswaren, D. M. Sonnenfroh, and M. A. Druy, “Precision and accuracy of miniature tunable diode laser absorption spectrometers,” Proc. SPIE 8032, 803209 (2011). [CrossRef]
32. M. Trincavelli, V. H. Bennetts, and A. J. Lilienthal, “A least squares approach for learning gas distribution maps from a set of integral gas concentration measurements obtained with a TDLAS sensor,” in Proceedings of IEEE Sensors (2012).
33. H. I. Schiff, G. I. Mackay, and J. Bechara, “The use of tunable diode laser absorption spectroscopy for atmospheric measurements,” Res. Chem. Intermed. 20(3-5), 525–556 (1994). [CrossRef]
34. M. B. Frish, “Current and emerging laser sensors for greenhouse gas sensing and leak detection,” Proc. SPIE 9101, 91010H (2014). [CrossRef]
35. F. Stritzke, O. Diemel, and S. Wagner, “TDLAS-based NH3 mole fraction measurement for exhaust diagnostics during selective catalytic reduction using a fiber-coupled 2.2-µm DFB diode laser,” Appl. Phys. B 119(1), 143–152 (2015). [CrossRef]
36. M. Gu, J. Chen, J. Mei, T. Tan, G. Wang, K. Liu, G. Liu, and X. Gao, “Open-path anti-pollution multi-pass cell-based TDLAS sensor for the online measurement of atmospheric H2O and CO2 fluxes,” Opt. Express 30(24), 43961 (2022). [CrossRef]
37. M. R. McCurdy, Y. Bakhirkin, G. Wysocki, R. Lewicki, and F. K. Tittel, “Recent advances of laser-spectroscopybased techniques for applications in breath analysis,” J Breath Res 1(1), 014001 (2007). [CrossRef]
38. B. Henderson, A. Khodabakhsh, M. Metsälä, I. Ventrillard, F. M. Schmidt, D. Romanini, G. A. D. Ritchie, S. te Lintel Hekkert, R. Briot, T. Risby, N. Marczin, F. J. M. Harren, and S. M. Cristescu, “Laser spectroscopy for breath analysis: towards clinical implementation,” Appl Phys B 124(8), 161 (2018). [CrossRef]
39. J. Reid and D. Labrie, “Second-Harmonic Detection with Tunable Diode Lasers - Comparison of Experiment and Theory,” Appl. Phys. B 26(3), 203–210 (1981). [CrossRef]
40. L. S. Rothman, I. E. Gordon, Y. Babikov, et al., “The HITRAN2012 molecular spectroscopic database,” J. Quant. Spectrosc. Radiat. Transfer 130, 4–50 (2013). [CrossRef]
41. K. Nishida, R. Nakauchi, Y. Tabata, T. Umekawa, and M. Kawasaki, “Evaluation of Dry-Wet Transition in PEFC Under Load Changes Based on Laser Spectroscopy Using Fiber-Optic Probe,” ECS Trans. 92(8), 119–124 (2019). [CrossRef]
