Robust Operation of Fuel Cell Systems in Subfreezing

Subscriber access provided by University of Glasgow Library

Article

Robust Operation of Fuel Cell Systems in Subfreezing Conditions: A Material-Based Solution to Achieve Better Anode Durability Leiming Hu, Bo Ki Hong, Jong-Gil Oh, and Shawn Litster ACS Appl. Energy Mater., Just Accepted Manuscript • DOI: 10.1021/acsaem.9b01108 • Publication Date (Web): 03 Sep 2019 Downloaded from pubs.acs.org on September 3, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Energy Materials

Robust Operation of Fuel Cell Systems in Subfreezing Conditions: A Material-Based Solution to Achieve Better Anode Durability Leiming Hu, a Bo Ki Hong, b Jong-Gil Oh,b and Shawn Litster. a * a Department of Mechanical Engineering, Carnegie Mellon University, Pittsburgh, PA 15213, USA b Fuel Cell Research Lab., R&D Division, Hyundai Motor Company, Yongin-si, Gyeonggi-do, 16891, Republic of Korea. * Corresponding author. Scott Hall, Room No. 5107, Carnegie Mellon University. 5000 Forbes Avenue, Pittsburgh, PA, 15213, USA, Phone: +1-412-268-3050, Fax: +1-412-268-3348. KEYWORDS: Polymer electrolyte membrane fuel cells, anode durability, reversal tolerant anode, hydrogen starvation, cell reversal, carbon corrosion, subfreezing condition, water electrolysis catalyst

ABSTRACT: Automotive fuel cells can suffer from cell voltage reversals due to hydrogen fuel starvation at the anode, which can be exacerbated by subfreezing temperatures because of extensive ice formation and its blockage of the hydrogen supply channels. Here we report on low 1 ACS Paragon Plus Environment

ACS Applied Energy Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

temperature (-15 to 45oC) reversal degradation due to hydrogen starvation and evaluate the use of a reversal-tolerant anode (RTA) with an oxygen evolution reaction (OER) catalyst (e.g., iridium oxide). During subfreezing cell reversal, we observed a water electrolysis cell voltage plateau, but the typically subsequent voltage plateau for carbon corrosion usually seen at high temperatures was not found. Repeated cell reversal tests at subfreezing temperature shows a much slower degradation rate compared with high temperature reversals. A series of isothermal reversal tests at different temperatures indicate that the degradation rate related with carbon corrosion is temperature dependent. As for the effect of RTA on the cell reversal behavior, we found at subfreezing temperatures that the voltage reversal duration time of the MEA with RTA is similar to that of the MEA without RTA. However, once the cell temperature was sufficiently above freezing, such as during the heating of the startup process, the MEA with RTA would facilitate water electrolysis reactions by consuming water and delaying the onset of carbon corrosion. Thus, our results show that cell reversal with subfreezing internal temperatures does not substantially degrade the anode by carbon corrosion, but that as the cell warms an RTA is one approach to extend the survival of the fuel cell stack.

INTRODUCTION For the past few decades, polymer electrolyte membrane fuel cells (PEMFCs) powered by hydrogen fuel have gained much attention as power sources for fuel cell electric vehicles (FCEVs) due to their high efficiency, high power density, and zero-emission features.1 For FCEVs to become commercially viable, however, three main challenges, i.e., performance, cost, and durability, must be resolved. To this end, the fuel cell components and systems need to be highly robust and durable under a variety of operating conditions, in particular, under freezing 2 ACS Paragon Plus Environment

Page 2 of 36

Page 3 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


conditions which frequently cause cold start-up/shut-down degradation,2,3 freeze/thaw cycling degradation, etc.4–6 The industry has recognized that degradation caused by hydrogen fuel starvation is one of the most critical degradation mechanisms for the membrane electrode assembly (MEA) in automotive fuel cells,7–11 as well as other fuel cell configurations like dead-ended fuel cells.12 Figure 1a shows the typical arrangement of an MEA.

Figure 1. (a) Schematic of the MEA structure. (b) Schematic of the anode during regular operation with HOR at high temperature. (c) Schematic of the anode during hydrogen starvation at high temperature. OER and COR could occur inside the catalyst layer. (d) Schematic of the anode during hydrogen starvation at subfreezing temperature. 3 ACS Paragon Plus Environment


Page 4 of 36

Under normal fuel cell operating conditions, hydrogen and oxygen (in air) gases are sufficiently supplied to the anode and cathode, respectively, in compliance with the load demand of fuel cells through the hydrogen oxidation reaction (HOR) at the anode (as shown in Figure 1b) and the oxygen reduction reaction (ORR) at the cathode.1,9,13 Under the abnormal condition of hydrogen starvation due to either hydrogen supply malfunction or hydrogen channel blockage by liquid water or ice, however, the anode potential (Ean) increases relative to the cathode’s (Eca) and the cell voltage (Ecell) eventually reverses, i.e., Ecell = Eca – Ean < 0 V, causing the voltage reversal degradation.7–11 If purging or other active controls do not promptly resolve the hydrogen starvation, the anode potential will keep increasing and become high enough to cause water electrolysis (Reaction 1) with an oxygen evolution reaction (OER) and a carbon oxidation reaction (COR) or carbon corrosion reaction (CCR) (represented by Reactions 2-3 and shown in Figure 1c and 1d).9,14,15

H 2O  1 O2  2 H   2e  , E o  1.23 V (vs. SHE , 25 oC ) 2

[1]

C  2 H 2O  CO2  4 H   4e  , E o  0.207 V (vs. SHE , 25 oC ) [2] C  H 2O  CO  2 H   2e  , E o  0.518 V (vs. SHE , 25 oC )

[3]

where E0 is the standard electrode potential versus the standard hydrogen electrode (SHE). The water electrolysis reaction is known to be kinetically favorable and thus proceeds faster than the COR, even though the COR is thermodynamically favorable.16,17 When the anode potential increases up to high voltages, i.e., > 1 V, however, the sluggish COR becomes fast 4 ACS Paragon Plus Environment

Page 5 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enough to cause the corrosion of carbon supports in the anodes of PEMFCs.8,9,18 During hydrogen starvation at high temperatures, the cell experienced an OER dominated period, which is followed by a COR dominated period, as has been shown in previous research works7–9,11. During our prior cell reversal tests at 65oC with a relative humidity (RH) of 100%

8,9,

the cell

voltage exhibited the first plateau at about -0.8 ~ -1.1 V, which corresponds to anode potentials of about 1.6 ~ 1.9 V, and water electrolysis reaction is the main reaction at this potential range. The water electrolysis plateau was then followed by the second plateau at about -1.7 ~ -2.0 V (anode potentials of about 2.5 ~ 2.8 V), a potential at which the Faradaic current is dominated by the COR.8 It is important to note that COR will be occurring at lower OER anode potentials, but at a much lower rate. The COR is known to cause significant damage to the anode, in the form of carbon catalyst support loss.8,19 The loss of the carbon causes a detachment and agglomeration of Pt catalyst particles and a corresponding decrease of the electrochemically active surface area (ECSA).14 The continuous corrosion of carbon results in thinner catalyst layers and increases of the electrical contact resistance. The surface oxidation also increases the hydrophilicity of carbon supports, which causes a higher mass transfer resistance.20,21 Furthermore, the loss of carbon at the membrane interface leads to a significant increase in the high-frequency resistance due to the low proton conductivity region formed between the membrane and the remaining active thickness of the anode next to the anode’s gas diffusion media. To mitigate the hydrogen starvation degradation, besides optimizing gas purging control strategies22,23 and modifying cell configurations,24,25 materials-based solutions have been proposed through adding OER catalysts (i.e., IrO2, RuO2, TiO2, IrRu, etc.) to the anode, resulting in a reversal-tolerant anode (RTA) that promotes the OER over the COR.7–11,17 Compared with 5 ACS Paragon Plus Environment


adding extra cell voltage monitor and purge control system, material-based solution is more preferable to the automotive industry since it will not increase the system complexity. Researchers have extensively studied RTAs and their effects on the hydrogen starvation degradation at typical operating temperatures, i.e., 60 ~ 90 oC.7–11,26–29 Taniguchi et al. examined the anode and cathode potential changes vs. the reversible hydrogen electrode (RHE) during cell reversal at 80oC, and the spatial distribution of the electrode’s degradation was analyzed.28 A segmented cell technique was adopted to study the uneven voltage and current density distribution during cell reversals at 60oC26 or 70oC.27 The reversal tests at a high temperature of 90oC and various RH conditions from 14 to 82% showed that the IrO2 RTA performance surprisingly dropped under not only water-deficient but also water-excess conditions due to a competition between water electrolysis and carbon corrosion reactions under the given RH conditions.9 Recently, You et al.10 reported that the IrRu4/C RTA also could increase the anode durability under cell reversal conditions of 65oC and 50% RH. Since channel-blocking ice formation in freezing conditions can exacerbate hydrogen starvation, it is of great interest to examine if the RTA still promotes the water electrolysis over carbon corrosion when the water was frozen into immobilized ice at subfreezing temperatures (i.e., -15oC  temperature < 0oC), as illustrated in Figure 1d. It is also of great importance to know how efficiently the RTA could work at low temperatures slightly higher than 0oC (i.e., 0oC < temperature  45oC) whose conditions are frequently encountered by automotive fuel cells as transient operating conditions, i.e., cold start processes. There are three main differences which differentiate the hydrogen starvation and cell performance deterioration at normal operating temperatures from those at subfreezing temperatures. First, the immobilized ice formed inside the fuel cells won’t be easily removed from the cells by a conventional technique like gas 6 ACS Paragon Plus Environment

Page 6 of 36

Page 7 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


purging. Second, the low temperature may significantly change the relative kinetics of the water electrolysis and carbon corrosion reactions at the anode during reversal. Third, the ice could be continuously formed at subzero temperatures because of product water generation through the ORR at the cathode, exacerbating the overall cell performance deterioration. A key challenge to studying the reversal behavior at subzero temperatures is the ability to measure the individual potentials of electrodes independently, which necessitates an accurate reference electrode (RE). Piela et al. adopted a dynamic hydrogen electrode (DHE) as an RE.30 They observed that placing the reference point above the gas diffusion layers (GDLs) and connecting it to the electrode through an ionomer bridge could be an ideal configuration for measuring the individual electrode potentials. Due to the difficulties of fabrication equipment, however, adopting this kind of RE is challenging.29 Another approach was to employ a detachable RE built externally and connected to the edge of a membrane in the MEA, which is easy to implement.31 However, this methodology used air as the working environment for the RE, meaning that the reference point would be vulnerable to a change in ambient conditions. To date, despite its great necessity and importance, systematic research on the cell voltage reversal behavior at subfreezing and low temperatures is very scarce. Here we report the effects of an IrO2 RTA addition on the cell reversal degradation from -15 to 45oC through electrochemical diagnostics and a reference electrode. In this study, the hydrogen starvation status was simulated by manually controlling the hydrogen gas supply to the anode. To mimic hydrogen starvation, we would quickly switch the hydrogen feed to the fuel cell test stand to nitrogen. Our tests include both hydrogen-air cell and hydrogen-pumping cell reversals. During hydrogen-air cell tests, we supplied the cathode with air and examined the effects of continuous ice formation due to ORR at the cathode in addition to the cell reversal behavior. We used a 7 ACS Paragon Plus Environment


specially designed hydrogen reference electrode in these experiments to extract the anode potential during reversal. The hydrogen-pumping cell reversal tests involve feeding the cathode side with nitrogen so it can serve as the counter and a dynamic hydrogen electrode, and thus the cell voltage primarily reflects the anode potential.

EXPERIMENTAL MEA and Fuel Cell

An in-house MEA with an active area of 4 cm2 was fabricated using a perfluorosulfonic acid (PFSA) ionomer membrane (Nafion® NRE211CS (H+) form, DuPont, USA) and electrodes consisting of carbon-supported platinum catalyst (Pt/C: HISPEC4000® grade, Johnson Matthey, UK) and an ionomer binder (Nafion® D2021 grade, DuPont, USA). The anode and cathode Pt loadings were 0.1 and 0.4 mg-Pt/cm2, respectively. The ionomer binder content in the electrode was 30 wt% with respect to the total solid content of the dried electrode. For preparing the RTA, an acid-stable water electrolysis catalyst, IrO2, was synthesized using the Adams fusion method.32,33 To compare the effectiveness of IrO2 during cell reversal in freezing conditions, two types of MEAs were fabricated, one with no IrO2 in the anode (i.e., 0 wt% RTA MEA) and the other one with 5 wt% IrO2 in the anode with respect to the amount of Pt catalyst in the anode (i.e., 5 wt% RTA MEA). Adhesive-backed polyimide Kapton (25 μm thickness, McMasterCarr, Aurora, USA) films were used as subgaskets. We assembled the MEAs in commercial fuel cell testing hardware (Dual-area fuel cell fixture, Scribner Associates Inc., North Carolina, USA) with commercial GDLs (25BC grade, SGL Technologies, GmbH, Germany). A commercial fuel cell test station (850 Test stand, Scribner Associates Inc., North Carolina, USA) was used to 8 ACS Paragon Plus Environment

Page 8 of 36

Page 9 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


record the gas conditions and the anode and cathode potentials. The cell potential during reversal testing was recorded using a potentiostat (VSP, Bio-Logic, Claix, France).

Reference Electrode

During hydrogen-air cell experiments, a hydrogen reference electrode was used to monitor the cathode and anode potential changes.28 Figure 2a shows a schematic of the cross-section of the reference electrode. We used a strip of the polymer electrolyte membrane (Nafion 211 grade, IonPower, New Castle, USA) as the electrolyte bridge. A polyimide Kapton film was used as a covering to prevent any loss of water from the Nafion strip. One end of the Nafion strip was in contact with a 1 cm2 gas diffusion electrode (GDE) (0.3 mg Pt/cm², 40 wt% Pt on Vulcan, carbon paper, Fuel Cells Etc., College Station, USA) inside a custom-made acrylic hardware that provided gas sealing and good contact between the GDE and the Nafion strip as well as connected the reference electrode to the fuel cell hardware. A Pt sputtered stainless steel plate was used to further ensure good electrical contact with the GDE of the reference electrode. The other end of the Nafion strip was in contact with an overhanging edge of the PFSA membrane in the MEA. Figure 2b shows a top view of the integration of the reference electrode with the MEA that was subsequently inserted into the fuel cell hardware. As we previously described,29 an electrode misalignment between the anode and the cathode with an overhang distance of  was intentionally introduced (as shown in Figure 2a). Adler et al.34 reported that if the ratio of the electrode misalignment divided by the membrane thickness d (i.e., /d) is greater than 4, then the RE would measure the ionic potential at the interface between the overhanging electrode and the membrane. The  and the membrane thickness d values were fixed to be 1 mm and 25 m, 9 ACS Paragon Plus Environment


respectively, resulting in the /d ratio value of 40 which is greatly larger than 4. Thus, our reference electrode is primarily isolating the potential of the anode in these measurements. It should be noted that the reference electrode measured the electrolyte potential on the anode side near the overhanging part, which can differ from the anode potential in other regions, particularly during gas switching. However, as shown in our previous work29, the measurement error is acceptable in terms of monitoring the anode potential change when the overall cell voltage change is highly complex due to several simultaneous processes in the hydrogen-air freezing reversal tests, as will be shown later.

Figure 2. (a) Cross-sectional schematic of the reference electrode and its connection with the MEA. (b) A top view image of the reference electrode hardware and its integration with the MEA. The dashed lines represent the Nafion sensing layer in between the Kapton subgaskets. (c) Cell and anode potential changes of 0 wt% RTA MEA during hydrogen-air cell reversal tests as a function of time at a subfreezing temperature of -15oC.

Preconditioning

10 ACS Paragon Plus Environment

Page 10 of 36

Page 11 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Before each hydrogen starvation test and the post-reversal electrochemical characterization, the cell was preconditioned at 65 oC and 100% RH, with an anode hydrogen flow rate of 134 ml/min and cathode air flow rate of 428 ml/min. During the preconditioning, the cell voltage was controlled to be 0.7 V, 0.5 V and 0.3 V each for 5 min and repeatedly until the cell current was stable. Before each freezing, the cell was purged with nitrogen gas at both the anode and the cathode with a flow rate of 200 ml/min with a cell temperature of 65oC and a gas relative humidity of 100% RH for 1 hr. After that purging, the cell was removed from the fuel cell test stand and placed in the environmental chamber for the cell reversal tests. Care was taken to remove condensed water from the inlets and outlets when connecting the fuel cell to the gas lines within the environmental chamber prior to freezing.

Hydrogen Starvation and Cell Reversal Testing Protocol

Hydrogen-air cell reversal experiment We performed these tests to investigate the simultaneous effects at the cathode and anode during reversal at subfreezing conditions when the cathode is generating water. We initially operated the cell at a constant current density of 0.2 A/cm2 with anode and cathode supplied with hydrogen (flow rate: 134 ml/min) and air (flow rate: 428 ml/min), respectively. Then the anode hydrogen supply was manually switched to nitrogen with the same flow rate. Both the anode and cathode potential changes were monitored using the reference electrode. We controlled cell at below room temperature using an environmental chamber (Model TJR, Lunaire Limited, USA). During reversal tests at temperatures below room temperature (i.e., 25oC), both the anode and the cathode gases passed through humidifier water tanks at 25oC before being supplied to the cell. 11 ACS Paragon Plus Environment


For cell temperatures above room temperature, we set humidifier temperatures to be the same as the cell temperature so the RH of the gases was 100%. The reference electrode was used for the hydrogen-air cell reversal tests. Hydrogen was supplied to the reference electrode hardware with a flow rate of 100 ml/min. The reference electrode’s electrolyte was pre-humidified at 65oC and 100% RH before insertion into the environmental chamber.

Hydrogen-pumping cell reversal experiment We designed our hydrogen-pumping experiments to specifically focus on the anode and isolate it from the freezing degradation at the cathode. We initially supplied the anode with hydrogen (flow rate: 134 ml/min) but the cathode was fed with nitrogen (flow rate: 428 ml/min) instead of air, resulting in a hydrogen evolution reaction (HER), i.e., 2H+ + 2e-  H2, at the cathode, which renders its potential close to zero allowing for its use as a reference point during freezing reversal tests. We also operated at a constant current density of 0.2 A/cm2 during these tests and the temperature and RH control were the same as the hydrogen-air cell reversal experiments.

Electrochemical Characterization

After the cell reversal experiment, we characterized the cell at a normal temperature of 65oC. The cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and polarization curve measurements were conducted to examine the electrode damage due to hydrogen starvation and cell reversal. For the cathode CV and EIS measurements, the cell temperature was 65oC and the gases were delivered at 100% RH. In both measurements, we supplied the anode 12 ACS Paragon Plus Environment

Page 12 of 36

Page 13 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with hydrogen at a flow rate of 134 ml/min. During CV, we supplied the cathode with nitrogen at a flow rate of 134 ml/min. The CV scan was conducted from 0.04 to 1.2 V at a scan rate of 50 mV/s. During EIS, we fed the cathode with air at 428 ml/min. The EIS spectra were measured at 0.1 and 0.4 A/cm2. For the polarization curve measurements, the temperature and gas conditions were the same as those of EIS. The cell voltage was recorded at constant current densities ranging from 0.0 to 1.6 A/cm2.

Characterization by Scanning Electron Microscope (SEM) and Energy Dispersive Spectroscopy (EDS)

Cross-sectional specimens of MEAs were obtained by fracturing in liquid nitrogen. The crosssectional morphology of anodes was analyzed using an SEM (Quanta 600 FEG, FEI Company, OR, USA) with its secondary electron detector and EDS (Noran Vantage DSI, Thermo Noran, WI, USA) detector. The GDL was carefully separated from the anode catalyst layer using tweezers to image the anodes top surface.

RESULTS AND DISCUSSION Hydrogen-Air Cell Reversal Experiment

Figure 2c shows the cell and anode potential changes during hydrogen-air cell reversal tests as a function of time at a subfreezing temperature of -15oC. The voltage reversal time shown on the 13 ACS Paragon Plus Environment


x-axis was the duration of time cell’s voltage stayed negative. The open circuit voltage (OCV) of the cell was recorded before the reversal test under normal supply conditions of hydrogen and air to the anode and the cathode, respectively. The OCV was 0.96 V while the anode potential was around 0 V. Then, a constant current density of 0.2 A/cm2 was applied to the cell with the hydrogen fed being simultaneously replaced by nitrogen, in order to simulate cell voltage reversal conditions. At 0.2 A/cm2, the cell voltage quickly dropped to a small negative value of 16 mV, for a short plateau of roughly 30 s. We surprisingly observed that during the freezing reversal at -15oC for about 130 s, the cell voltage of the 0 wt% RTA MEA decreased relatively gradually down to about -3 V, but the anode potential did not increase significantly and remained stable, indicating there was still hydrogen present from the slow displacement of hydrogen from the test stand. Thus, the observed decrease in cell voltage during freezing reversal tests for the first 130 s of reversal duration time is primarily a result of a decrease in the cathode potential. This unstable cathode potential was also observed in prior cold-start studies.35–37 The ORR at the cathode continuously generates water, and thus in freezing conditions, ice is formed inside the cathode and blocks the pores essential for oxygen transport. Local depletion of oxygen with applied currents leads to the HER instead of the ORR at the cathode, which is attributed to the short plateau formation of cell voltage at around -16 mV. As the ice kept forming and filling more pores in the cathode, the overpotential for the ORR increased and the cell potential slowly decreased to and plateaued at around -1 ~ -1.2 V, which was followed by a relatively rapid drop down to -3 V without causing any significant increase in the anode potential, suggesting an effect isolated to the cathode. After 130 s of cell reversal time, the cell presented a significant increase of anode potential and then a plateau at 1.6 ~ 1.8 V that is consistent with water electrolysis reaction. It should be noted 14 ACS Paragon Plus Environment

Page 14 of 36

Page 15 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


that, as shown in our previous work29, our reference electrode configuration has a measurement error of ~ 110 mV for the anode potential in reversal conditions. During this time, the cell voltage is a very negative value of -4 V. After 60 s duration of the water electrolysis plateau at the anode, the anode potential sharply increases, and cell voltage drops to -7 V at the termination of the experiment. This anode behavior is in clear contrast to previous observations at higher temperatures, e.g., 60 ~ 90oC, where the cell voltage dropped and the anode potential increased quickly during reversal tests to the first water electrolysis plateau followed by a second plateau due to carbon corrosion at the anode.7–11,26–29 No carbon corrosion plateau was evident in these subfreezing temperature tests. Instead, following the water electrolysis plateau, we generally saw significant decreases in cell voltage due to capacitive charging of the anode rather than Faradaic reactions. In cases where we allowed sustained current density at these large voltages (negative cell, positive anode voltage), we would see a voltage plateau that we have attributed that to selfheating of the cell and likely the melting of ice and thermal activation of carbon corrosion.

Hydrogen-Pumping Cell Reversal Experiments

To eliminate the ice formation effects due to the ORR at the cathode in well-controlled experiments, we performed the hydrogen-pumping cell reversal experiments using 0 wt% and 5 wt% RTA MEAs. When the current was applied to the hydrogen-pumping cell, there was only HER at the cathode instead of ORR. In this case, the cathode had a negligible overpotential due to the fast HER kinetics and no problematic water generation in the absence of oxygen. Thus, the anode is the primary source of any changes of cell voltage.



Repeated Hydrogen-Pumping Cell Reversal Tests of 0 wt% RTA MEA at -5oC.

Figure 3 shows the repeated hydrogen-pumping cell reversal test results of 0 wt% RTA MEA at -5oC. We repeated reversal tests on the same MEA at two different anode gas flow rates: 134 and 1000 ml/min. We studied the humidified gas flow rate since it changes the amount of water delivered to support OER and also shows whether freezing of the water in the supply gases affects the results.

Figure 3. Repeated hydrogen-pumping cell reversal test results at -5oC using 0 wt% RTA MEA with an alternating gas flow rate between 134 ml/min (R1, R3 and R5) and 1000 ml/min (R2, R4 and R6) and characterization at BOL and after each reversal. (a) The cell voltage changes during hydrogen-pumping cell tests. (b) Polarization curves. (c) Anode CV. (d) EIS Nyquist plots for current densities of 0.1 and 0.4 A/cm2.


Page 16 of 36

Page 17 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3a shows the cell voltage changes during the 0.2 A/cm2 hydrogen-pumping cell reversal tests as a function of time. At the beginning of the hydrogen-pumping cell tests, as mentioned before in the Experimental section, the anode and cathode were firstly supplied with hydrogen and nitrogen, respectively. In this case, the half-cell reactions are the HOR at the anode, and the HER at the cathode due to the absence of oxygen, and the overall cell voltage was around 0 V. The anode gas feed was then switched to nitrogen, which caused a sudden drop of the cell voltage to around -2 V. We attribute the voltage plateaued at -2 V to OER dominated reactions at the anode. The plateau lasted for roughly 50 s followed by a voltage drop down to -4.5 V when the reversal experiment terminated. It is noteworthy that no plateau representing the COR dominated region was observed during hydrogen-pumping cell reversal tests at -5oC, similar to the hydrogen-air cell experiments. Again, this is in contrast to COR dominated plateaus that usually observed during reversal tests at higher temperatures (e.g., 65oC and 90oC) in a cell voltage range of approximately -1.5 to -2.0 V.8,9 Although there was approximately a seven-fold difference of the anode gas flow rates between 134 and 1000 ml/min, no significant change in the voltage reversal time duration was observed. Thus, we are confident that the loss of OER current after 50 s and the lack of a COR voltage plateau is not due to water starvation from under-delivery of water vapor in the nitrogen. Also, though two different gas flow rates were used, there is no significant difference in the timing of the cell voltage’s sudden drop, indicating that the cell voltage change is not likely to be caused by icing of water vapor from inlet gases. The polarization curves were obtained at beginning-of-life (BOL) and after each hydrogenpumping cell reversal test as shown in Figure 3b. Though the cell experienced repeated reversal tests, the polarization curves after each reversal almost overlap each other. Only a slight decrease 17 ACS Paragon Plus Environment


of cell voltage in the mass transport loss region at high current densities, i.e., from 1.2 to 1.6 A/cm2, was observed. Prior work by Mandal et al.8 with the same type of MEA and cell reversal testing protocols at a high temperature, 65oC, showed that the cell voltage at 1.2 A/cm2 decreased by 70% within 2.2 min of accumulated cell voltage reversal time. In contrast, for the repeated cell reversal tests at -5 oC, from the polarization curves, the cell voltage degreased by less than 5 % after 5 min of cell voltage reversal time. Figure 3c shows the anode CV curves. As the hydrogen-pumping cell reversal tests proceeded, there was a continuous decrease of hydrogen desorption (anodic) and adsorption (cathodic) peaks at around 0.1 ~ 0.3 V as well as the Pt oxide formation (anodic) and oxide reduction (cathodic) peaks in the range of 0.7 ~ 0.9 V,38 indicating a minor loss of ECSA. However, this loss of ECSA with subfreezing temperature reversals is more likely to be caused by Pt catalyst degradation or freeze-thaw damage rather than massive carbon corrosion damage at high anode overpotentials, since the CV peak frequently observed at around 0.5 ~ 0.7 V due to the hydroquinone-quinone (HQ-Q) redox reaction8,9,19,39, which is commonly used as an indicator for drastic COR at high anode potentials, is hardly seen in these CV results. Figure 3d shows the corresponding EIS Nyquist plots recorded at current densities of 0.1 and 0.4 A/cm2. The Nyquist plots are almost all identical for each current density after successive reversals, showing a negligible increase in the charge transfer resistance (i.e., semi-circles’ apparent radius) and highfrequency resistance (HFR) (i.e., first real-axis intercept). In our previous study of cell reversal at higher temperature of 65 oC, the cell HFR increased from 0.07 Ohm.cm2 to 0.39 Ohm.cm2 within 2.2 min of accumulated voltage reversal time, and significant change in the anode charge transfer resistance (Rct, an) can be observed. In contrast, there is negligible change in HFR and Rct, an for cell voltage reversals at -5 oC. Admittedly, there might be coexistence of COR and OER at the 18 ACS Paragon Plus Environment

Page 18 of 36

Page 19 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


voltage potential plateau of -2 V during reversal tests at -5 oC. However, the minimal cell voltage decrease, HFR and Rct, an increases compared with high temperature reversals in our prior work8, indicate a heavily suppressed COR at subfreezing temperatures.

Temperature-Dependent Hydrogen-Pumping Cell Reversal Study

To better examine the effect of temperature on suppressing cell reversal degradation, we performed hydrogen-pumping cell reversal measurements with consecutively increasing temperature during each. Furthermore, to elucidate the efficiency of RTAs as a function of temperature, we carried out these experiments on both 0 wt% and 5 wt% RTA MEAs. Figure 4a shows the temperature-dependent hydrogen-pumping cell reversal test results of the 0 wt% RTA MEA. The reversal tests were conducted for cell temperatures ranging from -15 up to 20oC. For the test at -15oC, after the cell voltage reverses (i.e., Ecell < 0 V), the cell voltage plateaus at around -2 V, according to the water electrolysis reaction. As we increased the temperature in subsequent reversals, the voltage reversal time duration at the first plateau accordingly increases, from 18 s at -15oC to 63 s at 5oC. In this temperature range, we did not observe a secondary plateau due to carbon corrosion, and the anode voltage became very large following the electrolysis plateau. As the temperature further increases from 10 up to 20oC, however, the reversal time decreases with the advent of a second plateau in the range of -2.5 to -3.5 V, which we believe is due to the emergence of COR sustaining the applied current density. Post-reversal electrochemical characterizations are supporting this observation and will be discussed later.



Figure 4. Temperature-dependent cell voltage change during hydrogen-pumping cell reversal tests. (a) Cell reversal test results of 0 wt% RTA MEA from -15 to 20 oC. (b) Cell reversal test results of 5 wt% RTA MEA from -15 to 45 oC. (c) Comparison of the voltage reversal time between 0 wt% and 5 wt% RTA MEAs as a function of the reversal test temperature


Page 20 of 36

Page 21 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4. Temperature-dependent cell voltage change during hydrogen-pumping cell reversal tests. (a) Cell reversal test results of 0 wt% RTA MEA from -15 to 20 oC. (b) Cell reversal test results of 5 wt% RTA MEA from -15 to 45 oC. (c) Comparison of the voltage reversal time between 0 wt% and 5 wt% RTA MEAs as a function of the reversal test temperature



To examine the effectiveness of the IrO2 RTA in posponing the carbon corrosion reaction at low temperatures, we conducted the same study with a 5 wt% RTA MEA. Figure 4b shows the voltage time series for reversal tests that were conducted for temperatures ranging from -15 to 45oC. The cell voltage during reversal at -15oC shows the first plateau at around -2 V for 20 s. In the case of 5 wt% RTA MEA, the voltage reversal duration time at the first plateau representing the water electrolysis reaction generally increases with increasing the reversal temperature from 15 to 20oC, with the longest water electrolysis time of 140 s observed at 20oC. This behavior is quite distinct from cell reversals with the 0 wt% RTA MEA, which exhibited the second plateau of carbon corrosion from 10 to 20oC and maximum first plateau reversal times of 63 s. As the reversal temperature keeps increasing from 25 to 45oC, the water electrolysis time of the 5 wt% RTA MEA decreases from 100 s at 25oC to around 30 s at 45oC. Figure 4c shows the comparison of the voltage reversal times of 0 wt% and 5 wt% RTA MEAs as a function of the reversal test temperature. By adding IrO2 RTA, a longer water electrolysis time was achieved at low temperatures above freezing (i.e., 10 ~ 45oC) and the second plateau for the COR was not observed until reaching a temperature of 45oC after several consecutive reversals. It’s noteworthy that although a second plateau due to carbon corrosion of the 5 wt% RTA MEA was not observed for temperatures from 25 to 45oC, we cannot fully exclude the possibility of the minor contribution of carbon corrosion to the overall reversal degradation and reduced duration of the water electrolysis plateau in this temperature range. The prolonged water electrolysis time of the 5 wt% RTA MEA suggests that the IrO2 in the RTA enhanced the water electrolysis reaction, suppressing the onset of carbon corrosion reaction.


Page 22 of 36

Page 23 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The increasing voltage reversal time as the temperature is increased from -15 to 0oC (see Figure 4c) is an intriguing finding. Our present understanding from these results is that the end of the water electrolysis plateau is due to local water starvation in the freezing condition. As the temperature increases, there is a greater fraction of unfrozen water in the polymer electrolyte and greater water mobility in the anode that extends the period before the OER is water starved. Shown in previous research by Siu et al.40, as the temperature increases in the range from -15oC to 0oC, the amount of non-frozen water in Nafion® ionomer and membrane also increases. The subsequent decrease in water electrolysis reversal time at higher temperatures is likely due to initially minor poisoning of the Pt (i.e., 0 wt% RTA MEA) or Pt and IrO2 (i.e., 5 wt% RTA MEA) catalysts by small amounts of carbon corrosion products as we have postulated in our prior work.8,9 Figure 5a (1-3) present detailed electrochemical analyses that were carried to investigate the cell degradation mechanisms due to the carbon corrosion during hydrogen-pumping cell reversals at varying temperatures with and without IrO2 in the anode. Figure 5a-1 shows the CV curves of the 0 wt% RTA MEA at BOL and after each hydrogen-pumping cell reversal test. The CV curve in black represents the BOL state. From -15 to 5oC, both the hydrogen desorption peaks at around 0.1 ~ 0.3 V and Pt oxide formation peaks at around 0.7 ~ 0.9 V decreased. The decrease of these characteristic peaks shows a similar trend to the CV results of repeated hydrogen-pumping cell reversals at -5oC in Figure 3c, which could be attributed to freeze-thaw damage. In the case of the hydrogen-pumping cell reversal test at 10oC, another characteristic peak began to evolve in the range of 0.5 ~ 0.7 V, which corresponds to HQ-Q redox peak8,9,19,38,39 and indicating drastic carbon corrosion reaction. This carbon corrosion peak became larger after the reversal tests at 15 and 20oC, suggesting an increased extent of carbon 23 ACS Paragon Plus Environment


surface oxidation. Figure 5a-2 presents the polarization curves of the 0 wt% RTA MEA at BOL and after each hydrogen-pumping cell reversal test. After the hydrogen-pumping cell reversal tests in the temperature range of -15 to 5oC, we only observed slight voltage decreases at current densities higher than about 1.2 A/cm2 (mass transport loss region). At higher reversal temperatures of 10 ~ 20oC, however, more pronounced voltage decreases were observed over the entire polarization curve, including at lower (i.e., activation loss region) and medium (i.e., ohmic loss region) current densities. As shown in Figure 5a-3, the increase in ohmic resistances after the hydrogen-pumping cell reversal tests at 10 ~ 20oC was clearly observed in the EIS Nyquist plots, indicating an increased contact resistance in the MEA. Our prior work has shown that this is due to a depletion of carbon at the anode’s interface with the membrane.8,9


Page 24 of 36

Page 25 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Post-characterization results after the temperature-dependent hydrogen-pumping cell reversal. Anode CV curves for (a-1) 0 wt% and (b-1) 5 wt% RTA MEAs. Polarization curves for (a-2) 0 wt% and (b-2) 5 wt% RTA MEAs. EIS Nyquist plots at 0.1 A/cm2 for (a-3) 0 wt% and (b-3) 5 wt% RTA MEAs.

Figure 5b-1 shows the CV curves of the 5 wt% RTA MEA at BOL and after each hydrogenpumping cell reversal test. Similar to the characterization results of 0 wt% RTA MEA in Figure 5a-1, both the hydrogen desorption and Pt oxide formation peaks decreased. However, the carbon corrosion peaks at around 0.5 ~ 0.7 V due to the HQ-Q redox reaction did not substantially appear until the highest temperature of 45oC, implying that the IrO2 catalyst in the anode suppressed the COR up to this temperature. Figure 5b-2 shows the polarization curves of the 5 wt% RTA MEA at BOL and after each hydrogen-pumping cell reversal test. After the hydrogen-pumping cell reversal tests in the temperature range of -15 to 20oC, we only observed voltage decreases in the mass transport loss region. At higher reversal temperatures of 25 ~ 45oC, severe voltage losses were observed in both the activation and ohmic loss regions. As Figure 5b3 shows, the ohmic resistance increased significantly at the highest test temperature of 45oC, indicating a significant loss of carbon at the anode’s membrane interface. As the electrochemical diagnostics in Figure 5 show, the use of an RTA with an OER catalyst significantly reduces the rate of degradation, particularly at temperatures above freezing. Figure 6a-b summarize the changes of the cell voltage after each reversal for the 0 wt% and 5 wt% RTA MEAs. The cell voltage values were taken at a current density of 0.4 A/cm2 and were normalized based on the BOL cell voltage at the same current density. Figure 6a is the normalized cell 25 ACS Paragon Plus Environment


voltage vs. cell reversal temperature. We can see that for both 0 wt% and 5 wt% RTA MEAs, the normalized cell voltages after reversal tests in freezing conditions show no significant change. However, the cell voltage decay becomes evident when the cell reversal temperature is above 5oC for 0 wt% RTA MEA and above 20oC for 5 wt% RTA MEA. Figure 6b shows the normalized cell voltage vs. accumulated cell voltage reversal time. For the 0 wt% RTA MEA, the cell voltage starts to distinctly decrease at around 250 s of accumulated cell voltage reversal time. In contrast, the cell voltage of 5 wt% RTA MEA decreases significantly at about 590 s of accumulated voltage reversal time. Both Figure 6a and b indicate that there was negligible COR damage to anodes when the fuel cell’s voltage was reversed in freezing conditions for both 0 wt% and 5 wt% RTA MEAs. When exposed to temperatures sufficiently higher than 0oC, however, the fuel cell with the 5 wt% RTA MEA degraded significantly less than 0 wt% RTA MEA, implying the effectiveness of the IrO2 RTA catalyst for suppressing the COR at anodes. To examine the voltage reversal damage to the morphology of anode catalyst layers, SEM images and EDS spectra were taken from the cross sections of the pristine and reversal-tested MEAs, and are shown in Figure 6c-d. The average anode thickness of the BOL 0 wt% RTA MEA was 3.0 µm based on SEM images in Figure 6c. The 0 wt% RTA MEA after repeated reversal tests at -5oC showed a negligible anode thickness change. However, the anode thickness of 0 wt% RTA MEA after repeated reversal tests with increasing cell temperature decreased to 2. 5 µm, i.e., approximately 17% of reduction. The change in anode thicknesses is due to the carbon corrosion and partial collapse of the anode when the cell was reversed at temperatures above freezing. This observation is supported by EDS spectra and the change in the element weight percentage. As Figure 6c shows, there is no significant element weight percentage change for the 0 wt% RTA sample after repeated tests at -5oC. In contrast, there is a notable increase in the 26 ACS Paragon Plus Environment

Page 26 of 36

Page 27 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


relative weight of Pt vs. carbon when the reversal temperature was increasing due to the loss of carbon. This trend is much clearer in Figure 6d that shows the results for the 5 wt% RTA MEA sample after repeated reversal tests with increasing cell temperature, where the ratio of Pt to carbon is dramatically increased due to the anode being exposed to more facile carbon corrosion conditions for a more extended period.

Figure 6. The change of normalized cell voltage at 0.4 A/cm2 vs. cell reversal temperature (a) and accumulated voltage reversal time (b). (c) Cross-sectional SEM images and EDS spectra of 27 ACS Paragon Plus Environment


anodes of the 0 wt% RTA MEAs without any reversal test (0 wt% BOL), after repeated reversal tests at -5oC (0 wt% EOL -5oC) and after repeated reversal tests with increasing cell temperature (0 wt% EOL increasing T). (d) Cross-sectional SEM images and EDS spectra of anodes of the 5 wt% RTA MEAs without any reversal test (5 wt% BOL) and after repeated reversal tests with increasing cell temperature (5 wt% EOL increasing T).

CONCLUSIONS In this study, we have investigated cell reversal degradation from -15 to 45oC through electrochemical diagnostics, including the use of an integrated reference electrode. We examined both conventional anodes and RTAs with OER catalyst added. From the hydrogen-air cell freezing reversal tests at -15oC, we observed substantially a different cell voltage time series from previously reported high-temperature reversal tests. Using the reference electrode, we observed that the anode potential exhibited an increase to a single voltage plateau consistent with water electrolysis reaction. The single voltage plateau is followed by a significant voltage increase to > 2 V due to capacitive charging, without any indication of the carbon corrosion voltage plateau observed at normal operating temperatures. Repetitive hydrogen-pumping cell reversal tests of the MEA at a constant subfreezing temperature of -5oC revealed no significant changes in terms of fuel cell polarization curve and EIS spectra. In contrast, repeated hydrogenpumping cell reversal tests with consecutive temperature increases from -15 to 20oC (conventional anode MEA) or 45oC (for 5 wt% RTA MEA), we found the carbon corrosion was highly suppressed at temperatures of 0oC and lower for both MEAs. At 10oC and above, the carbon corrosion became apparent with the conventional anode. With the addition of IrO2 OER catalyst (5 wt% RTA MEA), we found significantly increased reversal times and suppression of 28 ACS Paragon Plus Environment

Page 28 of 36

Page 29 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


carbon corrosion at higher temperatures (> 40oC). We also observed that the duration of the water electrolysis plateau decreased at the lower temperatures below freezing. This is likely due to the limited availability of water that is governed by the unfrozen fraction of the water and its mobility. These findings are encouraging for the development of PEMFCs that can be robustly operated in subfreezing conditions. Despite the negligible carbon corrosion at subfreezing condition, the temperature dependence of that carbon corrosion indicates the need for an RTA to prevent carbon corrosion as the cells warm up during initial operation.

AUTHOR INFORMATION Corresponding Author

*Shawn Litster, E-mail: [email protected]

ACKNOWLEDGEMENT This work was financially supported by the collaborative research project (R - 164690.0001) between Hyundai Motor Company, Hyundai-NGV, and Carnegie Mellon University. The authors acknowledge the use of the Materials Characterization Facility at Carnegie Mellon University supported by grant MCF-677785.



References (1)

Kongkanand, A.; Mathias, M. F. The Priority and Challenge of High-Power Performance of Low-Platinum Proton-Exchange Membrane Fuel Cells. Journal of Physical Chemistry Letters. American Chemical Society April 7, 2016, pp 1127–1137.

(2)

Tajiri, K.; Tabuchi, Y.; Wang, C.-Y. Isothermal Cold Start of Polymer Electrolyte Fuel Cells. J. Electrochem. Soc. 2007, 154 (2), B147–B152.

(3)

Oszcipok, M.; Riemann, D.; Kronenwett, U.; Kreideweis, M.; Zedda, M. Statistic Analysis of Operational Influences on the Cold Start Behaviour of PEM Fuel Cells. J. Power Sources 2005, 145 (2), 407–415.

(4)

Han, K.; Hong, B. K.; Kim, S. H.; Ahn, B. K.; Lim, T. W. Influence of Anisotropic Bending Stiffness of Gas Diffusion Layers on the Degradation Behavior of Polymer Electrolyte Membrane Fuel Cells under Freezing Conditions. Int. J. Hydrogen Energy 2011, 36 (19), 12452–12464.

(5)

Guo, Q.; Qi, Z. Effect of Freeze-Thaw Cycles on the Properties and Performance of Membrane-Electrode Assemblies. J. Power Sources 2006, 160 (2), 1269–1274.

(6)

Borup, R.; Meyers, J.; Pivovar, B.; Kim, Y. S.; Mukundan, R.; Garland, N.; Myers, D.; Wilson, M.; Garzon, F.; Wood, D.; Zelenay, P.; More, K.; Stroh, K.; Zawodzinski, T.; Boncella, J.; McGrath, J.E.; Inaba, M.; Ota, K.; Ogumi, Z.; Miyata, S.; Nishikata, A.; Siroma, Z.; Uchimoto, Y.; Yasuda, K.; Kimijima, K.I.; Iwashita, N.; Scientific Aspects of 30 ACS Paragon Plus Environment

Page 30 of 36

Page 31 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Polymer Electrolyte Fuel Cell Durability and Degradation. Chemical Reviews. American Chemical Society 2007, pp 3904–3951. (7)

Ralph, T. R.; Hogarth, M. P. Catalysis for Low Temperature Fuel Cells. Platin. Met. Rev. 2002, 46 (3), 117–135.

(8)

Mandal, P.; Hong, B. K.; Oh, J.-G.; Litster, S. Understanding the Voltage Reversal Behavior of Automotive Fuel Cells. J. Power Sources 2018, 397, 397–404.

(9)

Hong, B. K.; Mandal, P.; Oh, J.-G.; Litster, S. On the Impact of Water Activity on Reversal Tolerant Fuel Cell Anode Performance and Durability. J. Power Sources 2016, 328, 280–288.

(10)

You, E.; Min, M.; Jin, S.-A.; Kim, T.; Pak, C. Highly Durable, Cost-Effective, and Multifunctional Carbon-Supported IrRu-Based Catalyst for Automotive Polymer Electrolyte Fuel Cell Anodes. J. Electrochem. Soc. 2018, 165 (6), F3094–F3099.

(11)

Lim, K. H.; Lee, W. H.; Jeong, Y.; Kim, H. Analysis of Carbon Corrosion in Anode under Fuel Starvation Using On-Line Mass Spectrometry in Polymer Electrolyte Membrane Fuel Cells. J. Electrochem. Soc. 2017, 164 (14), F1580–F1586.

(12)

Chen, B.; Cai, Y.; Shen, J.; Tu, Z.; Chan, S. H. Performance Degradation of a Proton Exchange Membrane Fuel Cell with Dead-Ended Cathode and Anode. Appl. Therm. Eng. 2018, 132, 80–86.

(13)

Ohs, J. H.; Sauter, U.; Maass, S.; Stolten, D. Modeling Hydrogen Starvation Conditions in Proton-Exchange Membrane Fuel Cells. J. Power Sources 2011, 196 (1), 255–263. 31 ACS Paragon Plus Environment


(14)

Knights, S. D.; Colbow, K. M.; St-Pierre, J.; Wilkinson, D. P. Aging Mechanisms and Lifetime of PEFC and DMFC. J. Power Sources 2004, 127 (1), 127–134.

(15)

Zhou, F.; Andreasen, S. J.; Kær, S. K.; Yu, D. Analysis of Accelerated Degradation of a HT-PEM Fuel Cell Caused by Cell Reversal in Fuel Starvation Condition. Int. J. Hydrogen Energy 2015, 40 (6), 2833–2839.

(16)

Stolten, D.; Emonts, B. Fuel Cell Science and Engineering: Materials, Processes, Systems and Technology; 2012.

(17)

Zhang, J. PEM Fuel Cell Electrocatalysts and Catalyst Layers: Fundamentals and Applications; Springer Science & Business Media, 2008.

(18)

Mandal, P.; Hong, B. K.; Oh, J.-G.; Litster, S. 3D Imaging of Fuel Cell Electrode Structure Degraded under Cell Voltage Reversal Conditions Using Nanoscale X-Ray Computed Tomography. ECS Trans. 2015, 69 (17), 443–453.

(19)

Kangasniemi, K. H.; Condit, D. A.; Jarvi, T. D. Characterization of Vulcan Electrochemically Oxidized under Simulated PEM Fuel Cell Conditions. J. Electrochem. Soc. 2004, 151 (4), E125–E132.

(20)

Yu, X.; Ye, S. Recent Advances in Activity and Durability Enhancement of Pt/C Catalytic Cathode in PEMFC: Part II: Degradation Mechanism and Durability Enhancement of Carbon Supported Platinum Catalyst. J. Power Sources 2007, 172 (1), 145–154.

(21)

Wang, X.; Li, W.; Chen, Z.; Waje, M.; Yan, Y. Durability Investigation of Carbon Nanotube as Catalyst Support for Proton Exchange Membrane Fuel Cell. J. Power Sources 2006, 158 (1), 154–159. 32 ACS Paragon Plus Environment

Page 32 of 36

Page 33 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(22)

Chen, B.; Cai, Y.; Tu, Z.; Chan, S. H.; Wang, J.; Yu, Y. Gas Purging Effect on the Degradation Characteristic of a Proton Exchange Membrane Fuel Cell with Dead-Ended Mode Operation I. With Different Electrolytes. Energy 2017, 141, 40–49.

(23)

Chen, B.; Cai, Y.; Yu, Y.; Wang, J.; Tu, Z.; Chan, S. H. Gas Purging Effect on the Degradation Characteristic of a Proton Exchange Membrane Fuel Cell with Dead-Ended Mode Operation II. Under Different Operation Pressures. Energy 2017, 131, 50–57.

(24)

Chen, B.; Wang, J.; Yang, T.; Cai, Y.; Pan, M.; Tu, Z.; Zhang, C.; Chan, S. H.; Yu, Y. Mitigation Studies of Carbon Corrosion by Optimizing the Opening Size of the Cathode Outlet in a Proton Exchange Membrane Fuel Cell with Dead-Ended Anode. Energy Convers. Manag. 2016, 119, 60–66.

(25)

Chen, B.; Ke, W.; Luo, M.; Wang, J.; Tu, Z.; Pan, M.; Zhang, H.; Liu, X.; Liu, W. Operation Characteristics and Carbon Corrosion of PEMFC (Proton Exchange Membrane Fuel Cell) with Dead-Ended Anode for High Hydrogen Utilization. Energy 2015, 91, 799– 806.

(26)

Liang, D.; Shen, Q.; Hou, M.; Shao, Z.; Yi, B. Study of the Cell Reversal Process of Large Area Proton Exchange Membrane Fuel Cells under Fuel Starvation. J. Power Sources 2009, 194 (2), 847–853.

(27)

Liu, Z.; Yang, L.; Mao, Z.; Zhuge, W.; Zhang, Y.; Wang, L. Behavior of PEMFC in Starvation. J. Power Sources 2006, 157 (1), 166–176.

(28)

Taniguchi, A.; Akita, T.; Yasuda, K.; Miyazaki, Y. Analysis of Electrocatalyst Degradation in PEMFC Caused by Cell Reversal during Fuel Starvation. J. Power Sources 33 ACS Paragon Plus Environment


2004, 130 (1–2), 42–49. (29)

Hu, L.; Hong, B. K.; Oh, J.-G.; Litster, S. Investigation of Hydrogen Starvation of Polymer Electrolyte Fuel Cells in Freezing Condition Using Reference Electrode. ECS Trans. 2017, 80 (8), 535–542.

(30)

Piela, P.; Springer, T. E.; Davey, J.; Zelenay, P. Direct Measurement of IR-Free Individual-Electrode Overpotentials in Polymer Electrolyte Fuel Cells. J. Phys. Chem. C 2007, 111 (17), 6512–6523.

(31)

He, W.; V. Nguyen, T. Edge Effects on Reference Electrode Measurements in PEM Fuel Cells. J. Electrochem. Soc. 2004, 151 (2), A185–A195.

(32)

Oh, J.-G.; Lee, W. H.; Kim, H. The Inhibition of Electrochemical Carbon Corrosion in Polymer Electrolyte Membrane Fuel Cells Using Iridium Nanodendrites. Int. J. Hydrogen Energy 2012, 37 (3), 2455–2461.

(33)

Adams, R.; Shriner, R. L. Platinum Oxide as a Catalyst in the Reduction of Organic Compounds. III. Preparation and Properties of the Oxide of Platinum Obtained by the Fusion of Chloroplatinic Acid with Sodium Nitrate1. J. Am. Chem. Soc. 1923, 45 (9), 2171–2179.

(34)

Adler, S. B.; Henderson, B. T.; Wilson, M. A.; Taylor, D. M.; Richards, R. E. Reference Electrode Placement and Seals in Electrochemical Oxygen Generators. Solid State Ionics 2000, 134 (1), 35–42.

(35)

Jiao, K.; Alaefour, I. E.; Karimi, G.; Li, X. Cold Start Characteristics of Proton Exchange Membrane Fuel Cells. Int. J. Hydrogen Energy 2011, 36 (18), 11832–11845. 34 ACS Paragon Plus Environment

Page 34 of 36

Page 35 of 36 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(36)

Ge, S.; Wang, C.-Y. In Situ Imaging of Liquid Water and Ice Formation in an Operating PEFC during Cold Start. Electrochem. Solid-State Lett. 2006, 9 (11), A499–A503.

(37)

Ko, J.; Ju, H. Comparison of Numerical Simulation Results and Experimental Data during Cold-Start of Polymer Electrolyte Fuel Cells. Appl. Energy 2012, 94, 364–374.

(38)

Cherevko, S.; Kulyk, N.; Mayrhofer, K. J. J. Durability of Platinum-Based Fuel Cell Electrocatalysts: Dissolution of Bulk and Nanoscale Platinum. Nano Energy 2016, 29, 275–298.

(39)

Wang, J.; Yin, G.; Shao, Y.; Zhang, S.; Wang, Z.; Gao, Y. Effect of Carbon Black Support Corrosion on the Durability of Pt/C Catalyst. J. Power Sources 2007, 171 (2), 331–339.



Table of content


Page 36 of 36

Robust Operation of Fuel Cell Systems in Subfreezing

Recommend Documents