Accelerated Stress Tests in PEM Fuel Cells: What can we learn from it? D.P. Wilkinson 1,3, W. Merida 2,3 1 st Workshop : Durability and Degradation Issues in PEM Electrolysis Cells and its Components Fraunhofer ISE, Freiburg, Germany (March 12 th - 13 th, 2013) 1. Department of Chemical and Biological Engineering, UBC 2. Department of Mechanical Engineering, UBC 3. Clean Energy Research Centre (CERC), UBC
Presentation Overview Introduction Examples of PEMFC durability and degradation issues In-situ diagnostic techniques Spatial diagnostics Accelerated stress tests Closing remarks / summary
PEM Fuel Cell Research Closely Related to PEM Electrolysis Many PEM fuel cell advances applicable to PEM electrolysis Some related areas between the PEM fuel cell and PEM electrolysis: Electrolysis anode has similar requirements to PEMFC anode in cell reversal (fuel starvation) More durable and less expensive PEM membranes required 2 phase PEMFC cathode/flow field similar to 2-phase PEM electrolysis anode/flow field PEMFC anode/ flow field similar to PEM electrolysis cathode/flow field Bipolar stacked flat plate approach similar in both cases Air bleed on PEMFC anode for reformate compared with O 2 crossover to PEM electrolysis cathode Diagnostics and failure / durability testing similar for the PEM fuel cell and PEM electrolysis Etc, etc
Lifetime Issues Operational conditions, materials and design affect fuel cell lifetime Some key operational areas include: Low humidification High/excess humidification Low reactant flows High temperature Low temperature Fuel composition Current density etc
Failure Modes (Low Humidity) Low humidity Dehydration Resistance increase Increases cross-over Membrane pinhole development Knights, Wilkinson et al, J. Power Sources, 127, pp 127-134 (2004)
Low humidity and hot spots can lead to failures such as shorting and membrane holes (gas crossover detection by IR) Gas crossover site Wilkinson, Lamont : US patent 5,763, 765 (1998) Cell 4b Stack 033
Failure Modes (Flooding) Inlet Reactants Flooding Cell Temperature = 75 o C High current issue Mass transport issues Purging required Cell design important
Effect of Cell Design for Water Management on Degradation 0.70 0.65 1 µv/h Average cell voltage (V) 0.60 0.55 0.50 operation with a temperature rise, Mk513 8 cell stack, air/h 2, 2/1.5 stoichiometry, 3.04/3.04 atma, internal humidifier, 75 C coolant inlet, 85 C coolant outlet, 1.076 Acm -2 60 µv/h 0.45 isothermal operation, Mk5 single cell, air/h 2, 2/1.5 stoichiometry, 3.04/3.04 atma, internal humidifier, 80 C, 0.538 Acm -2 0.40 0 1000 2000 3000 4000 5000 6000 Time (h) Pierre, Wilkinson et al J. New Materials for Electrochemical Systems,3, pp 99-106 (2000)
Failure Modes (Low Reactant Stoichiometry) Fuel Oxidant Normal reaction: 2H 2 4H + + 4e - In absence of H 2 and depending on potential: H 2 O 1/2 O 2 + 2H + + 2e - (E o 25 o C = - 1.23 V vs. NHE) H + O 2 + 4H + + 4e - H 2 O C + 2H 2 O CO 2 + 2H + + 4e - (E o 25 o C = - 0.21 V vs. NHE) e - Anode e - Cathode Thomas, Hudson, Wilkinson, Electrocatalyst Stability and the Role of Fuel Starvation and Cell Reversal Tolerant Anodes : ECS Transactions 1 (8) 67 (2006)
Currents consumed generating O 2 and CO 2 for fuel cell undergoing fuel starvation
Enhanced water supply at anode sustains water electrolysis during fuel starvation reversal
Extended fuel starvation reversal at 200 ma / cm 2 with different catalyst compositions and structures Anode Catalyst: 20% Pt/ / 10% Ru on Shawinigan Anode catalyst + 20% RuO 2, IrO 2 (Ru 0.9 Ir 0.1 ) / Shawinigan Anode catalyst + 20% RuO 2 / Shawinigan Anode catalyst + 20% RuO 2, TiO 2 (Ru 0.9 Ti 0.1 ) / Shawinigan Anode catalyst Wilkinson, Knights et al, US Patents: 6,527,943 (2003) ; 6,936,370 (2005), etc
Degradation of carbon catalyst support and other carbon components during fuel starvation When Pt anode catalyst is supported on carbon, degradation can occur due to loss of catalyst support. In addition to the O 2 evolution reaction, C oxidation can occur: C + 2H 2 O CO 2 + 4H + + 4e - Degradation of carbon Loss of Pt/Ru Knights, Wilkinson et al, J. Power Sources, 127, pp 127-134 (2004)
Average stack performance recovery vs time spent in fuel starvation reversal 7 7 Performance (mv) 6 5 Pulse Set 1 Pulse Set 2 Pulse Set 3 Pulse Set 4 Pulse Set 5 Pulse Set 6 Pulse Set 7 Pulse Set 8 Pulse Set 9 Pulse Set 10 6 5 4 Degradation In Performance 4 Baseline Cell Reversal Tolerant 3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Recovery Time (min)
In-Situ PEM Cell Diagnostics for Durability and Degradation Ideal approach should target: As many failure modes as possible Be specific to as many components as possible Flow field Membrane Electrodes Etc Very low invasiveness Low cost Provide accelerated insight into mechanism(s), i.e., predict longer term issues
Reference Electrodes E a E c Fully hydrated hydrogen at constant conditions 1. O.E. Herrera, W. Mérida, and D.P. Wilkinson. "Sensing Electrodes for Failure Diagnostics in Fuel Cells." Journal of Power Sources 190, 103-109 (2009). 2. O.E. Herrera, W. Mérida, and D.P. Wilkinson. "New reference approach for fuel cell performance evaluation." Transactions of the Electrochemical Society 16(2), 1915-1926 (2008).
Anodic and Cathodic Component Overpotentials O. Herrera, D. Wilkinson, W. Mérida. Electrode overpotentials and temperature profiles in a PEMFC. Journal of Power Sources. 198, 132-142 (2011).
Anodic and Cathodic Overpotential Ratios (effect of different operating conditions) Activation Ohmic η / η0 Concentration Mass Transport
In-Situ Half Cell Electrochemistry for Electrode/Catalyst Evaluation LOAD Example CVs after Lifetime Testing e - e - PtRu Anode after 50 hours operation CO 2 /H 2 O H 2 H + H + / H 2 PtRu Anode after >5000 Hour Operation CO / H 2 O H 2 i (A) - ELECTRODE (PtRu) + ELECTRODE (Pt) POLYMER MEMBRANE (PEM) E (mv vs H 2 electrode) CO Stripping Voltammetry for Catalyst Surface Area (CO + H 2 O CO 2 + 2H + + 2e - )
Impedance Analysis for Failure / Degradation Analysis Low Humidity Failure Mode Flooding Failure Mode Z"/Ω cm 2-12 -10-8 -6-4 j = 0.1 A cm -2 T stack = 70 o C E cell /V 0.8 0.6 0.4 0.2 Drying time/min -20 0 20 40 60 80 100 t o a b 0.0 0.0 0.2 0.4 0.6 0.8 1.0 j/a cm -2 c d e f g h Z"/Ω cm 2-8 -7-7 -6-6 -5-5 -5-4 -4-3 -3 Normal Flooded E cell /V j = 0.2-0.5 A cm -2 T stack = 70 o C 0.8 0.6 0.4 Time/hours 1.6 1.8 2.0 2.2 2.4 0.2 b 0.0 0.0 0.2 0.4 0.6 0.8 1.0 a j/a cm -2-2 f e a b c d h t 0 o g 0 2 4 6 8 10 12 Z'/Ω cm 2-2 -2-1 -1 0 ~ 1 khz ~ 50 Hz a 0 2 4 6 8 Z'/Ω cm 2 b W. Mérida, An Empirical Model for Proton Exchange Membrane Fuel Cell Diagnostics, Electrochemical Society Transactions 5(1), 229-239 (2007).
Performance Losses from Tri-Oxidant Polarization (O 2 (21%)/N 2, O 2 (21%)/He, O 2 (100%)) : D (O 2 /He) / D(O 2 /N 2 ) ~ 3; D(O 2 )D(O 2 /N 2 )~ 7 Multi-component gas analysis (experimental and theoretical analysis) (J. Electrochem. Soc., Vol. 141, No.8, August 1994, pp 2084-2088 ; ibid., pp 2089 2090)
2-Phase Flow Analysis in PEM Based Cells Diagnostic tools to understand two-phase flow and flow distributions in PEMFC flow fields Provides better solutions for water management in PEMFCs This approach can be applied to PEM electrolysis H 2 O (l) = 1/2O 2 + 2H + + 2e - 1. R. Anderson, D.P. Wilkinson et al, A critical review of two-phase flow in gas flow channels of proton exchange membrane fuel cells, J. Power Sources 195 (2010) 4531-4553 2. R. Anderson, D.P. Wilkinson et al, Two-phase flow pressure drop hysteresis in parallel channels of a proton exchange membrane fuel cell, J. Power Sources 195 (2010) 4168-4176 3. L.Zhang, D.P. Wilkinson et al, Gas-liquid two-phase flow behavior in minichannels bounded with permeable walls, Chem. Eng. Sci., 3377-3385 (2011)
Anode Purge Diagnostic based on Anode Water Removal Normal Operation Anode Water Removal Water Concentration Profile Cathode Anode Cathode Water Concentration Profile Anode 1.2 Water Production ( ) Electro-osmotic Drag + Back Diffusion Water Production ( ) Electro-osmotic Drag + Back Diffusion 1 0.8 Standard Cell Anode Water Removal = Net Flux = Net Flux V (V) 0.6 Membrane Membrane 0.4 0.2 Test Conditions: 4.40 atm a Air and Hydrogen Inlet, 70 o C Cell Temperature, Stoichiometric Flow 2.0 Air (both cells) 0 0.00 0.50 1.00 1.50 2.00 2.50 1. Wilkinson et al, Electrochimica Acta, 40, pp 321-328 (1995) 2. Blanco, Wilkinson et al, Int. J. of Hydrogen Energy, 3716093 (2012) i (Acm -2 )
In-Situ Characterization and Diagnostics at UBC Challenges by regions or components Spatial resolution Concurrent techniques impedance, electrochemical techniques anode water removal, transient operating conditions, etc High throughput (combinatorial analysis) Material performance and degradation
Spatially resolved current distribution mapping in an operating fuel cell using segmented cells UBC Fuel Cell: 4 x 4 segmentation over 50 cm 2 J. Stumper, D.P. Wilkinson et al, Electrochimica Acta, Vol. 43, No.24, 3373 (1998)
Spatially resolved current distribution mapping in an operating fuel cell using fully isolated subcells J. Stumper, D.P. Wilkinson et al, Electrochimica Acta, Vol. 43, No.24, 3373 (1998)
Effect of Failure Mode Conditions on Sensing Electrodes (example for sensing external to active area) 21 Fuel in Fuel out Differences of SEs at different conditions while REs remain constant O.E. Herrera, W. Mérida, and D.P. Wilkinson. "Sensing Electrodes for Failure Diagnostics in Fuel Cells." Journal of Power Sources 190, 103-109 (2009).
Accelerated Fuel Cell Testing and Product Life Voltage cycling Accelerated Stress Parameters voltage, current density temperature humidification reactant concentration catalyst loading Etc, etc
Transient Testing for Accelerated Performance Degradation (e.g., cell voltage reversal testing) 10 Performance Degradation Degredation vs. vs. Time Time Spent Spent in Reversal in Reversal 666 5 459 Average Voltage (V) 0-5 -10-15 -20 Each cycle:30 x 10 sec reversal pulses at 200 ma/cm 2 Baseline 40/20 Pt/Ru/Shaw in PTFE Cell Reversal Tolerant Catalyst 0 100 200 300 400 500 Time (min) Baseline Catalyst
1 Accelerated Spatially Resolved Stress Testing Gradients in operating conditions allow accelerated testing because of enhanced sensitivity in some regions A variety of in situ characterization techniques can be applied and spatially determined Dynamic/transient spatially resolved measurements Water transport Current density, etc reference electrode anode segment electrolyte Current density/a cm -2 2.0 1.6 1.2 0.8 0.4 0.0 4 2 Row 3 2 3 Column 4 1
Parameter Change as a Result of In-plane Gradients Inlet Outlet Dimensionless oxygen concentration, temperature, and relative humidity profiles across a PEMFC flow field, x = dimensionless distance D.P. Wilkinson and J. St. Pierre, J. Power Sources, 113, pp 101-108 (2003)
Increased Sensitivity with Lower Reactant Concentration Cell inlet reactant concentration decreased Effect of stoichiometry on concentration gradient 1.0 ν Ο =10 0.8 5 Mass Transport Loss and Sensitivity Increases c O /c O (0) 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 X Sensitivity Increases 1 1.5 2
Increased sensitivity to operating parameters H2 Stoichiometry Sensor Cell Response 0.7 0.6 Voltage response of other cells in stack Cell Voltage (V) 0.5 0.4 0.3 0.2 0.1 Sensor cell voltage response 4 runs, different rates of flow change Early detection of out of range operating conditions or degradation Wilkinson et al, US Patent 6,673,480 (2004) 0 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 H2 Stoichiometry
Summary Many diagnostics and failure/durability testing methods for the PEM fuel cell are applicable to PEM electrolysis In-situ PEM cell diagnostics can have low invasiveness and provide specific component and mechanistic information More sensitive operating conditions and transients can be used to provide accelerated test information Spatially resolved measurements can provide an important accelerated test platform for operating conditions component materials and designs
Wilkinson Electro/Photo-Chemical Research Group (since 2004) Fuel Cell and Battery Research New materials and new approaches Technology simplification New catalysis approaches Failure modes, durability, and accelerated testing methods Clean Energy Research Electrochemical approaches to clean energy and clean water CO 2 reduction and solar fuels Hydrogen production and storage Photochemical & Photoelectrochemical hydrogen production Advanced electrolysis Metal and chemical hydrides Water treatment Some Wilkinson Group Research Members Sources of Funding: NSERC NRC CMC PICS CRC CFI WED Industry
Thank You! Questions? We have not inherited this world from our parents, but we have it on loan from our children - Native British Columbia Haida Quote