Polymer Electrolyte Fuel Cells

Polymer Electrolyte Fuel Cells

1. 1 GAS CHANNELS & FLOW FIELD (8)

   • GEOMETRICAL CONSTRAINTS
     • Geometry of flow needs to EVENLY DISTRIBUTE REACTANT throughout field
     • e.g. Meander, square, parallel, interdigitated
     • At low T fuel cells can produce liq. H2O => form blockage if there is insufficient flow
     • Constraints
       • must connect air & H2 chambers
       • At same current density need to provide ~2.5x flow of air compared to H2
   • ELECTRICAL CONDUCTIVITY & CONTACT RESISTANCE
     • Bipolar plates have high conductivity BUT
       • low contact resistance to gas transport layer
         • => if the lateral conductivity isn't high => local reversal of potential
         • Contact resistance 5x effect of bulk resistance
     • Increasing clamping pressure may help, but results in “tenting”
       • gas transport layer is pushed into the channel, making gas distribution more difficult
   • CURRENT & REACTANT DISTRIBUTION
     • As reactant flows along channel it is consumed => Reactant conc varies along channel
       • want to achieve even distribution of reactivity rate across plate
         • Want distribution to be HOMOGENEOUS
     • Amount of ox has to = amount of red => [H+] is fixed
       • Rate of run limited by H+ production => flow of H+ is dominated by migration => driving force is electric field
     • Conductivity of electrolyte is low => anode & cathode runs are strongly coupled
       • Local current density may be controlled by rate of either anode or cathode => whichever is smaller
   • HEAT REMOVAL
     • If fuel cell at 40% => 60% heat loss
     • low grade/low temp => difficult to get rid off
     • Bipolar plates: integral channels for cooling H2O
   • CORROSION RESISTANT
     • Typical electrolyte: superacid pKa~-6
       • highly corrosive towards metals
     • Fe & other TMS used in degradation of membrane materials
     • Metal oxides films highly resistant to corrosion e.g. Al2O3
   • WATER REMOVAL
     • See paper notes
   • EXAMPLE & STRUCTURE
     • Bipolar Plates: used to conduct current from +ive electrode of one cell to -ive of the next
       • Plates should be gas impervious
         • thin and lwt
           • Power density critically effected by thickness
     • E.g.
       • e.g. Stainless steel or graphite (high Ts)
       • Injected moulded carbon/plastic composites (low Ts)
       • Stamped steel w/ anti-corrosion coatings
   • The gas channels and flow fields distribute reactants through the fuel cell and channel current through the stack

2. 2 GAS TRANSPORT LAYER (4)

   • ELECTRICAL CONDUCTIVTY
     • In plane conductivity >10x higher than through plane
       • All fibres are in plane
         • anistropy => current flows well through fibres but not b/w
     • Contact resistance is important
   • REACTANT & TRANSPORT PROPERTIES
     • Need gas to penetrate under ribs
       • Flow under rib b/c 1) diffusion or 2) pressure b/w channels
         • => need layer to be highly porous
   • MECHANICAL PROPERTIES
   • WATER TRANSPORT PROPERTIES
     • Hydrophobicity introduced by adding ptfe emulsion to gas trans layer & sintering at 360C
       • If totally imbibed w/ H2O => gas can't get through
         • But cathode layer produces H2O so need to allow H2O into chain
           • => need hydrophobic & hydrophilic pores
   • EXAMPLES
     • Composed of carbon fibre e.g. PAN
       • Fibre: thermally decomposed T 2500
       • Paper: thin, flat, brittle
       • Cloths: woven, stronger, thicker
       • Felt: layers powder bound together
   • IS RESPONSIBLE FOR COLLECTING CURRENT & TRANSPORTING IT TO THE RIBS OF THE FLOW FIELD
     • PROVIDES AN ELECTRICALLY CONDUCTIVE BACKING MATERIAL FOR THE ELECTRODES

3. 3 THE MICROPOROUS LAYER (3)

   • POROSITY
     • Needs to have pore size b/w cat layer (<1um) & gas transport layer (100um)
   • ELECTRICAL CONDUCTION
   • WATER TRANSPORT PROPERTIES
     • Difficult to measured as buried under GTL => can use neutron radiography or X-Ray tomography
     • Need compromise between hydrophobicity and hydrophilicty or there will be a H2O build up somewhere
   • EXAMPLES
     • Composed of carbon & hydrophobic material e.g. ptfe
     • MPL produced by mixing carbon powder, ptfe solution and a pore former to produce an ink
       • Ink is screen printed or sprayed on gas transport layer
   • FUTURE DESIGN
     • Use carbon nanotubes: MWNTs, expensive
       • Improve properties => better distribution of pores
   • B/W gas transport layer (del large pores) & cat layer (fine pores)
   • No reaction happening: acts as buffer layer & as support to keep things flat
     • prevent thin cat layer being extruded into GTL
   • Conduct current
   • Aid water management

4. 4 CATALYST LAYER

   • PROPERTIES
     • Good proton/e- transport
     • Good H2O handling properties
     • High porosity & good reactant transport properties
     • Good long term stability
   • Electrochemical rxns occur in catalyst layer
   • FUTURE DESIGN
     • Production of ultra thin cat layers using SPUTTERING
     • Nanostructural delevopment of cat layers using regular structured substrates
     • Ultralow loading electrodes
   • ISSUES TO BE CONSIDERED
     • Issues with producing catalyst layers
       • interface is buried
       • No clear relationship b/w fabrication methods & layer properties
     • Amounts
       • Amount of soluble Nafion
       • Amount of catalyst
       • Amount of porous volume
     • OTHER
       • Delamination of cat layer from membrane => leads to local regions where no reactions occur
       • Thickness variation leads to greater gas crossover & thermal runaway => leads to pinholes
       • Cracks can also lead to pinhole formation in the membrane material
   • TWO TYPES: b/c in water/Nafion permeability of reactants is low but much higher in gas phase
     • ULTRA THIN
       • Difficult to produce but uniformly polarised
       • Proton conduction is fast and we only have to worry about local potential & diffusion of reactants
       • Water handling: flooding decreases the available pore volume
         • Can be modelled as a change in the effective diffusion coefficient of air
           • As diffusion coefficient approaches that found in pure H2O the performance becomes v poor
     • THICK
       • Thicker layers with gaseous pores throughout the layer to increase mass transport
       • Effect of finite conductivity of carbon => non-uniform potential distribution
         • Want low cathode potential
       • Higher porosity leads to better distribution of reactant

5. 4a CATALYST SUPPORT (5)

   • Improves dispersion of catalyst
   • Have high surface area
     • => distribute cat as uniformly as poss
     • Particles have para-crystalline structures
       • Morphology & particle size distribution is dependent on the source material & the process of thermal decomposition
   • Be electrically conductive
   • Be pure (2)
     • Addition of H3PO4 or ZnCl2 to precursor gets rid of halides
     • Steam &/ CO2 at 800-1100C gets rid of amorphous C
   • Be corrosion resistant
     • Loss of cat SA => strong effect on performance
       • Highest loss seen on highest SA carbon
         • Effect is smaller when high Pt loading used
   • Common structure
     • ~40nm C primary spherical particles
   • Future design
     • Aerogel
       • v. open support => v. sharp pore size distribution
       • Small pores, high Pt
       • High electrical conductivity
     • Carbon nanotube
       • 10% wt loading; MWNTs w/ d=30nm
       • May show better oxidative stability & electrical conductivity
       • hydrophobic, hard to get Pt to stick

6. 4b THE CATALYST (8)

   • Catalyse the electrochemical reactions leading to hydrogen oxidation & oxygen reduction
   • HIGH ACTIVITY FOR REACTIONS
     • HOR
       • H2 -> 2H+ +2e-
       • fast kinetics
         • Tafel-Volmer kinetics (see notes)
       • Little economic incentive to replace Pt on anode if using pure H2 as a fuel
     • ORR
       • H2O2 produced => bad b/c strong oxidant
       • Activity for oxygen reduction is modified because formation of oxide at high potential
       • BIG LIMITATION: Pt starts activity at too low a potential
       • Volcano plot
   • LARGE SURFACE AREA & SA/WEIGHT
     • smaller not necessarily better
       • trade off b/w surface activity & SA
         • Try to disperse cat as much as poss => want large SA
         • decrease area => activity/unit area starts dropping off
       • Too small can force particles to expose 'wrong' crystal face
   • RESISTANCE TO POISONING: FUEL DEPENDENT
     • Major poisons: CO, Cl & S containing compounds
       • See notes for CO removal
         • Use PtRu
     • can act revers/irreversibly w/ cat
     • Poisons bind stronger to cat than reactant
       • leads to a decrease in effective SA
         • Rate of loss dependent upon composition
           • PtCo retains SA better than Pt
         • Take into account both surface loss & activity loss
     • only a small amount req as only monolayer needed to block reactant
   • GOOD CHEMICAL STABILITY: CORROSION RESISTANT
   • RETAIN SA OVER TIME
   • SHOW A LONG LIFE
   • INEXPENSIVE
   • EXAMPLE OF STRUCTURE
     • Pt or Pt alloys => high activity & stability
       • Usually dispersed on high surface area carbon to increase the specific SA
       • Pt w/ H2; PtRu w/ reformate
       • Cathode: Pt, PtCo or PtCr
   • CORE-SHELL CATALYSTS
     • Put Pt on surface where it is most valuable & place other metal in core
       • Core metal mustn't corrode
       • Pt should surface segregate
       • Possibility of modifying electronic structure
       • defects poss
   • FUTURE MATERIALS
     • Mesoporous catalysts
       • unsupported so less problems w/ C corrosion
       • SA comes from internal porous structure
       • Produced by utilising a templating material

7. 5 PROTON TRANSPORT MEMBRANE MATERIALS (6)

   • SHOW GOOD PROTON CONDUCTIVITY BUT NO ELECTRICAL CONDUCTIVITY (3)
     • EQUIVALENT WEIGHT
       • The lower the EW the larger the no. of ion exchange sites
       • Polymer stability is reduced at low EW - dissolution occurs
       • Conductivity extremely sensitive to water content
         • Rapid transfer of H2O through Nafion b/c interconnection b/w sulphonic groups
     • PERCOLATION NETWORK
       • Proton conductivity due to the formation of a percolation network through the membrane
         • Strongly dependent upon amount of H2O present
       • Nafion proton density ~1M H2SO4
         • Proton mobility ~1/10th that of 1M H2SO4
     • Grottius vs Vehicular see notes
       • VEHICULAR MECHANISM
         • Proton attached to solvent e.g. H3O+
           • Moves at rate of vehicular diffusion
             • Vehicle counter diffusion possible
       • GROTTIUS MECHANISM
         • Stationary vehicles, only local motion, hopping
           • Always w/i H bond environment
             • Solvent reorientation provides H+ pathway
         • At low T FCS; Higher conductivity
   • PREVENT CROSSOVER OF GASES BUT HAVE GOOD WATER TRANSPORT
     • GAS CROSSOVER IN MEMBRANE NOT BENEFICIAL (4)
       • Permeability related to conc & diffusion coefficient of gases in membrane
       • H2 & O2 may chemically recombine
         • Permeation b/w anode & cathode is bad => chemically react/recombine => use resources & produces heat
         • Leads to pinhole formation & direct recombination
           • Flux dependent on thickness => thin section => hot spot => snow-ball effect
       • Current associated w/ gas crossover is a few mAcm-2
       • Nitrogen crossover into anode side still req purging
     • WATER ISSUE
       • Enters cell on anode & cathode => humidified gas
       • Transported from anode to cathode => ELECTROSMOTIC DRAG
         • Anode side depleting => conductivity by anode drops
         • Electrosmotic drag proportional to current
         • Electrosmotic drag drys out anode => leads to excess H2O at cathode
           • Too much H2O will flood cathode
       • Diffuses from the cathode to anode => BACK DIFFUSION
       • Water is produced at cathode
       • SOLUTION: make membrane thiner & increase back diffusion
         • Decrease dx: Fick's Law J=-D dc/dx
         • Also reduces resistance & material req
         • PROBLEMS: easier to tear & gas transport depends on thickness
   • BE CHEMICALLY & ELECTROCHEMICALLY COMPATIBLE W/ ELECTROCATALYST
     • Operating conditions are chemically hostile
       • acidic pH, pKa~-7
       • High potential on cathode side: oxidising
       • Low potential on anode side: reducing
       • Production of H2O2, .OH & .OOH poss (see notes)
         • Leads to degradation of solid polymer electrolyte and other MEAs
           • H2O2 undergoes autocatalytic decomp. in presence of Pt
             • => most damage occurs in membrane adj to cat layer
               • 2H2O2 --Pt--> 2H2O + O2
         • Chemical degrade of pFSA materials proceeds via H2O2 radical attack
           • Produces F- & low wt perfluorosulfonic acid species
         • Aliphatic hydrogens are v susceptible to degrad
   • SHOW GOOD MECHANICAL PROPERTIES
   • SHOW A LONG LIFE
   • BE INEXPENSIVE
     • Aims
       • Reduce cost of materials
       • Improve operating temperature
     • Future design
       • Use Nafion w/ some inorganic material to prevent dehydration e.g. SiO2
         • use new inorganic materials which are proton conducting themselves
       • Use materials which don't req H2O to show proton conduction
       • Use anion conducting membrane materials
   • COMMON STRUCTURE
     • Nafion: like ptfe w/ attached ion exchange group
       • Superacid ~-6 H+ from SO3H v labile
       • Conductivity highly dependent on H2O content
       • Phase separation causes formation of hydrophilic & hydrophobic domains

Bilder und Dateien der Mindmap