Membranes water content

For the NaCI electrolysis, finite-gap operation proved initially to be the most promising way to realise good and reproducible cell voltages. Predominantly the reasons were a proper control of the membrane water content, the relatively simple retrofit of existing membrane electrolysers and the ongoing utilisation of the peripheral caustic equipment with the ODC technique. 

For HCI electrolysis the cathodic reaction product is water, which is easily drained through the ODC without affecting the membrane water content. Consequently, the ODC can be attached directly to the membrane and pressure compensation is not necessary. The cell concept, which was developed in another co-operation with DeNora, could not be simpler - the basic cell principle is shown in Fig. 4.6. Initial laboratory tests conducted in 1994 at Bayer on the basis of old GE developments [4] demonstrated the feasibility of HCI electrolysis with ODC and the potential for a reduction of the cell voltage to about one-third of present values. 

Unfortunately, the value for depends upon the type of measurement and also varies according to the chemical nature of the membrane, its acid content (lEC), temperature, and water content. Increasing temperature leads to higher values, whereas reducing membrane water content via lower RH values results in decreasing values. With regard to membrane-based... 

Membrane Water Content. Whether the dilute solution or concentrated solution theory equations are used to model the membrane system, functional forms for the transport parameters and the concentration of water are needed. The properties are functions of temperature and the water content, In the models, empirical fits are... 

Figure 18. Pseudo-2-D simulation results at 0.4 A/cm where the feed gases are dry and countercurrent, (a) Water partial pressure profiles at four positions in the fuel-cell sandwich as a function of distance along the channel the positions are at the anode and cathode gas channels (I and IV) and catalyst layers (II and III), respectively. Also plotted is the value of fS, the net flux of water per proton flux, as a function of position. The data are from Janssen. (Reproduced with permission from ref 55. Copyright 2001 The Electrochemical Society, Inc.) (b) Membrane water content as a function of position both along the gas channel and through the thickness of the membrane for the same simulation conditions as above. The data are from Weber and Newman. (Reproduced with permission from ref 55 and 134. Copyright 2004 The Electrochemical Society, Inc.)...

Figure 9. Membrane water content (A) distribution in a three-straight channel fuel cell (only the half cell is shown due to symmetry) with anode and cathode counter-flow at 0.65 V or average current density of 1.1 A/cm. Anode and cathode feed conditions are humidification temperature = 50 °C, pressure = 2 atm, stoichiometry = 2, and the cell temperature = 80 °C.

Effect of Hydrophilic Bentonites. All membrane models imply a direct relation between flux and membrane water content. The gross water content of the membranes can be increased by incorporating pre-gelled hydrophilic bentonites into the membranes. The useful bentonite concentration is limited by the fact that pre-gelation introduces water into the casting solution (J ). 

During cell/stack operation, water content in the membrane is affected by the local intensive variables, such as local temperature, water vapor concentration in the gas phase, gas temperature and velocity in the channel, and the properties of the electrode and gas diffusion media. The power fluctuation can result in temperature variation inside the cell/stack, which will subsequently change the local membrane water content. As the water content in the membrane tends to be non-uniform and unsteady, this results in operation stresses. When the membrane uptakes water from a dry state, it tends to expand as there is no space for it to extend in plane and it can wrinkle up as schematically shown in Fig. 4 when the membrane dries out, the wrinkled part may not flatten out, and this ratcheting effect can cause the pile up of wrinkles at regions where membrane can find space to fold. The operation stress is typically cyclic in nature due to startup-shutdown cycles, freeze-thaw cycles, and power output cycles. 

The cool-down process of the cold-start experiment also provides an opportunity to obtain the membrane proton conductivity as a function of temperature at a known water content. Note that the temperature dependence of proton conductivity with low membrane water content is of particular interest here as PEFC cold start rarely involves fully hydrated membranes after gas purge. In addition, unlike PEFCs operated under normal temperatures, the membrane resistance under low water content and low temperature typical of cold start conditions is much greater than the contact resistance, making in-situ measurements of the membrane proton conductivity in a PEFC a simple but accurate method. 

Figure 1. Proton conductivity as a function of temperature for various membrane water content k between room temperature and —30°C. (reproduced with permission...

Effects of Key Parameters 2.4.1. Initial Membrane Water Content... 

During the initial current ramp period the cell voltage is higher as the initial membrane water content increases, as expected. In the... 

Figure 4. Effect of current density on voltage evolution during cold start from —30° C with initial membrane water content of /.j = 6.2. (reproduced with permission from Tajiri et al.18)...

Detailed investigation of ORR kinetics at subzero temperature was first reported by Thompson et al.9 Fig. 8 shows the Tafel plots measured at -20°C with various membrane water content. At high... 

The temperature-compensated HFR is then essentially indicative of membrane water content only, and the temperature dependence is removed from actually measured HFR data. 

Fundamental mechanisms of this relaxation phenomenon remain unknown and need future investigation. However, correlating the T-compensated HFR after relaxation with that after purge is of practical interest, because the initial membrane water content critically important for PEFC cold-start performance corresponds to the HFR after relaxation (during cool down), not the FIFR immediately after purge. For this reason, an empirical correlation is attempted between the HFR after purge and that after relaxation, as shown in Fig. 21. It can be seen that a reasonable correlation exists over a... 

The correlation shown in Fig. 21 provides a practical means to estimate the HFR or membrane water content as the important input to evaluate cold-start performance. That is, one can estimate the HFR after purge from Eq. (10) based on the purge conditions, and subsequently correct for HFR relaxation using Fig. 21. Based on the HFR value after relaxation or prior to cold start, one can use the analytical models and performance data developed in previous work to estimate the cold-start performance. 

Galvanostatic discharge of a fuel cell (MRED method) provided information related to liquid water in a fuel cell in a minimally invasive manner.157 Stumper et al.158 showed that through a combination of this MRED method with a current mapping (segmented fuel cell similar to the one discussed in Stumper et al.135), it was possible to obtain the local membrane water content distribution across the cell area. The test cell was operated with a current collection plate segmented on the cathode along the reactant flow direction. In addition to the pure ohmic resistance, this experimental setup allowed the determination of the free gas volume of the unit cell (between the inlet and outlet valves). Furthermore, the total amount of liquid water presented in the anode or cathode compartment was obtained. 

---