PHOSPHORIC ACID FUEL CELLS
5.2 SPECIAL FEATURES OF AQUEOUS PHOSPHORIC ACID
Like other acids, phosphoric acid (H3PO4) in aqueous solutions dissociates into ions according to
H3PO4!HỵỵH2PO4 ð5:1ị Unlike other acids, in concentrated phosphoric acid solutions (where the water concentration is low) the hydrogen ions exist not as hydrated ions H+nH2O but as ions solvated by phosphoric acid molecules: H+nH3PO4. For this reason, the conductivity of these solutions is a complex, nonmonotonous function of their concentration. As a conduction mechanism, these ions do not move like a spherical particle in a viscous medium when an electric field is applied to the solution (the Stokes mechanism), but rather, the protons alone jump in the field from one acid molecule to another (the Grotthuss mechanism, suggested as long ago as 1806 as an explanation for the conductivity behavior in aqueous solutions).
Another special feature of phosphoric acid is its dimerization to H4P2O7
in aqueous solutions at a concentration of about 85 wt% and higher 102 PHOSPHORIC ACID FUEL CELLS
temperatures:
2H3PO4!H4P2O7ỵH2O ð5:2ị This change is very important for fuel cell operation. Phosphate ions (H2PO4) adsorb well on a platinum catalyst surface, displacing the electrochemical reactants, which leads to an appreciably slower reaction. The pyrophosphate ions, H3P2O7, adsorb much less, so when they are present, the reaction goes much faster than in the presence of phosphate ions, reducing the polarization of the electrodes and raising the cell voltage at high current densities.
Another feature of concentrated phosphoric acid solutions that is very important for fuel cells is the water vapor pressure, which decreases drastically with increasing acid concentration. This feature allows the phosphoric acid solution to be immobilized in a porous solid matrix, greatly simplifying the elimination of water as a reaction product from the fuel cell’s cathode space by gas (oxygen or air) circulation. It is safe to adjust this circulation to the maximum current load (maximum rate of water production) without the need to readjust it at lower loads, as because of the immobilization, there is no risk of excessive drying of the matrix. In this way, water elimination has a peculiar self- regulation. No such feature exists in sulfuric acid solution, where for water elimination, the acid itself would have to be circulated, which would cause problems of sealing and of corrosion.
5.3 CONSTRUCTION OF PAFCs
Basically, the construction of PAFCs differs little from what was said in Section 1.4 about fuel cells with liquid acidic electrolyte. In the development of PAFCs and two decades later in the development of PEMFCs (described in Chapter 3), many similar steps can be distinguished, such as the change from pure platinum catalysts to catalysts consisting of highly disperse platinum deposited on a carbon support with a gradual reduction of platinum content in the catalyst from 4 to 0.4 and then to 0.25 mg/cm2, and the change from pure platinum to Pt–Ru catalysts. The bipolar graphite plates that have special channels for reactant supply and distribution over the entire electrode surface now used widely in PEMFC stacks were first used in PAFCs.
The concentrated phosphoric acid solution in a PAFC is absorbed into the pores of a porous matrix with fine pores and a total thickness of about 50mm.
From the outside, this matrix electrolyte behaves like a solid electrolyte (like the membrane in PEMFCs), preventing the reactant gases hydrogen and oxygen from getting to the ‘‘foreign’’ electrode and mixing.
In early work, Kynar poly(vinylidene fluoride), a thermoplastic material, was used to make the matrix. It was soon discovered that in concentrated phosphoric acid at high temperatures, it is not sufficiently stable chemically and
5.3 CONSTRUCTION OF PAFCs 103
produces fluorine-containing impurities tending to adsorb on the catalyst surface, lowering the catalyst’s activity and with it the fuel cell’s performance.
Among new materials suggested for the porous electrolyte matrix in PAFCs, we mention a mixture of silicon carbide (SiC) and PTFE (Mori et al., 1998). A suspension of the components is mixed in a ball mill for a long time, then spread onto the surfaces of the cathode and anode. This assures good contact between the electrodes and the electrolyte immobilized in the matrix.
Song et al. (2002) suggested using a mixture of fine and coarse silicon carbide particles to improve electrolyte and gas management in the matrix. Similar results can be obtained with a combination of silicon carbide and zirconium silicate particles (Neergat and Shukla, 2001).
The first hydrogen–oxygen PAFCs in the mid-1960s had 85% phosphoric acid and were operated at temperatures not higher than 1001C. Relative to the results obtained with alkaline hydrogen–oxygen fuel cells, the performance was poor. For this reason, subsequently the phosphoric acid concentration was gradually raised, first to 95% and then to 100%, and the temperature was brought up to 2001C.
During the decade between 1975 and 1985, research in this field was widespread, and large industrial organizations gradually joined the efforts.
5.4 COMMERCIAL PRODUCTION OF PAFCs
Due to the special features mentioned above, of the partial water vapor pressure decreasing with increasing phosphoric acid concentration, the elim- ination of product water from these fuel cells becomes markedly simpler. It was thus possible to make electrodes as large as 1 m2. In fuel cells of the PEMFC type, electrodes of such a size would give rise to unmanageably large difficulties in water elimination and the maintenance of water balance within MEAs. Thus, the first models of large-scale power plants producing a power of hundreds of kilowatts and more, and thus having industrial importance, were built from PAFCs.
United Technologies Corporation (UTC) built a large plant for PAFC production in 1969. Together with the Japanese company Toshiba, a special enterprise, Fuel Cells International (FCI), was created for mass production of these fuel cells and for their further improvement. A little later the FCI subsidiary ONSI was set up. In these industries, mass production of PC-25 power plants with an output of 200 kW was begun. These power plants are designated for the combined on-site heat and power supply of individual residential and municipal structures such as hospitals. These power plants were operated autonomously with natural gas, and in addition to the PAFC battery, included equipment for the conversion of natural gas to hydrogen and for subsequent hydrogen purification. Such a plant had a total weight of about 16 tons and occupied 4 m2of floor space. The electrical efficiency was 35 to 40%. Including the thermal energy produced, the total energy conversion 104 PHOSPHORIC ACID FUEL CELLS
efficiency was as high as 85%. The thermal energy was produced in the form of hot water with a temperature around 801C or as superheated steam at 1201C (the total thermal power was about 800 MJ/h). Unlike ordinary power plants of similar size, a PC-25 operates without producing noise or vibrations and without noticeable output of contaminants into the ambient air. Power plants of the PC-25 type have had a relatively large commercial success. In the United States, Japan, and various European countries, about 300 such plants have been installed.