A 75kW class pilot scale MCFC stack composed of 128 unit cells was assembled and operated. 10,000cm2 class MCFC stack components are also developed to fabricate the MCFC stack. The separators with the effective area of 10,000 cm2 were adopted new design concepts such as 3-piece architecture, simple channel structure, and flat mask plate to improve the gas flow behavior and wet seal performance. The electrolyte impregnated anode and cathode with the effective area of 10,000 cm2 were fabricated by the ex-situ process to enhance the reliability of the stack during the heat-treatment. Finally, the gas distribution manifold embedded light-weight stack structure was employed. The stack showed the average open circuit voltage of 1.065V that near meet theoretical one and 78.5kW DC in power output. A 125kW class MCFC stack using the same components will be assembled and started to operate by the end of 2009.

In Korea, the research for the development of MCFC stack has been carried out since 1993. The research objective of the ongoing project started in 2004 is the development of 250kW class external reforming type MCFC power generation plant composed of two 125kW class internal-manifold type MCFC sub-stacks. Recently, some remarkable research results including the development of electrolyte impregnated electrodes and 3-piece architecture separators with 7,500cm2 and 10,000cm2 in effective electrode area were achieved. Several 5kw class stack operation tests proved new separators and stack components provided improved stack performances. A 100kW class MCFC stack composed of 130 unit cells with 10,000cm2 in effective electrode area was fabricated to evaluate the new stack concepts. These results will be reflected in the construction and operation of 125 kW class sub-stack and 250 kW class MCFC power generation system.

We report the first observation of the 13C nuclear magnetic resonance spectroscopy (NMR) of 13CO, adsorbed from 13CO saturated 0.5 M sulfuric acid solutions, onto the surfaces of commercial Ru-black nanoparticles. The 13C NMR spectra consist of a symmetrically broadened peak having a large isotropic shift as compared to CO adsorbed onto supported Ru catalysts. The variation of the spin-lattice relaxation rate follows Korringa behavior, indicating the metallic nature of adsorbed CO, in addition to varying across the spectrum in a Korringa-like manner. Motional narrowing of the NMR spectrum at higher temperatures, together with an additional contribution to the spin-lattice relaxation rate, indicate that adsorbed CO undergoes rapid diffusion on the particle surfaces. A two-band model analysis of the NMR results indicates that the CO adsorption bond is weaker on Ru as compared to either Pt or Pd. This is also supported by a reduction in the activation energy for CO diffusion on Ru vs either Pt or Pd nanoparticles.

(195)Pt NMR spectroscopic and electrochemical measurements were carried out on commercial Pt-Ru alloy nanoparticle samples to investigate the effect of high-temperature annealing in different vacuum/gas-phase environments. Samples annealed at 220 degrees C in Ar gas, or in a vacuum, did not show any demonstrable change in catalytic activity vs electrochemically reduced, room-temperature samples. In contrast, annealing at 220 degrees C in H(2) gas led to a 3-fold increase in reactivity toward methanol oxidation (per surface site). NMR experiments show that annealing at 220 degrees C (in both Ar and H(2)) leads to a slight reduction in the Fermi level local density of states (E(F)-LDOS) at the Pt sites, which we attribute to surface enrichment of Ru. This electronic effect alone, however, appears to be too small to account for the increase in the catalytic activity observed for the H-treated catalyst. By comparing the electrochemical and NMR data of the H- and Ar-treated samples, we conclude that annealing at 220 degrees C in the hydrogen atmosphere reduces surface Ru oxides into metallic Ru, and consequently, the presence of metallic Ru and its enrichment on the surface are essential for the enhanced catalytic activity. In contrast, heat treatment at 600 degrees C in both vacuum and argon atmosphere increases the particle size and reduces the amount of platinum on the nanoparticle surface, thereby increasing the surface Ru content beyond the optimum surface composition values. This causes a large reduction in catalytic activity. Our results suggest that optimizing the amount of surface Ru by heat treatment at temperatures near 200 degrees C, in a hydrogen atmosphere, can be utilized to produce Pt-Ru alloy nanoparticles with high methanol oxidation activity. Finally, our NMR and electrochemical data, taken together with the lattice parameter measurements, lead to a novel model of Pt-Ru alloy nanoparticles having a Ru-rich core and a Pt-Ru alloy overlayer.

Using an inverted spontaneous deposition method, unsupported Ru and carbon-supported 40% Ru nanoparticles were decorated with platinum after surface Ru oxides were reduced in the hydrogen atmosphere at 25°C for 1h. All samples that we produced in this way yielded high catalytic activity towards methanol oxidation at an electrode potential of interest to DMFC fuel cells. In the chronoamperometric experiment, the activity increased with the uptake of platinum to high Pt packing density values. The 40% C/Ru/Pt catalyst with an optimized packing density shows higher reactivity than that of an equivalent commercial, carbon-supported Pt/Ru catalyst. Reduction of Ru (and Ru/Pt) nanoparticles at higher than ambient temperature induced unacceptable level of sintering, and was avoided.

Lithiated NiO cathode dissolution has been a major problem for the development of molten carbonate fuel cells (MCFCs). Many studies have been contributed to find new alternative cathodes; here, lithium cobalt oxide, LiCoO2, was coated onto the commonly used NiO cathode by the electroplating method and the resulting cathode showed much reduced solubility compared with that of the common nickel oxide cathode. Thin film lithium cobalt oxide was prepared by the oxidation of Co metal deposited on a nickel plate in molten (Li,K)(2)CO3 at 650 degreesC under a CO2-O-2 (2 : 1 vol%) atmosphere. When this coated nickel plate was oxidized, the open circuit potential (OCP) decayed gradually, indicating two well-defined potential plateaux; the oxide films produced at each potential plateau were identified by X-ray diffraction methods. The surface product at the first plateau was CoO. LiCoO2 was formed at the second OCP plateau [around -0.47 V vs. CO2-O-2 (2 : 1 vol%) reference electrode]. By the same method, LiCoO2 was coated onto a porous nickel cathode in order to produce a MCFC.For 300 h steady operation of the cells, the mean voltages of the cells were 0.80 V using a NiO cathode and 0.85 V for a LC-NiO(EP) cathode at a current density of 150 mA cm(-2). In addition to the cell efficiency improvement, the solubility of the LC-NiO(EP) cathode was much lower than that of the NiO cathode.

The solubility of a nickel oxide cathode in molten carbonate fuel cell (MCFC) electrolyte is one of the major technical obstacles to the commercialization of such fuel cells. Lithium cobalt oxide, LiCoO2, has been selected as a candidate material for MCFC cathodes because its solubility is small and the rate of dissolution into the melt is slower than that for nickel oxide. On the other hand, the electrical conductivity of LiCoO2 is lower than that of nickel oxide. Thus, nickel oxide has been coated with stable LiCoO2 in carbonate by a PVA-assisted sol–gel method to give a LiCoO2-coated NiO (LC-NiO) cathode. Raman spectra show that the structure of LC-NiO is different from that of nickel oxide, and that a LiCo1−yNiyO2 phase is formed during heat-treatment of the LC-NiO cathode. The coating of LiCoO2 on NiO electrode increases with increase in the dipping and heating times. The performance of unit cells show that the mean voltage of the cells is 0.80 V using a NiO cathode and 0.85 V with a LC-NiO cathode at a current density of 150 mA cm−2. The solubility of the LC-NiO cathode in molten carbonate electrolyte is half that of NiO cathode after 300 h at 650°C.

Single phase La1-xSrxCoO3 (x less than or equal to 0.2) was synthesized as a uniform sized 100 nm particulates with relatively high surface area of 20-30 m(2)/g, at low temperature (greater than or equal to 600 degrees C), from a polymeric gel precursors prepared by using poly(vinyl alcohol) as homogenizer. No minor phase developed during the crystallization when polymer/metal mole ratio was higher than 3. As the polymer/metal mole ratio was raised in the gel, the amount of carbonaceous residues in the amorphous solid precursor prepared by heating the gel at 300 degrees C increased. Most of the residues were eliminated by exothermic thermal decomposition around 400 degrees C. The amount of residual carbon (less than 1%) left in the crystalline La1-xSrxCoO3 decreased as more polymer was used, eliminating detrimental effect which might be posed by using large amount of organic homogenizer. The crystal structure of La1-xSrxCoO3 synthesized at temperature lower than 800 degrees C was observed to be shifted from rhombohedral to more symmetric cubic. The structure shifted back to rhombohedral as the cubic sample was annealed at 1000 degrees C.

In order to determine the burnt-out condition of polyvinyl butyral as a binder in the fuel cell, thermal gravimetric analysis, gas chromatography and gas chromatography/mass spectrometry are used to analyse decomposed products during the thermal decomposition process in the matrix-green sheet and electrolyte-green sheet. Most of thermal degradation takes place under 400 °C, but degradation-resistant structures still remain up to 700 °C. Adding water vapour to the atmosphere gas could be one method to promote thermal degradation. Butyraldehyde and butene peaks among the released gases show characteristic decomposition behaviour. Thus, the butyraldehyde and butene peaks can be used as an index to check the extent of decomposition in the thermal decomposition process.