We entered into the study of the photoelectrochernical processes of various semiconductor electrodes primarily because of its charming aspect in the pure scientific view point. But we were also determined to pursue the way to the industrial utilization of solar energy which should become the main future energy resources of mankind.
For an efficient solar energy conversion, it is substantial to use semiconductors having appropriately small bandgaps, and the main problem in this field is the fact that most of the semiconductor materials having such bandgaps are corrosive in aqueous solutions. Among various methods proposed for the stabilization, we have demonstrated that the application of ultrathin metal layers is very effective. Duration tests of over 5000 h were cleared for n-Si electrodes coated, with several nm layers of Pt and other metals.
Coverage of semiconductor electrodes with continuous metal layers, however, brought about another problem in that the photovoltage arising from the semiconductor-liquid junction is spoiled by the metal coating. We have been working on the effect of discontinuous metal coating on semiconductor photoelectrodes since the discovery of the “misterious” behavior of the metal-coated GaP electrode. The theory for the effect was proposed and supporting experimental results were accumulated. Various trials were made to really construct Ptisland (several nm in size) coated n- or p-Si electrodes and they successfully showed the predicted behavior. It was also shown that these electrodes generate photovoltages of the order of 0.65 V, which are much higher than those of the commercial single crystalline silicon solar cells. The solar to chemical conversion efficiencies of 10.8% for the decomposition of hydrogen iodide and the solar to electrical conversion efficiency of 14.0% in hydrogen iodide have been achieved with these electrodes.
These data, together with the fact that the photoelctrochemical cells do not need energyconsuming p, or n doping processes, indicate the possibility that the photoelectrochemcal cells can be a strong candidate for the industrial application of photovoltaic solar energy converters. The application of our technique to amorphous silicon materials and to new type solid photoelectric converters are also suggested.

View full abstract

The concentration of carbon dioxide in the atmosphere continues to increase with an increase of fossil fuel consumption. Methane is the dominant ingredient of natural gas, and is produced more widely throughout the world than crude oil. However, these gases are too inert to be used as raw materials for the chemical industry.
In the present study, an investigation has been made to convert CO₂ and CH₄ into more valuable compounds by using electric energy and photoenergy. CO₂ has been electrochemically reduced to methanol in the presence of homogeneous catalysts with a mediated electrode. The surface-confined compounds capable of functioning as a mediator were Everitt′s salt, metal porphyrins, and quinones. Methane has been converted into oxygen-containing compounds by photochemical reactions., Irradiation to gas mixtures of methane, water, and air with a low-pressure mercury lamp gave formaldehyde, methanol and acids as major products. The selectivity of this reaction was dependent on the molar ratio of methane to water vapor, the feeding rate of air and the rate of gas circulation. The methane conversion is initiated by the hydroxyl radical and hydrogen atom which are formed by photolysis of water vapor. The hydroxyl radical reacts with methane to produce methanol, and the hydrogen atom abstracts a hydrogen atom from methane to give a methyl radical. Further reactions of the methyl radical with methane and/or methanol yield species of higher order.

View full abstract

Photocurrent spectra, current-potential curves and photoelectrochemical properties of Pb, Zn or Au coated p-GaP photocathodes have been investigated to clarify their behaviors as the photocathode for photoelectrochemical reduction of carbon dioxide [CO₂] in aqueous electrolytes. In the case of Pb-p-GaP, silvery luster was observed on the electrode surface during the photoelectrolytic reduction of CO₂. Such a silvery state of the surface, probably metallic Pb, might prevent the hydrogen evolution reaction, and resulted in a remarkable increase of the faradaic efficiencies for the formation of HCOOH and CO, although the spectral photocurrents were depressed. On the other hand, Au coated p-GaP photocathode became more stable than the bare one, and the photocurrent was kept almost constant for many hours. In the aqueous solutions of tetraalkylammonium salts, the size of cation did not affect the faradaic efficiency for HCOOH formation. The larger cation, however, promoted the formation of CO.

View full abstract

High-efficiency crystalline silicon solar cells have been developed by optimizing the cell processing technologies including surface reflection, back-surface-field and surface passivation. The surface reflection could be reduced to less than 2% by applying a double-layer antireflection film on a V-grooved surface. The back-surface-field structure has been optimized by alloying a printed thick Al layer on the back in a lamp furnace. The process enabled the opencircuit voltage to increase by about 7%. The variation of carrier profiles after the surface oxidation has been investigated to examine the effect on open-circuit voltage.
Based on the developed technologies mentioned above, a 20.5 %-ef ficient crystalline Si solar cell has been fabricated. Surface recombination velocity and minority-carrier lifetime were estimated by fitting the experimental data to calculated ones from computer simulation.

View full abstract

We previously made it clear theoretically that semiconductor electrodes coated with minute metal islands were stable and generated very high open-circuit photovoltages (V_OC), and showed that n-type single crystal silicon electrodes coated with minute platinum (Pt) islands actually generated remarkably high Vc of 0.68 V. In the present paper we have tried to apply this method to amorphous silicon (a-Si) electrodes having n-i junctions (cf. Fig.1 (b)).
The a-Si electrodes were prepared with an rf plasma CVD apparatus, by depositing n -type a-Si layers and then i-type a-Si layers on metal 'plates or conductive oxide films deposited on glass plates. Four methods (cf. the first column of Table 1) were employed to deposit minute Pt islands. All the a-Si electrodes thus coated with Pt were stable and showed Voc's in the range from O.57 to 0.76 V, much higher than those for a-Si electrodes coated with continuous Pt layers (O.37 to 0.47 V) (Fig.3 and Talbe 1). The a-Si electrodes, coated with 3-nm thick Pt layers and etched in hot alkali solutions, showed the highest Voc of O.76 V (Table 1). This Vc is nearly equal to that of an n-i-p junction a-Si solid-state solar cell (O.72 to 0.87 V). These results are interesting in that they point to a new direction of the development of highly efficient photoelectrochemical solar cells using a-Si electrodes.

View full abstract

The electrochemical behavior of van der Waals face of n-InSe single crystals in acidic and alkaline solutions with and without illumination was investigated.
Photovoltage, Vph, depended linearly on the redox potential of the solution, Vr, when Vr was more positive than -0.35 V (vs. SCE) but Vo, was nearly zero when Vr was more negative than -O.35 V. These results suggest that the flat band potential, VFB, of n-InSe is fixed at -O.35 V. However, large Vph was observed in solutions containing S2-/S2couple, although Vr of these solutions were much more negative than -O.35 V, suggesting the negative shift of V_FB. Impedancp analysis confirmed the shift and AES measurements showed that sulfur was adsorbed on InSe surface. Thus, the specific adsorption of S^2- (or SH^-) onto InSe causes the negative shift of the conduction band edge, i. e., V_FB:
Photoanodic dissolution process was studied based on AES measurement, detection of dissolved species by in situ monitor electrode and UV spectra of solution. S^2- suppresses the dissolution quite significantly.

View full abstract

An anodic photocurrent was observed when an SnO₂ electrode, immersed in chromatophore suspension containing L-ascorbate and 1-methoxy-5-methylphenazinium methyl sulfate (m-PMS), was irradiated under potentiostatic conditions. The magnitude of photocurrent induced depended on chromatophore concentration, m-PMS concentration, ascorbate concentration, intensity of light, wavelength of the incident light and electric potential applied. A possible mechanism for induction of the anodic photocurrent was discussed.

View full abstract

Polycrystalline WO₃ was synthesized by the thermal decomposition of ammonium paratung: state. The optical property and the photoelectrochemical behavior of the WO₃ and WO₃ added with MoO₃ or metalic tungsten were investigated. The effects of added MoO3 and tungsten on the anodic oxygen evolution reaction were measured in acidic aqueous solutio ns under illumination with a high-pressure mercury lamp. The additive/WO₃ molar ra tios for MoO₃ and tungsten were 0.1 and 0.05, respectively. The optical band gap was measured by an UV-VIS spectrophotometer with an integrating sphere, and the values of both MoO₃/WO₃ and W/WO₃ systems were ca.2.6 eV.
For WO₃ electrodes of W/WO₃ molar ratio 0.005, the photocurrent of oxygen evolution had a maximum value and it was 1.5 times larger than that for the pure WO₃ polycry stalline electrode. In a concentration ranges of below 0.01, the increase in photocurrent corresponded to the change in their donor concentration obtained from Mott-Schottky plo ts. In the higher concentration ranges of additives, the photocurrent was decreased by the formation of carrier traps and recombination centers.
These results suggest that the anodic photocurrent of oxygen evolution is due to bandband excitation of the impurity bands arising from the lattice defects of oxygen and the addition of tungsten or MoO₃.

View full abstract

In order to survey the universality of the photocatalytic activity of titanium dioxide (TiO₂) as well as its relationship with the characteristics of the crystal type, various photo-catalytic reactions were carried out. The rates of six types of photocatalytic reactions were measured using twelve different TiO₂ catalysts. Catalysts made by chlorine method showed superior activity than those made by sulfuric acid method irrespective of the crystal structure, i. e. anatase type or rutile type. This is due to the fact that the catalyst made by former method has less quantity of hydroxyl groups on the TiO₂ surface. For the catalysts obtained by chlorine method, the rutile type shows superior photocatalytic activity than the anatase one towards some photocatalytic reactions.

View full abstract

Photodecomposition of water over TiO₂ powders loaded with Pt or NiOx was carried out in a top-illumination type flat cell. TiO₂ samples used were commercial anatase (TP-2 and P-25). Stoichiometric water photosplitting over Pt/TiO₂ (TP-2) was observed when the amount of NaOH solution was reduced to less than ca.0.3 ml (less than O.2 mm solution thickness), and the formal yield was maximized at ca.0.12 ml of solution volume. A similar dependence on solution volume was also observed on NiOxiTiO, (TP-2), but the stoichiometric formation of H₂2and O₂ was observed even in a large amount of solution. These results can be explained in terms of the recombination of H₂ and O₂ in bubbles formed on the catalyst immersed in solution. In H₂SO₄ solution, stoichiometric water photolysis over Pt/TiO₂ (TP-2)was declined rapidly and evantually stopped. Pt-loaded fine powdered TiO₂ (P-25) showed no activity for water photolysis in NaOH solution, but NiOx-loaded one as well as oxidized Pt-loaded one did. To explain these results, a mechanism was proposed in which hydrogen spillover from, the Pt to the TiO₂ surface occurs, leading to the recombination of hydrogen adatoms with oxygerr species produced by the photooxidation of water.

View full abstract

Photocatalytic properties of ternary niobate have been measured using the following methods: 1) measurement of photocatalytic hydrogen evolution rate in the pres ence of electron donor (HCOOH, CH₃OH); 2) measurement of photocatalytic oxygen evolution rat e in the presence of electron acceptor (KBrO₃); 3) determination of bandgap energy (E_g)using UV-Vis diffuse reflectance spectrum. Flatband potential (V_fb) and potential at the top of valence band (V_{top of VB}) have been estimated on the basis of the E_g value and of Mulliken′s electronegativity of niobate. In a manner similar to that has been observed for ternary titanates, the V_{top of VB} shifts towards th e negative direction and the V_{top of VB} towards the positive direction with the increase of E_g. The hydrogen evolution rate increases as the V_fb shifts towards the negative direction.

View full abstract

A few metal alkoxide polymers exhibit photocatalytic activities. In the present report, alkoxide polymers of various other metal elements were examined to see if such photocatalytic activities are possessed by metal alkoxide polymers in general. ESR observation has revealed that the metal alkoxide polymers also respond to visible light. The ESR response in a vacuum and in methanol vapor showed that there are two types of ESR intensity response; one type is featured with the weakened ESR signal appeared in alkoxide polymers of Cu, Mn, V, etc. and another is the strengthened ESR signal observed in alkoxide polymers of Ti, Zr, Sn, Nb, Ta. The formation of new paramagnetic species in the alkoxide polymer, when illuminated by visible light, apparently has a proportional relation to the photocatalytic activity of these metal alkoxide polymers, except for the polymers containing Ti and Zr.

View full abstract

Details in the reaction mechanism for ethylation of ammonia with PtiTiO₂ photocatalysts were described. The sequential ethylation in Fig.9 was clearly shown experime ntally (Fig.4). The ratio of bimolecular rate constants obtained from a simulation (Fig.5) was 1: 417: 175 for ethylation of ammonia: ethylamine: diethylamine. Reaction mixture in NMR sam ple tube with catalyst was irradiated with the apparaitus shown in Fig.2 and analyzed by using NMR spectroscopy (Fig.8). Quantum yield for total amount of ethylated products was 6% on 360-nm irradiation (Fig.10).

View full abstract

Methylviologen (MV²⁺) was \ educed with ascorbate (Asc^-) in inner solution of egg-phosphatidylcholine vesicle in the course of reaction photosensitized by Acridine Orange (AOH⁺). Electron was transported against thermodynamical potential gradient. Accumulation of MV^+. slowed down as the reaction proceeded and the concentration of MV^+. reached a constant value after an hour (Fig.1). Accumulated MV^+. was reoxidized slowly in the dark. This reoxidation must be the effect of small residual amount of oxygen which cannot be purged by our deaeration procedure and/or some other oxidizing agents.
In homogeneous aqueous solution containing Asc^-, MV²⁺ and AOH⁺, accumulation of MV^+. was not observed. In the homogeneous system, electron is transferred rapidly from MV^+. to the oxidized dye (AOH^2.+), and from MVt to the oxidized form of. Asc^- (sAsc^-). But, in the vesicle solution, these backward electron-transfer reactions were intercepted by adsorption of the oxidized dye onto the vesicle surface and by immediate reduction of the oxidized dye by ascorbate in the vesicle interior. Ascorbate radical (sAsc^-.) generated by one-electron process disproportionates into ascorbate and dehydroascorbic acid, the latter can hardly be reduced by one-electron donor such as MV^+..
Continuous illumination was performed under vari ous conditions. A reaction under a specific condition may be characterized by two parameters, initial rate of MV^+. formation (ν₀) and MV^+. concentration at the steady state (cs). Dependence of these param eters on the reaction conditions is shown in Figs.4, 7 and 8.
These results are explained by the kinetic model for the steady state shown in Fig.10. On the basis of this model, one obtains eqs. (8) and (14). Partitioning of the charge-separation products between bulk and membraneous phase is essential for efficient accumulation of MV^+., because it affects the value of k′s in these equations.
In conclusion, although photoreduction of MV^+. in the vesicle solution is irreversible, the existence of bilayers is essential for effective charge separation.

View full abstract

Norbornadiene-quadricyclane photo-thermal isomerization system is one of the most important and popular solar energy storage system. However, norbornadiene -itself does not possess any absorption band in visible light region where the solar light energy is the most denseregion.
For efficient utilization of solar light energy, norbornadienes [1] which possess a chalcone chromophore were synthesized by Diels-Alder reaction of cyclopentadiene with dehydrochalcones. Photoisomerization reactions of [1] gave corresponding quadricyclane derivatives [2] in good yields. The quantum yields of the isomerizations are O.06-0.60. The properties of [1] and [2] and the results of the reactions are shown in Table 1. Solvent effects and wavelength dependency of the isomerizations are also presented.
Isomerizations of [2] to [1] were carried out by contact with silver perchlorate, with trifluoroacetic acid, and even with silica gel. All the isomerizations proceeded under very mild conditions.

View full abstract

Incoherent solar radiation, collected by the solar concentrator installed in the Solar Energy Laboratory of our Institute, is successfully converted to a coherent laser radiation, i. e. the concentrated solar radiation is used to pump a water-cooled Nd: YAG laser rod 10 mm in diameter and 100 mm in length. The maximum output of the laser is 40 W in multimode, the highest output power reported so far in solar-pumped laser.
The high-temperature steam was electrolyzed using zircon ia ceramics as an electrolyte and hydrogen was produced with high efficiency of about 92%. The overvoltage of Pt electrode is analysed from temperature dependence of the relationship between electrolytic current and applied voltage. SEM observations of Pt electrode revealed that porosity of electrode greatly decreased after electrolysis. The decrease in porosity is considered as an origin of overvoltage.
From these fundamental investigations, it is concluded that solar-pumped laser and hightemperature steam electrolysis have a potential ability as new energy plants and that the construction of a solar hydrogen system is expected.

View full abstract

A methanol fuel cell combined with a photocatalytic reaction apparatus is proposed. This new type cell works as a methanol fuel cell in the dark and as a hydrogen-oxygen fuel cell under illumination. Methanol is reformed into hydrogen by the photocatalytic reaction using Pt/TiO₂, and evolved hydrogen is provided for the anode of the fuel cell. The performa nce of this cell under illumination is better than that of the methanol fuel cell.

View full abstract

Photocatalytic decomposition of acetic acid in TiO₂ suspensions is accelerated by addition of reducible species such as Cu²⁺, Fe³⁺, CuO and MnO₂. This process is proposed to involve the capture of the electron photogenerated in the TiO₂ by these species, which suppresses the undesirable electron-hole recombination. The decomposition is retarded when less reducible species such as Ti⁴⁺ and Sn⁴⁺ are added instead. Significant enhancement of the decompositi on is observed when copper powder is added to the suspension. This process is also considered to stem from the trapping of photogeneratedelectron in the TiO₂ by copper particles.

View full abstract

The photoassisted hydrogen storage in rare earth intermetallic compounds (R₂Co₇ and RCo₃)has been investigated using methanol or 2-propanol as a hydrogen carrier. Though the hydriding kinetics was negligibly small at room temperature, the irradiation by a highpressure mercury vapor lamp accelerated the reaction much rapidly. When the powders of rare earth intermetallic compounds were dispersed in a solution of alcohol with CCl₄ and irradiated at room temperature, the alloy was transfer-hydrogenated to form metal hydrides by the photodehydrogenation of alcohol. Irradiation of Sm₂Co₇ (ca. O.3 g) for 3 h in a solution of 2-propanol (220 cm³) with CCl₄ (0.05 cm³) resulted in the formation of α+β hydrides. It is concluded that upon irradiation of alloys in alcohol-CCl₄, the photodehydro genation of alcohol occurred to produce dehydrogenated products and hydrogen in the form of hydride.

View full abstract

Amphiphilic surfactants containing porphyrin with one or four poly(oxyethylene) chains were prepared by an immortal_polymerization. The photosensitizing effect of these polymers on the oxidation-reduction of L-ascorbic acid and Fast Red A(FRA) was examined in acetate buffer solution. The polymer having four poly(oxethylene) chains photosensitized the reduction better than the polymer with one poly(oxyethylene) chain which formed association. The effects of pH and of surfactant concentration on the photosensitized reaction were examined.

View full abstract

As a result of an investigation for the effects of various kind of metal oxides on separation of the HI-H₂SO₄ mixed solution produced in the S-I scheme, the Mg-S-I thermochemical water-splitting cycle was proposed as follows.
Sufficient rate and yield were experimentally confirmed for each of the constituent reactions. By use of the pyrolysis temperature difference between MgSO₄ and MgI₂, a batch system demonstration of the whole cycle was performed by repetition of 38 times of cycle operation with roughly constant production of O.3 dm³ H₂ and O.15 dm³ O₂ per hour, where the solid reactants were unmoved. More detailed study revealed that MgI₂ and MgSO₄ can be separated by the common ion (Mg²⁺) effect under the highly concentrated conditions of the MgI² MgSO⁴ mixed solution.
By use of the MgI²-MgSO⁴ separated process, a continuous flow system of the whole cycle was demonstrated for 33 h with roughly constant production of 0.5 dm³ H₂ and O.25 dm³ O₂ per hour, where all the reactants including solid ones were smoothly circulated in the cycle apparatus. Thermal efficiency of the cycle was estimated to be 17-39% for 65-85% of the overall heat recovery.

View full abstract

Thermal decomposition of metal sulfates was studied as an oxygen-generating reaction in the thermochemical water splitting process. The cycle using metal (M) sulfates is generally expressed as follows:
I₂+SO₂+2H₂O=H₂SO₄+2HI.
H₂SO₄+2HI+2MO=MSO₄+MI₂+2H₂O
MSO₄=MO+So₂+1/2O₂
MI₂+H2O=MO+2HI
2HI=H₂+I₂.
The initial decomposition temperatures obtained by thermogravimetry were related to thermodynamic functions: The equilibrium SO₃ pressure at the decomposition temperature was roughly 1×10^-4 atm at a heating rate of 2°Cfmin in flowing nitrogen. The decomposition rate of the sulfates is proportional to the surface area of a sample pellet, and not to that of a separate particle. The decomposition rate was in the order: Fe₂(SO₄)₃>CuSO₄>CoSO₄>NiSO₄>ZnSO₄>CdSO₄, which was nearly th e same as the decreasing order of decomposition temperature. Isothermal kinetic results indicated that sulfates of Fe, Cu, Co and Ni could be candidates for the thermochemical process.
In order to convert the produced oxides into sulfates, Cu and Fe(III) oxides were dissolved into sulfuric acid. The dissolution time, t, obeys and equation: log t= 1/2logE1-[H₂SO₄]+const. The dissolution rate was proportional to the surface area of a particle.
Thermodynamics show that iodides of Mg, Al and Fe(II) are hydrolyz ed easier than other iodides at low temperatures. For the purpose of composing a cycle using iron, sulfur and iodine, the products of the reaction I₂+2S0₂+H₂O+Fe₂O₂=2FeSO₄+2HI were investigated.

View full abstract

Thermal decomposition of potassium nitrate has been investigated over the temperature range from 850 to 1100 K as a removing step of oxygen in the KNO₃-I₂ hydrogen production cycle proposed by the present authors. Required amount of oxygen, i. e.0.25 mol/mol-KNO₃, was successfully removed in a plane alumina tube during the preheating of KNO, at 925 K for 1 h in a stream of argon, prior to its reaction with I₂ at 1000 K.
Formation of N₂ on the thermal decomposition of KNO₃ was measured by gas chromatography and nitrogen balance analysis. It was negligible up to 1100 K in a plane alumina tube, whereas significant amount of N₂ was formed in the presence of Pt, Au, stainless steel and MgO. It was considered that these materials catalyzed in formation of N-N bond.
As a result, alumina and zirconia were found to be the most suitable material as a reaction vessel since they reacted with neither I₂ nor KNO₃, and formation of N₂ on the decomposition of KNO₃ was negligible, while acidic oxides such as TiO₂ and SiO₂ reacted with KNO₃ forming double oxides and basic oxide such as MgO or metals caused formation of N₂.

View full abstract

As a stage of the study for a high-efficiency, high-temperature thermal energy transport, experiments were carried out on the heat transport by SO₂/SO₃ reversible thermochemical reaction in a labscale hermetic gas-circulating reaction system. In this system, the thermal energy is converted into the chemical energy in the form of SO₂ and O₂ by absorbing the reaction heat through the deoxidization process of SO₃ in the higher temperature side. The produced SO₂ and O₂ are transported to the lower temperature side by a gas-circulating pump. Then, the thermal energy is recovered by the exothermic oxidation of SO₂ with O₂. Subsequently, the produced SO₃ is fed back to the higher temperature side. In this study, the feasibility of this heat transport characteristics were investigated fundamentally in terms of both the change in reactant gas composition and the temperature change in a Pt-catalyst particle bed during SO₂-oxidiation and SO₃-deoxidization.
Under the conditions employed in the present experiments, SO₂ and SO₃ of nearly equilibrium concentration were found to be attained in both the oxidation and deoxidization reactors. It was verified that the thermal energy transport could be achieved continuously. Further, it was recognized that the catalyst-bed temperatures showed the maximum and the minimum in the vicinity of the bed inlet in the process of SO₂-oxidation and SO₃-deoxidization, respectively. These maximum and minimum temperatures became higher with increases in both the molar ratio of SO₂ to O₂ and the circulating rate of the reactant gas.

View full abstract

Chemical heat pump systems are useful for thermal energy storage. The most important factor for the actual operation of chemical heat pump is reaction rate. In the present work, liquid ammoniate NH₄NO₃⋅nNH₃ was selected as a low temperature material, because the dissociation rate of liquid ammoniate is relatively fast even at low temperatures. NiCl₂⋅6NH₃ has been adopted as a high temperature material. The present chemical heat pump system consists of the following reactions;
High temperature reaction:
Low temperature reaction:
On the basis of preliminary experiments, a bench scale system was constructed. NH₄NO₃(677 g), charged in a low-temperature vessel made of Pyrex glass, was ammoniated at 0°C until NH₄NO₃⋅2.4NH₃, was formed. The low temperature vessel was connected to high temperature vessels (280 mm in length and 50 mm diameter, made of SUS 304) containing NiCl₂⋅6NH₃ (480 g in total weight in 5 vessels). At the heat storage stage, NH₃ gas generated from the high-temperature vessels by heating at 195°C in a silicone oil bath was successfully absorbed by the liquid ammoniate within 8 h. At the heat-releasing stage, the reaction of NiCl₂⋅2NH₃ with NH₃ proceeded in the high-temperatture, vessels, which were immersed in a water bath, whereas the NH₃ gas was generated from NH₄NO₃⋅nNH₃ at 0°C. The flow rate of NH₃ gas was well suited for NH₃ absorption with Ni salt, and the reaction was completed in 2 h. The amount of released heat was calculated from the temperature increase of the water bath. The recovered thermal energy was 388 kJ amounting to 75% of the ideal value 520 kJ. Similar results were obtained even after 20 cycles of operation.

View full abstract

The controllable ranges of hydrogen equilibrium pressure, pressure hysteresis and plateau characteristics of Ti(Zr)-Mn hydrogen storage alloys with C 14-type (MgZn₂ type) Laves phase were improved by optimizing the alloy composition and the fabrication process. It was found that the ratio of, Zr to Ti controls the equilibrium pressure in a wide range (e, g.10-10⁸ Pa at 20°C, corresponding to a temperature range -50+ 250°C at 1.013×10⁵ Pa). Modification of the alloy composition by partial substitution, for example Cr, Cu and Ni for Mn, and the alloy homogenization by annealing at a high temperature, 1000-1200°C, were effective for improving the hysteresis and plateau characteristics in all range of hydrogen equilibrium pressure.
Ti(Zr)-Mn alloys react with hydrogen more rapidly than La-Ni and Fe-Ti alloys, and exhibit no degradation in cycle life tests in a closed system.
Based on these results we chose three alloy system s with different equilibrium pressure, and applied them to a double effect heat pump system operated at a low pressure. The system achieved high coefficency of performance (COP), 0.65 in cooling and 1.78 in heating, and proved to be promising for effective use of high temperature waste heat.

View full abstract

In order to improve the efficiency of electrolysis in the hybrid sulfur cycle for hydrogen production, the feasibility of molten salt electrolysis for the electrochemical step has been investigated, The process designed and examined in this paper is concerned with SO₂ electrooxidation in the hydrogen sulfate melts, which is related to the SO₂ depolarized process developed by Westinghouse Electric Corporation. Using a porous carbon anode, the electrolysis was performed at the cell voltage of 1.2 V at 100 A. m^-2. It is expected that the efficiency of the electrolysis will be improved by modification of gas (SO₂ and H₂O)supplying method.

View full abstract

Pretreatment conditions of nickel fine-particle catalysts have been studied by changing the temperatures of hydrogen reduction. The catalysts were used for an endothermic lowtemperature reaction in a newly-proposed chemical heat pump. Surface oxides of the catalyst were removed by hydrogen between 150 and 180°C according to TG analysis (Fig.1). Dehydrogenation rates of 2-propanol without acetone (82.4°C) were relatively high for the catalysts reduced at high pretreatment temperatures, whereas those of 2-propanol containing 5.5 % acetone (80.0°C) became rather unfavored (Table 1). Proper pretreatment conditions for high suspension stability and catalytic activity at high acetone concentrations are important for improving heat pump performance.

View full abstract

The catalytic adtivities, promotive effect and carrier effect of sulfided Mo based hydrodesulfurization catalysts and supported MoS₂ catalysts in the decomposition of hydrogen sulfide were examined at 773 K using a closed circulation system. Sulfided CoMo/Al₂O₃ catalyst and MoS₂ catalyst supported on γ-Al₂O₃ showed the highest activities among the Mo based hydrodesulfurization catalysts and supported MoS₂ catalysts, respectively. Cobalt sulfid e in sulfided CoMo/Al₂O₃ catalyst acted as the promoter for Mo/Al₂O₃ catalyst, whereas nickel sulfide in sulfided NiMo/Al₂O₃ catalyst acted as desulfurization reagent for hydrogen sulfide. Furthermore, acid properties of carriers in supported MoS₂ catalysts was supposed to play an important role for the decomposition of hydrogen sulfide.

View full abstract

Electrode kinetics of hydrogen oxidation and oxygen reduction in molten alkali carbonate have been studied extensively from the viewpoints of i) kinetic techniques, ii) electrode materials and iii) reaction order analyses of gas partial pressures. Comparative studies of kinetic parameters obtained by the AC impedance (AC) method, the coulostatic relaxation (CS) method, the potential step (PS) method and the chronocoulometry of PS transients have been carried out at various electrode materials, and kinetic features of both reactions in Li/K melts at 650°C have been disclosed. In order to reveal the reaction mechanism of hydrogen oxidation, the stoichiometric number (ν) was determined by the chronocoulometry. We obtained ν=2 on Ni, Pt, Ir, Au and Ag electrodes, supporting a CEC mechanism; H₂+2M_??_2MH, 2MH+2CO₃^2-→2M+2CO₂+2OH^-+2e, 2OH^-+CO₂_??_CO₃^2-+H₂O. The Allen-Hicklign plots with nlν=1 resulted in well-linear plots over the whole range of polarization and gave reasonable values of the exchange current density. It was concluded that the metal electrodes examined here have the same rate determining step. As to the oxygen reduction, two processes are known, the superoxide path and the peroxide path, but the actual path in Li/K melts is still an arguing point. In order to resolve this point, we made the reaction order analysis of the Warburg coefficients obtained from AC method, and reached the conclusion that the mass transfer is a mixed diffusion process of superoxide ions and dissolved carbon dioxide.

View full abstract

High-performance gas-diffusion electrodes are essential for practical fuel cells. Our efforts approaching to high performance have been devoted to the improvement of micro-structure in the porous gas diffusion electrode, based on the fundamental requirements to the electrode processes, i. e. high utilization of catalyst clusters and enough gas supply to the clusters, and also have been done to the development of electrocatalysts of high catalytic activity. The concepts, the preparation methods of electrodes and the evaluation technique are presented in the paper. It was found that a hot pressing is very important to meet these requirements resulting in the formation of fine and continuous electrolyte, gas and carbon black networks, respectively. Two preparation methods of high-performance catalyst layer and a gas-supplying layer were proposed. In the former one uses a single powder of semihydrophobic catalyzed carbon black and another one uses a mixed powder of a hydrophilic, catalyzed carbon black and a wet-proofed carbon black. Resulting electrodes exhibited larger values both in the utilization and the performance than that prepared by the conventional method by a factor more than 3. The gas-supplying layer exhibited high gas permeability and high hydrophobic property, suggesting the application of the materials to the gas network in the catalyst layer, which may promise long life to the electrode. The micro-structure of the reaction layer or the gas-supplying layer were analyzed experimentally in connection with the electrode performance. A more practical image of the structure has resulted in the comparison with any of the other models postulated previously. The importance of use of high surface area catalyst supports is demonstrated in achieving much higher activity of electrocatalysts than the existing level.

View full abstract

Ceria-based oxides with high ionic conductivity were investigated for the solid oxide fuel cell. The ionic conductivity of the ceria-samarium oxide system was higher than that of yttria stabilized zirconia and was the highest among the ceria-based oxides (Fig.3). The ceria-samarium oxides were used as the electrolytes of oxygen-hydrogen fuel cell. Althou gh the open circuit voltage of the cell for the ceria-samarium oxide system was low er than that for the zirconia-yttria system, the current density for the ceria-samarium oxide system was higher than that for the zirconia-calcia system at low temperatures. The overall overvoltage of the ceria-samarium oxide fuel cell is caused by the resistance and activation overvoltages (Figs.4 and 5). When the fuel cell is operated at 600-700°C, the output current is largely affected by the electrode material as well as the electrolyte. The perovskite-type, oxide, of La_0.6Sr_0.4CoO₃ exhibited a good performance as an oxygen electrode (Fig.6). The current interruption method revealed that the overvoltage at the oxygen electrode was low for La_0.6Sr_0.4CoO₃ at a small current density, but it was high at a high current density in comparison with Pt electrode (Fig.5). To suppress the reduction near the hydrogen electrode, a thin film of stabilized zirconia was coated on the fuel side of a ceria-samarium oxide disk by theion plating method. The resistance of the fuel cell slightly increased by zirconia coating. However, the fuel cell with zirconia-coated ceria-samarium oxide exhibited a high durability of the cell output and stable open circuit voltage.

View full abstract

The cathodic polarization of the La_1-xSr_xMO₃ (M=Mn, Fe, Co) electrodes sputtered on yttria-stabilized zirconia electrolyte was studied. The La_1-xSr_xCO₃ system showed high electrode activity for oxygen reduction. The electrode resistance was measured as a functions of P02 and temperature by, ac methods. The ac impedance behavior revealed that the rate determining steps for oxygen reduction were clasified as the charge-transfer process for La_1-xSr_xCO₃ the dissociation of oxygen molecules on the surface for La_1-xSr_xFeO₃ a nd the diffusion of oxide ions in the electrode for La_1-xSr_xMnO₃ The marked electrode activity of La_1-xSr_xCoO₃ was explained by the high dissociation ability of oxygen molecules.

View full abstract

Planar thin-film solid electrolyte cells were investigated to develop a small size portable power supply. Problems caused by the reduction of electrolyte thickness are analyzed to find effective counter measures. Thin-film cells are damaged by the thermal stresses generated from the mismatch of thermal expansions between electrolyte film and substrate. The stresses decreased (1) by the use of low thermal expansion coefficient materials for the electrolyte support and (2) by inserting a porous nickel layer between the electrolyte and the support. The solid phase reaction between cathode and electrolyte must also be described in the development of thin-film solid electrolyte cells. The cell performance was lowered by the electron conduction induced by the reduction of electrolyte film by the fuel gas at elevated temperatures. The grain boundaries enhance the solid-phase reaction and the expansion of the reduced area toward the cathode. Therefore, it is preferable to suppress the grain growth in the electrolyte film. A stack design of the planar thin film cell and a conceptual flow diagram for a power generation system are presented.

View full abstract

The authors have developed a diagnostic method for monitoring the degrees of gas and current leakage under the operation of molten carbonate fuel cell (MCFC). This method is based on the atomic mass balance and exhaust gas composition analyses. The fractions of water contained in the exhaust gases are obtained easily from measured dry-based exhaust gas compositions. Therefore, exact wet-based gas compositions can be applied to atomic mass balance analysis. Computer codes using this method are also developed. This method enables quantitative diagnosis of gas and current leakage in the MCFC under power generation without regard to supplied gas composition and operating pressure.

View full abstract

Electrolyte plates for molten carbonate fuel cells have been prepared by paper-making method (Fig.1), and dependence of cell I-V performance, cell life performance, and electrolyte plate strength on the porosity of the electrolyte plate matrix were investigated. It was found that the matrix of 60-70% porosity shows good I-V and life performance (Figs.4, 7), and that the matrix of about 70% porosity exhibits resistance against cracking (Fig, 11). These results suggested that the I-V and life performances are related to the relative reduction of the thickness of electrolyte plate which depends on the porosity (Fig.5), and that the strength of electrolyte plate is related to the bending strength of the matrix, the bending elastic modulus of the matrix, and the linear expansion coefficient of the electrolyte plate all of which depend on the porosity (Figs.8, 9, 12).
To confirm these results, a thermal cycle test of a 200 cm² single cell was carried out by using about 70% porosity matrix. No performance reduction occured for 1800 h test including 25 times thermal cycle (Fig.13).

View full abstract

The dissolution of the cathode (in-site type NiO) in co-flow type molten carbonate fuel cells (MCFC) was investigated. The anode was made of Ni-10% Cr powder, and the electrolyte tile consisted of LiAlO₂ and carbonate (Li/K=62/38) powder. These components were manufactured by a doctor-brade process. These cells were operated several times (500-4000h) at 650°C. The amount of dissolved Ni was estimated by chemical analysis. Deposited Ni particles were observed in the tile by the characteristic X-ray technique. These particles had two types of shape (grain and column). The content of Ni in the tile was higher near the gas inlet than that near the outlet. The dissolution rate of Ni depended on the CO₂ partial pressure in cathode gas. The higher dissolution rate was observed at low er temperatures. This tendency might be caused by the equilibrium solubility of NiO in the carbonate. It is supposed that the fraction occupied by carbonate in the cathode′s pore had influences the NiO dissolution rate.

View full abstract

NiO has been utilized as the cathode of molten carbonate fuel cells (MCFC′s). But NiO is reported to be unstable in long-time operation, since it dissolves in the carbonate melt. In order to improve the stability of a cathode of MCFC, a new cathode material of low solubility should be developed. In this study, the solubilities of nickel ferrites (Ni_δFe_3-δO₄)of several compositions (=0.33, O.53, O.75, 1.00) have been measured in a Li-K binary eutectic carbonate melt (Li₂CO₃: K₂CO₃=62: 38 mol%) under a carbon dioxide pressure ranging from O.1 to 1 atm in a temperature range from 873 to 1023 K, in order to estimate their stability when used as a cathode of MCFC.
The solubilities of nickel and iron in th e ferrites increased as the carbon dioxide pressure increased. This might be explained by the acid fluxing mechanism. The dependence of the solubilities of nickel ferrites on the pressure was smaller than that of nickel oxide. The solubilities of nickel decreased with a raise in temperature in any composition. In the ferrites of lower nickel content the solubility of iron decreased at higher temperatures.
The solubilities of iron in the higher nickel ferrites increased at higher temperatures. The solubilities of nickel ferrites were much smaller than that of nickel oxide by a factor of 3-18 under any conditions of this work. Considering the solubilities, the nickel ferrite might be a more stable cathode material of MCFC than nickel oxide. However, the ferrites reacted with the carbonate melt and the reaction products had a low electrical conductivity, although the nickel ferrites themselves have a good electrical conductivity, especially for the low nickel ferrites. These points should be improved in utilizing these materials as the cathode of MCFC.

View full abstract

We have been engaged in the development of a phosphoric acid fuel cell with an air-cooled system. This paper discribed the cell structure, cell performance and cooling system, DIGAS cooling method and SGC cooling method, especially the evaluation of cooling air temperature of the cell stack inlet. Based on this study, the 50 kW fuel cell test system and 200 kW fuel cell cogeneration demonstration power system were developed and operated.

View full abstract

A 220 kW air-cooled phosphoric acid fuel cell demonstration plant has been developed with a view to apply to a cogeneration system for in-city use, and installed at Shin-Tokyo power station in Tokyo, Japan. A preliminary test has been carried out to define the system performance and to assess the operationability of individual components.
The system can be characterized by its newly-designed air-cooled fuel cell, which operates at an atmospheric pressure, and a compact rectangular reformer with a high thermal efficiency. The test results revealed that the system was capable of generating power at a rate of 207 kW, though the designed value at full load; 220 kW; could not be achieved because of minor technical problems, that could be easily improved after this test phase. The fuel cells gave satisfactory results in the voltage-current characteristic curve and the uniformity of the cell voltage. The reformer could be heated up at a rate of 350°C/h, so that the rapid cold-start operation was possible. The initial operation test was, thus, successfully carried out, and the necessity of further experimental work was supported.

View full abstract

Experimental and theoretical investigations on the increase of hydrogen concentration at the cathode outlet during maintenance operation of a phosphoric acid fuel cell plant have been carried out using full size cells (Fig.1). The study clarified the two processes by which hydrogen is transferred into a cathode: one is gas diffusion based on a concentration gradient and the other is production of hydrogen by the electrode reaction of a concentration cell.
Even when cathode pressure is higher than anode pressure, some hydrogen diffused into the cathode. In this case, the pressure gradient does not influence the diffusion rate significantly (Fig.6). Conversely, when anode pressure is higher, the amount of transferred hydrogen increases remarkably according to Hagen-Poiseuille flow. Under a negative pressure gradient, the amount of transferred hydrogen is mainly determined by gaseous diffusion. The diffusion rate seems to be dependent strictly on the size and distribution of the micropores at the cell edge seals. On the other hand, when a hydrogen concentration gradient exists, hydrogen is produced at the cathode, according to the electrode reaction of a concentration cell when a low dummy load is connected between the electrodes. The hydrogen increases linearly with increasing load current (Fig.11), and the amount can be estimated from the two-electron reaction of a hydrogen concentration cell.
Theoretical cell voltage can be calculated by using Nernst′s equation. The calculated value agrees well with the experiments. The amount of hydrogen, based on a concentration cell reaction, is much larger than that based on hydrogen diffusion through cell edge seals. Therefore, it is important to disconnect the dummy load at the time when the actual cell voltage drops to the theoretical voltage of the concentration cell, or residual oxygen at the cathode is effectively consumed by the electrochemical reaction during the maintenance operation.

View full abstract

This study aims at a solidification of fuel cell electrolyte using polyphosphoric acids. When the electrolyte of phosphoric acid cells was replaced by metaphosphoric acid, the density of limiting current was greatly lowered without an essential change 'in the polarization. This marked performance drop was attributed to the high viscosity of polyphosphoric acids. Phosphoric acid was solidified when copolymerized with silicic acid, becoming a solid proton conductor which could be used at 200°C for a fuel cell electrolyte. The all-solid cell installed with this solid electrolyte sustained a load of 400 mA/cm² with approximately the same anode polarization as for the phosphoric acid cells, but with much greater potential drops of the cathode. The cell performance has been discussed in relation to the effective reaction area of the electrode-electrolyte interface.

View full abstract

The solubility of oxygen is usually high in fluorinated organic solvents such as trifluoromethanesulfonic acid (TFMSA). The kinetic of the oxygen reduction in TFMSA and 85%phosphoric acid solutions\ is of interest from the stand point of searching for electrolytes alternative to the phosphoric acidspecifically used in a practical acid fuel cell. The main aims of the present work are to compare the behavior of oxygen reduction in the TFMSA and 85% phosphoric acid solutions on a rotating platinum disk electrode (Pt-R. D. E. ). The species generated in the mixed solutions have been analyzed with a laser Raman spectroscopy.
The band at 1042 cm^-1 assigned to S-O in CF₃S0₃^- and the band at 777 cm^-1 assigned to C-S in both CF₃S0₃^- and CF₃S0₃H were observed in the laser Raman spectra. The intensity ratio of the band at 1042 cm^-1 to that of 777 cm^-1 (I₁₀₄₂/I₇₇₇) was dependent on the degree of electrolytic dissociation of TFMSA. From the plots of the ratio against the mole fraction of TFMSA (X), the TFMSA may be nearly completely dissociated in the range of X<<0.45. However in X >0.45, a part of the TFMSA remained in the form of molecule in spite of the interaction between a part of the proton produced by the electrolytic dissociation of TFMSA and oxygen of P=0 bond in molecular H₃PO₄.
In the Pt-R. D. E. method, the diffusion-limiting current density (i_d) was proportional to the Levich′s B value, where B is evaluated from the reciprocal of the slope of 1/i_d vs.1/ω^1/2 plots. As X increased, the B value increased due to the enhancement of oxygen solubility and diffusion coefficient. Simultaneously, in the range of X>>O.2, the exchange current density (i₀) in the mixed solutions was larger than that in 85% H₃PO₄. The addition of TFMSA in 85% H₃PO₄ is effective to increase io and the cell voltage of this acid fuel cell.

View full abstract

A platinum/carbon catalyst of the phosphoric acid fuel cell was evaluated to obtain high stability and activity. Among three carbons (Denka Black (acetylene black, Denkikagaku Kogyo), Regal 660 R, Vulcan XC-72 R (these two are furnace blacks, Cabot Corp. )), Denka Black had the lowest rate constant for the reaction with oxygen and steam (Figs.2, 4 and 7, Tables 2-4). Denka Black was considered to be stable for oxygen and steam, because the results of X-ray diffraction show that the spacing of graphitic layers is smaller and the crystallite size is larger than in the other carbons (Figs.3, 5 and 6).
The oxygen reduction was measured in about 98 wt% phosphoric acid at 190°C. Platinum/Denka Black also showed the highest activity from the polarization curves (Fig.8). Such activity seemed to be affected by the platinum particle dispersion, though the platinum crystallite (particle) sizes were similar in all the catalysts. Platinum particles supported on Denka Black were highly dispersed, as judged from the observation by transmission electron microscopy (Fig.9).

View full abstract

In order to commercialize alkaline fuel cells, it is neccessary to reduce the cost without degrading the efficiency and life. In this study, the quantities of Ag and Pt catalysts for air electrodes were reduced by loading them on various carbon powders.
For the Ag catalyst, carbon powders with large surface area and high corrosion resistance were selected as the supports.50 wt% of Ag-colloidal graphite electrodes (9 mgAg/cm²) and 66.7 wt% of Ag-furnace black A electrodes (6 mg Ag/cm²) revealed small polarizations (Fig.7)and long lives (Fig.9).
For the Pt cataly st, acetylene black with high corrosion resistance was selected as the support. It was found that 14-20 cm² of Pt surface area per 1 cm²×1 μm of catalyst layer and more than 12 pm of the layer thickness are neccessary to get low polarization electrodes (Figs.12, 13). The Pt-acetylene black electrodes (0.3 mg Pt/cm²) showed better performance than the Ag-carbon electrodes (Table 3).

View full abstract

Characteristics of oxygen electrode of an alkaline fuel cell in concentric cylindrical fluidized bed electrode (CFBE) were studied in 6 mol⋅dm^-3 KOH solution at 60°C. A cylindrical nickel screen of 300 mm high and 40 mm in diameter was installed in the CFBE as a collector electrode.
Silver-activated carbon catalyst and alkaline electrolyte in the cell were fluidized by the introduction of oxygen gas.
The polarization charac t eristics were measured galvanostatically. The observed polarization curves were analyzed on the basis of a two-phase model. The performance of the CFBE was markedly improved with increasing catalyst loading. Apparent exchange current density, i₀, increased with the amount of the catalyst loaded. The effective specific resista nce of catalyst particle phase, ρm, decreased with increasing amount of catalyst.
An effective distance of electron transfer was approximately c onstant regardless of the amount of catalyst loaded.

View full abstract

A fluidized bed was used as a hydrogen electrode in an alkaline fuel cell. A collector electrode of a concentric tube made of a nikel net was set up in the fluidized bed. Best electrode characteristics were obtained when a flow pattern in the fluidized bed attained vigorous vortexes in the collector electrode within 5 cm high from the bottom. In order to obtain the vortical flow, it was necessary to regulate the catalyst amount, catalyst particle size and superficial velocity of hydrogen gas. A hydrogen storage alloy CaNi₅ was suitable for the catalyst in the fluidized bed electrode.

View full abstract

Various modifications were carxied out for a Pt-solid polymer electrolyte (SPE) electrode catalyst to increase the catalytic activity of a methanol fuel cell anode. The electrodeposition of Sn or Ru on Pt and the alloy formation of Pt with Ru and/or Sn were found to lead to the enhancement of the activity of Pt for methanol electrooxidation. The highest activity of Pt itself was estimated as high as 5 mA⋅cm^-2 at 0.6 V (vs. RHE) and 23°C for a true Pt surface area by the measurement of the transient method. At present, a Pt Ru Sn alloy-SPE gave a catalytic qtivity of the apparent current density, 50 mA⋅cm^-2, at 0.4V and 60°C in 1 mol⋅dm^-3 CH₃OH+0.5mol⋅dm^-3 H₂SO₄, which was the highest among the Pt-SPE′s prepared.

View full abstract

Electrolysis of aqueous HBr was investigated by use of an electrolytic cell equipped with a membrane-electrocatalyst composite (Ir-Pt/Nafion 117/Pt-Ir) of 50 cm². The cell voltages increased almost linearly with an increase in current density in the range from 10 to 100A/dm² and were strongly affected by the difference of the theoretical decomposition voltage (Fig.2, Table 1). The IR losses were mainly due to the membrane resistance and decreased with the rise of temperature and with decreasing HBr concentration (Fig.3). The sums of the overvoltages for hydrogen and bromine evolution reactions at 100A/dm² were 0.21 and O.18 V at 25 and 50°C, respectively, and were independent of HBr and Br₂ concentrations (Fig.4). The current efficiencies higher than 97% at 25°C and than 96% at 50°C were obtained above 40 A/dm² (Fig.5). In addition, the electrolytic performances were also examined under such conditions as HBr concentration in the catholyte was different from that in the anolyte (Tables 2 and 3, Fig.6), as external supply -of the catholyte to the cathode was omitted (Fig.7), and as the composite with nonplated anode electrocatalysts was used (Fig.8).

View full abstract

Platinum cluster-supported electrodes were prepared from Pt and Ru carbonyl complexes. [Pt₃(CO)₆]_n2M(n=3, 5), [PtCl₂(SnCl₃)₂]2M, [Pt₃Sn₈Cl₂₀]4M, [Pt₃Fe₃(CO)₁₅]2M and [HRu₃(CO)₁₁]M (M=Na⁺, NMe₄⁺, NEt₄⁺, NMe₃(CH₂ph)⁺) as precursors by an ion-exchange tech nique on an anion type solid polymer electrolyte membrane or an graphite which was surfacemodified with a silanizer of quarternary ammonium salt. The cluster-supported electrodes with Pt₉/C Pt₁₅/C but not with Pt or Pt₃, revealed a high level of electrocatalytic activity towards the anodic methanol oxidation in acidic Media. The specific activity was 0.5-1orders of magnitude higher than that of Pt electrode of ordinary type. Mixing of Pt and Ru clusters give improved activity on C but not on SPE. In Pt-Sn clusters, Pt₃Sn₃/C give noticeable activity only after strong anodic treatments. The amount of Pt required for methanol fuel cell may be decreased by this technique of preparing Pt in a highly dispersed state. It was found that catalytic activity towards the methanol electrooxidation of the Pt cluster-supported electrodes greatly depended upon the Pt size of clusters.

View full abstract