The dividing-wall column, a compact structure of the fully thermally-coupled distillation or the Petlyuk column, has been known for around half a decade. Typically, it can claim up to 30% savings in terms of both capital and energy costs when compared to the other conventional column arrangements. The dividing wall column is thermodynamically equivalent to the Petlyuk column on the condition that no heat transfer is allowed across the dividing wall. However, better energy efficiency of the column may be obtained if heat transfer occurs within a certain part of the wall. The effects of heat transfer across the dividing wall can be analyzed by using the Column Grand Composite Curve (CGCC) [Lestak, et al., 1994]. The heat transfer potential across the wall can be observed by looking at the CGCC of both column sections alongside the dividing wall. However, the possibility of whether heat should be added or rejected at any stage is not clearly known ahead of the CGCC. Consequently, in this work, the exergy analysis is applied to the dividing wall column in order to determine whether heat should be added or rejected at any particular stage. The heat load targets at any stage, plotted as a T-H profile similar to the CGCC, can then be determined using the Method of Pinto (1998). This method was reported to successfully apply to the column with multiple feeds and products. After having identified the locations and quantities of the feasible heat transfer across the dividing wall, the benefits are discussed via a case study.

View full abstract

The isomerization of glucose to fructose is an important industrial process in obtaining high fructose syrup, a sweetener widely used in food industry. Traditional process in producing 55% high fructose syrup (HFS55) needs a large amount of desorbent. In this work, a hybrid simulated moving bed reactor (SMBR) system is optimized using experimentally verified dynamic SMB model to maximize the net productivity of HFS55 using minimum solvent. An adaptation of the state-of-the-art AI-based robust optimization technique, non-dominated sorting genetic algorithm with jumping genes is used in finding the Pareto (non-dominated) solutions for both the existing as well as SMBR system at the design stage. Finally, SMBR configuration was modified to further improve the system performance. Systematic multiobjective optimization resulted in significant performance improvement. Moreover, the new optimization technique gives much faster, smoother and larger spread of the Pareto-optimal solutions.

View full abstract

A thermo- and steam- stable silica membrane was prepared by a counter diffusion chemical vapor deposition method using tetramethyl orthosilicate (TMOS) and O₂ as reactants at 873 K. H₂ permeance at 873 K permeation test was 1.5 x 10^-7 mol m^-2 s^-1 Pa^-1, and H₂/N₂ permeance ratio was ca. 1000. H₂/N₂ permeance ratio was kept for 21 h under H₂O/N₂ ratio = 3 at 773 K that is the typical methane steam reforming conditions for a membrane reactor. 3 pieces of alumina tube substrates were welded by amorphous glass for a module and treated by the CVD at the same time. H₂/N₂ permeance ratio was over 3000 for the 3 membrane modules with the membrane gaps of 1.8 mm. This shows that uniform treatment was conducted on 3 substrates in the module by the counter diffusion CVD.

View full abstract

The IS process is one of water splitting methods to produce hydrogen using I₂ and SO₂ as catalysts. Three kinds of membrane techniques (Hydrogen permselective membrane reactor for HI decomposition reaction (HPMR), Electro-electrodialysis for concentration of HI (EED), Electrochemical cell for the reaction of SO₂, I₂ and H₂O (EC))were investigated by viewpoints of the total thermal efficiency. The HI concentration procedure by EED was the best method to improve thermal efficiency. The maximum thermal efficiency was 40.8 % at 12.5 mol kg_-H2O^-1 of HI concentration after EED. Thermal efficiency for the EC was slightly lower than that for EED by using the electrochemical data we had obtained. However, this method has a potential for the simple process operation compared with EED. In order to obtain higher thermal efficiency by featuring the EC, about 30 % of the required electric energy demand should be reduced. Total thermal efficiency can be improved only by 1 % for HPMR, and this technique should be applied to help material stabilities after HI decomposition reaction.

View full abstract

Amorphous alloys are expected as new cheep membrane materials for hydrogen separation, which is substitutable to expensive palladium alloys, maybe resulting in enlargement of application field. In this study, Zr-Ni based amorphous alloys, typical hydrogen permeable amorphous alloys, were focused on, whose diffusivity was investigated in detail especially with respect to hydrogen concentration dependency. Amorphous alloy membranes 20-40 µm thick were directly prepared from the melt by rapid quenching. All the membranes were coated with palladium using RF sputtering method to remove the difference in surface activity of each alloy. Hydrogen permeation rate was measured for pure hydrogen at elevated temperatures mainly in the range of 473-573 K. Hydrogen solubility was investigated using the Sieverts method. Hydrogen diffusivity at these temperatures was derived from the pressure dependence of permeation rate and solubility. Hydrogen permeation rate was proportional to the square root difference of hydrogen pressures on both sides as is often the case with other metal membranes. On the other hand, hydrogen solubility was found to be proportional to about the quarter power of equilibrium hydrogen pressure. This pressure dependency of hydrogen solubility could be explained using Kirchheim's theory, where potential energy distribution for hydrogen sites in amorphous alloys was taken into account. Using the theory, concentration dependence of diffusivity could be also derived, which was well consistent with experimental diffusivity determined by permeability and solubility.

View full abstract

Zr-Ni based amorphous alloy membranes are potentially alternative materials to Pd alloys which are having been utilized for hydrogen separation. We investigate on various characteristics of Zr-Ni based amorphous membranes, assuming the utilizations for hydrogen separation in the steam-reforming process of hydrocarbons. Gas permeation measurements are carried out with introducing the simulated steam-reforming gas mixture of methanol, as well as with a pure H2 gas, at 573K and 0.2-0.7MPa in pressure difference (introduced gas pressure: 0.3-0.8MPa, permeated gas pressure: 0.1MPa). Some of Zr-Ni based amorphous membranes, e.g. Zr₅₀Ti₁₅Ni₁₀Al₁₀Cu₁₅ membrane show excellent hydrogen permeability. As a result of the gas separation measurement from the gas mixture, comprised of approximately 23vol% CO₂, 2.3vol% CO, 8vol% steam and H₂ in balance, we confirm that Zr₃₆Ni₆₄ amorphous membrane keeps its good hydrogen permeability and has excellent hydrogen selectivity, and demonstrate that separation of H₂ gas by the membrane. We have investigated the relationships between alloy compositions and the permeability and the structural durability of those alloys', the latter of which is estimated by isothermal aging test conducted at 523K in the air. As a result, we find that both of the characteristics are strongly dependent on alloy compositions, especially on Zr-content. We find that several membranes including Zr₃₆Ni₆₄ show excellent structural durability at 523K over 17,000h. As for a Zr₃₆Ni₆₄ membrane, the structural durability at 573K in 0.5MPa H₂ atmosphere is also investigated. It shows good durability even after more than 5,500h hydrogenation aging.

View full abstract

Zeolite NaA membrane composed of a continuous intergrowth of zeolite NaA crystals had synthesized hydrothermally on the surface of a porous tubular support (12mm outer diameter, 800mm length and 1µm average pore size). The membrane was highly selective for permeating water preferentially with the high permeation flux in pervaporation (PV) and vapor permeation (VP). VP is proposed to be one of the most challenging techniques for dehydration of organics. The tubular-type module was designed in order to make the zeolite membrane viable for dehydration application. Thus, the largest zeolite membrane module has put into an existing beverage alcoholic manufacturing plant which produces 1,250kgh^-1 of ethanol at less than 0.2wt% of water from 84.8wt% biomass ethanol solution at 135°C. The plant is equipped with 6 modules, each of which consists of 550 pieces of zeolite NaA membrane tubes. The successful fabrication of the plant and the high level of VP properties of the module are discussed.

View full abstract

Separation properties of mordenite membrane for water-methanol-hydrogen mixtures were studied in the temperature range from 150 to 250°C under pressurized conditions. Mordenite is a kind of large pore zeolites with a pore diameter of 0.65 x 0.70 nm. Mordenite membrane was prepared on the outer surface of a porous alumina tubular support by a secondary-growth method. We found that water was selectively permeated through the membrane. The separation factors of water/hydrogen and water/methanol were 49 and 73, respectively, at 250°C. When only hydrogen was fed at 5 atm, its permeance was as low as 10^-9 mol m^-2 s^-1 Pa^-1 up to about 220°C, possibly suggesting that water pre-adsorbed in the micropores of mordenite hindered the permeation of hydrogen. The hydrogen permeance, however, dramatically increased to 6.5 x 10^-7 mol m^-2 s^-1 Pa^-1 at 230°C and reached to 1.4 x 10^-6 mol m^-2 s^-1 Pa^-1 at 250°C owing to the formation of cracks in the membrane. It was, however, found that the membrane was thermally stabilized in the presence of steam and/or methanol.

View full abstract

We have proposed selective reactions using catalyst particles covered with a permselective membrane. In this study, we have developed the coating method of silicalite-1 on spherical Pt/TiO₂ particles (silicalite/Pt/TiO₂). Silicalite-1 membranes were synthesized on spherical Pt/TiO₂ particles with a diameter of 0.5 mm from an aqueous solution of fumed silica, tetrapropylammonium hydroxide (TPAOH) and ethanol under hydrothermal conditions. A two-step hydrothermal synthesis (method B) was performed as well as a conventional one-step synthesis (method A). The silicalite/Pt/TiO₂ (B) particles were composed of a thinner silicalite layer with smaller crystals than the silicalite/Pt/TiO₂ (A). Hydrogenation of a mixture of linear 1-hexene (1-Hex) and dibranched 3,3-dimethylbut-1-ene (3,3-DMB) was performed using a fixed-bed reactor with a continuous flow system. The composite silicalite-1/Pt/TiO₂ (B) catalyst showed high 1-Hex/3,3-DMB hydrogenation selectivities of 20-35 at 323 K and 22-70 at 373 K due to the selective permeation of the reactant 1-Hex into Pt/TiO₂ particles through the silicalite-1 layer. The conversions on the silicalite/Pt/TiO₂ (B) were higher than those on the silicalite/Pt/TiO₂ (A), suggesting that the effective membrane thickness of the silicalite layer prepared by the tow-step synthesis was thinner than those by the one-step synthesis. The high selectivity for the silicalite/Pt/TiO₂ (B) indicates that the membrane quality was improved by the two-step synthesis. The feasibility of the application of catalyst particles coated with a permselective membrane to achieve reactant selectivities on a particle level has been demonstrated.

View full abstract

A titanium-substituted silicalite (TS-1) membrane with high pervaporation separation performance has been synthesized onto 10-cm-long mullite tube by in-situ crystallization using tetraethyl orthosilicate, titanium tetrabutoxide and tetrapropylammonium hydroxide, respectively, from the sol with a composition of SiO₂: 0.02 TiO₂: 0.17 TPAOH: 120 H₂O for the hydrothermal treatment time of 20 h at 200°C. The membrane and the accompanied powders were characterized by XRD, FT-IR and UV spectroscopy, SEM and ICP analysis. The separation performance was evaluated by pervaporation test. For a feed concentration of 5 wt% ethanol/water mixtures at 60°C, the separation factor for ethanol over water and total flux were 127 and 0.77 kg/m².h, respectively.

View full abstract

Computational fluid dynamics (CFD) calculation was successfully demonstrated for H₂/CO permeation through the membranes. CFD calculation was performed using the tube geometry model with the same conditions as the previously reported experiment. The concentration polarization observed in CFD simulation compared well with that reported experimentally. This suggests the CFD simulator can be used to design a membrane module involving prediction of selectivity and cut with consideration of concentration of polarization.

View full abstract

Boundary-Driven type Non-Equilibrium Molecular Dynamics (BD-NEMD) simulations have been carried out to investigate effusion processes of pure gas from the inside of silicalite membrane pores to the gas phase. Two types of model outer surfaces, SC and IS models, were devised to see whether any structural change in the outer surface affects the effusion flux of permeating gas. An SC model is made to have the outer surface terminated at the cross section passing through the center of straight channels of silicalite, while the outer surface of an IS model is terminated to pass through the center of intersections. Propane was chosen as a permeating gas. In all simulations, the silicalite crystal atoms and the propane molecules were represented by flexible (movable) models so as to exchange their kinetic energy without any artificial velocity scaling near the membrane exit. The present simulations have provided an interesting result in effusion flux; that is, the effusion flux of propane across the IS surface model is larger than that across the SC surface model. This enhanced effusion of propane is ascribed to the local density enhancement at the surface of the IS model; that is, according to energetic analyses, the larger the local density at the outer surface, the larger the opportunity for molecules to take extra kinetic energy necessary for desorption. This simulation result will be informative for molecular design of the structure of membrane exit.

View full abstract

Uphill transport of boric acids in aqueous solutions through PVA membrane was investigated. The PVA membranes were prepared by casting aqueous solution of PVA, subsequent drying at 294 K for 1 day, and treated at 383 K for 15 min to form cross-link in membrane. The permeation experiments were carried out under pH control, and the concentrations of boron were analyzed by the ICP-AES technique. When both sides were kept at an equal pH, a normal permeation due to the concentration gradient of boron was observed, but the flux of boron was higher in acid solutions than in alkaline solutions. For the condition where one side was kept at pH = 5.0 (acid side) and other side at pH = 10.0 (alkaline side), permeation of boron from the acid side to the alkaline side was observed under equal initial concentration of both sides at 100 ppm. Such an uphill transport was observed even under the condition of which the initial concentration of the alkaline side (200 ppm) was higher than that of the acid side (100 ppm). The uphill transport could be explained in terms of the pH sensitivity of a complex formation equilibria of boron with PVA; B(OH)₃, the dominant form under lower pH could form the complex with PVA more easily than [B(OH)₄]-, the dominant form under higher pH. As a result, transport of the complex due to their concentration gradient would take place against the concentration gradient of boron between the acid side and the alkaline side. We also investigated uphill transport of boric acids through modified PVA membrane, but we couldn't observe uphill transport with these membranes.

View full abstract

We have prepared gating membranes that are sensitive to ethanol concentration using a plasma graft filling polymerization technique. poly- n-isopropylacrylamide-co- n-isopropylmethacrylamide (poly-NIPAM-co-NIPMAM) were grafted onto porous polyethylene (PE) substrates, and their gating response to the ethanol concentration in the feed was investigated. A copolymer of NIPAM and NIPMAM-grafted porous polyethylene membranes prepared from a monomer solution that contained 60 mol% of NIPMAM exhibited large changes in flux (from almost zero up to about 100 kg/m²/h) at around 8 wt% of ethanol in the feed. The membranes that we have developed would be applicable to ethanol production by a biomass fermentation process accompanied by a purification stage using membranes, where it is important to control the ethanol concentration in the fermentation tank for efficient production.

View full abstract

In recent years, bio-reactors and bio-artificial organs utilizing cellular functions have been developed, although there are still unsolved problems such as that they cannot be used for a long period. One of the substantial differences between synthetic materials and biomaterials is their ability to self-repair. When some cells die on the artificial material, inflammatory substances are released from the dead cells and damage adjacent living cells, leading to the expansion of the inflammation. In this study, we developed a rapid regenerable cell culture system. Namely, the material can selectively recognize local cell death and subsequently remove the dead cells actively from the material surface, so that the resulting vacant spaces can be regenerated voluntarily and quickly by the growth of surrounding cells. Cells are detached from N-isopropylacrylamide (NIPAM) grafted surface by temperature decreasing below 32°C, because NIPAM changes its properties. In this study, a crown is used as a sensor that recognizes signals from dead cells, and NIPAM is used as an actuator. A designed polymer is a copolymer of NIPAM and Benzo-18-Crown- 6-acrylamide (BCAm) with a pendant crown ether receptor. When the crown ether captures specific ion, the LCST shifts to higher temperature, because guest-host complex makes the polymer chain more hydrophilic. Therefore, when the cells die, the polymer surface recognizes potassium ion released from the dead cells, NIPAM hydrates, and selectively detach the dead cells. Moreover, because the diffusion of inflammatory substances from the dead cells are prevented, rapid tissue repair occurs. This study is a pioneering approach that recognizes cell signals and which works in living systems in the novel field of tissue engineering.

View full abstract

The performance of the microwave-irradiated membrane reactor for methanol synthesis from CO₂/H₂ mixture gas was theoretically predicted using reaction-diffusing equations. Our previous study reported that microwave (MW) irradiation for membrane module can enhance flux and selectivity in pervaporation using silicalite membrane. The flux of MWadsorbed species increased by the MW irradiation for membrane module, while the flux of MW-inactive species does not change by the MW irradiation. We modeld MW irradiation for a membrane reactor of methanol synthesis from CO₂/H₂ mixture gas. The products (methanol and water) are MW-actives and the reactants (CO, CO₂ and H₂) are MW-inactive in this process. We calculated the methanol conversion at various temperatures and pressures with assumption that the MW irradiation increases the methanol and water permeance through the silicalite membrane. Calculated methanol conversion increased significantly by the MW irradiation and which is much higher than that for conventional membrane reactor.

View full abstract

Study was initiated to develop polymer inclusion membranes (PIMs) which will be applied for the treatment of radioactive wastewater. The PIMs were prepared using cellulose triacetate (CTA) as a polymer matrix, 2-nitrophenyl n-octyl ether (NPOE) as a solvent and octyl(phenyl)-N,N-diisobutyl carbamoylmethylphosphine oxide (CMPO) or N,N,N',N'-tetraoctyl-3-oxapentanediamide (TODGA) as a carrier. An aqueous solution containing cerium(III)nitrate, nitric acid and sodium nitrate was used as a model for the radioactive wastewater, and the transport of cerium ions through the PIMs was investigated experimentally. The experimental results showed that a carrier-mediated transport of cerium ions was obtained using the PIMs. A relatively short time was needed to remove cerium ions from the feed phase to the strip phase using the present PIMs. The experimental results revealed that the rate-limiting step of the transport process is the diffusion of the ion-carrier complex in the membrane.

View full abstract

A novel facilitated transport membrane module for gas separation is proposed in which a carrier solution is forced to permeate hollow fiber membranes. Both a feed gas and a carrier solution are supplied to the lumen side (high pressure side, feed side) of a hollow fiber ultrafiltration membrane module and flow upward. The carrier solution, which contains dissolved gas, permeates the membrane to the shell side (low pressure side, permeate side), where the solution liberates dissolved gas to become a lean solution, and it is recycled to the lumen side. It was confirmed by the experiment on the CO₂/N₂ separation using water as absorbent that the efficiency of this membrane system is very high. This membrane system was applied to the separation and concentration of CO₂ in a model flue gas consisting of CO₂ and N₂ using various amines as carriers. The feed side pressure was about 1 atm and the permeate side pressure was about 0.1 atm. CO₂ in the feed gas was successfully enriched from 1.5-15 % to 98.5-99.8% by one-stage membrane separation. The CO₂ permeance and the CO₂/N₂ selectivity were as high as 4.1x10^-4 mol m^-2 s^-1 kPa^-1 and 4800, respectively, when the CO₂ mole fraction in the feed was 0.1. The energy required for CO₂ separation was much lower those by a polymeric membrane process and a conventional gas absorption process consisting of absorption and stripping column. The present membrane system was also applied to the separation and concentration of C₂H₄ in a C₂H₄/C₂H₆ mixture using AgNO₃ as the carrier of C₂H₄. The feed gas was supplied to the module at high pressures (<500kPa) and the permeate side was atmospheric. Ethylene could be enriched from 80% to 99.8%. The membrane modules were stable over a discontinuous 2-4 month testing period.

View full abstract

Osmotic distillation is a newly developed membrane separation process that can concentrate aqueous solution without elevating the system temperature or pressure; hence, it is suitable for separation of biological products. To perform osmotic distillation, porous hydrophobic membranes are needed. In the present work, poly(4-methyl-1-pentene) (TPX) was used to prepare porous hydrophobic membranes, and the application of the prepared membranes to osmotic distillation was also investigated. With TPX/cyclohexane as the casting solution, by using suitable coagulant, such as propanol, particulate TPX membranes with inter-connected pores can be successfully fabricated. By adjusting the coagulation environment, the pore size of the porous TPX membrane can be tailored. Experiments of osmotic distillation, to concentrate BSA, raffinose, and aspartame, were performed to evaluate the performance of the prepared membranes. The results indicate that the performance of the TPX membranes prepared in the present work is comparable to the commercial PTFE membranes. In addition, ultrafiltration and nanofiltration were also carried out to compare with osmotic distillation to evaluate the advantages of osmotic distillation.

View full abstract

Membrane permeation and separation characteristics of mixed gas/membrane systems are typically calculated from single-component transport parameters, namely, diffusion coefficients and solubility constants. As a operating condition, temperature has great impact on permeability and selectivity of membrane for various gases and therefore the membrane efficiency.In this work, transport of oxygen and nitrogen through Poly(ethylene-co-vinyl acetate) (EVA) membranes prepared from polymeric solutions of EVA in tetrahydrofuran (THF), was studied by permeation measurements (variable pressure method) in different temperatures. Permeabilities, diffusivities and solubilities were determined for EVA films containing 28% (w/w) VA. Then these transport properties together with ideal selectivity were compared in four temperatures. By analyzing the acquired experimental data, we found out that permeability and diffusivity both increase with increasing temperature and also gas solubility increases slightly with temperature. Meanwhile we observed that temperature dependence of these transport properties obeys an Arrhenius-type relation. In the end, we calculated the activation energies of gas permeation (E_p) and diffusion (E_d) processes and the heat of gas solution (ΔH_s).

View full abstract

Although many experiments on membrane reactor use sweep gas for establishing partial pressure, only few works have been dealt with the measurement of permeability constant by using sweep gas. Conventionally, gas permeability constant is determined by the following steps: i) flowing of a single or mixed gas through permeation cell, ii) measuring variables and parameters such as volumetric rate of permeated gas and pressure difference, and iii) solving flux equation using measured variables and parameters to obtain permeability constant. In this study, a method for determining gas permeability constant through porous membrane was developed by fitting experimental and calculated data of binary system. A double cylindrical type of permeation cell with porous membrane as inner tube was used in the experiments, in which two different gases were fed into inlets of the cell (shell and tube sides). Then compositions and volumetric rates of gas exhausting from outlets were measured. Separately, compositions and volumetric rates were also calculated by integrating a set of differential equations derived from species balances in the cell. By using minimization technique, permeability constants were determined when sum of squares of the differences between measured and calculated values reach minimum. This method was tested for determining permeability constants of N₂, CO₂, and H₂ through Vycor glass. At 298 K it was obtained that permeability constants for N₂, CO₂, and H₂ are 2.24 x 10^-11, 2.23 x 10^-11 and 6.82 x 10^-11 mol.s^-1.m^-1.Pa^-1 respectively. Furthermore, experiments and calculations were also conducted for ternary system in which N₂ was fed into shell side while mixture of H₂ /CO₂ was fed into tube side. Calculations of ternary system used the constants determined from binary system. Work on ternary system showed that this method has better results than the conventional one.

View full abstract

A model, on the basis of incorporation of probabilities of particle adhesion in combined with pore-blocking and cake-filtration models, was developed in the present work to describe protein fouling during cross-flow microfiltration. It was verified first by scanning electron micrographs (SEM) that the deposition of protein aggregates is responsible for the resistance growth during cross-flow microfiltration of the solution of bovine serum albumin (BSA). Previous researches indicate that BSA aggregates can block membrane pores via the mechanisms of pore blockage and cake filtration during dead-end microfiltration, and a combined model of these two blocking mechanisms was developed accordingly. Such a combined model was modified in the present work in two folds. First, a combined model of intermediate blocking and cake filtration was developed which can describe the growth of filtration resistance more satisfactorily than the original model when microfiltration membranes with interconnected pores are used. Second, probabilities of particle adhesion were incorporated in the combined models to quantitatively describe the resistance growth during cross-flow microfiltration. The probability of particle adhesion on membranes, which can be calculated from estimations of the forces on particles along the flow and permeation directions, was introduced to take into account the effect of particle sweep due to cross flow. The developed models have been proven to be in good agreement with the experimental data obtained. It was found that the adhesion probability and the pore blockage parameter used in the model govern the behavior of resistance growth. With such models, the effects of cross-flow rate, protein concentration, solution pH, and membrane hydrophilicity on particle adhesion and pore blocking during cross-flow microfiltration can be quantitatively investigated.

View full abstract

Flux decline behaviors on ultrafiltration of Corynebacterium glutamicum (C. glutamicum) slurry were investigated with and without protein, BSA, using washed cell slurry to eliminate the effect of metabolites and medium ingredients. A dead-end cell with UF membrane was employed, and the filterability was evaluated by average specific resistance α_av, which is inversely proportional to filtration rate. First, the effect of pH on ultrafiltration behaviors of C. glutamicum slurry was investigated without BSA. The resistance α_av of C. glutamicum slurry rapidly decreased with decrease of pH below pH 4.5, and minimized at pH 3.0. According to the measurement of particle size distribution, cells tended to aggregate in such pH range. Thus aggregation of cells caused decrease of α_av. Secondly, the effect of the mass fraction s_b of BSA on α_av was investigated under a constant total mass fraction s of BSA and cells at pH 4.5, where cells and BSA were oppositely charged. Addition of BSA was effective for reducing α_av via aggregation of cells below 0.08 of s_b/s. However, α_av increased due to the increase of free BSA, of which α_av was much higher than that of cells, over 0.08. The interaction manner of BSA and cells was found mono-layer adsorption of BSA to cells, following Langmuir isotherm, according to the measurement of adsorption amount of BSA to cells.

View full abstract

Filtration and sedimentation properties have been investigated using various activated sludges prepared under the artificially controlled culture condition, to clarify the effects of organic loading on solid-liquid separation properties of activated sludge in wastewater treatment process. In microfiltration of mixed liquor containing activated sludge, the dynamic behaviors were well described by the Ruth filtration rate equation. It was revealed that the filter cake composed of activated sludge has extremely high compressibility. Therefore, membrane filtration of activated sludge should be conducted at relatively low pressure, because increase in the filtration pressure does not necessarily lead to increase in the filtration rate. Also, the filtration rate and the initial sedimentation velocity of mixed liquor are highly dependent on the characteristics of activated sludge. On the basis of the experimental data of the average specific filtration resistance, sludge volume index (SVI) and the area mean diameter of activated sludge, a correlation between solid-liquid separation properties and floc size of activated sludge was clarified. In addition, floc size of activated sludge was greatly influenced by biochemical oxygen demand (BOD) loading of the wastewater. From these results, it is clear that solid-liquid separation properties of activated sludge are closely related to the BOD loading of wastewater.

View full abstract

This presentation covers two areas using facilitated transport membranes: (1) the removal and recovery of heavy metals from waste waters by supported liquid membranes (SLMs) with strip dispersion and (2) fuel processing with carbon dioxide-selective membranes for fuel cells. New membrane technology based on SLMs with strip dispersion for the removal and recovery of metals, including chromium, copper, zinc, and strontium, from waste waters has been developed. The technology not only removes the targeted metal in the treated effluent allowable for discharge or recycle, but also recovers the metal at high concentration and purity suitable for resale or reuse. In other words, the goals of zero discharge and no sludge have been achievable. The SLMs contain selected complexing agents/carriers for the facilitated transport of the target species. The stability of the SLM has been ensured by engineering the modified SLM with strip dispersion. Recently, new membranes for the removal of carbon dioxide from hydrogen-containing reformed gases have been synthesized by incorporating amino groups into polymer networks. The membranes are selective to carbon dioxide preferentially versus hydrogen since carbon dioxide permeates through the amine-containing membranes via the facilitated transport mechanism due to its reaction with the amine. This type of membranes has the potential for fuel processing for environmentally friendly fuel cells, including the use of the membranes both in the membrane reactor configuration to enhance water gas shift reaction and in the purification of hydrogen for the reformed gas generated from the partial oxidation of liquid fuel (e.g., gasoline or diesel). Our modeling results of the membrane reactor show that significant hydrogen enhancement via carbon dioxide removal and CO reduction to 10 ppm or lower are achievable.

View full abstract

A new centrifugal ultrafiltration theory under centripetal-flow was developed to predict the filtration flux and the local filtration characteristics in the protein solute filter cake. The centripetal-flow cell is constructed whereby the centrifugal force vector is perpendicular to the membrane surface but opposite to the filtration flux vector for transmembrane flow. Therefore, this process makes it possible to maintain a high filtration flux by minimizing the cake formation on the membrane surface during ultrafiltration. By using a new centrifugal ultrafiltration equation for the filtration flux data, the values of the solute mass in the centrifugal filter cake were determined experimentally and the centripetal-flow ultrafiltration effects were quantitatively evaluated. The experimental results of bovine serum albumin solution in the pH4 to 9 and soy protein solution in the pH6.7 coincided favourably with the calculated results based on the new centrifugal ultrafiltration theory under centripetal-flow and the data of dead-end ultrafiltration.

View full abstract

Synthetic wastewater was treated in a bench scale Submerged Membrane Activated Sludge Process (SMASP) unit at different operating conditions (SRT and BOD loading) to study the behavior of change in suction pressure and filtration resistance. It was observed that the influence of SRT and BOD loading on stabilized MLSS was significantly high. At long SRT (500 d) and low BOD loading (≤0.5 kg/m³·d), the organic metabolism was good and the supernatant TOC was low in the bioreactor throughout the operation. Accumulation of unconverted organic matters and metabolites increased the suction pressure rapidly and the operation was unstabilized. At long SRT (500d) and low BOD loading (≤0.5 kg/m³·d), low and steady suction pressure and filtration resistance were observed for long time. At short SRT (≤90d), both clogging and cake resistance were almost equally responsible to develop high filtration resistance. On the other hand, high MLSS concentration (>20000 mg/L) produced high cake resistance at long SRT (500 d) and high BOD loading (0.6 kg/m³·d). Cake resistance was increased with an increase of mass of adhered sludge on the membrane surface. At long SRT (500d) and low BOD loading (≤0.5 kg/m³·d), the adhered sludge quantity was less and the cake resistance was minimized.

View full abstract

In order to clarify the relationship between the conditions of crystallization and the characteristics of solid-liquid separation in a batch crystallization process, the anti-solvent crystallization was conducted for various conditions, and subsequently dead-end filtration of the produced crystal suspensions was carried out under constant pressure. Glycine, a type of amino acid, was used as a target material, and methanol and ethanol were used as the anti-solvents. In addition to determining the size distributions of produced crystals by a laser granulometer, the photomicrographs of them were observed directly using a digital microscope. The experimental data clearly demonstrated that the filtration properties of glycine crystal suspensions were strongly affected by the crystallization conditions. Ethanol produced the large crystals compared to methanol, and consequently led to the high filterability. The glycine crystals became smaller and the specific filtration resistance of the filter cake consisting of glycine crystals became larger with an increase of the anti-solvent concentration in either case of methanol or ethanol. Moreover, the glycine crystal suspensions with higher filterability were produced by adding the anti-solvent to the glycine solution than by adding the glycine solution to the anti-solvent. And also, the slow addition of the anti-solvent achieved a higher filterability than the instantaneous addition. It was also found that the filter cake of the glycine crystal was moderately compressible and the filtration rate increased with an increase of the applied pressure.

View full abstract

The specific filtration resistance of a slurry was predicted from several fundamental properties of the particles in the slurry using neural network (NN) modeling. Polyethylene beads, polymethyl methacrylate beads etc. of several to several hundred micrometers in volume-based median diameter were dispersed in water or methanol to obtain the compression-permeability data. The porosity of the sediment obtained from the slurry in gravitational and centrifugal fields was newly added as one of the input parameters of the NN. The other input parameters were the particle size distribution and the porosity of the compressed cake. The product of the specific filtration resistance and the particle true density of the cake was chosen as the output parameter of the network. The NN trained with these data was found to be fairly effective in predicting the filtration property of another slurry containing particles used to obtain the training data. The porosity of the compressed cake could also be predicted successfully by another NN trained with the sedimentation results, the particle size distribution and the loaded pressure in compression-permeability tests. The predicted cake porosity could be utilized as the input value to predict the filtration property. With the method, after the database of filtration resistance and the properties of various slurries is developed, the filtration resistance of a slurry can be predicted with much less amount of the test slurry compared with conventional filtration experiments.

View full abstract

Superabsorbent hydrogels display a drastic change in volume by absorbing several hundred times their own weight of water, and they have attracted a considerable amount of attention in recent years in widely diversified fields. It is well-known that the dynamic behaviors of hydrogel particles swelling in the solution are markedly influenced by the ionic environment of the solution. In this study, both expression and expansion properties of the granular bed of the superabsorbent hydrogels under the action of the mechanical loads were investigated at various ionic environments using the KCl solution, in order to clarify the role of the osmotic pressure of the solution on the deformation behaviors of hydrogel particles. The expression behaviors were described accurately via the simplified computation method based upon the modified Terzaghi model. As the concentration of the KCl solution was increased, the expression rate was increased, and the packed hydrogels became dense. Interestingly, these results were similar to the tendency obtained under the absence of the mechanical loads. It was revealed that the data were analyzed well on the basis of the idea of the effective osmotic pressure of the solution. Moreover, in the expansion process of the compressed bed of hydrogel particles, it was clarified that the secondary expansion effect known as the creep effect plays an important role on the expansion behavior. The dynamic behaviors of the gel expansion were evaluated well by the simplified computation method in which the creep effect was taken into account.

View full abstract

The Rotary Press Filter is a sludge dewatering device equipped with metallic filter elements. It is attracting much attention in industry and is used for sludge dewatering in various applications such as sewage treatment plants, food processing factories, chemical plants, and paper mills. The metallic filter element is the key component of the Rotary Press Filter. The author has conducted filtration performance tests to compare the performance of two filter elements of different filter pore sizes (0.38 mm and 0.18 mm) using digested sludge at a sewage treatment plant. The results showed that the cake moisture content and SS recovery rate of the 0.38-mm filter were lower than those of the 0.18-mm filter, which is thought to be due to the lower filtration resistance of the 0.38-mm filter.

View full abstract

A constant-current electro-osmotic dewatering is analyzed by use of the Terzaghi-Voigt combined model for considering a creep deformation of the material. The basic differential equation based on the model is solved analytically by assuming that both an electro-osmotic pressure gradient E_pg and a modified consolidation coefficient C_e of the material are constant. The solution can explain the time course of changes in a solid compressive pressure distribution. The theory can also explain the final moisture distribution of the material. The progress of electro-osmotic dewatering can be represented by an average consolidation ratio U_c as in mechanical expression. Both C_e and E_pg increase with an electric current density i. A larger C_e leads to a higher dewatering rate; a larger E_pg results in a higher solid compressive pressure distribution in the material. It is found that the amount of creep deformation depends upon the dewatering rate; the faster the dewatering rate, the larger the creep deformation, i.e., the creep deformation is more remarkable when a higher current density is applied to the material.

View full abstract

Froth flotation is one of surfactant based separation processes which is suitable for dilute wastewater treatment. There are several advantages such as low space requirement, high removal efficiency, flexibility for various pollutants at different scales, and low cost. To achieve high performance for froth flotation, the combination of ultra-low interfacial tensions between oil and water and stable foam production must be achieved. To get the ultra-low interfacial tensions, Winsor type's III microemulsion or middle phase has to be formed. In this study, branched alcohol propoxylate sulfate sodium salt with 14 - 15 carbon and 4 PO groups (Alfoterra 145 - 4PO) was used to form microemulsion formation with diesel. The effects of surfactant concentration and NaCl concentration on phase study, foam characteristics, and performance of froth flotation operation were investigated in this work. An increase in surfactant concentration decreases interfacial tension (IFT), but increases foam stability. For the effect of NaCl concentration, the minimum IFT was achieved at 5 wt% NaCl. However, this optimum salinity cannot be operated in froth flotation experiment due to poor foam characteristics. Therefore, both IFT and foam characteristics should be optimized to achieve high efficiency of froth flotation.

View full abstract

This paper deals with an improvement of settling tank performance. The new equipment is proposed and the usefulness of this method is confirmed. In combined sewerage systems, storm waters often exceed the capacity of the treatment plant, which may cause environmental problems. To meet this problem, the enhancement of the settling tank capacity is an important and urgent issue to the sewage treatment plant in the city. Although the lamella settler is widely used to improve the settling performance of the space limited settling tank, the capacity of this method is limited by the surface area of the tank, since the parallel plates are arranged horizontally. The newly developed settler is an improvement of the lamella settler, which arranges the inclined parallel plates in the vertical direction. In this method, the separated clear water is removed directly by suction from the top end of the parallel plates. For the removal of clear water, the right and left edges of the plates are closed, to make a tube with rectangular cross section. These are the unique and original points of our equipment. Since each settling tube acts as a small settling tank, the treatment power is proportional to the number of the settling tube. Furthermore, this system is less restricted by the surface area of the tank, since the tubes are arranged vertically. It was shown from our laboratory and onsite experiments, that the new system can enhance the capacity of the settling tank more than 5 times grater than the conventional one.

View full abstract

The capture of solute into freezing part is difficult due to vigorous agitation of the freezing interface during the freezing of solution. We have been studying the applicability of supersonic radiation to the agitating method, and found that the freeze concentration efficiency of solutes is improved greatly by this radiation. In this paper, the effect of the freezing rate on the freeze concentration characteristics is examined. We modified the experimental apparatus where the distance between a supersonic transducer and freezing interface can keep constant (40 mm). Freezing columns (55 mm inner diameter and 320 mm height) made of stainless steel were descended in a refrigerator (-16.5°C) at some prescribed rates (20, 40, 80, 120 mm/h) with/without supersonic radiation (54 W). This means freezing rate can be maintained almost prescribed constant values. From these experiments, it was found that the solute could hardly be concentrated in unfrozen part and the distribution factors of solutes were about 1.0 for all freezing rates without supersonic radiation. On the other hand, with supersonic radiation the distribution factors were much smaller than 1.0 at slower freezing rate (20, 40 mm/h) and were small up to 0.5 at faster freezing rate (80, 120 mm/h).

View full abstract

Hydrofluoroethers have been applied as diluent for phosphorus acid extractants in the solvent extraction of zinc(II). They are promising alternatives to conventional organic diluents because of non-flammability, chemical stability and non-toxicity. The hydrofluoroethers used in this study are a mixture of ethylperfluorobutylether and ethylperfluoroisobutylether (Novec^TM HFE-7200, SUMITOMO 3M Limited.) and a mixture of methylperfluorobutylether and methylperfluoroisobutylether (HFE-7100). Among the extractants examined, two phosphorus acid extractants, di-(2-ethylhexyl)phosphoric acid (D2EHPA) and 2-ehtylhexyl phosphoric acid mono-2-ethylhexyl ester (PC88A), were soluble in both hydrofluoroethers. Because of the hydrofluoroethers' high specific gravity and low solubility in water, phase-separation was rapid and the resulting phases were clear. In the extraction of zinc(II) using phosphorus acids, the system with D2EHPA gave higher extraction than that with PC88A, while there was little difference in the extraction behavior between the two hydrofluoroethers. The combination of these extractants and diluents showed a little higher extractability of zinc(II) than such conventional chlorinated organic solvents as chloroform and carbon tetrachloride. Effects of both concentrations of zinc(II) and the extractants on the zinc extraction were examined, and the stoichiometric relationship of the extraction was proposed with the equilibrium constants determined. Moreover, the back extraction of zinc(II) and the selective separation of heavy metals were satisfactorily attained.

View full abstract

Waste generating from photographic industry contains metals, which basically need to recover due to either economic reasons or to treat and remove the toxic metals before it can discharge into environment. Extraction of precious metals such as silver from this waste is very attractive. However, selection of the right technique is crucial in order to obtain relatively pure products. Solvent extraction may offer some advantages over other techniques. In solvent extraction process, selection of the right extractant, which offers high selectivity and capacity is important especially when involves extraction of metal from mixed diluted metals solution like photographic wastes. Therefore, different types of extractants were examined in this study in term of their capability and selectivity in extracting metals from photographic waste. It was found that the Cyanex 302, dithizone and tetramethylthiuram disulfide are most promising extractants amongst others in extracting silver. It was found that also more than 90% of silver was extracted using only 0.05M extractant. This result shows that extractants containing sulfur donor atoms have high selectivity towards silver compared to other metals. Stripping of extracted metals from organic-loaded metals solution using various stripping agents such as nitric, hydrochloric and sulphuric acids shows that nitric acid gives a good capability to strip silver from metal loaded tetramethylthiuram disulfide and Cyanex 302. Metal loaded dithizone could not be stripped using these acids. Further study shows that the extent of extraction process depends on, not only on extractant concentration but also on the initial pH of the liquid photographic waste.

View full abstract

A synergistic extraction system of trivalent gold ion (Au(III)) with an acidic extractant D2EHPA (di(2-ethylhexyl)phosphoric acid) and a hexadentate ligand TPEN(N,N,N',N'- tetrakis(2-pyridylmethyl)- 1,2-ethylenediamine) was investigated and its applicability to the simultaneous recovery process of noble metals, Pd(II), Pt(II) and Au(III), from waste electric devices was discussed. Since Au(III) is present stably as an anionic complex, AuCl₄-, in aqueous chloride solution, Au(III) and noble metals cannot be recovered simultaneously by an acidic extractants such as D2EHPA. However, by the addition of TPEN in amounts equimolar with Au(III) in the aqueous phase, the extractability of Au(III) with D2EHPA was improved drastically. AuCl₄- is complexed strongly with TPEN in the aqueous phase and a cationic complex, Au(TPEN)³⁺ is formed by the ligand exchange between chloride ion and nitrogen donor. The formation constant of Au(TPEN)³⁺ was evaluated as about 10¹⁰ by the potentiometric titration technique. From these results, both the stoichiometry for the synergistic extraction of Au(III) with TPEN and D2EHPA (HR) and its extraction constant were determined as
AuCl₄^- + TPEN + 3(HR)₂ ↔ Au(TPEN)R₃(HR)₃ + 3H⁺ + 4Cl^- K_ex=10^5.7TPEN acts as a powerful synergist for the extraction of Au(III) with D2EHPA. Thus, by introducing the proposed synergistic extraction system, Au(III) and other noble metals can be recovered simultaneously and effectively from aqueous chloride solution.

View full abstract

Application of supercritical carbon dioxide (SC-CO₂) to fractionate shorter-medium (C8-C14) and longer chain (C16-C18:2) triglycerides in terms of fatty acids constituent in palm kernel oil (PKO) was studied for formulating possible cocoa butter replacer blends. The extraction of dehulled ground PKO was carried out at 40 and 80°C and pressures ranging from 20.7 to 48.3 MPa with a flow type apparatus. The PKO was extracted successively into four fractions. At higher temperatures (80°C) the total yield of PKO increased with pressure. The highest yield (99.6%) was obtained at 48.3 MPa and 80°C with minimum CO₂ used. Much shorter chain C8, C10, C12 fatty acids constituents were extracted fraction 1. Longer chain 16, C18:0, C18:1 and C18:2 fatty acid constituents were extracted in fraction 4. The slip melting points were reduced to 6 - 8°C for fractionated PKO in the later fractions. The latter fractions had composition and slip melting points close to that of cocoa butter replacer blends.

View full abstract

Entrainment on a distillation tray affects efficiency, pressure drop and the capacity of the distillation columns. Entrainment on a sieve tray increases with the increase of vapor load at constant liquid load. However, at constant vapor load, entrainment behaves somewhat differently. It has a minimum value. This phenomenon was uncovered by systematic study on a wide range of trays and systems at FRI (Fractionation Research Incorporated). Many arguments have been given for this phenomenon. Of these, the dominant one seems to be that the minimum is caused by the transition of vapor liquid contact from spray to froth on the tray. We, however, argue from an entirely different point of view. We have made an assumption that the longer the residence time of the liquid on the tray, the greater the chances of generation of entrainment, The reason is that the tray specification, especially the height of the outlet weir, affects the phenomenon of minimum entrainment. We discerned, from the data released by FRI, this new relationship between minimum entrainment and the newly defined minimum residence time at constant vapor rate and increasing liquid rate for sieve tray. The liquid load at which the entrainment becomes minimum at constant vapor load operation coincides with the liquid load at which the residence time of liquid on the tray becomes minimum. In the newly defined residence time, the definition of volume is the product of the bubbling area and the observed clear liquid height. The observed clear liquid height has a significant effect on the residence time on the tray. The clear liquid height as reported by FRI was obtained from a series of experiments for pressure drop study. This afforded us an opportunity to predict the minimum of entrainment from only the value of clear liquid height. This relation can be applied to predict the minimum entrainment, if the clear liquid height is known.

View full abstract

An analysis of energy savings by an internally Heat-Integrated Distillation Column (HIDiC) is conducted for the separation of multicomponent mixtures. The mixtures to be separated in this study are benzene-toluene-p-xylene (BTX) ternary system and twelve-component hydrocarbons system. The design and the operating variables are defined for both a conventional distillation column and a HIDiC, and the effect of each variable on the energy consumption is investigated by sensitivity analysis. Including the compressor duty, the energy consumption for BTX separation with HIDiC is only about 63 % of the conventional distillation column. For twelve-component hydrocarbons system, the HIDiC can save the energy of about 65 % compared to the conventional distillation column. These results show that the HIDiC can significantly reduce the energy consumption for the distillation processes.

View full abstract

This paper presents results of continuous feed and total reflux distillation tests carried out with a common type and size structured packing using two- and three component mixtures of common alcohol's and water. With the binary mixture, packing performed slightly better than with a three component mixture, and, surprisingly, this performance appeared slightly better than in the case of the total reflux operation. Under total reflux a test was carried out also with a water rich mixture, however without any consequences for the mass transfer performance of the packing. Delft model predictions of the overall mass transfer efficiency appeared to be conservative enough around the operating/design point load of the packing in question. The composition profiles measured with the three components mixture were used to validate the rate-based (non-equilibrium) model developed at the Nagoya Institute of Technology, which appeared to be highly accurate, but also sensitive to the choice of the predictive method for the interfacial area.

View full abstract

The aim of this work is to separate the benzene and to purify the hexane fraction of Bandar Imam Petrochemical Complex-Iran (BIPC). Benzene content of existing fraction is around 6 wt% and it is required to remove the benzene content to ppm level. Benzene and n-hexane make azeotropic mixture and their separation by distillation method at normal atmospheric pressure is impossible. By using suitable solvent like N-methyl pyrrolidone (NMP) the relative volatility of components could be modified and enhanced to some extent. For the selection of thermodynamic model to predict the liquid activity coefficients, isobaric vapor-liquid equilibrium of n-hexane-NMP and benzene-NMP were determined at 101.3 kPa. VLE data were used to optimize the interaction parameters and thermodynamic coefficients of Wilson, NRTL and UNIQUAC activity models. Results showed excellent consistency of binary VLE data with Wilson model. VLE data for the mixture of hexane fraction and NMP as a multi-component system were determined at 101.3 kPa. Different methods of modeling were evaluated for multi-component system of hexane fraction and NMP. It was found that by using optimized binary interaction coefficients for n-hexane, NMP and benzene and calculation of all other binaries by UNIFAC model, the best result with minimum deviations from experimental data could be obtained.To evaluate the application of this thermodynamic modeling in an actual extractive distillation process, an existing bench scale column was used. Hexane fraction of BIPC was fed to the column packed with LDX structure packing (Sulzer Chemtech Co.) at its optimum operation conditions of hold up and boilup rate. NMP is fed as entrainer to the top of the column and benzene free fraction is obtained from the top of this column with less than 80 ppm-wt of benzene. The experimental results were simulated using HYSYS software and optimized thermodynamic model of Wilson.

View full abstract

A reactive distillation (RD) for the production of ethyl acetate (EtAc) from acetic acid (HAc) and ethanol (EtOH) using sulfuric acid as homogeneous catalyst is presented. To meet the industrial specification of HAc(≤0.01 wt%) and EtOH (≤0.2 wt%) in the product stream, a complete esterification process including a RD column with an overhead decanter and a stripping column are synthesized and studied by using the Aspen Plus® simulator. In a basic case, (Tang et al., 2003) perfect Murphree tray efficiency was assumed. To be more practical, the effect of the Murphree tray efficiency is considered in this study. Because of the effects of the non-ideality in Murphree tray efficiency, the heat duty of stripper and the organic reflux flow of RD column are adjusted to meet EtOH and HAc requirements in the product stream, respectively. It was found, output multiplicity occurred while this organic reflux flow rate is between 41 to 45 moles/sec and the Murphree tray efficiency is lower than 70%. As a result, the Murphree tray efficiency becomes a bifurcation parameter. As the tray efficiency is even lower, the existence of multiple steady-states is getting more obvious.

View full abstract

The use of hydrogen for automotive power is made attractive by its potential unlimited availability and its low emissions. The main obstacles to the use of hydrogen in road vehicles have been, and remain, the storage of hydrogen on board the vehicle, hydrogen handling, hydrogen distribution and cost. For various reasons, such as safety, overall energy considerations and fuel handling, alternatives to compressed or liquid hydrogen must be seriously considered in future applications. It is possible that liquid organic hydrides will become increasingly important for their role as hydrogen carriers in any future hydrogen economy as the transportation and storage of such organic hydrides would exploit existing fuel-distribution infrastructures. For instance, a hydrogen source could be employed to hydrogenate toluene (C₇H₈) to methylcyclohexane, MCH (C₇H₁₄), while, onboard any vehicle employing the system, hydrogen could be liberated through the endothermic reverse reaction.Recently there has been increased interest in developing proton-conducting solid oxide fuel cell (SOFC) technology. Such proton-conducting systems will potentially be able to operate at reduced temperatures compared to more conventional oxygen-ion conducting systems. Furthermore, such systems would be able to exploit any move towards hydrogen-based fuels.Here we report results from the integration of this MTH (methylcyclohexane, toluene, hydrogen) fuel storage system with a novel proton-conducting SOFC operating at 750°C. MCH was vapourised and heated prior to being fed to a dehydrogenation reactor followed by a separation stage where toluene and unreacted MCH were condensed. The vapour phase product from the condenser was fed to the fuel cell. The solid oxide fuel cell (SOFC) was based on a BaCe_0.9Y_0.1O₃ (BCY) proton-conducting electrolyte material with porous platinum electrodes. The performance of the fuel cell was evaluated under hydrogen generated entirely from MCH. At 750°C power outputs of more than 20 mW cm^-2 were obtained.

View full abstract

The simultaneous evaluation of the methanol crossover (MCO) and the carbon dioxide crossover (CCO) in a DMFC during operation was conducted with simple measurements of the CO₂ and methanol at the cathode and the anode, respectively, in order to derive an equation to estimate the MCO flux and to quantify the magnitude of the CCO flux at the DMFC. The effects of temperature (293-363K), current density (0-400mA/cm²) and the methanol concentration (1, 2 and 3M), as well as the gas/liquid flow rates on the crossover rates were quantitatively studied. With an increase in the current density, the MCO flux linearly decreased, whereas the CCO flux increased to a certain value (about 20% of the MCO flux) and was not affected by it at high current densities. The MCO flux strongly affected by the methanol concentration, temperature and current density, whereas the flow rate of the methanol aqueous solution to the anode and that of the oxygen gas to the cathode weakly affected the MCO in the range measured. From the results, an equation estimate the MCO flux under operation was proposed.

View full abstract

In the development of more efficient and stable Polymer Electrolyte Fuel Cell (PEFC), it is important to propose the optimal operating condition and the optimal shape that can uniform the distribution of current density and relative humidity in a single cell. In this study, the effect of changing operation conditions on the i-V characteristic with a small PEFC was examined experimentally. And the model was created that can express these influences on PEFC reaction characteristics. Furthermore, a PEFC reaction and thermal flow analysis model was constructed. Various operation conditions (membrane thickness, gas diffusion layer thickness, etc.) were evaluated from the viewpoint of average current density and uniformity of current density and relative humidity. As a result, by using the thin membrane, high power density was obtained, and the water management was simplified, but the current density distribution increased. The information that decided the optimal operation conditions was obtained.

View full abstract

To develop the new electrically heated catalyst supported by anodized alumina layers for preparing high heat-resistance and quickly start-up catalyst systems, we have used a commercial aluminum /NiCrAlloy/ aluminum clad as a raw material and anodized them in oxalic acid then, activating these anodized layers according to the former method for preparation of alumite catalysts. However, the problem is the detachment of alumina layers from Ni-Cr alloy when they were burned at high temperature (650°C and over) and suddenly cooled them down to low temperature, because of the difference in thermal expansion. Therefore, our objective is to develop a way for reducing the difference in thermal expansion rate of anodized alumina layers and Ni-Cr alloy to prevent the detachment of alumina layers from base metal. A preparation method was proposed according to a three-steps procedure involving: pre-heating, anodizing and subdivision of alumina layers. An aluminum clad was calcined at 500°C for forming a diffusion layer of Al, Ni and Cr between aluminum layers and Ni-Cr Alloy. Then, It was anodized into oxalic acid solution to obtain about 50µm of alumina layers. And, the plate was soaked into oxalic acid, then, soaked into deionized water, sequentially; it was calcined at 500°C for 3 h in air. The experimental results showed that the alumina layers have not detached from the base metal even if they were calcined at 1000°C. Moreover, the detachment of alumina layers had not occurred during the electrically heating test at about 800°C for over 5000 times of ON-OFF repeating test. We present herein data concerning the effects of major steps on the detachment of alumina layers from Ni-Cr alloy and results of thermal-shock tests as well as the thermal-electro characters, physical characters of the new electrically heated catalyst support.

View full abstract

Fuel cell (FC) offers the possibility of expanding the electricity utilization market. One of the key technologies that will make the widespread use of fuel cells possible is a hydrogen supply system. The use of chemical reactant as a hydrogen career medium has the possibility of a safe hydrogen supply system. The possibility of a FC vehicle system, which utilized chemical reactants and capable to realize carbon dioxide zero-emission, was discussed experimentally in this study. The system uses a portable thermally regenerative fuel reformer of carbon dioxide fixation type. The reactivity of calcium oxide to carbon dioxide was used for the carbon dioxide fixation and also for heat source of fuel reforming. The reaction realizes high-purity hydrogen production under mild operation conditions, and is regenerative thermally. To estimate the efficiency of the fuel reforming system using the reaction system, the reactivity of hydrogen production was examined experimentally. The proposed reforming system was expected to be applicable as a hydrogen career system in carbon dioxide zero-emission FC vehicles. The contribution of the zero-emission hydrogen career system on the global environment was evaluated based on the experimental results.

View full abstract

Iodine-sulfur (IS or S_I) process has been researched as one of the promising thermochemical hydrogen production methods. Process design based on heat/mass balance through the process considering the operation conditions and apparatus properties is important for an effective hydrogen production. In this study, preliminary analysis of the total IS process was carried out. Sensitivity of parameters in H₂SO₄ decomposition and in HI decomposition procedure was discussed by total thermal efficiency. Temperature difference at heat exchangers in the HI decomposition procedure changed thermal efficiency by 3.0 %. Thermal efficiency was improved by 1.4-2.5 % for one stage in the multi-stage H₂SO₄ evaporator. Heat exchange in the H₂SO₄ decomposition procedure should be optimized to obtain optimum thermal efficiency. HI concentration after the electro-electrodialysis cell for HI concentration changed thermal efficiency by 13.9 %. Effect of HI one-pass conversion rate at HI decomposition membrane reactor was about 1 %. The most important parameter in the IS process was HI concentration after the EED.

View full abstract