Oxygen permeation study of perovskite hollow fiber membranes

The first oxygen permeation data of a dense hollow fiber perovskite membrane based on BaCo_xFe_yZr_zO_3 − δ are reported. The hollow fiber was prepared by a phase inversion process. Dense fibers were obtained with the following typical geometries: outer diameter, 800–900 μm; inner diameter, 500–600 μm; length, 30 cm. The O₂-permeation through the hollow fiber perovskite membrane was studied in a high-temperature gas permeation cell under different operation conditions. The increase of the helium gas flow rate reduces the oxygen partial pressure (p_O2) on the core side and a higher oxygen permeation flux is observed. High oxygen flux of 0.73 m³ (O₂)/(m² (membrane) h) was achieved at 850 °C under the operation parameters F_air (shell side) = 150 ml/min and F_He (core side) = 30 ml/min. The oxygen partial pressure dependence of the O₂ permeation flux indicated an interplay of both surface reaction and bulk diffusion as rate limiting steps. During 5 days of permeation a high and stable oxygen flux was observed. X-ray diffraction patterns of fresh and spent membranes after the permeation measurements revealed that no degradation after oxygen permeation appears.     

Novel zirconia-supported catalysts for low-temperature oxidative steam reforming of ethanol

Three zirconia-supported platinum group metal (Pt, Ru and Pt–Ru) catalysts were prepared by impregnation. The activity of these catalysts toward the oxidative steam reforming of ethanol (OSRE) was examined in a fixed-bed reactor in the temperature range of 260–380 °C. The catalysts were characterized by X-ray diffraction (XRD), temperature-programmed reduction (TPR), transmission electron microscopy (TEM) and nitrogen adsorption at −196 °C. Activity results indicated that the optimized experimental conditions involved a reforming temperature of close to 300 °C and the molar ratios of O₂/EtOH and H₂O/EtOH of 0.44 and 4.9, respectively. An ethanol conversion (C_EtOH) approaching 100% and a hydrogen yield (Y_H2) exceeding 3.0 mole/mole ethanol were noticed at 280 °C over all the catalysts. Among these catalysts, the Pt–Ru/ZrO₂ catalyst was an excellent OSRE catalyst at low temperature. The maximum Y_H2 was 4.4 and the CO distribution was 3.3 mol% at 340 °C.     

Low-temperature H2 and N2 transport through thin Pd66Cu34Hx layers

The single gas H₂ and N₂ permeability of a 4 μm thick dense fcc-Pd₆₆Cu₃₄ layer has been studied between room temperature and 510 °C and at pressure differences up to 400 kPa. Above 50 °C the H₂ flux exhibits an Arrhenius-type temperature dependence with J_H2=(5.2±0.3) mol m⁻² s⁻¹ exp[(−21.3 ± 0.2) kJ mol⁻¹/(R·T)]. The hydrogen transport rate is controlled by the bulk diffusion although the pressure dependence of the H₂ flux deviates slightly from Sieverts’ law. A sudden increase of the H₂ flux below 50 °C is attributed to embrittlement.     

The beneficial role of use of ultrasound in heterogeneous Fenton-like system over supported copper catalysts for degradation of p-chlorophenol

A heterogeneous sono-catalytic system with addition of hydrogen peroxide (US_H2O2+Cat.) was employed for the degradation of 200 ppm of p-chlorophenol (4-CP) at 25 °C and 100 W of ultrasound power. One thousand and six hundred parts per million of initial hydrogen peroxide (H₂O₂) concentration and 1 g/L of catalyst loading over three heterogeneous copper catalysts, CuO, Cu/Al₂O₃ (Cu/Al) and CuO·ZnO/Al₂O₃ (Cu/Zn) was used. The benefits of ultrasound in a heterogeneous catalytic system were evaluated. A considerable synergistic effect of the US_H2O2+Cat. system was only achieved with supported catalysts (Cu/Al and Cu/Zn) possibly due to good dispersion of catalysts as a result of catalyst size reduction during ultrasound irradiation. Moreover, between the two supported copper catalysts, the Cu/Al provided promising catalytic performance by giving higher 4-CP and TOC removal accompanied with efficient H₂O₂ consumption. Experiments with a homogeneous copper catalyst revealed that use of ultrasound in a homogeneous system shows an adverse effect on decomposition of 4-CP.     

Performance of a catalytic membrane reactor for the Fischer–Tropsch synthesis

The study of permeable composite monolith (PCM) membranes for the Fischer–Tropsch synthesis is continued. On the scale of membranes with outer diameter of 42 mm, it is proved that PCM can combine high productivity of hydrocarbons (>55 kg_C5+ ( h)⁻¹ at 0.6 MPa, 484 K), high selectivity towards heavy hydrocarbons (_ASF > 0.85, C₅₊ upto 0.9) as well as high heat-conductivity and high mechanical strength.     

High temperature proton exchange membrane fuel cell using a sulfonated membrane obtained via H2SO4 treatment of PEEK-WC

A method for the sulfonation of PEEK-WC, a glassy poly(ether ether ketone) with sulphuric acid is presented. Depending on the reaction time, polymers with ion exchange capacity (IEC) from 0.30 to 0.76 meq_H⁺/g are obtained, as determined by titration with NaOH solutions. The thermal properties of the polymers were studied by differential scanning calorimetry, showing that the glass transition temperature increases with increasing degree of sulfonation, from 224 °C for pure PEEK-WC to 246 °C for the polymer having an IEC of 0.76 meq_H⁺/g. The sulfonated polymers were used to prepare proton exchange membranes for possible application in fuel cells. Dense membranes were prepared by solvent evaporation, using DMA as the solvent. The transport properties of the membranes were determined in terms of water uptake and permeability for hydrogen and oxygen. Electrochemical characterization was performed by measuring cell voltage and power density curves as a function of current density at different working temperatures and the results were compared with those of a commercial Nafion membrane. A power density of 284 mW/cm² was obtained for S-PEEK-WC membrane at 120 °C in H₂/air fuel cell, slightly above the corresponding value found for Nafion.     

Synthesis of ceria nanoparticles by flame electrospray pyrolysis

A flame electrospray pyrolysis is presented for synthesizing CeO₂ nanoparticles with a dense morphology, high crystallinity and nanometer size. Hydrated cerium nitrate precursor dissolved in an ethanol/diethylene glycol butyl ether mixture was injected into a CH₄/air premixed flame using an electrospray method. The number size distributions of the as-prepared particles were trimodal. It is suggested that the particles for the fine mode were formed by a Rayleigh disintegration of the charged precursor droplets during the droplet evaporation. The particles for the coarse and middle modes are surmised to come from primary and secondary droplets, respectively, which were formed simultaneously during the atomization processes. The CeO₂ nanoparticles for the coarse mode were nonspherical and composed of few crystallites. The nanoparticles for the fine and middle modes were nearly spherical and nonagglomerated. The as-prepared CeO₂ nanoparticles showed highly crystallinity.     

Study of the reverse water gas shift (RWGS) reaction over Pt in a solid oxide fuel cell (SOFC) operating under open and closed-circuit conditions

The (electro-)kinetics of the reverse water gas shift (RWGS) reaction was studied in a solid oxide fuel cell (SOFC) of the type Pt/YSZ/Pt. The effect of imposed potentials, cell temperature (650–800 °C), H₂ (1–10 kPa) and CO₂ (1–10 kPa) partial pressures on the kinetics and mechanism of the catalytic and electrocatalytic RWGS reaction, were systematically examined. The apparent catalytic activation energy was found equal to 15.6 kcal/mol, while H₂ and CO₂ apparent reaction orders were equal to 0.5 and 0.7, respectively. At both open and closed circuit operation, the associative formate decomposition reaction mechanism was considered to describe kinetics. Under closed circuit operation, rate enhancement factor, |Λ|, values up to 10 were achieved. Finally, current density–voltage and current density–power density characteristics of the cell were recorded at various temperatures and gas mixtures of CO₂ and H₂. It was found that electrical power output of the cell was optimized by increasing temperature and decreasing CO₂/H₂ feed ratio. Maximum power density obtained was 9 mW/cm² (at 520 mV cell voltage and a current density of 17.3 mA/cm², at 800 °C and P_CO2/P_H2=0.16).     

Study of the water gas shift reaction on Fe in a high temperature proton conducting cell

The water gas shift (WGS) reaction was studied in a double-chamber high temperature proton conducting cell (HTPC). The proton conductor was a strontia–ceria–ytterbia (SCY) disk of the form: SrCe_0.95Yb_0.05O₃₋ and the working electrode was a polycrystalline Fe film. The reaction temperature and the inlet partial pressure of CO varied between 823 and 973 K, and between 1.0 and 10.6 kPa, respectively. The inlet partial pressure of steam (P_H2O) was kept constant at 2.3 kPa. An increase in the production of H₂ was observed upon “pumping” protons away from the catalyst surface. The Faradaic efficiency (Λ) was lower than unity, indicating a sub-Faradaic effect. The highest value of rate enhancement ratio (ρ) was approximately 3.2, at T = 823 K. The proton transport number (PTN) varied between 0.45 and 1.0. An up to 99% of the produced H₂ was electrochemically separated from the reaction mixture.     

Water effect on the conductivity behavior of NH4PO3-based electrolytes for intermediate temperature fuel cells

The electrical conductivity at intermediate temperature of 150–250 °C and the activation energy for conductivity of composite proton conductors, 2NH₄PO₃–(NH₄)₂Mn(PO₃)₄ and 2NH₄PO₃–(NH₄)₂SiP₄O₁₃, were investigated as a function of water vapor pressure, P_H2O. The activation energy decreased linearly with the natural logarithm of P_H2O, indicating that water is chemically adsorbed to the electrolytes. The decrease in activation energy is possibly caused by formation of hydrogen bonds between the adsorbed water and the electrolytes. In addition, the pre-exponential factor of Arrhenius equation, σ₀, increased with P_H2O. This suggests that the adsorbed water may generate additional mobile protons for the composite electrolyte. Therefore, the enhancement in the electrical conductivity of a NH₄PO₃-based electrolyte in a water-vapor rich atmosphere originates possibly from the decrease in activation energy as well as the increase in mobile proton concentration.     

Cobalt Fischer-Tropsch synthesis: Deactivation by oxidation?

Cobalt catalysts as used in the Fischer-Tropsch synthesis (FTS) are relatively expensive (as compared to iron) and need to have a high metal dispersion and long life to be able to offer a good balance between cost and performance. The oxidation of nano-sized metallic cobalt to cobalt oxide during Fischer-Tropsch synthesis has long been postulated as a major deactivation mechanism. However, to date there is no consistent picture. This paper presents an extensive overview of the literature on this topic of deactivation by means of oxidation for unsupported as well as silica-, alumina- and titania-supported cobalt catalysts. Furthermore, it presents results on the deactivation of an industrial Co/Al₂O₃ catalyst as obtained by pseudo in situ X-ray diffraction, magnetic measurements and X-ray absorption near-edge spectroscopy. These analyses were performed to study the oxidation state of spent industrial Co/Al₂O₃ catalyst samples withdrawn from a slurry reactor operating under realistic FTS conditions, and it was concluded that oxidation can be ruled out as a major deactivation mechanism. Finally, these data together with all relevant literature were used to create a common view on the oxidation behaviour of metallic cobalt during FTS. The apparent discrepancies in literature on the oxidation behaviour of cobalt are most likely due to the lack of direct characterisation of the cobalt oxidation state and due to the comparison of catalysts with varying cobalt crystallites sizes, compared at different reactor partial pressures of hydrogen and water (P_H2O/P_H2). It was shown that the oxidation of cobalt can be prevented by selecting the correct combination of the reactor partial pressures of hydrogen and water (P_H2O/P_H2) and the cobalt crystallite size.     

Characterization of composite nanofiltration membrane using two-parameters model of Extended Nernst–Planck Equation

The characteristics of polyamide membranes with respect to interfacial polymerization of diamine mixtures with trimesoyl chloride (TMC) are studied using two-parameter model of Extended Nernst–Planck Equation. The investigation provided the information about the effect of TMC content and reaction time on the diffusive and convective flow of ions through the membrane. These indirectly reflected structural properties such as effective skin thickness, pore size and structural integrity of membrane. Membrane flux and rejection are related to the TMC content and reaction time, when NaCl and CuSO₄ are used as testing solutes. The diffusive transport, and convective transport, J_vC_1,0(1−R^′₁) contributions are successfully determined by fitting the model to the experimental data to get f^′₁ and R^′₁ parameters. It was found that at high TMC content the contribution of convective transport over diffusion transport is increased due to the increase of effective thickness. However, for smaller size and higher diffusive solute like Na⁺, the ratio of diffusive flow over convective flow is increased at high TMC and high reaction time. This indicated that numbers of tightened pores membrane are increased. An optimum membrane with high flux and high copper ion rejection could be obtained by incorporating 0.1% (w/v) of TMC in the polymerization reaction mixture under reaction time period of 5 s.     

Spin design of iron complexes on Fe-ZSM-5 zeolites

The spin state of iron ions in Fe-ZSM-5 zeolites can be purposefully varied by adsorption of gaseous probe molecules. The resulting Fe complexes with half-integer spin (, and ) can be reliably identified by electron paramagnetic resonance (EPR). A good correlation has been found between the concentration of surface sites active in low-temperature nitrous oxide decomposition and the concentration of low-spin () nitrosyl complexes of Fe formed after adsorption of NO molecules. Based on the analysis of the formation of such complexes under varying conditions, we conclude that these active sites contain a binuclear iron complex with S = 0 and three adsorbed NO molecules. An approach to investigate various Fe-containing sites in oxidation catalysts is discussed.     

Effect of spacers on the electrostatic charge properties of metered dose inhaler aerosols

The electrostatic charge properties of commercial metered dose inhaler (MDI) aerosols, including Ventolin^®, Flixotide^®, Tilade^® and QVAR^®, sampled through new and detergent-coated AeroChamber^® Plus spacers were studied using a modified 13-stage electrical low pressure impactor (ELPI) with aerodynamic cutoff diameters ranging from 0.028 to . Aerosol particles deposited on the impactor stages according to their aerodynamic diameters and their charges were simultaneously measured by the electrometers. The deposited drug mass was assayed chemically using HPLC. The surface potential on the inner spacer wall was measured with an electrostatic probe before and after aerosol actuation. High surface potentials were found on the new spacers whereas the detergent-coated spacers had minimal charges due to the conductive coating. MDI aerosol charges were decreased when spacers were used but the charge profiles of the aerosols were not altered qualitatively. New spacers had the lowest throat deposition, fine particle dose, and net aerosol and fine particle charges as a result of high spacer retention. These trends were partially reversed by the detergent-coated spacers. In general, the charge per mass of drug (charge-to-mass ratio) for particles from detergent-coated spacers was higher than those from new spacers. This was thought to be due to the reduction of electrostatic deposition inside the spacer thus leading to particles carrying higher charges being sampled. The calculated number of elementary charges per drug particle ranged from zero to several hundred, which is sufficiently high to potentially affect lung deposition. The ELPI provided high resolution charge profiles on MDI aerosols delivered through spacers.     

On heat conduction between laser-heated nanoparticles and a surrounding gas

Uncertainties in modeling heat conduction in connection with the application of laser-induced incandescence (LII) to primary particle sizing are discussed. Comparing two models widely used in this context, namely those of Fuchs [(1963). On the stationary charge distribution on aerosol particles in a bipolar ionic atmosphere. Pure and Applied Geophysics 56, 185–193] and McCoy/Cha [(1974). Transport phenomena in the rarefied gas transition regime. Chemical Engineering Science 29, 381–388], it is demonstrated that arising differences may be accounted for by the choice of a proper “effective” thermal accommodation coefficient _eff. In experiments on a large number of carbon blacks an overally good agreement between LII results and specified values for particle sizes based on electron-microscopy (EM) is obtained with a choice of _eff=0.25 (based on the McCoy/Cha-model). As aggregate size is expected to influence heat transfer from primary particles, the experimental data are analyzed by a model for an effective heat transfer surface of fractal aggregates. Based on values for the average number of primary particles per aggregate as derived from photocentrifuge measurements the data yield an extrapolated value for the physical accommodation coefficient for isolated particles of ₁=0.43.     

About randomness of aerosol size distributions

All aerosol formation and evolution processes, such as nucleation, condensation, fragmentation, etc., are understood and rationalized via fundamental probabilistic concepts such as probabilities of collision, coagulation, dispersion, etc. Therefore all theoretical size distribution functions (lognormal, modified gamma distribution, self-preserving particle size distribution for Brownian coagulation, etc.) are in fact size probability density functions pdf(r). Any (e.g., measured) size distribution f(r) of an aerosol system is some random realization of its pertinent size probability density function pdf(r). When pdf(r) is viewed as a continuous function, the corresponding size distribution vanishes almost everywhere excluding some randomly set of sizes where f(r)=1. We investigate proximity between f(r) and pdf(r) in finite size intervals and derive expressions for estimation of the standard deviations of several aerosol size-dependent properties arising from randomness of f(r).     

Methanol reformate treatment in a PEM fuel cell-reactor

The treatment of methanol reformate, containing up to 2500 ppm CO, by the anode of a PEM fuel cell, operating as a preferential oxidation (PROX) reactor, was investigated in order to examine the possibility of electrochemically promoting the water–gas-shift (WGS) reaction and thus making the gas mixture suitable for anodic oxidation. It was found that the electrochemical promotion effect plays a significant role in a normal fuel cell operation (air at the cathode) but not in a hydrogen pumping operation (H₂ at the cathode). This implies that the role of oxygen crossover in the electropromotion (EP) of the WGS reaction and in the CO oxidation is vital. During fuel cell operation, the increase in the rate of CO consumption over a Pt/C anode is 2.5 times larger than the electrochemical rate, I/2F of CO consumption, while for oxygen bleeding conditions (fuel mixture + 1% O₂ at the anode) the increase is up to five times larger than I/2F, i.e. the Faradaic efficiency is up to 5. This shows that the catalytic properties of the Pt anode are significantly modified by varying catalyst potential and by the extent of O₂ crossover.

The effect of temperature, gas composition, membrane thickness and Pt anode alloying with Cu was studied. It was found that the rate of CO consumption is significantly enhanced by increasing T, p_H2 and increasing O₂ crossover rate. Also the Faradaic efficiency reaches even higher values (up to 9) when using PtCu/C anodes. However, the Faradaic efficiency drops in general below 100% at high current densities and CO conversion levels.     

Direct hydrothermal synthesis of mesoporous Sn-SBA-15 materials under weak acidic conditions

A direct synthetic route for the preparation of Sn-SBA-15 materials with n_Si/n_Sn ratios ranging from 100 to 10 under milder acidic conditions than normally employed for the preparation of Si-SBA-15 is reported. The changes in the pH conditions of the gel were made through an adjustment of the molar ratio of n_H2O to n_HCl (<1 pH <2) during preparation. The samples prepared under three different acidic conditions have been characterized by various techniques. An expansion of the lattice (powder XRD) and an increase in mesopore area (low temperature N₂ adsorption) indicate that the hexagonal structure of the SBA-15 is maintained with no loss of long range ordering. The UV–vis reflectance spectra of Sn-SBA-15 samples show the presence of Sn⁴⁺ ions both in tetrahedral and octahedral environment. ²⁹Si MAS NMR spectra of samples prepared under an intermediate acid condition show the presence of Q₃ and Q₄ species. Their ratio increases with a decrease in tin content. The presence of Si in (2Si, 2Sn) i.e., Q₂ environment may point to incorporation of considerable Sn⁴⁺ ions in tetrahedral positions. Sn-Mössbauer spectroscopic studies reveal that Sn²⁺ species form upon reductive treatments and can probably be stabilized in the pore wall upon reoxidation. This to some extent is an indication of the formation and stabilization of Si–O–Sn–O–Si linkages in Sn-SBA-15. A progressive increase in the pH of the medium (increasing the n_H2O to n_HCl ratio) results in the location of Sn⁴⁺ ions, (a) at the surface of the mesopores (surface of the corona region) as a thin film of SnO₂ or small aggregates (loss in mesopore area) depending on the concentration of Sn; (b) at the walls of the mesopore structure, substituting Si⁴⁺ ions (some lattice expansion and tetrahedral Sn⁴⁺ ions); and/or (c) as a part of the corona region, neutralizing the resulting Si–OH groups (a loss of micropore area and octahedral Sn⁴⁺ ions). The studies reveal that the method of preparation, n_H2O/n_HCl ratio and the n_Si/n_Sn ratio (concentration of SnCl₄) of the gel significantly influence the type of tin species in the resulting Sn-SBA-15 samples.     

Solubility of carbon monoxide and hydrogen in propylene carbonate and thermomorphic multicomponent hydroformylation solvent

Solubilities of carbon monoxide and hydrogen in propylene carbonate (PC), biphasic mixture of PC and dodecane (1:1 v/v) and thermomorphic (or temperature dependent) multicomponent solvent (TMS)-system consisting of PC, dodecane and 1,4-dioxane were measured over the temperature and pressure range of 298–343 K and 0.1–1.5 MPa, respectively, in a high pressure solubility cell. The measured solubilities were correlated by a temperature-dependent Henry's law constant and interpreted by activity coefficient models based on the regular solution theory (RST) with Yen and McKetta extension for polar solvents as well as by the UNIFAC group contribution method. The experimental data showed a very good fit in terms of Henry's law constant except for H₂–PC and CO–PC binaries. The RST-based model, that did not involve any adjustable constant, could predict the experimental solubility to within ±11.0% error. The UNIFAC model worked better with the interaction parameters computed as a linear function of temperature using a part of the experimental solubility data set. The accuracy of prediction was found to be within a maximum error of ±8.5%. The TMS system shows higher affinity for CO and H₂, which is comparable to the single phase PC. The experimental solubilities in the liquids are substantially larger than those in most other hydroformylation solvents thereby establishing its advantage over the alternative solvents for industrial use. Liquid–liquid equilibrium for the TMS system consisting of PC, dodecene and 1,4-dioxane system was also measured at 298, 353 and 373 K.     

Direct decomposition of NO into N2 and O2 on BaMnO3-based perovskite oxides

Effects of dopant in BaMnO₃ perovskite oxide on the NO direct decomposition activity were investigated. NO direct decomposition activity was greatly elevated by doping La and Mg for Ba and Mn site in BaMnO₃, respectively. The highest N₂ yield was achieved on Ba_0.8La_0.2Mn_0.8Mg_0.2O₃. The NO decomposition rate increased with increasing NO partial pressure with P_NO^1.19. Coexistence of oxygen lowered the N₂ yield with P_O2^−0.18; however, N₂ yield of 40% was sustained even under coexisting of 5% O₂ at 1123 K. Adsorption state of oxygen was also studied with temperature programmed desorption (TPD) method and the desorption temperature of oxygen was lowered by doping Mg for Mn site in BaMnO₃.