Enthalpy recovery of gases issued from H2 production processes: Activity and stability of oxide and noble metal catalysts in oxidation reaction under highly severe conditions

Oxidation activity and stability under reaction was investigated for a series of mixed oxide catalysts, doped or not by a precious metal (Pd, Pt). The reaction feedstock, containing CO, H₂, CH₄, CO₂ and H₂O, simulated gases issued from H₂ production processes for fuel cells. Contrarily to conventional noble metal catalysts, mixed oxide samples present generally good stability under reaction at high temperature. The activities measured for the perovskite and hexaaluminate catalysts, are however largely lower than that of the reference Pd/Al₂O₃ catalyst. High activities were obtained after impregnation of 1.1 wt.% Pd or 0.8 wt.% Pt on the hexaaluminates samples. Even if Pd/Al₂O₃ was found to present a high activity, this sample suffered from drastic deactivation at 700 °C. Better stability were obtained on perovskite. Furthermore, doping hexaaluminate by Pt led to samples with good activities and high stability. Even if better activities were obtained by doping the hexaaluminate samples by Pd, the Pd/BaAl₁₂O₁₉ strongly deactivated, as it was previously observed for the reference catalyst. Interestingly, this Pd deactivation was not observed when Pd was impregnated on the Mn substituted hexaaluminate, leading to a stable and active catalyst. This suggests that it is possible to stabilize the palladium in its oxidized form at high temperature (700 °C) on the surface of some supports.     

Inhibition of neuronal (type 1) nitric oxide synthase prevents hyperaemia and hippocampal lesions resulting from kainate-induced seizures

The possible roles for nitric oxide produced by neurons in epileptic conditions have been investigated from two different aspects: microcirculation and delayed damage. Our aim was to determine whether the selective inhibition of neuronal (type 1) nitric oxide synthase by 7-nitroindazole, during seizures induced by systemic kainate, modifies hippocampal blood flow and oxygen supply and influences the subsequent hippocampal damage. Experiments were performed in conscious Wistar rats whose electroencephalogram was recorded. 7-Nitroindazole (25 mg/kg, i.p.) or its vehicle was injected 30 min before kainate administration (10 mg/kg, i.p.) and then twice at 1-h intervals. Kainate triggered typical limbic seizures evolving into status epilepticus, identified by uninterrupted electroencephalographic spike activity. The seizures were stopped by diazepam (5 mg/kg, i.p.) after 1 h of status epilepticus. Three types of experiments were performed in vehicle- and 7-nitroindazole-treated rats. (1) Hippocampal nitric oxide synthase activity was measured under basal conditions, at 1 h after the onset of the status epilepticus and at 24 h after its termination (n = 4-6 per group). (2) Hippocampal blood flow and tissue partial pressure of oxygen were measured simultaneously by mass spectrometry for the whole duration of the experiment, while systemic variables and body temperature were monitored (n = 6 per group). (3) Hippocampal damage was revealed by Cresyl Violet staining and evaluated with a lesion score seven days after status epilepticus (n = 12 per group). Hippocampal nitric oxide synthase activity was not significantly modified during status epilepticus or the following day in vehicle-treated rats. In contrast, it was inhibited by 57% in 7-nitroindazole-treated rats, both in basal conditions and after 1 h of status epilepticus, but was not different from its basal level 24 h later. 7-Nitroindazole significantly decreased basal hippocampal blood flow and tissue partial pressure in oxygen by 30% and 35%, respectively without affecting any systemic or thermal variable. During status epilepticus, 7-nitroindazole significantly reduced the increase in hippocampal blood flow by 70% and prevented any increase in the tissue partial pressure of oxygen. Seven days later, the hippocampal damage in the CA1 and CA3 layers was significantly less in 7-nitroindazole-treated rats than in vehicle-treated rats. These results indicate that the inhibition of neuronal nitric oxide synthase by 7-nitroindazole protects neurons from seizure-induced toxicity despite reducing blood flow and oxygen supply to the hippocampus.     

A multiscale virtual element method for the analysis of heterogeneous media

We introduce a novel heterogeneous multiscale method for the elastic analysis of two-dimensional domains with a complex microstructure. To this end, the multiscale finite element method is revisited and originally upgraded by introducing virtual element discretizations at the microscale, hence allowing for generalized polygonal and nonconvex elements. The microscale is upscaled through the numerical evaluation of a set of multiscale basis functions. The solution of the equilibrium equations is performed at the coarse scale at a reduced computational cost. We discuss the computation of the multiscale basis functions and corresponding virtual projection operators. The performance of the method in terms of accuracy and computational efficiency is evaluated through a set of numerical examples.     

On the tube hydroforming process using rectangular,trapezoidal, and trapezoid-sectional dies: modeling and experiments

It is generally known that the contact between tube and die, in the case of tube hydroforming process, leads to the appearance of friction effects. In this context, there are many different models for representing friction and many different tests to evaluate it. In the present paper, the pin-on-disk test has been used and the theoretical model of Orban-2007 has been chosen and developed to evaluate friction coefficient. The main goal is to prove the capacity of theoretical model to present the friction conditions in comparison with the pin-on-disk test. From the Orban model, values of 0.05 and 0.25 of friction coefficient have been found under lubricated and dry tests, respectively. On the other hand, by the classical pin-on-disk test, other values were experimentally obtained as friction coefficient at the copper/steel interface. In the case of pure expansion hydroforming, based on an internal pressure loading only, a “corner filling” test has been run for tube hydroforming. Both dry and lubricated contacts have been considered. Various configurations and shapes have been studied such as the rectangular, trapezoidal, and trapezoid-sectional dies. Finite element simulations with 3D shell and 3D solid models have been performed with different values of friction coefficients. From the main results, it was found that the critical thinning occurs in the transition zone for the square and rectangular section die and in the sharp angle for the trapezoidal and trapezoid-sectional die. The comparison between numerical data and experimental results shows a good agreement. Moreover, the thickness distribution along the cross section is relatively consistent with those measured for the 3D shell model; however, the 3D solid models do not provide a realistic representation of the thickness distribution in the shaped tube. Finally, the results obtained from the theoretical model were more efficient than the results obtained from the pin-on-disk test.     

Coke deposition on supported palladium catalysts

Palladium/silica catalysts of varying dispersities were submitted to the coking reaction with cyclopentane at 500°C. Temperature programmed oxidation curves reveal the presence of two peaks for carbon deposition on the metallic function: the first one near 250°C would correspond to coke deposited on palladium atoms of high coordination and the second one near 400°C would involve palladium atoms of low coordination.     

Effect of Pd precursor salt on the activity and stability of Pd-doped hexaaluminate catalysts for the CH4 catalytic combustion

This study reports the influence of palladium salt precursor on the catalytic activity of palladium-doped hexaaluminate catalysts for the combustion of 1 vol% CH₄ in the presence of CO₂ and H₂O as inhibitors. Thermal stability of the catalysts is evaluated in long-term catalytic test at 700 °C. The hexaaluminate supports were synthesized using two different procedures: conventional coprecipitation and solid/solid diffusion procedure. Palladium impregnation was carried out by two different routes using Pd(NO₃)₂ in water or Pd(acac)₂ in toluene as impregnation solution. It was observed that using Pd(acac)₂ as precursor allows to attain higher dispersion of the active phase (Pd particles size <3 nm). Compared to the catalysts obtained by impregnation of Pd(NO₃)₂, higher catalytic activities are then obtained. Nevertheless, a deactivation of the samples obtained using Pd(acac)₂ is observed. At the end of the stability test, almost similar catalytic activity is obtained whatever the palladium precursor. Reduction–reoxidation experiment showed that this deactivation is irreversible, and TEM analysis suggest that this deactivation is related to the sintering of Pd particles under reaction over samples synthesized using Pd(acac)₂ as precursor.     

A direct simulation Monte Carlo scheme and uniformly distributed sequences for solving the Boltzmann equation

The Direct Simulation Monte Carlo (DSMC) scheme of Nanbu is considered for the solution of the Boltzmann equation in a simplified case. It is interpreted as a one-step method using particles combined with numerical quadratures after each step. A modified scheme in which the particles are ordered after each step is proposed. It is called the Low Discrepancy (LD) method. The error of the LD method is defined as the discrepancy of the set of particles relative to the exact solution. This error is estimated by means of other discrepancies, namely those of the sequences which perform the quadratures. The replacement of pseudo-random numbers used in the quadratures by uniformly distributed sequences is consequently suggested. Numerical comparisons are given between the DSMC scheme and the LD method that repeatedly uses the Hammersley sequence in the quadratures (LDH method).     

Stability of supported platinum catalysts in aqueous phase under hydrogen atmosphere

Two supported platinum catalysts (1.0 wt.% Pt/alumina (D=51%) and 1.0 wt.% Pt/silica (D=41%)) were treated at room temperature under nitrogen or hydrogen in aqueous media of various pH (pH=0.7, HCl; pH=6, water; pH=10, NH₄OH). Then, the catalysts were filtered, dried and activated following various procedures. The dispersions of the different samples were determined from transmission electron microscopy (TEM) and/or chemisorption measurements.

On preoxidized samples, a sintering of platinum is observed under hydrogen bubbling whatever the pH of the solution and it is higher on the silica support. An in situ reduction allows to prevent the metal accessibility loss on alumina and to restrict that observed on silica. Otherwise, the activation mode (direct reduction or calcination–reduction) can modify the platinum accessibility of the Pt/alumina catalyst treated in HCl medium.