Hydrotreatment of high aliphatic and low sulfur atmospheric residue (Residue A) over a combination of demet-allation and desulfurization catalysts caused greater catalyst deactivation than low aliphatic and high sulfur residue (Residue B). Since the catalytic dissociative reaction of the aliphatic group was initiated by the adsorption of the residue on the alumina support of the desulfurization catalyst, olefins produced by acid sites on the alumina are probably responsible for the catalyst deactivation through the deposition of coke. Use of an amorphous silica-alumina with stronger acid sites than alumina as the catalyst support for the desulfurization catalyst reduced the reaction temperature necessary for the desired conversion (WAT) by 20°C compared to the alumina suppolt catalyst at 1500h after the beginning of the reaction. However, catalyst deactivation of the catalyst supported on silica-alumina was significantly faster than of the catalyst supported on alumina. Feeding a mixture of Residue A (60vol%) and Residue B (40vol%) into the desulfurization process decreased catalyst deactivation of the combination of conventional catalysts supported on alumina. Although sulfur compounds in Residue B are important to reduce coke formation during the hydrotreatment of atmospheric residue, effective hydrotreatment of Residue A with long aliphatic chains was not achieved with the present catalyst combination.
The catalytic transformation of petroleum-derived asphaltene with Co loaded smectite and CoMo loaded alumina based catalysts was studied in the range from 40 to 420°C using proton magnetic resonance thermal analysis (PMRTA). Residual hydrogen as a qualitative measure of the hydrogen remaining in the sample and second moment (M2T16), which is sensitive to molecular mobility, were obtained as a function of temperature. Both residual hydrogen and M2T16 of the asphaltene were dependent on the properties of the coexisting catalyst. Metal oxides in the Co porous saponite and CoMo alumina based catalysts, which are active for the transformation of asphaltene, were probably reduced to the metals and became ferromagnetic at about 350°C, because the M2T40 values of the char derived from the asphaltene with the active catalysts after cooling to room temperature were over 250kHz2, compared to typical values of 150kHz2 for chars without ferromagnetic material. Hydrodesulfurization and hydrocracking with the catalysts were correlated with the values of M2T16 at 400°C. Moderate interaction of the metal oxide with the support material resulted in an intermediate reduction state with high HDS acctivity. The PMRTA method with 1H as a probe of the state of hydrocarbons such as asphaltene provides useful information about the catalyst activity and is helpful for the design and preparation of the catalyst.
The utilization of the Pt/ZnO catalyst to a hydrogen permeation type membrane reactor is expected, because the Pt/ZnO catalyst has high catalytic activity and selectivity for the low-temperature dehydrogenation of isobutane to isobutene. However, the catalytic activity of Pt/ZnO catalyst gradually deteriorates due to coke deposition in the course of reaction. The introduction of steam into the reaction system, which expected the prevention of coke deposition, brought about the decline of the catalytic performance. Then, various experiments were carried out in order to clarify this cause, and the following conclusions were obtained. As the reason why the Pt/ZnO catalyst showed the high performance in the reducing atmosphere under the steam nonexistence, it was inferred that the free electrons, which occur by reducing the part of ZnO to Zn, is donated to Pt, and that the resulting increase of the electron density of Pt brings about the increase in the ability of pulling out hydrogen. On the other hand, it was considered that the Pt/ZnO catalyst does not show high catalytic activity, because the free electrons in ZnO does not occur in the oxidizing atmosphere under the steam existence.
In the low-temperature dehydrogenation of isobutane under the steam coexistence, the performance of Pt/ZnO catalyst deteriorates. Then, in order to improve the performance of Pt/ZnO catalyst, Pt/ZnO-M2O3 catalysts that added various third components were prepared, and the characteristics of those catalysts were examined. The addition of Cr, Al, and Ga to the Pt/ZnO catalyst markedly improved the catalytic performance, especially the catalytic activity. As this cause, it was inferred that the free electrons, which occur in the ZnO lattice by the substitution of these elements with Zn, are donated to Pt, and that they promote the dehydrogenation ability of Pt. Further, it was found that the addition of the appropriate third component brings about the increase in the BET surface area by the suppression of the crystal growth of ZnO. In addition, it was considered that the appropriate third component is the element whose ionic radius is close to that of Zn in order to substitute Zn, and is also the element that is easy to release an electron, namely the element whose electronegativity is small.
The single-continuum approach employing effective permeability is one of the practical methods for simulating naturally fractured reservoirs. Sensitivities of the flow behavior to the assumed region for effective permeability calculations, and to the scale of simulation grid-blocks were studied to examine the equivalent single-continuum system. A flux-continuous full-tensor model was used to deal with permeability tensors resulting from upscaling of fractured systems. The representative elementary volume (REV) for stochastic fracture distributions was evaluated to examine the behavior of the effective permeability with change in the area of the calculating region. The REV for effective permeability was established for fracture systems of mean fracture length (ml) 0.02, 0.06, and 0.2 within the domain of unit area. Variations in effective permeability are small even for the upscaling sizes lower than the REV in the case of short fractures with ml=0.02. Therefore, homogeneous models can correctly simulate tracer test performances. For medium and long fractures of mean length 0.06 and 0.2 respectively, effective permeability corresponding to the upscaling sizes lower than the REV indicates large fluctuations reflecting significant local heterogeneity. Local heterogeneity of small scale must be incorporated in permeability distributions to obtain good simulation results. For modeling local heterogeneity, a grid-block should be assigned a specific effective permeability tensor and should be as coarse as the order of ml, followed by refinement of the coarse grids to avoid numerical dispersion.
The authors have developed CuO-ZnO-Al2O3-Ga2O3-MgO catalyst for methanol synthesis from CO2 and H2, and have been studying a technology for the direct synthesis of dimethyl ether (DME) from CO2 and H2 using the combination of the methanol synthesis catalyst above mentioned and a methanol dehydration catalyst. In the present study, the catalytic activities of three types of γ-Al2O3 with different specific surface areas, four types of compound oxides (ZrO2•Al2O3, SiO2•Al2O3, SiO2•TiO2, ZrO2•TiO2), and a ZSM-5 zeolite for DME synthesis by methanol dehydration were tested. DME synthesis activity increased with higher specific surface area of γ-Al2O3 catalyst. Compound oxide catalysts containing Al2O3 showed higher DME synthesis activity than catalysts without Al2O3, and ZrO2•Al2O3, the best compound oxide, showed higher DME synthesis activity than the best γ-Al2O3 with the largest specific surface area (250m2/g). Addition of a metal oxide as the promoter is effective for improving the DME synthesis activity of Al2O3 by methanol dehydration. ZSM-5 zeolite produced more olefins rather than DME. The presence of water in the methanol feed suppressed the DME synthesis reaction by methanol dehydration.
This paper describes experimental and modeling studies to investigate the behavior of gas-liquid two-phase flow in horizontal and slightly inclined pipelines. The studies were performed with experimental data newly obtained at over 1000 flow conditions that were designed to cover all the flow patterns. The experiments were conducted at a test loop of a 54.9mm diameter, 105.0m long pipeline, with 0, 5, and 10 degree inclination of the test section. A new mechanistic model for horizontal and slightly inclined flow was developed implementing transition criteria between the flow patterns, correlation for liquid holdup, and flow models for pressure drop computation. The criteria for the transition boundaries between flow regions were determined for eight flow patterns matching the predicted flow patterns with the experimental observations. For stable dispersed bubble flow, constraints on the bubble size were additionally applied. At the transition between stratified and non-stratified flow, we added such a criterion that slug flow can remain stable at much lower flow rates of liquid. The froth flow region was newly modeled to reproduce the experimental observations between annular and slug flow. The pressure drops were computed for individual conditions of the experimental measurements. Excellent performance of the model was demonstrated by good agreements between the computed results and the experimental data. Correct identification of the flow pattern is crucial for accurate estimation of pressure drops, particularly for those near the transition boundaries. Use of appropriate friction factors is of primary importance for pressure drop calculations in any of the flow patterns.
The effects of CeO2 addition to supported Pd catalysts were investigated in the decomposition of MeOH to H2 and CO. The catalytic activity of SiO2-supported Pd was lower than those of Al2O3-, MgO- and ZrO2-supported Pd catalysts. Carbon dioxide and methane were produced in a temperature region higher than 350°C over Pd/Al2O3, Pd/MgO, and Pd/ZrO2 but not over Pd/SiO2. Addition of CeO2 promoted the reaction for all catalysts. The activity of Pd-CeO2/SiO2 was high as that of Pd-CeO2/ZrO2 without losing the high selectivity for CO and H2. XAFS analysis indicated that most of the Pd is metallic particles which are highly dispersed on CeO2/SiO2 and a small part of the Pd is in the oxidized state.
Genetic algorithm was applied to optimize the Cu/Zn/Al ratio of mixed oxide catalyst for methanol synthesis from syngas. A layered neural network was used instead of experiments to evaluate the "fitness" of the catalyst code. This procedure eliminated laborious steps, such as catalyst preparation and activity testing, from the optimization loop. The calculated activity (STY) was almost identical to the original one and could be used as an evaluation function in the genetic algorithm program. The combination of catalyst design by genetic algorithm and activity evaluation by a layered neural network is a promising method for highly efficient catalyst screening.