The behavior of adsorbed nitrogen monoxide (NO) and the reactivity of NO with reducing agents were investigated. Two species with N-O stretching frequencies at 1, 790 and 1, 850 cm^-1, respectively, were found from the infrared studies of adsorbed NO on a cobalt-silica catalyst. The former arose from the strongly chemisorbed species and it was the intermediate species in the reaction between NO and reduced surface. Bases such as NH₃, NH₂CH₃, NH (CH₃)₂ and N(CH₃)₃ resulted in the lowering of the N-O stretching frequencies of preadsorbed NO, indicating weakening of the N-O bond. This was explained on the basis of a change in the electron density of metal-N-O molecular orbitals when a second gas was admitted.
The roles of reducing agents such as H₂, CO and NH₃ in the catalytic reduction of NO over supported metal or metal oxide catalysts were studied from the viewpoint of reaction mechanism. The reaction of NO with H₂ or CO was shown to proceed through a redox cycle of the catalyst, that is, the catalyst was oxidized by NO and then it was reduced by the reducing agents to complete the catalytic sequence. On the other hand, it was confirmed by means of isotope labelling techniques that NH₃ or fragments of NH₃ directly reacted with the adsorbed NO. Also, the reaction of NO with NH₃ was greatly enhanced by preoxidation of a Cr₂O₃-Al₂O₃ catalyst and this effect was due to the oxygen with a high oxidation power which was produced on the catalyst surface.
These differences in the roles of reducing agents in the reaction mechanism satisfactorily explain some aspects of catalytic removal of NO in the practical process where oxygen is present.

View full abstract

The kinetic study of the oxidative dehydrogenation of isobutyraldehyde, IBA, by the periodic pulse technique has been made to obtain some information on the improvement of the catalyst for the periodic pulse technique. The reaction rate to each product was experimentally and theoretically examined as a function of air pulse width, IBA pulse width and IBA partial pressure, and the rate was explained by assuming the following reaction scheme. IBA reacts with the oxygen retained by the catalyst to form methacrolein, MA. The IBA adsorbed on the oxidized surface is oxidized by the oxygen coming from the gas phase to form carbon monoxide and carbon dioxide. The adsorbed IBA desorbs as another oxidized product.

View full abstract

For calculation of gas volume factor (B_g), a method in which not only the reservoir pressure but also the composition of the remaining gas in the reservoir was used for making a reservoir model of a water drive type gas condensate reservoir. Several analyses were carried out using the actual field data, and the results, including performance prediction, were found satisfactory.

View full abstract

The surface course in the pavement structures must carry varying wheel loads and spread loads to the lower courses. Concerning the material properties, the main factors that effect the above behavior are the modulus and fatigue properties of the paving materials.
In this research, two kinds of testing methods, dynamic loading testing and fatigue testing, were used to compare the mechanical properties of the various paving materials. The former testing was employed to evaluate the dynamic properties of the paving materials and the latter testing was employed to evaluate fatigue behavior-mainly the resistance to repetitions of loading-of the materials.
In this experiment, four kinds of binders, straight asphalt cement, modified asphalt cement, epoxy asphalt cement and epoxy resin, were used by mixing each of them with dense graded aggregates. Conclusions drawn from this experiment are as follows:
1. Epoxy resin system mixtures (epoxy resin mixtures and epoxy asphalt mixtures) showed less temperature and time susceptibility than that of the asphaltic mixtures. They also showed lower limits of complex modulus that mean descending limits of load spreading effects (Figs. 1-10).
2. The upper limit of the loss tangent was found in the epoxy resin system mixtures at a certain specified loading time, and a greater time susceptibility of the complex modulus was found at the same loading time (Figs. 5, 7).
3. Specific temperatures T_c, of paving mixtures used in this experiment were obtained from the reduced variable method and they were about 45-50°C higher than the glass transition point of the same paving mixtures (Figs. 2, 4, 6, 8, 9).
4. The fatigue resistance of the epoxy resin system mixtures to load repetitions was greater than that of the asphaltic mixture; moreover, the former mixtures also showed a smaller strain susceptibility to fatigue life than the latter ones (Fig. 14).

View full abstract

Based on the analysis of the results obtained with the used catalysts used in a previous paper, the deactivation mechanism of HDN catalyst was studied. The analytical results of the catalysts, before and after reaction, are shown in Table 1.
The quantities of carbon deposited on the catalyst at various positions in the reactor are shown in Fig. 1. The quantity of carbon is found to be uniform at 400°C but if increases at 430°C. The quantity of carbon deposited at reactor inlet differs with the catalyst used, but Fig. 2 shows that the quantity deposited per unit surface area is well related to the pore diameter of the catalyst. As shown in Fig. 3 the average amount of carbon deposited per unit surface area for the whole reactor is in a good linear relationship with deactivation of the catalyst. These results suggest that a catalyst with a larger pore diameter can capture carbon precursors more easily and, it is therefore poisened by more extensive coking and its activity is declined more rapidly.
The N/C (atomic ratio) values shown in Table 1 suggest that there are two kinds of cokes which break out in different mechanisms from each other. One of them is formed directly from the feed (probably through dehydrogenation) and the other is formed from the cracked fragments (probably through polymerization). A model of carbon deposition is illustrated in Fig. 4.
By analysis of metal compounds deposited on the catalyst, V and Ni are probably in the forms of V₃S₄ and Ni₃S₂. From the densities of these sulfides and from the pore volume of the catalyst used, the specific gravity of coke was estimated.
By use of these values and effectiveness factors calculated from the deposition curves of V and Ni shown in Fig. 6, the degree of pore mouth plugging was obtained. The results are shown in Table 2.
The activities of the used catalyst sampled from the various parts of the reactor were examined, and they are shown in Table 3. The influence of carbon deposition on degradation of the catalyst can be recognized clearly but not that of vanadium. The AP catalyst taken out from the reactor inlet shows very low activity despite the deposition of only small quantities of carbon and metals on the catalyst, because the degree of pore mouth plugging of the catalyst is about 1.0.

View full abstract

Dehydrogenation of n-hexane and other normal paraffins (C₃-C₈) was studied on active carbons impregnated with iron and copper. The active carbon which contained 5wt% of copper (C-Cu) showed higher catalytic activities compared to metal free active carbon (C) for all hydrocarbons tested (Table 1). The maximum activity difference between the two catalysts was observed in the case of n-pentane or n-butane dehydrogenation. The reactivity of n-paraffin increases with the number of carbon atoms in the molecule (Fig. 1). The major dehydrogenated product was the thermodynamically-equilibrated mixture of mono-olefins (in the case of C₃-C₅ paraffins). From normal paraffins with 6 or more carbon atoms, aromatic hydrocarbons were formed as well as mono-olefins and di-olefins (Table 2). Ethylbenzene and o-xylene were the only aromatic hydrocarbons formed from n-octane, suggesting that the six membered ring is directly formed from a straight chain. The activity of iron supported (5wt%) active carbon (C-Fe) was higher than that of C-Cu (Table 4). On C-Fe, hydrogenolysis of n-hexane (forming methane) proceeded besides dehydrogenation. The hydrogenolysis on C-Fe was suppressed by pretreating it with H₂S or alkali carbonate or by alloying it with copper (Table 5).
On C-Cu and C-Fe, the adsorption rates of hydrogen at 400°C were far higher than the rate on C (Fig. 3). The temperature-programmed desorption of the adsorbed hydrogen showed that the hydrogen on C-Fe began to desorb at a temperature level lower by about 200°C than that on C (Fig. 4). The amount of desorbed hydrogen from C-Fe in the temperature range up to 550°C was about ten times of that from C. These results lead us to the conclusion that, on the metal supported carbons, the movement of hydrogen between carbon surface and gas phase is far more rapid than on C (because of the spillover phenomenon); thus, they exhibit higher catalytic activities.

View full abstract

The effects of non-cylindrical extrudates consisting of MoO₃-CoO-γ-Alumina on hydrodesulfurization of vacuum gas oil and atmospheric residue have been investigated. In the experiments, non-cylindrical extrudates with cross sections of symmetrical quadrulobes, asymmetrical quadrulobes and three-lobes and cylindrical extrudates with nominal diameters of 1/32, 1/16 and 1/12 inch were tested in a small bench scale unit with diluted catalyst bed. Some of them were tested also in a pilot plant scale unit with undiluted catalyst bed.
A good relation between activity and particle size expressed in terms of volume and surface ratio have been observed. This relation is independent of variation in catalyst shape.

View full abstract

The selectivity of hydrogenation and hydrodesulfurization of dibenzothiophene on a presulfided molybdena-alumina catalyst has been examined by selective poisoning studies, using acridine, carbazole, dicyclohexylamine, hydrogen sulfide, and phenanthrene. The experiments were carried out for a xylene solution containing 5wt% dibenzothiophene under the following conditions: temperature, 300°C; total pressure, 100atm; weight hourly space velocity, 7; and H₂/HC (molar) 6.1.
Cyclohexylbenzene and biphenyl were the major products of dibenzothiophene hydrodesulfurization, with the formation of minor amounts of tetrahydrodibenzothiophene, hexahydrodibenzothiophenes, bicyclohexyl and ethylbicyclo [4.4.0]decane. Increasing the concentration of the nitrogen compounds or phenanthrene increased the concentration of biphenyl, but it decreased the concentration of the other products (Figs. 1, 2, 3, 5). In the case of acridine and dicyclohexylamine, the decrease in cyclohexylbenzene corresponded exactly to the increase in biphenyl. The poisoning effect decreased in the order of acridine>carbazole>dicyclohexylamine>phenanthrene. The addition of hydrogen sulfide increased the concentrations of tetrahydro- and hexahydrodibenzothiophenes but decreased the concentrations of cyclohexylbenzene, biphenyl and bicyclohexyl (Fig. 4).
Consequently, hydrogenation of dibenzothiophene was inhibited by a small addition of the nitrogen compounds and it was suppressed by hydrogen sulfide. Hydrodesulfurization of hexahydrodibenzothiophene to cyclohexylbenzene, of dibenzothiophene to biphenyl, and of perhydrodibenzothiophene to bicyclohexyl took place on the same active sites which were poisoned by hydrogen sulfide. Phenanthrene seems to inhibit weakly not only the hydrogenation of dibenzothiophene but also the hydrodesulfurization of dibenzothiophene. Hydrogenation and hydrodesulfurization of dibenzothiophene proceeded on different sites. The mechanism of selective poisoning of dibenzothiophene hydrodesulfurization on a presulfided molybdena-alumina catalyst was summarized in Fig. 6. The course of poisoning effects of the additions of acridine, carbazole, or dicyclohexylamine on the formation of cyclohexylbenzene followed the Langmuir adsorption equation.

View full abstract

Hydrocracking of Gach Saran-VR, Kuwait-VR & RC, Ta-Ching-VR & RC with a discharged catalyst, which contained 9.6wt% vanadium and 12.9wt% carbon and which had been used in a commercial hydrodesulfurization plant, was carried out in a bench-scale suspended bed reactor in order to examine its hydrocracking activity.
1) Hydrocracking of reduced crudes and vacuum residua
Table 2 shows the results of hydrocracking of Gach Saran-VR, Kuwait-VR & RC, Ta-Ching-VR & RC with the discharged catalyst at 450°C; there were no coking problems. Those results showed that it would be possible to use the discharged catalyst as a hydrocracking catalyst.
1.1) Hydrocracking of vacuum residua
Gach Saran-VR, Kuwait-VR and Ta-Ching-VR were hydrocracked at 100kg/cm². The percentage of Ta-Ching-VR was the lowest among the three residua, when they were cracked under the same conditions, and the yield of vacuum gas oil from the cracked oil of Ta-Ching-VR was higher than that from Gach Saran-VR and Kuwait-VR. These results showed that Ta-Ching-VR was not cracked easily as compared with Gach Saran-VR and Kuwait-VR
1.2) Hydrocracking of reduced crudes
Fig. 1 shows the relationship between V/F^*a) and yields of distillates in hydrocracking of Kuwait-RC. As shown in Fig. 1, the yield of distillates increased with decreasing operating pressure. This tendency was more remarkable in hydrocracking of Ta-Ching-RC (Fig. 2). As shown in Fgi. 3, the hydrocracking rate constant of Kuwait-RC was about 2.5 times as large as that of Ta-Ching-RC under the same conditions.
2) Properties of fractions of cracked oils
Cracked oils (GVR-32, KVR-1, KRC-1, 2, 3, TRC-3, 6), Gach Saran crude, Kuwait crude and Ta-Ching crude were distilled into seven fractions (LN^*b), HN^*c), K^*d), LGO^*e), HGO^*f), LVGO^*g) and VR^*h)), and analytical data of these fractions are summarized in Tables 3 to 9.
*a) (reactor volume, l)/(feed rate, kg/hr)
*b) light naphtha (IBP-90°C)
*c) heavy naphtha (90°C-170°C)
*d) kerosene (170°C-230°C)
*e) light gas oil (230°C-290°C)
*f) heavy gas oil (290°C-340°C)
*g) light vacuum gas oil (340°C-400°C)
*h) vacuum residue (400°C⁺)

View full abstract

The catalytic activity of 30% NiO-Al₂O₃ catalyst for hydrocracking of perhydroindene was investigated under a hydrogen pressure of 100atm and at the temperature range of 300°C to 375°C. The catalyst was reduced with hydrogen at 450°C for 12hr. The products were analyzed by means of gas chromatograph and GC-MS system. The product distribution observed is shown in Tables 1 and 2. When the reaction was carried out at lower temperatures, 1-ethyl-2-methylcyclohexane was produced selectively, but a small amount of propylcyclohexane was also produced. These data show that the CH₂-CH₂ bond in the pentane ring of perhydroindene cleavages predominantly at lower temperatures and that successive demethylation of the ringopened products takes place as the reaction proceeds.
The reaction scheme of hydrocracking of perhydroindene over the nickel-alumina catalyst is given in Fig. 1. It is shown that the main reaction path is the ring opening to 1-ethyl-2-methylcyclohexane and successive demethylation to 1, 2-dimethylcyclohexane, methylcyclohexane and cyclohexane, and that side reaction path is the ring opening to propylcyclohexane which is then successively demethylation to ethylcyclohexane, methylcyclohexane and cyclohexane.

View full abstract