Coke Layer Research Articles

Burden Distribution Analysis by Digital Image Processing in a Scale Model of a Blast Furnace Shaft.

During the realigning of Aceralia's blast furnace B in Gijon, a 1/10 scale half-section, three-dimensional cold model of the BF shaft was built to test charging patterns and the effect of gas flow in burden distribution. The front side of this half model is closed by a methacrylate sheet that allows images of the burden distribution inside the model to be obtained, during and after the charging process. Image processing techniques were applied to obtain useful information from burden profile images. A complete study of a charging pattern is presented in this paper. The model has allowed study of some local effects of gas flow in burden distribution, such as changes in the final coke layer during ore charging, and the formation of a narrow window in the centre of the model that connects the successive coke layers.

Heat transfer in intumescent heat- and fire-insulating coatings

The physicochemical and thermophysical processes that jointly occur in intumescent heat- and fire-insulating coatings are studied numerically with the use of a model based on the hypothesis that the gaseous products of thermal expansion of an initial material are transparent. It is found that with the radiative mechanism of heat transfer in a layer of intumescent-coating coke taken into account, the temperature at the surface of a structure to be protected is almost equal to a temperature reached at this surface if a nonintumescent coating is used.

Hydrotalcite-like compounds as precursors for mixed oxides catalysts in the oxydehydrogenation of ethylbenzene

The dehydrogenation of ethylbenzene to styrene is a very important industrial reaction due to the multiple applications of styrene as a monomer for synthetic polymers. Reported catalysts that are active and selective in the oxidative dehydrogenation of alkylaromatics include supported metals through both main group and transition metal oxides to polymers and activated carbons. However, most of these catalysts are acid oxides, inducing the formation of an active coke layer that is the actual catalyst. We report in this work the use of various hydrotalcite-like compounds as precursors for the preparation of mixed oxides which are then used as basic catalysts in the oxydehydrogenation reaction. A series of materials has been synthesised by coprecipitation at constant pH and fully characterised by usual analytical techniques. The catalytic screening tests were carried out in a fixed bed quartz reactor at a temperature of 450°C and the liquid products were analysed off-line by gas chromatography. Some particularities of these materials are brought forward, like the absence of coking which however, does not seem to affect the activity, thus suggesting a different mechanism; the results show high selectivity in styrene, while the activity of the catalysts needs some improvements. For example, with vanadium containing oxides a selectivity of 98% in styrene is achieved with a conversion of 38% in ethylbenzene.

98/04087 Blast-furnace charging with ore and coke layers and peripheral fuel briquets

98/04087 Blast-furnace charging with ore and coke layers and peripheral fuel briquets

Kinetic studies of coke formation and removal on HP40 in cycled atmospheres at high temperatures

An austenitic FeNiCr alloy, HP40Nb, has been preoxidized and subsequently exposed to an alternating carburizing/oxidizing /carburizing atmosphere. During the oxidation at 1000°C a thick Cr 2 O 3 layer was formed which partly spalled off during cooling to room temperature, in this way chromium depleted areas resulted at the surface. The carburizing and reducing condition was established by a C 2 H 6 /C 2 H 4 /H 2 mixture at a temperature of 850 °C while the oxidation for decoking was conducted in air at 800°C. The exposure times were relatively short, respectively 90/30/180 minutes. During the first exposure of the preoxidized alloy to the carburizing atmosphere, coke formation took place, and underneath the coke layer the alloy was carburized, however, only locally. After the decoking in air at 800°C, during the second exposure to the carburizing atmosphere much more catalytic coke formation was observed compared to the first exposure. The coke formation was initiated by the reduction of (Fe,Ni,Cr)-spinels formed in the oxidizing atmosphere. The reduction of the oxides gives rise to the formation of (Fe,Ni)-particles which show strong catalytic activity towards coke formation.

Simulation of decoking of thermal cracking coils by steam/air‐mixtures

AbstractA rigorous approach was developed for the simulation of the decoking of an industrial cracking furnace. A one‐dimensional heterogeneous reactor model, which accounts for the interfacial gradients between the process gas and the coke surface, was used to simulate reactor coils. Both the combustion and steam gasification of the coke layer were taken into account. The reactor model for the decoking was coupled with a detailed firebox simulation model. The initial profile of the coke layer thickness, required for the decoking calculations, was obtained by a run‐length simulation. The evolution with time of the temperature distribution inside the cracking coil and in the furnace was generated simultaneously, which made it possible to understand the decoking operation in detail and to predict its duration accurately.

96/06414 Thermal stress and deformation of coke layer during carbonization

96/06414 Thermal stress and deformation of coke layer during carbonization

コークス生成過程における熱応力と塑性変形挙動解析

コークス生成過程における熱応力と塑性変形挙動解析

Characterization and Deactivation Studies of Spent Resid Catalyst from Ebullating Bed Service

Two series of bimodal CoMo/Al{sub 2}O{sub 3} catalyst samples have been analyzed and characterized using porosimetry, temperature-programmed oxidation and NMR spectroscopy. The samples are withdrawn after varying times on stream of up to 120 days. The results show that coke and metals are deposited rather rapidly and that the pore volume of the catalyst at equilibrium is almost filled with coke and/or metals. With increasing time on stream, the coke in both the first- and third-stage samples becomes more hydrogen and sulfur deficient as evidenced by increasing concentration of aromatic carbon and a lowering in concentration of S associated with the coke. There is evidence that the sulfur in the coke is associated with the upper layers of coke and that nitrogen may adsorb preferentially during the initial coke laydown. Furthermore, it is shown that vanadium catalyzes the oxidation of coke. Model compound activity measurements show that the HDS activity is less sensitive to the effects of coke deposition than the HDN and hydrogenation activities. It is also shown that the initial metals deposits have a stronger effect on the loss of activity for HDS than for HDN and hydrogenation, indicating that different sites are involved in these hydrotreating reactions.

Deactivation of a multimetal supported catalyst for aniline N-alkylation with alcohol

A multicomponent catalyst, containing platinum, tin and calcium on a silica gel support was tested for N-alkylation of MeEt-aniline (MEA) with methoxyisopropanol in a continuous fixed bed reactor at atmospheric pressure in the presence of hydrogen. It was found that the active Pt-Sn-Ca phase is located in the macro pores of the silica carrier; it has an average crystallite diameter of 55–165Åand a crystallinity degree of 30–90%, depending upon the activation-working conditions, and after reduction it contains Pt 3Sn alloy. Catalyst deactivation is caused by the formation of a strongly bonded coke layer containing nitrogen, which interacts with the Pt-component leading to the formation of a surface complex Pt...N with electron transfer from nitrogen to Pt. ‘ Pure ’ coke (without nitrogen) up to 4 wt.-% does not affect the catalyst activity/selectivity. The amount of N-containing coke depends upon MEA purity and operational conditions controlling the reaction selectivity. The strongly poisoning N-containing coke is produced by the red-colored oxidation-condensation products present in crude MEA and/or by a relatively high concentration of intermediate N-isopropylidenemethoxy-MeEt-aniline (IM) in the reactor. The IM concentration depends upon operational conditions and catalyst composition. Cleaning of crude MEA by distillation and adsorption of heavy impurities on silica and operating at proper conditions yielded low IM concentration with a stable yield of N-isopropylmethoxy -MeEt-aniline 63–65% and selectivity about 95% within a 200 h period of operation. The N-containing coke cannot be removed by treatment of the catalyst with nitrogen or hydrogen, but oxidative regeneration allows to reestablish the active phase with the same initial activity / selectivity patterns.

Determination of Gas Flow Pattern in a Test Coke Oven Charge and Its Time Variation by Tracer Technique.

The travel path of produced gas in a coke oven has always been of interest. However, few studies have been reported on this subject. In the present study, a tracer technique was employed for the determination of the path of gas in a test coke oven with width of 45 cm, and charge height of 85 cm. Three tubes were inserted horizontally from the coke side at 15.6 cm high (L) and at 0, 8 and 16 cm from the center of the width. Two kinds of tracer gas were uniformly introudced along the tubes. Other six tubes were also inserted at 36 (M) and 54 (H) cm high and at 0, 8 and 22 cm from the center, and the gas in the oven was withdrawn from their ends located at the center of the length of the oven. The gas sampling was conducted every two hours through the sampling tubes and from the exit ascension pipe, where the total flow rate of the produced gas was determined from the concentrations of tracers. Ar injected at L16 was hardly detected inside the charge (0 or 8 cm from the center) at the middle stage of carbonization. The flow rate of produced gas at the vicinity of the wall (22 cm from the center) was determined from mass balance of Ar assuming that all Ar flowed there. Rates of gas cross from the center side to the wall side was evaluated from the concentration profile of He introduced into the coal layer. At the middle stage, most of the produced gas passed upward in the coke layer and no cross flow was found. This phenomena were to be explained by large flow resistance of the plastic zone. At the last stage, both of the tracers were uniformly distributed all over the coke layer.

Flow Direction and Resistance of Produced Gas in a One-dimensional Carbonization Apparatus.

The travel path of the gases in the coke oven has always been of interest. However, few studies have been reported. In the present investigation, the time variations of flow direction and flow resistance of the produced gas in the coal and coke layers were measured in an experimental one-dimensional carbonization apparatus simulating a commercial coke oven.The coal packed in a quartz tube was heated from one side simulating the wall side of a commercial coke oven at a constant heating rate and then kept at constant temperature (1274 K). The other side corresponding to the center side of the commercial coke oven entered into the furnace lastly. The first series of runs were conducted by opening both sides of the tube. Most of the produced gas was evolved from the wall side except the last stage of t he carbonization. Small increase in the wall side pressure caused the drastic increase in the evolution rate from the center side at the first and latter middle stages. However, at the earlier middle stage of carbonization, most of the gas was evolved from the wall side even under this condition. In the second series of runs, one side of the tube was closed. The steep increase in the wall side pressure and the interruption of gas evolution were observed at the earlier middle stage of the carbonization, only when the wall side was closed. In the last series of runs, nitrogen was forcedly introduced through a thin layer from one side. Steep pressure increase was randomly observed when the layer was at around 773 K, irrespective of the flow direction of the nitrogen. In conclusion, at the earlier middle stage of the carbonization, a plastic zone at around 773 K existed between the main gas production zone and the end of the center side and the flow resistance of the zone was found to be large.

Coke deposits on H-ZSM-5 zeolite

Carbonaceous deposits (coke) formed on zeolites H-ZSM-5 and Linde L during methanol conversion at 460°C have been studied with transmission electron microscopy (TEM). At ca. 6% (w/w) of coke, deposits could just be detected, and at ca. 10% (w/w) the zeolite was enveloped in coke layers > 5 nm thick. Catalyst performance was not affected by coke deposits which were void of any structure in the transmission electron microscope (amorphous), but catalysts became deactivated when coke deposits contained packets of aromatic sheets of carbon parallel to the catalyst surface.

乾留過程におけるコークス層内熱応力の推算

Fissures in the coke layer have a great influence on the coke qualities, for instance, its strength and size, and the production cost. It is well-known that the thermal stress caused by the drastic contraction in the coke layer would be the major cause of fissures. But few analytical studies have been carried out on the thermal stress in the coke layer. We have then tried to estimate the thermal stress distribution in the coke layer, using the finite element method. We have applied an incremental theory to our FEM formulation. As a result, we have obtained the following results.1. Comparing with x-ray photographs taken through the coking process, we find a good correspondence between the fissure formation behavior on the photograph and the estimated principal tensile stress distribution.2. With the increase of the heating rate on the wall, we estimate the greater principal tensile stress in the coke layer, which indicates that higher heating rates lead to more and greater fissures in the coke layer.3. The estimated results of the thermal stress distribution depend on temperature-dependencies of thermal property, contraction coefficient and Young's modulus.

Millisecond pyrolysis – Mathematic model and parametric sensitivity

Ethane-water vapour pyrolysis and the influence of the system parameters on the pyrolysis furnace behaviour were studied using mathematical model describing the reactor constructed from Field's pipes. The parameters under discussion are following: reaction mixture feed rate, inlet pressure, heat transferrer temperature (furnace temperature), addition of overheated water vapour, thickness of coke layer on the reactor walls and radiation heat transfer inside the reactor. All parameters were found to be important. For feed rate, furnace temperature and ethane-water ratio an optimum values set can be evaluated. Coke formation on the reactor walls in case the coke layer is very thin can improve the heat transfer due to surface emissivity increasing. Otherwise the coke layer effect is negative.

A study on burden distribution of blast furnace with a bell‐less top

Abstract The so‐called coke layer collapse phenomenon was observed on a blast furnace top model which was 1/4‐scale of No. 2 blast furnace of China Steel. A simulation program based on the Bishop simplified method of soil mechanics and the material balance was developed to identify the collapse occurrence and to predict the burden distribution after the collapse. A close agreement between the experimental results and the prediction of simulation program was found.

Microwave Burden Sensor for Blast Furnaces

Measurement of the kinematic movement of burden materials is important for the stable and efficient operation of the blast furnaces. Without devising several types of burden sensor, few of them are put to practical use because of limitation of the proposed principles which are insufficient for measuring burden materials having complexity of chemical and physical change during the deoxidation process in furnaces and bad surroundings in addition. A microwave sensor is newly developed, which is able to use at almost every part in a furnace. This sensor can get simultaneously signals relating to the descent velocity, the thickness of ore and coke layer and the particle size of burden by use of the reflecting and scattering X-band wave. Experimental results at laboratory showed that these variables can be measured by the microwave burden sensor within the accuracy of ± 10%.Various types of microwave sensor have been installed in the shaft, belly and boshof several furnaces. Even though a sensor installed in the level of about 1.5 m above a tuyere is surrounded by hot burden of about 1500°C, it indicates stable signal for more than half a year. Interesting signals such as stagnation of descending burden in the lower level of a furnace, disorder of burden size and mixing of ore and coke in the neighbour of a furnace shell could be detected.

Coke formation on metal foils from heavy oils

The coke deposition on nickel and steel foils during pyrolysis of heavy oils has been studied. The amount of coke deposited on the foils was found to be dependent on foil type, reactor history, feed type and reaction conditions. On nickel, an asphaltene-containing feed gave more coke than an asphaltene-free feed. The coking on the metal foils increased in the order SS-302 < Inconel 718, Inconel X < Nickel. Introducing a rougher surface by rubbing the foils increased the coking. Scanning electron micrographs of the coked foils are presented and reveal details of two types of deposit. Type i is a catalytically formed deposit found only on nickel. The surface is covered in grains, crystalline in appearance. The grains are either irregular or needle-shaped, depending on the feed. The structure of the grains is not affected by the reaction conditions. Type ii is a deposited layer of coke, often containing droplets or mesophase spheres. The thickness and structure of the layer depends on the feed and the reaction conditions. This coke is formed in the liquid phase and deposited onto the surface. Surface roughness is important in determining the amount of coke deposited. A mechanism explaining these phenomena is suggested.

高炉内層頂部におけるコークス層崩れ現象の装入物分布に及ぼす影響

The phenomenon of coke layer-collapse occurs when ore is dumped onto the inclined coke bed. The coke layer-collapse causes to change the distribution of ore and coke layer thickness in the radial direction. Therefore, the effects of charging condition and distribution of gas flow and burden descending velocity on the coke layer-collapse have been studied by using a reduced-size experimental apparatus in which the conditions of gas flow and burden descent have been simulated to those of actual furnace top.The experimental results show that the magnitude of coke layer-collapse is mainly dominated by the charging mode, and the state of coke stacking in the central region after the collapse is also influenced by gas flow and descending velocity distribution. The changes in inclination angle of coke layer and coke movement accompanied by the collapse have been investigated. It is confirmed from these investigations, that the scale of coke layer-collapse depends on the stability of inclined coke layer. On the base of these findings, a mathematical model for quantitative evaluation of coke layer-collapse is developed by introducing the theory of soil mechanics. The calculated results by the model consist with experimental results for coke layer-collapse.

Investigation of coke and cohesive layer behaviors near the raceway using endoscope and ultrahigh speed camera.

The photographing system with an ultrahigh speed shutter camera has been developed at Kimitsu No. 3 blast furnace. In this system, the shutter speed can be raised up to about 1μs, and the actual raceway can be observed over a long time with high accuracy in on-line real time.The raceway images obtained with this system contain two sorts of the basic information, i.e., brightness and particle size of coke. By using these two informations, the two phenomena occurring near the raceway were elucidated; melting down behavior of the root of cohesive zone and change in coke properties. As a result, the mechanism for the periodical change of raceway brightness has been considered, and the degradation mechanism of coke in the furnace has been elucidated by analyzing the results of test operations using coke with various high temperature properties. Besides, this system has been used effectively in the blowing-out operation of blast furnace by lowering the burden level to the tuyere level.Based on the observation results obtained with this system, change in coke properties in the furnace has been investigated by using the model experiments and mathematical calculations. As a result, the following facts were guessed to be occurred.(1) Coke has already been deteriorated before running into the raceway.(2) Impact energy working on coke circulating in the raceway was not so great.(3) There was few cases where coke was degraded in the raceway by the thermal stress generated in coke particle when it came into the raceway.