The gasification of coke by carbon dioxide or steam is one of the important reactions which take place in the blast furnace. But the rate equation, which may be applied to such an atmosphere of CO2-H2O-O-H2-N2-mixture as in the blast furnace, has not been reported yet. In order to determine the rate equation applicable to such gas mixtures, the gasification rates of pulverized metallurgical coke (24-35 mesh) by CO2-N2, CO2-CO, H2O-N2and H2O-H2systems were measured between 1000 and 1100°C. The experimental results led to the following expression, (Rate of gasification)=k1pCO2+k4pH2O/l+k2pCO2+k3pCO+k5pH2O where PCO2, PH2O and Pco were the partial pressures of CO2, H2O and CO respectively, and the constants were functions of temperature. Some workers have reported the retardation of gasification reaction by CO and H2, but hydrogen did not retard the reaction under our experimental conditions. The above equation may be available for the mathematical simulation of the blast furnace.
To analyze reaction in a blast furnace shaft and other reducing furnaces, a reduction rate of an are bed must be calculated. Recently many mathematical models to describe a kinetics of iron are reduction have been proposed byEl-MEHAIRY, W. M. MCKEWAN, K. Motu, W.O. PHILBROOK and others. The are bed was divided into several unit layers. The reduction of unit layer was divided into three steps, those were wustite, magnetite and hematite reductions. The mathematical model that was proposed by K. Mori with mixed controlling step by both chemical reaction and gas diffusion, was used for each reduction step. The calculation for the first unit time was done step by step until reaching to desired heap. Then the calculation for the second unit time was done in the same way using the degree of reduction thet had been calculated at first unit time, and so on. The calculated data was compared with the experimental data, and these data arrived at an essential agreement. The reduction by H2was controlled by interface reaction, but that by CO was controlled by both interface reaction and diffusion.
Some experiments on the rate of silica reduction by carbon saturated iron have been carried out in the range of 1550-1620°C and kinetics of silica reduction has been considered combining the present results with already published data by many investigators. The results obtained are as follows: 1. The reduction of silica in a graphite crucible is promoted by the increase of slag-graphite interface area, Al addition to metal and application of external electric potential. 2. Present result and the results already published by many investigators on the silica reduction have been put in order by considering the ratio γ=Ss-G/Ss-M-. In small value of γ, the rate of reaction is pro-portional to activity of silica in CaO-SiO2binary slag and apparent activation energy of silica reduction is about 90-100 kcal/mol. In this condition, silica reduction is controlled by chemical reaction in general but diffusion sometimes affect the rate according to the composition and transport properties such as viscosity of slag. On the contrary, in the range of γ>30 up to the present experiment diffusion step of silicate ion in slag is considered to be the limiting step.
An electrochemical approach to the kinetics of multicomponent reactions occurring at the molten slagmetal interface has been proposed and applied to kinetic study of silica reduction. 1. Polarization curves of several elements dissolved in iron were evaluated from the formation free energy of oxides and available electrochemical data. 2. The limiting current density for the diffusion of SiO2 in slag was estimated as nearly 150-400 rnA/cm2 from the experimental results. 3. The facts that the rate of silica reduction is accelerated with the increase of the slag-graphite interface area was clearly explained using the polarization curves. 4. The rate of silica reduction approaches to the maximum, that is, the limiting current, under the large cathodic polarization which could be realized by the increase of slag-graphite interface area, decrease of oxygen activity or applying the external electric potential. In this condition, the rate of silicd-(Si4+) diffusion in slag is the rate-limiting step. Such an electrochemical approach as proposed here might be much useful for the treatments of the multicomponent reaction kinetics in iron-steelmaking processes.
The cooling process of plain carbon steel plate was studied. A plate of 28 × 220 × 220 mms size was heated up to 930°C and then cooled by water spray from both sides; the temperature of cooling water was 38°C. The results obtained are as follows; 1. The heat transfer coefficient α' decreases gradually as the temperature of cooled surface θsis raised when water flow rate of spray W is large, while in the case whereWis small, the coefficient decreases abruptly with the elevation of θsover 300°C. 2. The relation between α′ and W is described as “α′∝Wn” in the case of W>5 × 10-4 1/cm.2rnin, and values of n are divided into two groups according to the boiling condition of cooling water, i. e.groups of value having 03-0.5 and0.5-0.8 3. The relation between the mean heat transfer coefficient and W is described as “α∝W0.65-0.75” in the case of W>5 × 10-41/cm2.min. 4. The relation between a and cooling water temperature θW(°C) is described as “α∝ [1-(5-8)×10-3θw]” in the case of W=0.060-%-0.062 1/crn2.min. 5. When W is large, the cooling rate V is hardly influenced by Os, but increases abruptly for 0, lower than about 500°C when W is small. Furthermore, V' decreases regardless of W, in such a temperature region that the transformation heat is evolved. 6. The relation between the mean cooling rate V- andWis described as “vcx 1470.-0.7” in the case of W=5 ×10-4-8 × 10-2 l/cm2-min. 7. The relation between V and θw(°C) is described as “V∝ [1 (4.5-5.5)× 10-3θw]” in the case of W=0.060-0.062l/cm2.min. 8. It is deduced that “spray cooling is a sort of turbulent flow cooling” from the quantitative relation between the water flow rate of spray W and heat transfer coefficient a, i. e.“α∝Wn” where values ofnis about0.5-0.8.
A heat of type 304H steel which shows the highest creep-rupture strength in the tests of our laboratory has considerably iigh creep-rupture strength at long times and seems to have even higher strength than that of type 316H steel. If such 403H steel is always available, the profit m ould be expected very much to reduce the construction cost of boiler plants. The difference of the strength in commercial heats can be explained by that of the electron microscopic structure, showing that the heat of higher strength has very fine and good distribution of M23C6type carbides and they have a resistance to coalesce to longer times-due to the presence of small amounts of titanium and niobium. An experimental heat of type 304 steel containing 0.16%C, 0-10%Ti and 015% Nb had much higher creep-rupture strength than that of type 316H steel and showed itself to be very much economical new stainless steel for high temperature use. And also it was found that at C/(Ti+Nb) atomic ratio of about 4 the maximum high temperature strength is obtained and precipitated carbides distribute themselves very finely and uniformly.