3/s and 800 cm 3 /s as well as bath heights ranging from 106 to 314 mm. The mixing times were also calculated based on an expression involving the Strouhal and Reynolds numbers. The experimentally determined mixing times were found to be within Ϯ20 % of the theoretical values, which is considered to be good in physical modelling. Overall, the mixing time was found to be influenced by the gas flow rate and the vessel diameter, but not by the bath height.KEY WORDS: physical modeling; mixing time; converter; side-blown; conductivity measurements. 663© 2010 ISIJ used to simulate steel and air to simulate argon/oxygen. The gas flow rates were determined by using the Froude number to scale the rate from a full scale converter to a physical model.The different flow rates were calculated by using the Froude number similarity criteria: (2) where r g is the density of air, Q g is the gas flow rate, r l is the density of water, g is the gravitational constant and d i is the nozzle inner diameter. In addition the following Froude number was used to scale the gas flow rate: Equation (3) was used to scale the gas flow rate, compared to a real converter. When this is determined, Eq. (2) was used to scale the size of the nozzle. The size of the nozzle was determined to be 3.2 mm for the smaller physical model. From experience it was determined that the nozzle size has a very small influence on the mixing time in the same manner as observed in a bottom blown bath. The mixing time is mainly governed by the gas flow rate, although the penetration depth of bubbles in the horizontal direction is affected by the nozzle size as well as the gas flow rate. Thus, the same nozzle was used in the larger physical model.A mass flow controller was used to regulate the gas flow rate during the experiments. Also, a conductivity probe was used to measure the change in conductivity. The probe was placed 30 mm from the bottom of the bath and near the side wall, as illustrated in Fig. 1. The output of the probe was fed to a conductivity meter and provided a 0 to 1 V DC output to a computer interface. In addition, the fluid flow was fully developed before the tracer injection of the KCl-solution was made. The solution had a concentration of 1 M KCl, and the amount added was 1 mL solution per 1 000 mL of liquid in the bath. Furthermore, the injection was made by a syringe close to the wall of the vessel. The conductivity of the bath was measured with an interval of 100 ms.The experimental mixing time was defined as the period required for instantaneous tracer concentration to settle within Ϯ5 % deviation around the final tracer concentration in the bath. Results and DiscussionExperiments were done using two vessel diameters of 200 mm and 300 mm, respectively. In addition, experiments were done for bath heights varying from 106 to 314 mm.Air was injected at the bottom of a side wall, as illustrated in Fig. 1. Flow rates between 30 cm 3 /s and 800 cm 3 /s were tested during the trials. Based on Eq. (2) the inner diameter of the side nozzle was determi...

There is no known example of an AOD converter with oblong cross sections in the literature. Changing the geometry of the converter vessel, from the traditional circular cross sections, to increase converter volume could potentially influence the performance of the converter and in particular the decarburisation rate. To study the feasibility of implementing an oblong converter, physical modelling was used to study the fluid flow of the proposed converter configuration, geometry and number of tuyeres, and the potential influence on the decarburisation rate. Two water models were employed using water containing NaOH and gas injected through six or eight tuyeres as fluids. In the model, CO 2 gas was injected and the reaction of CO 2 and NaOH was indirectly measured by detecting the pH value of the water. The mixing time is considered to be a good indicator of the decarburisation as kinetics will be diffusion controlled in the latter period of the process. The following three configurations were studied: (i) a circular converter with six tuyeres, (ii) an oblong converter with six tuyeres, and (iii) an oblong converter with eight tuyeres. The mixing time can be used to evaluate the different converter configurations. The average CO 2 concentrations based on several experiments, differed by less than 5% between the circular and oblong models after 165 s of injection of air and CO 2 . The results also showed that no difference in mixing time could be found when using 6 and 8 tuyeres, respectively in the oblong model, where the CO 2 concentrations differed by less than 2% after 165 s of injection time of air and CO 2 . Based on the findings, it has been observed that the influence of converter geometry on mixing time is small, it was concluded that decarburisation rate is likely to be the same irrespectively of converter geometry.

A 1:4.6 scale physical model of a production argon oxygen decarburisation (AOD) converter was used to study the influence of top slag on the AOD process. Specifically, the gas penetration depth, fluid flow and slag behaviour under different nozzle diameters, nozzle numbers and gas flow rates were studied. The results show that the relative gas penetration depth generally increases linearly with an increased gas flow rate and a decreased nozzle size. Furthermore, the slag thickness increases linearly with an increased gas flow rate. In addition, the open-eye size was found to increase exponentially with an increased gas flow rate. Overall, three kinds of fluid flow patterns were found in the experiments: (i) a counter-clockwise rotation, (ii) a clockwise rotation and (iii) a double circulation with the plume in the middle of the converter. A counter-clockwise rotation was most common for the current experimental conditions.

Argon Oxygen Decarburization (AOD) converter slags are known to consist of both liquid and solid phases, but limited information on the slag characteristics has been published in the open literature. Therefore, a new methodology to study the characteristics of slag samples taken from the AOD converter process during production was developed based on petrography. The results show that the preparations of the slag samples using the borax method are suitable to use when determining the chemical composition of AOD slag samples using the X-ray fluorescence (XRF) method. The results also showed that both the light optical microscopy (LOM) method and a method combining scanning electron microscopy (SEM) with energy dispersive spectroscopy (EDS) can be used to characterize the slag samples and that the correlation between the methods was found to be good. This means that it is possible to use the faster LOM method instead of the more complicated SEM-EDS method to characterize AOD slag samples. Finally, the results show that the difference between calculated values based on stoichiometry and measured data for Ca and Cr in AOD slags are 11.7 mass% and 11.3 mass%, respectively. This is considered to be a good agreement for industrial samples.