Abstract Due to the high-quality steel demand, especially for ultra-low Sulfur steel, RH desulfurization has been practiced. Based on this, mathematical and physical modeling have been applied to study steel desulfurization by reagent addition in the RH degasser vacuum chamber. The main result of cold modeling, using water and oil emulating steel and slag, respectively, was to assess the influence of density difference between the continuous and disperse phases on oil droplet behavior. It is shown that its flow tends to be more restricted near the down snorkel when the density difference increases. Moreover, these results provide the basis for CFD modeling validation, which enabled the determination of slag drop residence time inside steel on RH and the average value of the rate of dissipation of turbulent kinetic energy inside the RH ladle. These two parameters were used to develop a kinetic model, which reaches a good agreement with industrial trial results available in literature. The optimum desulfurization degree of 31.44% was achieved for a gas flow rate of 90 Nm3/h, according to the kinetic model. This value can be useful in some steel grade production, where the required S content is less than 10 ppm. Even in common steel grade production, if some punctual chemical adjustment is necessary, this technique is also useful. The main kinetic parameter for steel desulfurization is the steel circulation rate. For best results, it should be reduced in the desulfurization stage, and after that, the normal operation can be resumed, so that the production cycle is not affected.

Abstract In order to improve steel quality, it is necessary to increase the cleanliness of the liquid steel; that is, to reduce the number and size of the inclusions in the liquid steel, as well as to control their chemical composition. For this purpose, processes (with different operating costs) are used, such as the bubbling of inert gas in the steel ladle and its treatment in the vacuum degasser RH. This article deals with inclusion removal through an argon purging process and RH treatment, and provides a comparison of their effectiveness using industrial data and a mathematical model. The inclusion count shows a strong exponential decay for RH treated heats with chemical heating, as expected. The same behavior is not seen for RH treated heats without chemical heating as well as an argon purging process. It is suggested that in the later cases, there could exist competition between inclusion removal and inclusion generation from sources, such as refractory-slag-metal interaction and open eye atmosphere metal interaction. It has been concluded that the RH and the argon purging processes are both able to reduce the amount of inclusions. However, the bubbling process would require a much smaller gas flow rate to avoid open eye formation.

Informações digitais tornam possível um elevado grau de redundância das informações disponíveis para o ajuste de modelos preditores aplicados ao Mapeamento Digital de Solos (MDS). Entre esses modelos, a técnica de Árvores de Decisão (AD) tem aplicação crescente, em razão da sua potência no tratamento de grandes volumes de dados. Objetivou-se com este trabalho avaliar o impacto do volume de dados utilizados para gerar os modelos por AD, na qualidade dos mapas de solos gerados pela técnica de MDS. Uma área de estudo com 889,33 km² foi escolhida na região do Planalto Médio do Rio Grande do Sul. As relações solo-paisagem foram obtidas a partir de reambulação da área de estudo e delineamento das unidades de mapeamento em cartas topográficas de escala 1:50.000. Seis covariáveis preditoras ligadas aos fatores de formação do solo, relevo e organismos, juntamente com os conjuntos de dados de um, três, cinco, 10, 15, 20 e 25 % do volume total de dados, foram usadas para gerar os modelos preditivos por AD no programa WEKA. Neste estudo, densidades de amostragem menores do que 5 % resultaram em modelos com menor poder de capturar a complexidade da distribuição espacial do solo da área estudada. Amostragens entre cinco e 15 % conduziram a uma melhor relação entre o volume de dados a ser manipulado e a capacidade preditiva dos modelos gerados. Dados coletados no campo indicaram acurácia dos mapas preditos próxima a 70 %, para os modelos oriundos dessas densidades de amostragem.

Digital information generates the possibility of a high degree of redundancy in the data available for fitting predictive models used for Digital Soil Mapping (DSM). Among these models, the Decision Tree (DT) technique has been increasingly applied due to its capacity of dealing with large datasets. The purpose of this study was to evaluate the impact of the data volume used to generate the DT models on the quality of soil maps. An area of 889.33 km² was chosen in the Northern region of the State of Rio Grande do Sul. The soil-landscape relationship was obtained from reambulation of the studied area and the alignment of the units in the 1:50,000 scale topographic mapping. Six predictive covariates linked to the factors soil formation, relief and organisms, together with data sets of 1, 3, 5, 10, 15, 20 and 25 % of the total data volume, were used to generate the predictive DT models in the data mining program Waikato Environment for Knowledge Analysis (WEKA). In this study, sample densities below 5 % resulted in models with lower power of capturing the complexity of the spatial distribution of the soil in the study area. The relation between the data volume to be handled and the predictive capacity of the models was best for samples between 5 and 15 %. For the models based on these sample densities, the collected field data indicated an accuracy of predictive mapping close to 70 %.

Nesse trabalho, são apresentados os resultados de simulações bidimensionais da formação de dendritas durante a solidificação do níquel e do ferro puros, utilizando um modelo de campo de fase. A morfologia obtida mostrou-se qualitativamente semelhante à forma dendrítica, com braços primários e secundários bem desenvolvidos. Além disso, as velocidades de crescimento em regime permanente, calculadas no presente trabalho, apresentaram boa conformidade com os dados experimentais em uma grande faixa de super-resfriamentos. Foi observado que a espessura da interface apresenta um forte efeito sobre a morfologia e a velocidade de solidificação. Interfaces espessas reduzem o tempo computacional; entretanto podem gerar resultados inconsistentes. A espessura que apresentou os melhores resultados sem aumento acentuado do tempo de processamento foi 4x10-8m.

Numerical results for the solidification morphologies of pure Nickel and Iron, obtained by a two dimensional Phase Field model, conformed qualitatively well with the dendritic form, clearly depicting developed primary and secondary branches. Additionally, the steady state solidification velocity obtained in the present simulations fitted the experimental data over a wide range of undercoolings. It has been observed that the solid-liquid interface thickness has a strong effect on both the morphology and steady state solidification velocity. A thicker interface which might reduce computation time may produce inconsistent results. The best results were obtained using an interface 4x10-8m thick.

Resinas troca iônica base forte destacam-se como uma alternativa para reduzir o impacto do uso de CN- pela indústria do ouro. Embora a adsorção com resinas possa ser facilmente realizada, a eluição enfrenta alguns problemas, sendo a etapa-chave do processo. O objetivo desse trabalho é estudar o carregamento e a eluição das resinas Purolite A500® e Bayer AP247®, carregadas com cianocomplexo de ouro e cobre. As resinas foram carregadas em ensaios batelada a 25ºC e depois eluídas, em coluna, com soluções 1,0mol/L dos eluentes NaSCN e NaNO3. As soluções 1mol/L de íons nitrato de sódio não foram capazes de eluir o cianeto de ouro de resinas com matriz de poliestireno (Purolite A500®). Por outro lado, eluíram ambos os complexos da resina poliacrílica (Bayer AP247®). Tanto na resina Purolite A500®, quanto na resina Bayer AP247®, o eluente NaSCN aumenta a recuperação e a velocidade de eluição dos cianocomplexos metálicos, quando comparado com o eluente NaNO3.

Ion exchange resins can be applied in cyanide recycling to address growing environmental concerns about the use of cyanide during gold extraction. Although sorption with resins can be easily accomplished, elution is the key step and presents some problems. This work aims to study both loading and elution of two resins with different matrixes, Purolite A500® and Bayer AP247®, loaded with copper- and gold cyanocomplexes. Both resins were loaded at 25ºC (batchwise) and column elutions were performed at 50°C with either 1mol/L NaNO3 or NaSCN solutions. It has been found that 1mol/L nitrate solutions could not elute gold from polysterene resins(Purolite A500®). Conversely, both cyanocomplexes were removed from the Bayer AP247® resin. For both resins, NaSCN (1mol/L) solutions increased metals yield as well as elution rates of both complexes.

O objetivo desse trabalho é a utilização de resinas de troca-iônica, base forte, na adsorção de cianocomplexos metálicos, visando a sua reciclagem. São utilizadas soluções sintéticas que simulam diferentes efluentes da lixiviação do minério de ouro, contendo concentrações de ferro, de zinco e de cobre, variando entre 50 e 500mg/l. São realizados ensaios de adsorção, em batelada, utilizando dois tipos de resinas comerciais: Ionac SR-7 (matriz de poliestireno, tripropilamônio quaternário como grupo de troca) e Dowex 1X8 (matriz de poliestireno, trimetilamônio quaternário como grupo de troca). Em ambas resinas houve uma adsorção maior do cianocomplexo de zinco com relação aos outros cianocomplexos metálicos existentes em solução. A ordem do carregamento, nesse caso, foi Zn>Cu>Fe. A resina Dowex 1X8 apresentou um maior carregamento dos cianocomplexos de zinco e de cobre, quando comparados com os carregamentos da outra resina. As diferentes afinidades são discutidas em função das características físico-químicas dos complexos e das estruturas das resinas.

This work aims the recycling of cyanide applying strong base ion exchange resins for the adsorption of metallic cyanocomplexes. The experiments were carried out with synthetic solutions simulating those found in the cyanidation of gold ores. The solutions are made up of copper, iron and zinc with concentrations raging from 50mg/L to 500mg/L. Batch adsorption experiments were carried out with two commercial strong base resins: Ionac SR-7 (polystyrene, tripropylammonium) and Dowex 1X8 (gel type, polystyrene, trimethylammonium). The tripropilammonium resin Ionac SR-7 presents a larger loading of zinc cyanide compared with those of copper and iron. The affinity sequence observed was Zn>Cu>Fe. Dowex 1X8 resin shows the highest uptake of both zinc and copper due to its high exchange capacity. The affinity sequence is similar to that of Ionac SR-7 resin. The different affinities achieved are discussed based on both the physic-chemical features of the complexes and the resin structure.

Esse trabalho estuda a redução do MnO por banhos de ferro saturado em carbono, nas temperaturas de 1500°C, 1550°C e 1600°C. São determinados os valores da energia de ativação aparente para essa reação, através do método das velocidades iniciais, para teores de manganês iniciais no banho de 0%, 10% e 40%. Os valores encontrados foram, respectivamente, 64,8 kcal/mol, 63,13 kcal/mol e 63,25 kcal/mol .

The objective of this work was to investigate the effect of temperature on the velocity of MnO reduction by a carbon saturated liquid iron bath. The tests were performed in a laboratory apparatus specially designed to simulate smelting reduction reactions. The reactions were carried out respectively at 1500°C, 1550°C and 1600°C, and also the initial manganese concentration in the bath were 0%, 10% e 40%. The apparent activation energies were graphically determined for the overall reduction process through the initial velocity method. The overall reaction rate increases as the temperature increases, and the apparent activation energies were 64,8, 63,13 and 63,25 kcal/mol, respectively for 0%, 10% and 40% of initial manganese content in the carbon saturated iron bath.