Modelamento computacional das transformações de fase durante os ciclos térmicos de processamento de um aço inoxidável superdúplex
Carregando...
Arquivos
Citações na Scopus
Tipo de produção
Dissertação
Data
2021
Autores
Fiorante, Mariana Tortella Merli
Orientador
Magnabosco, R.
Periódico
Título da Revista
ISSN da Revista
Título de Volume
Citação
FIORANTE, Mariana Tortella Merli. Modelamento computacional das transformações de fase durante os ciclos térmicos de processamento de um aço inoxidável superdúplex. 2021. 123 f. Dissertação (Mestrado em Engenharia Mecânica) - Centro Universitário FEI, São Bernardo do Campo, 2021. Disponível em: https://doi.org/10.31414/EM.2021.D.131324.
Texto completo (DOI)
Palavras-chave
Aço inoxidável superdúplex,Conformação plástica dos metais,Transformação de fase,Modelamento computacional
Resumo
O presente projeto tem por objetivo realizar a simulação computacional da cinética de transformação e evolução das fases durante os ciclos térmicos de processamento de um aço inoxidável superdúplex, considerando as etapas de aquecimento, conformação plástica e
resfriamento, com uso do software DICTRA®. Os dados de entrada para as simulações foram composição química e tamanho das fases, temperatura pretendida de simulação, e taxas de aquecimento e resfriamento quando necessárias a descrição do ciclo térmico. Foram utilizadas as bases de dados termodinâmica TCFE9 e de mobilidade atômica MOBFE4, com intuito de se obter resultados para diferentes modelos, possibilitando encontrar o que melhor descreve a
cinética de transformação das fases. Foram simuladas diferentes taxas durante o aquecimento do material a partir de 950 °C, considerando condição microestrutural inicial de 50 % de ferrita [alfa] e 50 % austenita [gama], até 1250 °C, temperatura típica de conformação. No aquecimento, obteve-se fração máxima de 66,6% a à taxa de 0,30 °C/s, valor próximo aos 70% esperados pela simulação de equilíbrio em Thermo-Calc®. Após 1000 s de patamar a 1250 °C e resfriamento até a temperatura de solubilização, 1090 °C, à taxa de 0,30 °C/s, a fração de [alfa] reduziu para valores de 58,7%. Na sequência, também simulou-se diferentes taxas de resfriamento com ou sem a presença de patamar de solubilização. Considerando-se 3600 s de patamar a 1090 °C, foi possível recuperar a condição dúplex desejada, atingindo-se 55,5% [alfa], mas não atingindo os 50% esperados pelo equilíbrio, visto que ainda há gradiente de composição em [alfa] e [gama]nas simulações em DICTRA®. A fim de se manter a microestrutura dúplex anterior, realizou-se resfriamentos até 790 °C à taxa crítica de 3,0 °C/s, obtendo-se frações volumétrica de 56% [alfa], 43% [gama] e frações de sigma (s) iguais ou inferiores a 1%. Por outro lado, ao não se considerar o patamar à 1090 °C, isto é, promovendo o resfriamento até 790 °C à taxa de 3,0 °C/s a partir da condição de 58,7% [alfa], obteve-se frações volumétricas de 59,6% [alfa], 39,4% [gama] e 0,9% [sigma]. O DICTRA® mostrou-se inapto a simular a precipitação de nitretos de cromo (Cr2N) no resfriamento, ou por não haver supersaturação de nitrogênio (N) em [alfa] ou por ser incapaz de prever esta supersaturação. A partir dos resultados de cinética e evolução da fração volumétrica das fases no ciclo térmico de processamento do aço UNS S32750, foi possível obter o modelo computacional que melhor descreve o comportamento real do aço estudado
This project aims to perform the computer simulation of the transformation’s kinetics and phase evolution during the thermal processing cycles of a superduplex stainless steel, considering the stages of heating, hot working and cooling, using DICTRA® software. The input data for the simulations were chemical composition and phase size, desired simulation temperature, and heating and cooling rates when necessary to describe the thermal cycle. TCFE9 thermodynamic database and MOBFE4 atomic mobility database were used in order to obtain results for different models, determining the one that best describes the phase transformation kinetics. Different rates were simulated during the heating of the material from 950 ° C, considering the initial microstructural condition of 50% of ferrite [alpha] and 50% austenite [gamma], up to 1250 ° C, typical forming temperature. In heating, a maximum fraction of 66.6% [alpha] was obtained at a rate of 0.30 ° C/s, [alpha] value close to the 70% expected by the equilibrium simulation in Thermocalc®. After 1000 s of plateau at 1250 ° C and cooling to the solubilization temperature, 1090 ° C, at the rate of 0.30 ° C/s, the fraction of [alpha] reduced to values of 58.7%. In the sequence, different cooling rates were also simulated with or without the presence of solubilizations plateau. Considering 3600 s of plateau at 1090 ° C, it was possible to recover the desired duplex condition, reaching 55.5% [alpha], but not reaching the 50% expected by the equilibrium balance, since there is still a composition gradient in [alpha] and [gamma] by DICTRA® simulations. Seeking the maintenance of the duplex microstructure, a cooling was performed from 1090 °C to 790 ° C at a critical rate of 3.0 ° C/s, obtaining volumetric fractions of 56% [alpha], 43% [gamma] and sigma fractions equal to or less than 1%. If the plateau at 1090 ° C was not considered, that is, promoting cooling from 1250 °C, where the condition of 58.7% [alpha] was reached, to 790 ° C at the rate of 3.0 ° C/s volumetric fractions of 59.6% alpha], 39.4% [gamma] and 0.9% [sigma] were obtained. DICTRA® was unable to simulate the precipitation of chromium nitrides (Cr2N) during cooling, either because there was no nitrogen supersaturation (N) in [alpha] or because it was unable to predict this supersaturation. From the results of kinetics and evolution of the phases’ volumetric fraction obtained in the thermal cycle of steel processing UNS S32750, it was possible to obtain the computational model that best describes the real behavior of the studied steel
This project aims to perform the computer simulation of the transformation’s kinetics and phase evolution during the thermal processing cycles of a superduplex stainless steel, considering the stages of heating, hot working and cooling, using DICTRA® software. The input data for the simulations were chemical composition and phase size, desired simulation temperature, and heating and cooling rates when necessary to describe the thermal cycle. TCFE9 thermodynamic database and MOBFE4 atomic mobility database were used in order to obtain results for different models, determining the one that best describes the phase transformation kinetics. Different rates were simulated during the heating of the material from 950 ° C, considering the initial microstructural condition of 50% of ferrite [alpha] and 50% austenite [gamma], up to 1250 ° C, typical forming temperature. In heating, a maximum fraction of 66.6% [alpha] was obtained at a rate of 0.30 ° C/s, [alpha] value close to the 70% expected by the equilibrium simulation in Thermocalc®. After 1000 s of plateau at 1250 ° C and cooling to the solubilization temperature, 1090 ° C, at the rate of 0.30 ° C/s, the fraction of [alpha] reduced to values of 58.7%. In the sequence, different cooling rates were also simulated with or without the presence of solubilizations plateau. Considering 3600 s of plateau at 1090 ° C, it was possible to recover the desired duplex condition, reaching 55.5% [alpha], but not reaching the 50% expected by the equilibrium balance, since there is still a composition gradient in [alpha] and [gamma] by DICTRA® simulations. Seeking the maintenance of the duplex microstructure, a cooling was performed from 1090 °C to 790 ° C at a critical rate of 3.0 ° C/s, obtaining volumetric fractions of 56% [alpha], 43% [gamma] and sigma fractions equal to or less than 1%. If the plateau at 1090 ° C was not considered, that is, promoting cooling from 1250 °C, where the condition of 58.7% [alpha] was reached, to 790 ° C at the rate of 3.0 ° C/s volumetric fractions of 59.6% alpha], 39.4% [gamma] and 0.9% [sigma] were obtained. DICTRA® was unable to simulate the precipitation of chromium nitrides (Cr2N) during cooling, either because there was no nitrogen supersaturation (N) in [alpha] or because it was unable to predict this supersaturation. From the results of kinetics and evolution of the phases’ volumetric fraction obtained in the thermal cycle of steel processing UNS S32750, it was possible to obtain the computational model that best describes the real behavior of the studied steel