Name: | Description: | Size: | Format: | |
---|---|---|---|---|
3.98 MB | Adobe PDF |
Advisor(s)
Abstract(s)
A unidade de desparafinação por cristalização/filtração da Fábrica de Óleos Base da refinaria de Matosinhos tem como objetivo remover as parafinas contidas nos óleos base com recurso a um solvente. No entanto, alguns problemas relacionados com o ajuste de proporções dos componentes do solvente (metil-etil-cetona – MEK e tolueno) ou até o teor de água contido no mesmo deram origem a este projeto. Este trabalho teve como objetivos encontrar uma tecnologia alternativa para a recuperação de solvente isento de óleo desparafinado tanto quanto possível e que permitisse, a jusante da recuperação do óleo, diminuir o teor de água no solvente. A implementação de membranas poliméricas foi a alternativa escolhida para a realização deste estudo. Assim, optou-se por avaliar o desempenho de membranas de poliimidas com um refinado tipicamente tratado na UN 2300, o óleo 150 SN. Para este caso, o caudal que é alimentado ao sistema de recuperação do óleo desparafinado é de 98360 kg/h com uma composição em massa de 27,2% em óleo e 72,8% em solvente. Verificou-se, através de uma otimização numérica, que o fluxo de componentes através da membrana resultava num transporte seletivo, ou seja, a permeação do MEK era muito maior em relação à do tolueno e esta última muito maior relativamente à do óleo. Obteve-se uma rejeição do óleo de 98,8% o que se traduz numa recuperação de 42% do solvente inicialmente alimentado ao conjunto de membranas com pureza de cerca 97% distribuído por 63,1% em MEK e 34,1 % em tolueno. Esta corrente de solvente recuperado pode ser utilizada diretamente nas diluições efetuadas a montante da filtração e, uma vez que é rica em MEK, poderá permitir um maior ajuste de proporções no solvente consoante o óleo que se pretende tratar. Verificou-se também que seriam necessários 3 conjuntos de membranas com uma área total de 784 m2 . A corrente de retentado possui 58% do solvente alimentado sendo este mais rico em tolueno, com 48% (m/m). Realizou-se uma simulação em Aspen Plus do sistema de recuperação do óleo desparafinado cuja alimentação seria esta corrente de retentado. Observou-se que é possível obter uma corrente de óleo desparafinado puro. Assim, 99,1% do solvente alimentado a este sistema é recuperado na proporção de 32÷68 (m/m) em MEK/tolueno. O teor de água na corrente de solvente húmido recuperado foi reduzido de 25,4% para 15%, o que se traduz numa redução do caudal de vapor de retificação de 1000 para 135 kg/h. Verificaram-se ganhos energéticos relativamente ao consumo de energia da fornalha, tendo havido uma poupança de 4,4% em relação ao valor do consumo energético da fornalha do sistema atual, 4,35 Gcal/h. Realizou-se um estudo económico do projeto para se verificar a atratividade do mesmo. Avaliou-se a possibilidade de aquisição dos modelos PuraMem ® Selective/Performance/Flux e PuraMem ® S600 da marca Evonik. No caso do modelo PuraMem ® S600, cuja área/módulo é de 24 m2 seriam necessários 33 módulos, o que representaria um investimento total de 743 k€. Por outro lado, seriam precisos 25 módulos do modelo PuraMem ® Selective/Performance/Flux, em que a área/módulo é de 32 m2 . Neste caso, o investimento seria de 563k€. A margem bruta resultante da redução do consumo energético na fornalha e do caudal de vapor de retificação foram avaliadas em 225 k€/ano. Apenas o modelo PuraMem ® Selective/Performance/Flux apresenta atratividade de investimento apresentando os seguintes valores dos indicadores de viabilidade: VAL de 139k€, TIR de 22% e PR de 2,9 anos. Em suma, implementando uma bateria de membranas de poliimida poderá resultar, para além de todos os ganhos mencionados, num aumento de carga à unidade devido a uma maior taxa de diluição, ou seja, aumentando a diluição do óleo efetuada a montante dos filtros, irá permitir uma maior taxa de filtração com menor colmatação do meio filtrante. Ainda assim, não se descarta a possibilidade de adquirir uma coluna de separação para que possam haver ganhos maiores no ajuste de proporções do solvente.
The crystallization/filtration dewaxing unit of Matosinhos refinery aims to remove the paraffins contained in the base oils using a solvent. However, some problems related to the adjustment of proportions of solvent components (methyl ethyl ketone - MEK, and toluene) or even the water content of the solvent led to the realization of this project. The aim of this work was to find an alternative solvent recovery technology free from oil and to reduce the water content in the solvent downstream of the oil recovery. The implementation of polymer-based membranes was the chosen alternative for this study. The performance of polyimide membranes was studied with a typically treated base oil of UN 2300, 150 SN. In this case, the flow rate that is fed to the dewaxed oil recovery system is 98360 kg/h with a mass composition of 27.2% in oil and 72.8% in solvent. After numerical optimization, it was verified that the flow of components through the membrane resulted in a selective transport, that is, the permeation of the MEK was greater than the toluene and the latter one much greater than the oil. A 98.8% oil rejection resulted in a recovery of 42% of the solvent initially fed to the train of membranes with purity of about 97% distributed by 63.1% in MEK and 34.1% in toluene. This recovered solvent stream can be used directly in the dilutions made upstream of the filtration and, since it is rich in MEK, may allow a greater adjustment of proportions in the solvent depending on the oil to be treated. It was also found that 3 sets of membranes with a total area of 784 m2 would be required. The retentate stream has 58% of the solvent fed and is richer in toluene, with 48% (w/w). A simulation for the dewaxed oil recovery system was performed in Aspen Plus. It has been observed that a pure dewaxed oil stream can be obtained. Thus, 99.1% of the solvent fed to this system is recovered with the ratio of 32÷68 (w/w) in MEK/toluene. The water content in the recovered wet solvent stream was reduced from 25.4% to 15%, which results in a reduction of the rectifying steam flow rate from 1000 to 135 kg/hr. There was energy savings related to the energy consumption of the furnace, and that resulted in an economy of 4.4% in relation to the energy consumption of the furnace of the current system, 4.35 Gcal/h. An economic study of the project was carried out to verify the attractiveness of the project. The possibility of purchasing the PuraMem ® Selective / Performance / Flux and PuraMem ® S600 models from the Evonik brand was evaluated. In the case of the PuraMem ® S600 model, whose area / module is 24 m2 would require 33 modules, which would represent a total investment of 743k€. On the other hand, it would be required 25 modules of the PuraMem ® Selective / Performance / Flux model, where the area / module is 32 m2 . In this case the investment would be 563k€. The gross margin resulting from the reduction of the energy consumption in the furnace and the rectification steam flow was evaluated at 225k€ / year. Only the PuraMem ® Selective / Performance / Flux model shows investment attractiveness, with the following results of viability indicators; VAL of 139k€, IRR of 22% and PR of 2.9 years. In summary, by implementing a polyimide membrane battery, it may result, in addition to all the gains mentioned, in an increase in unit load due to a higher dilution rate, that is, by increasing the dilution of the oil carried out upstream of the filters, it will allow a higher rate of filtration with less fouling of the filter fabrics. Still, it is not ruled out the possibility of acquiring a separation column so that there may be greater gains in adjusting solvent proportions.
The crystallization/filtration dewaxing unit of Matosinhos refinery aims to remove the paraffins contained in the base oils using a solvent. However, some problems related to the adjustment of proportions of solvent components (methyl ethyl ketone - MEK, and toluene) or even the water content of the solvent led to the realization of this project. The aim of this work was to find an alternative solvent recovery technology free from oil and to reduce the water content in the solvent downstream of the oil recovery. The implementation of polymer-based membranes was the chosen alternative for this study. The performance of polyimide membranes was studied with a typically treated base oil of UN 2300, 150 SN. In this case, the flow rate that is fed to the dewaxed oil recovery system is 98360 kg/h with a mass composition of 27.2% in oil and 72.8% in solvent. After numerical optimization, it was verified that the flow of components through the membrane resulted in a selective transport, that is, the permeation of the MEK was greater than the toluene and the latter one much greater than the oil. A 98.8% oil rejection resulted in a recovery of 42% of the solvent initially fed to the train of membranes with purity of about 97% distributed by 63.1% in MEK and 34.1% in toluene. This recovered solvent stream can be used directly in the dilutions made upstream of the filtration and, since it is rich in MEK, may allow a greater adjustment of proportions in the solvent depending on the oil to be treated. It was also found that 3 sets of membranes with a total area of 784 m2 would be required. The retentate stream has 58% of the solvent fed and is richer in toluene, with 48% (w/w). A simulation for the dewaxed oil recovery system was performed in Aspen Plus. It has been observed that a pure dewaxed oil stream can be obtained. Thus, 99.1% of the solvent fed to this system is recovered with the ratio of 32÷68 (w/w) in MEK/toluene. The water content in the recovered wet solvent stream was reduced from 25.4% to 15%, which results in a reduction of the rectifying steam flow rate from 1000 to 135 kg/hr. There was energy savings related to the energy consumption of the furnace, and that resulted in an economy of 4.4% in relation to the energy consumption of the furnace of the current system, 4.35 Gcal/h. An economic study of the project was carried out to verify the attractiveness of the project. The possibility of purchasing the PuraMem ® Selective / Performance / Flux and PuraMem ® S600 models from the Evonik brand was evaluated. In the case of the PuraMem ® S600 model, whose area / module is 24 m2 would require 33 modules, which would represent a total investment of 743k€. On the other hand, it would be required 25 modules of the PuraMem ® Selective / Performance / Flux model, where the area / module is 32 m2 . In this case the investment would be 563k€. The gross margin resulting from the reduction of the energy consumption in the furnace and the rectification steam flow was evaluated at 225k€ / year. Only the PuraMem ® Selective / Performance / Flux model shows investment attractiveness, with the following results of viability indicators; VAL of 139k€, IRR of 22% and PR of 2.9 years. In summary, by implementing a polyimide membrane battery, it may result, in addition to all the gains mentioned, in an increase in unit load due to a higher dilution rate, that is, by increasing the dilution of the oil carried out upstream of the filters, it will allow a higher rate of filtration with less fouling of the filter fabrics. Still, it is not ruled out the possibility of acquiring a separation column so that there may be greater gains in adjusting solvent proportions.
Description
Keywords
Óleos base Membranas Poliimida Desparafinação OSN Base oil Membranes Polyimide Dewaxing