Research On Computer Aided Molecular Design Method Of Solvent In The Cooling Crystallization Process Of Organic Compounds

The traditional trial and error method selective crystallization solvent requires large amounts of manpower and resources,computer-aided molecular design(CAMD)provides a new idea for the selection of crystallization solvent.In this paper,the computer-aided molecular design of the solvents in the cooling crystallization process of organic compounds was carried out to form the corresponding CAMD problem.The CAMD method with high practicability was researched and developed to design and select the solvents with better performance for the cooling crystallization process of organic compounds.The main research contents are as follows:Firstly,the toxicity,melting point,boiling point,solubility of the solvent and potential recovery rate were determined as the target properties for designing the solvent of the organic compounds cooling crystallization process,and the corresponding property prediction models were selected based on literature research and analysis.Secondly,the CAMD method based on single objective optimization was studied.In this paper,the CAMD problem of the solvent in the cooling crystallization process of organic compounds was regarded as a single objective optimization problem,with the groups of molecules as variables,the melting point,boiling point,toxicity,solubility and molecular structure feasibility of the solvent as constraints,and the potential recovery as the objective function.The problem was further transformed into an integer nonlinear programming(INLP)model.A CAMD method based on enumeration method was proposed to solve the model to obtain candidate molecules that meet the design objectives.Based on this method,the CAMD problem of the solvent in the cooling crystallization process of 2-mercaptobenzothiazole(MBT)was solved,and the best solvent toluene was obtained within the given design range.Furthermore,the CAMD method based on multi-objective optimization was studied.In order to solve the CAMD problem of the solvent in the cooling crystallization process of organic compounds,a multi-objective optimization model was established with the toxicity,solubility parameter difference and potential recovery of solvent as three objectives and the feasibility of molecular structure as constraints.A CAMD method based on fast non dominated sorting genetic algorithm(NSGA-II)was proposed to solve the model to obtain Pareo optimal solution set,that is,a series of candidate solvent molecules.This method was applied to solve the CAMD problem of the solvent in the cooling crystallization process of sebacic acid.The results of the design were compared with the results of other studies using CAMD method and searching the database.It was found that this method had better applicability.Then,based on this method,the CAMD problem of the solvent in the cooling crystallization process of MBT was solved.It was found that the best solvent toluene selected by enumeration method for the cooling crystallization process of MBT was also included in the Pareto optimal solution set obtained by this method.Finally,in view of the problem of MBT cooling crystallization solvent design and selection,the reliability of the solvent solubility prediction method and the UNIFAC model were verified.The cooling crystallization experiment of MBT was carried out in toluene solvent to explore the effects of crystallization end point temperature,cooling rate and initial concentration of solute on the yield and purity of the product.It was found that when the cooling rate was 0.3K/min,the crystallization end point temperature was 313.15 K and the initial concentration of solute was 0.046 g/g,a product with a yield of 72.1% and a purity of 99.7% was obtained.