Optimization of the process of phenol alkylation by isoprene cyclodimer in the presence of KU-23 catalyst at a continuous operation plant

Determining the theoretical optimal conditions for the phenol alkylation process creates the basis for evaluating the prospects of this process as well. For the alkylation process used phenol and isoprene cyclodimers as feedstock. To determine the optimal reaction conditions for the catalytic alkylation of phenol with isoprene cyclodimers in a continuous pilot plant, the effects of temperature, molar ratio of initial reagents, and space velocity on the yield and selectivity of the target product were studied. The study was carried out in the temperature range of 80 – 1500C, the molar ratio of phenol to CDI was 0.5:1 – 2:1 and the space velocity was within 0.25 – 1.0 h-1. To develop a regression model of the process, it is necessary to identify the functional relationship between the process parameters and use it for further process prediction. Considering that the number of experiments is m=13, and the output variables are n=3, the functional relationship can be represented as a non-linear polynomial. To determine the coefficients of the equation, the S-pluse 2000 Professional program was used, which allows us to automatically calculate statistical analysis data: regression model coefficients and pair correlation coefficients, as well as quadratic effect coefficients. Applying Student's criterion, significant and insignificant coefficients of the equation were found. To test the adequacy of the model, the Fisher criterion was used, which makes it possible to prove the adequacy of the description of the response surface by regression equations. When comparing the found values of Fes of the criterion with the tabular ones at the chosen confidence probability of 95% and the numbers of degrees of freedom f1=6 and f2=2, it can be seen that the calculated values of Fes are less than Ft=19.3, and this indicates the adequacy of the description of the response surface by regression equations. Using the developed regression model on a PC, calculations were made to study the influence of each input factor on the output parameters. To solve the optimization problem, the Matlab-6.5 program was used, which contains modern algorithms for solving the linear programming problem. As a result, the solution of the optimization problem was found that at a temperature of 1400C, a ratio of components of 1:0.5 and a space velocity of 0.25 h-1, the selectivity value is 91%, while the yield of the target product reaches 69%.