Modeling critical properties of organic materials

Dissertation

To obtain a master's degree

in the field of chemical engineering

Abstract

Critical characteristics for materials such as critical temperature, critical volume and critical pressure They are important characteristics for predicting many thermodynamic properties of different materials. Knowing the critical properties plays an essential role in all hydrocarbon production and process operations. Because these operations take place in conditions very close to the areas of dew points and bubbles and are often associated with the same temperature or same pressure phenomena.

Until now, various methods have been presented to estimate the critical properties of organic materials, the basis of which are different from each other.

In this research, with 7000 critical characteristics of organic materials, new semi-empirical models for critical properties have been presented. is In the following, models derived from artificial intelligence, i.e. artificial neural network and neural-fuzzy adaptive inference system, are presented.

In the semi-empirical models presented, we can estimate the critical properties by having the normal boiling point and molecular mass of the substance (critical temperature inputs), the number of atoms and molecular mass of the substance (critical volume inputs), and the critical temperature and critical volume of the substance (critical pressure inputs).

The suggested models have little error while being simple. Other characteristics of the models include the generality of the equations and the availability of the input parameters.

At the end of the research, by comparing the proposed models with the models derived from artificial intelligence as well as 4 semi-empirical relationships, it is determined that the proposed models have good accuracy for estimating the critical properties of materials.

The average relative error of the final model for critical temperature, critical volume, and critical pressure is respectively 3/86, 5/06 and 5/57, which indicates the sufficient accuracy of the models.

Key words:

critical properties, semi-empirical models, artificial neural network

Chapter One

Critical characteristics: theoretical issues

- Introduction

The correct prediction of critical properties is very important in determining the fuzzy properties of systems. The critical state is the only condition that identifies liquid and vapor phases and is equally important in theory and practice. Knowing the critical conditions plays an essential role in all hydrocarbon production and process operations. Because these operations take place in conditions very close to the areas of dew points and bubbles and are often accompanied by the phenomena of equal temperature or equal back pressure [1] [1]. Predicting fluid properties and designing calculations in this area is very difficult, and knowing the critical point helps us solve this problem.

From a theoretical point of view, many thermodynamic properties are determined using critical properties, and from a practical point of view, many experimental relationships are based on these properties of the studied systems.

Although having critical properties is very important in theory and practice, but determining These properties are very difficult to test, and from an economic point of view, the laboratory method for determining critical properties is not suitable] 2. [

Critical characteristics for pure materials such as critical pressure, critical temperature, and critical volume are important characteristics for predicting the phase behavior of materials. Also, these characteristics are measured to estimate the equilibrium between gas and liquid phase (VLE)) along with the equation of state. For example, these specifications are important parameters for the gas, oil, and petrochemical industry, and they are also necessary to describe the process of undefined petroleum components. 1-1-1- The purpose of the research, considering the importance of the critical characteristics of pure materials including critical temperature, critical pressure, and critical volume in the industry, especially in the oil and petrochemical industry, and also considering that measuring these parameters by laboratory is a difficult task.

1-1-1- The purpose of the research

Given the importance of the critical characteristics of pure materials including critical temperature, critical pressure and critical volume in the industry, especially in the oil and petrochemical industry, and also considering that measuring these parameters by a laboratory is a difficult and costly task, so we decided to provide different models to estimate the critical characteristics of pure materials. Also, due to the small volume of previous works in providing comprehensive models, this research has done modeling based on 7000 organic substances. Despite the generality of the models, the existing models have simplicity and little error.

1-2- History

The first methods of finding critical properties were experimental, which were used in the case of hydrocarbon systems. Among these methods, Katz[2] and Karata[3] are better known in 1942] 3. Also, the mathematical conditions of the critical point were first presented by Gibbs, and later in 1954, corrections were made by Defay[4] and Prigagin[5] and in 1977 by Reed[6] and Beagle[7] 4-5.[

Physical characteristics of pure substances have been measured during recent years. These characteristics include specific density, normal boiling point, molecular mass, critical characteristics, and eccentricity coefficient.

To estimate the critical characteristics of materials that have a simple molecular group, methods such as Jobek[8], Pitzer[9], Ledersen[10], Fedors[11], Klinsevich[12] and Halm[13] are the most important methods. One of the most important features of these methods is that it is possible to estimate the critical characteristics without using significant computing resources. Among the above methods, Jobek's method is easier and has less error] 6-11. [

Many properties of pure materials have been measured and collected over many years. These properties provide basic information to study the volumetric behavior and determine the thermodynamic properties of pure substances and their mixtures. The most important of these properties are:

Critical pressure, pc

Critical temperature, Tc

Critical volume, Vc

Critical compressibility coefficient, Zc

Excentric coefficient, ?

Molecular weight, MW.

Usually, the properties of hydrocarbon mixtures are more important to petroleum engineers than the properties of pure substances. Of course, these specific constants of pure materials can be used with independent variables [14] such as pressure, temperature and composition to determine and define the physical properties and phase behavior of mixtures [12]. rtl;"> There are many relationships to estimate the critical properties of pure materials. Most of these relationships use relative density (?) and boiling point temperature (Tb) as relationship parameters [15]. Choosing the right values ??for the above parameters is very important, because small changes in these parameters can cause a significant deviation in the expected results. Some of these relations are presented below:

1-3-1- Kaut's relations [16]

Kaut (1962) expressed relations to estimate the critical pressure and temperature of materials. These relationships have found wide acceptance in the oil industry, the reason for which is the ability to extrapolate in situations where the information used is outside the scope of other relationships. The proposed relationship is expressed analytically as functions of normal boiling point and API density. Kaut proposed the following models to estimate the temperature, pressure, and critical volume of materials: style="direction: rtl;">Log(Pc) = b0 + b1 (Tb) + b2(Tb)2 + b3 (API) (Tb) + b4 (Tb)3 + b5 (API) (Tb)2 + b6 (API)2 (Tb) + b7 (API)2 (Tb)2

Abstract

Critical properties for the pure materials such as critical temperature, critical pressure and critical volume are important properties to predict many different thermodynamic properties. Also critical properties play an important role in unit operations and chemical engineering processes. Several methods have been proposed to estimate the critical properties of organic materials.