Study On The Model Of Covalent Bonds' Cleavage And Coupling In Reaction Of Volatiles During Coal Pyrolysis

The pyrolysis is the generic and basic process during the thermos conversion of coal, thus the experiments and simulations on pyrolysis are paid much attention by researchers. The reaction of volatiles, which is the main factor that influences the quality of coal tar, is an important part in pyrolysis. It noted that, however, the mechanism of volatiles is still unclear, and the reaction of volatliles should be paid much attention.The researches on pyrolysis simulation are important parts among researches on coal pyrolysis. There were many models of reaction network on pyrolysis of coal or biomass representd in past 50 years. These models were all based on the experimental facts and therefore represented some characters of reactions during pyrolysis and may simulate volatile reactions to some extent. However, the models were lump and confused because they were built based on macro structure of reactant such as coal and biomass, and they therefore were not universal enough. Furthermore, one can not acquire sufficient information and mechanism of reactions of volatiles.As develop of computational chemistry and the computing power of computers, the simulations of properties of macromolecule clusters were feasible, among which the dynamic simulations were developed rapidly in recent years. The reactive force field simulations, such as ReaxFF, were paid much attention as they can simulate the dynamic processes such as pyrolysis of coal or volatile reactions. But unfortunately, because the scale of these simulations was too small, they can only give the results of high temperature and short residence time. That is to say, the comptation chemistry cannot tell the full story of a true pyrolysis also.Based on the informations of the two fields mentioned above, the most important premise of building a good model is to find a proper scale, which is neither too large nor too small, and therefore one can present a simulation model can both give sufficient chemical informations that ensure the scalable and have the enough speed to simulate the actual pyrolysis process.Although the chemical structure of fuel such as coal and biomass are very complex, but from covalent bond point of view, the sorts of bonds in fuel are simple, that is, there are tens kind of bonds in fuel. That is to say if we build a simulation model aimed at covalent bond in fuel, the information and speed may be enough to cover the need. Futhermore, the proper inversion may give some more information of pyrolysis products which can do some predictions.As the general understanding, the pyrolysis of coal can be divided into two steps, dissociation or cracking of covalent bonds to generate free radical fragments ("dissociation" step) and reaction or combination of the free radical fragments to form products ("coupling" step). The two steps may be the base of the model on simulation of covalent bonds.In pyrolysis processes, the bond dissociation energy might be affected by the long life radicals. The number and molecular weight of radicals might affect the probabilities of coupling step. To make improvements of the model or even judge the importance of "dissociation" and "coupling" steps, the second assumptions based on chemical principle is needed.In order to obtain the reliable conclusions, we conducted many relevant researches and the major conclusions obtained are as follows:1 The technologies of pyrolysis are various and have a various tar yield. The factors which influence the qualities and quantities of tars obtained by these technologies seem to be much on the surface, but from pyrolysis mechanism, the temerpature and residence time of volatiles generated from feed is crucial. By calculating the temperatures and residence times of volatiles generated from techno;ogies such as coking, fixed bed reactor, moving bed reactor, fluidized bed reactor, solid heat carrier pyrolysis, rotary kiln, tube furnace pyrolysis and Gray-King assay, we combained the tar yield and environments volatile suffered. It is found that the temperature increases of volatiles in reactors influence the qualities and quantities more and reduce the temperature increases may be effective.2 Build the model of bonds in fuel pyrolysis based on "dissociation-coupling" mechanism. For "dissociation" step, we give the formulation of the fraction of bond which dissociate in a relaxation time. For "coupling" step, we assump it follows Markov process and can be simulated by Monte Carlo algorithm. Thus the Boltzmann-Monte Carlo model is built to simulate the evolutionary trend of covalent bonds in pyrolysis.3 Based on pyrolysis mechanism of hydrocarbons and statisitical mechanics, some proper second assumptions are introduced that make the simulation results more realistic. In "dissociation" step, the mechanisms of formation of unsaturated bond such as Car=Car bond and C=O bond are introduced. Furthermore, the barriers of these reactions are calculated by thermodynamic research results. In "coupling" step, the collision frequencies of different radicals are calculated by Maxwell-Boltzmann formulation. The results of models using different second assumptions are then compared to the experimental data. It is found that the second assumptions of the two steps are both useful and make the results agree the experimental data more. It is also found that the second assumptions in "dissociation" step are more important.4 The coordination numbers of atoms involved in pyrolysis are introduced to Boltzmann-Monte Carlo model. By combaining the bond structure during pyrolysis and percolation formulation, we can give more information. Beyond the bond populations, we can obtain the char yield in different simulation cycles and temperatures by inputing bonding structure of volatiles to program. The trend obtained by simulation reaches a good agreement of experimental data. The valid and scalabitlity of Boltzmann-Monte Carlo model is proved from this point of view.