چكيده به لاتين
During the last three decades, silicon nitride (Si3N4) has received considerable attention
due to its lightness, strength and toughness especially in aerospace, medical and automotive
applications. This ceramic material is mainly produced by the carbothermal silica reaction in
the nitrogen atmosphere at high temperatures, while its preparation through direct reaction
(reactive bonding) of Si and nitrogen has been controversial due to the effect of various factors.
In the present work, the effect of various factors such as nitrogen pressure, Si particle size,
optimum amount of Fe catalyst on Si3N4 formation kinetics was investigated with different
experiments and the influence of parameters was optimized using modeling. The new model
presented in this study has enabled the modification of previous models and industrial
production of this material through reactive bonding. Nitridation test was performed on the
specimens in the form of compact pellet of primary silicon powder. The samples were exposed
to nitrogen gas as powder and compact in the tube furnace under different experimental
conditions such as temperature, pressure, particle size and initial silicon composition. The rate
of reaction progression at different times was determined by weight change of the samples. As
the nitrogen pressure increased from 0.4 atm to 0.95, the kinetic curve deviation phenomenon
was observed in the pressed samples due to the high number of initial nuclei and faster closure
of the nitrogen penetration channels at higher pressures. In the case of the particle size
parameter, by decreasing the size of the initial silicon particle, the reaction surface exposed to
the reactant gas increased and the reaction rate increased. The mechanisms of formation of α
and β silicon nitride phases in the possible mechanisms were investigated and the free energy
of each mechanism was compared. Addition of iron powder to the initial silicon powder
resulted in the formation of the liquid phase of FeSi2 and this phase is a suitable route for
nitrogen penetration into the sample and increased the rate of reaction. The optimum amount
of iron was determined to be 0.8 wt%. The effective factors on the synthesis and kinetics of
silicon nitride reaction bonding were investigated and identified and the behavior of the
effective factors was expressed as mathematical models. . The basis of this model is based on
the theory of Sharp Interface Modeling (SIM). This model tried to eliminate the deficiencies
and shortcomings of the previous models and to have a higher comprehensiveness. In this
study, the mechanism of the effects of temperature, time, particle size, pressure, and iron
addition on high detail of silicon nitration kinetics is presented and modeled analytically. The
exprimental results were compared with the results of the model and the performance of the
model was confirmed by good agreement with the results of the expriments. A secondary model
was also developed using the GMDH method. This model is based on neural network and
enables the prediction and investigation of effective parameters of nitriding reaction. Using this
method, the influence and priority of the effectiveness of the reaction parameters on the
nitration kinetics were determined
