The research work presented in this thesis focused on the feasibility of the Selective Catalytic Reduction of NOx with ammonia in forced unsteady state reactors. A reverse flow reactor (RFR) and, as a better alternative, a simulated moving bed (SMB) consisting in a network of three reactors with periodical variations of the feeding position have been explored by means of numerical simulations.
The study consisted in mathematical modeling and simulation, reactor design and experimental investigation.
The analysis begins with introspection in the modern modalities of treating high volume of diluted emissions of noxious gases. The attention was focused on the NOx emissions removal. The available technologies for NOx reduction have been suggested, highlighting the advantages, disadvantages and the possibility of their use in treating large volumes of waste gases.
Due to its perspective as highly effective for reducing NOx, from the point of view of economical efficiency, selectivity and yields predicted, the selective catalytic reduction has been the process employed for the analysis. In this respect a study of the available catalysts have been done resulting that the process performances should be investigated under the circumstances of using the vanadium based catalysts and metal-exchanged zeolites materials, using an Elay-Rideal mechanism in reaction description.
Because favorable temperature and composition distributions, which cannot be obtained in any steady-state regime, can be obtained by means of forced unsteady-state operations, the RFR and RN have been subjected to the study.
In order to gain information in a fast and easy way, a CBR application was initially implemented. Such information is related to the mode of forced unsteady state processes issue concerning the reactor design, the important system parameters and their values, the mathematical description of the process, the mathematical method of solving the system of partial differential equations etc.
The solutions suggested by the CBR system enabled forward reasoning about the way of dealing with reactors in forced unsteady-state operation and selective catalytic reduction of NOx with ammonia, taking into account the organization of the feature in the most similar retrieved cases. Nevertheless, the final decisions in approaching the selective catalytic reduction of NOx with ammonia in forced unsteady state reactors were not taken based just on the CBR suggestions. The solutions obtained in this way and their reliability has been tested, combining the information provided by the CBR system with a rigorous literature review concerning the topic of this thesis.
As a consequence, for these devices, first of all a sensitivity analysis has been applied in order to obtain qualitative and quantitative information about their complex behavior over a wide range of parameters.
First, isothermal operation has been investigated in order to focus on the consequence of trapping of one reactant on the catalyst surface, emphasizing in this way the complex interaction between the chemical reaction and the mass transfer processes.
The analysis assessed by means of simulations covered the influence of reaction kinetics, catalyst activity and switching time.
It was found that when adsorption and reaction rates are not very high, the RN is the only device which ensures the fulfillment on the emissions limits for the two reactants while when the adsorption and the reaction rates are very high not only the RN but also the RFR may fulfill the emissions limits because the problem of wash out is bypassed by the high adsorption and the reaction rate.
The switching time was identified as the most important variable, and its effect on overall reactors performances was studied. In isothermal conditions the sensitivity parameter analysis revealed the fact that there is a wide range of switching times where the RN exhibits almost no ammonia emissions and the NOx emissions are lower than those obtainable in the RFR. This effect is enabled not only by the absence of wash-out phenomena, as a consequence of single sense of gas flow circulation, but also by the most uniform exploitation of the catalyst length. Nevertheless, the extent of this “optimal”
range of switching times is a function of the value of system parameters. When the simulation have been performed with the parameters experimentally obtained by the Tronconi at al. [295], the influence of the switching time on the mean outlet concentration of NOx and NH3 was very different from that above presented one: it was found a maximum value of switching time beyond which conversion decreases both in the RFR and in the RN. With respect to the outlet emissions in the RFR two zones of high conversions, at low and high switching times, can be found, the performance of the RN with different switching strategy being similar. Very low NOx emissions are obtained at low switching times while for ammonia high switching times are required to decrease the emissions. If low emissions of both NOx and ammonia are required, a narrow range of switching times can be found, this range being a function of switching time strategy considered.
Finally, the non-isothermal conditions have been considered; in this case it was taken into account the complex influence of the dynamics of the heat wave on overall reactor performances. The analysis assessed by means of simulations covered the influence of the switching time, the feed flow rate, the initial catalyst temperature on the composition of product stream.
The range of switching times for which high performances are achieved differ significantly from the isothermal case; more important, in the RN just a narrow domain of
switching times enable auto-thermal operations. As far as the ammonia outlet concentration is concerned its value decreases when the value of the switching time was increased in the RFR, while in the various RN considered a minimum appeared. In addition, only the RN with the switching time strategy 1-2-33-1-2 allowed high performances regarding the NH3 emissions. As far as the emissions of NOx are concerned the RFR exhibits stable behavior in a wide range of the switching time. At higher values of the switching time the outlet concentration increases due to the lower temperature in the system which is a consequence of the heat removal form the catalyst. As expected, the RN has a different behavior: auto-thermal operation, with low NOx emissions is achieved at low value of the switching time and just in a narrower range.
Another important aspect highlighted by the different modes of operation is the shape of the temperature profiles obtained in the RFR and the RN. In the RFR the solid temperature profile has a typical bell shape and in the RN the profile is more uniform and tends to form a platform. The uniformity of the temperature profile in the RN is a function not only of the reactors number that forms the network but also a function of mode of operation (fast or slow switching operation). The more uniform temperature profile obtained in the RN enables a much more efficient exploitation of the total catalyst length with direct effect on increasing the reactant conversion. Even if the temperature level obtained in the RN is lower than in the RFR it is enough to sustain the chemical reaction. Also the presence both in the inlet and in the outlet section of the RN of almost the same temperature level as in the middle of the catalyst bed explains the performance of this device in achieving almost no emission of unconverted reactants and almost uniform concentrations in the outlet stream when the feeding position is changed. The uniformity of the temperature profile along the reactor length also allows an easier implementation of a control system.
When the influence of gas velocity was addressed it was found that when the superficial velocity changes it determine the modification of the range of switching time for which high reactant conversions are obtained. The decrease of the velocity determines an increase in the domain of switching times which enables high reactor performances.
When the gas velocity is increased the higher the inlet flow rate is the narrower is the range of switching times where auto-thermal operation with high conversion is obtained
both in the RFR and in the RN. The effect of the flow velocity is more important for the RN operation because applying an appropriate flow velocity the problems generated by the narrow range of switching times may be solved. As a recommendation, the gas residence time must be higher in the RN than in the RFR in order to enable a stable operation. This inconvenient is generated by the one way gas circulation in the RN.
When the influence of initial catalyst temperature was addressed the simulation results revealed the fact that once the condition for reaction ignition is fulfilled in the RFR the maximum temperature achieved is maintained at the same level. Also, the mean value outlet concentration of NOx and ammonia are maintained at almost the same level for any initial catalyst temperature that allows an auto-thermal operation. In the RN the maximum temperature achieved is a function of the initial temperature of the catalyst; it increases when initial temperature increases. As a consequence of this effect the mean value outlet concentration of NOx and ammonia in the RN decreases when the initial temperature of the catalyst increases. Even so, any initial catalyst temperature that allow for an auto-thermal operation gives higher conversions of both NOx and NH3. As a consequence the level of the initially catalyst temperature is important only as long it enables the ignition of the chemical reaction and it belongs to the catalyst temperature working domain.
The level of the conversions achieved, the more uniform temperature profiles, the uniformity of catalyst exploitation and the much simpler mode of operation impose the RN as a much more suitable device for the SCR of NOx with ammonia in usual operation and also in the perspective of control strategy implementation.
The investigation of systems response to disturbances in the feed composition evidenced the robustness of both forced unsteady state reactors and their ability to face against perturbations in the feeding for a long time interval without any control action.
Even so the control perspectives have been suggested and in this respect simplified models have been developed both for the RFR and for the RN.
In case of the RFR, the counter-current reactor has been studied as a simplified model. Simulation results revealed asymptotic thermal behavior of the CCR related to the RFR and almost the same concentration profiles. This analogy provides a simple basis for
short-cut calculations since the steady-state profile of a counter-current reactor can be computed much easier than the periodic steady-state of a reverse flow reactor.
In case of the RN, the fast switching model (FAS) generates asymptotes related to RN model simulation. This makes from the FAS model a feasible tool in the estimation of complex thermal behavior of the RN, in fast switching operation, especially in terms of calculating the maximum temperature level.
Experimental investigation of the process
The final analysis of this thesis was subjected to model validation. The model simulations enabled to evaluate and characterize the complex dynamic behavior of the RFR and the RN and in the same time to impose the RN as the more suitable device for the SCR of NOx with ammonia. Even if the simulations covered a wide range of possible scenarios a real behavior was imposed to be evaluated. The experimental investigation of RN was realized in order to determine the correctness of model predictions. The model validation was performed only in conditions of an isothermal system. New kinetic parameters have been obtained by applying transient experiments. The simulations performed with the new kinetic parameters revealed a good concordance between the mathematical model predictions and the experimental runs.