We have investigated CO2 adsorption on mixed metal oxides using a fixed bed reactor by measuring breakthrough time at constant pressure of 1 bar with premixed gas mixture (10% CO2 + 90% He by volume) at different temperatures. Many kinetic models have been described in literature for CO2 adsorption, including Virial model, Langmuir isotherm model, Freundlich isotherm, Toth isotherm, Bidisperse model, Diffusion model, Deactivation model, EOS (equation of state) model, LDF model, and Yoon-nelson model. Apparently, Yoon-nelson model is quite simple and best applied for liquid systems. Further, Pressure and CO2 volume fraction were fixed in our case and cannot be varied due to some constraints. Consequently, most of other models cannot be applied. We can only vary adsorption temperature and flow rate. Can anyone help us by suggesting a fairly simple, reliable kinetic model for our system with references or calculations? Thanks in advance.
I suggest that you can use any model you want as long as you appreciate the physical consequences of that model. For example, Langmuir has a single adsorption energy while Freundlich has a coverage dependent adsorption energy. I also think that tests exists to determine the confidence of one model over another. It could be as simple as comparing R^2 values from linear regression fitting, although I think more robust methods must exist too. Finally, I am vaguely aware of once reading a publication that applied a range of isotherm models to their data, discussed the implications of each to the system they had, and made a predication of confidence as to which model was the most realistic from the physical standpoint.
In summary, you should decide what model best suits your system. You should not pick a model just because it is "simple and reliable". You may find that you get "simple and reliable" results from it but that they make no sense to understand the true nature of your system.
Usually the investigation of adsorption and desorption kinetics is done by TPD experiments as well as DRIFTS studies in order to get an idea how many different adsorption sites there actually are. This can hardly done by just measuring the breakthrough curves, though you can validate your kinetic model with them as well. I guess you won't get around to construct your own kinetic model for your system as well as to conduct TPD experiments in order to elucidate the kinetic parameters.
Thank you Dr. Christoph Hahn for ansewr. We have done TPD and the result for a sample is shown in the file attached. The graph (data) is qualitative in the sense that we can only find the weak, medium and strong adsorption sites depending on appearence of peaks as a function of desorption temperature, and hence can conclude whether sorption is physical or chemical (of course, this is what we did). Even quantification of adsortpion sites in terms of mmol/g or mmol/m2 is tricky considering the universal nature of Thermal Conductivity Detector (TCD) of TPD system. I really don't know how this graph/data can be used to interprete kinetics in terms of rate constants, in spite of literature survey done to my level best. Could you kindly throw more light on this?
You need a calibration for your TCD curve or as a better way, you use IR spectroscopy for measuring the CO2 traces. With DRIFTS you can see which surface species are present (e.g. carbonates) and which species has the highest thermal stability. Based upon this you can formulate adsorption and desorption reactions (e.g. CO2(g) + * --> CO2*) for each surface species and further describe them with a kinetic model equation (e.g. Arrhenius-type rate expression). Than you just need a suitable reator model in order to balance your single species. If you went through this you "only" have to perform curve fitting to gain your kinetic parameters. That is just a rough description of how to construct a kinetic model for adsorption and desorption kinetics. Of course you should have a closer look in literature for this topic.
Dear Lakshminarayana
Langmuir, Freundlich, Toth and ... isotherms cannot describe kinetic behavior of a adsorption bed. These isotherms are used to estimate the equilibrium loading of adsorbates on adsorbent. So if you want to evaluate adsorption kinetics, you should use kinetic models like LDF (Linear Driving Force) model and I afraid for applying these models you should write material balance equations and solve resulted partial differential equations analytically or numerically.
I think this article could be helpful:
http://repository.um.edu.my/33636/1/A%20review%20modeling.PDF.pdf
There is no best model. It all depends on the criteria adopted. It may be that for two mathematically different functions, with the assumed measure of fit, we obtain a similar result. The calculations here do not concern a single point, but a range of changes in the adsorption capacity. In addition, it is important to be able to distinguish between adsorption isotherms and equations for the description of kinetic data.
- Badges
- Science topic
- Similar topics
- Econ
- Manufacturing
- Industry