| Author |
Message |
Stacey Stoffregen (Sstoffre)
Advanced Member
Username: Sstoffre
Post Number: 17
Registered: 2-2009
| | Posted on Wednesday, February 23, 2011 - 12:33 am: | |
Hi Diane and Katie,
Thanks for the interesting poster. I was wondering what you intend the composition of the gas mixtures to be in your future studies? Do you expect that the selectivity of gas adsorption will be largely governed by the size of the molecules that you choose to include in your mixture. For example, you mentioned that the LTA4A zeolites have large alpha cages and smaller beta cages, it seems that the selectivity in adsorption, at least at low pressures of the gas mixture, would be largely governed by molecular size.
Stacey |
James Mario Carbajal (Carbajalj)
Junior Member
Username: Carbajalj
Post Number: 3
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 9:16 am: | |
Hi,
You mentioned that aluminsilicate zeolites could be used for industrial applications such as reducing green house gas emissions. What other applications could adsorbing CO2 be used for?
- James |
Erin Dahlke (Erindahlke)
Intermediate Member
Username: Erindahlke
Post Number: 14
Registered: 2-2009
| | Posted on Wednesday, February 23, 2011 - 10:52 am: | |
Hi Diane and Katie,
Have you looked at snapshots of your simulations to see where the CO2 is absorbing relative to the positions on the Al?
Erin |
James Gorman (Jmgorman13)
Junior Member
Username: Jmgorman13
Post Number: 5
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 4:23 pm: | |
Diane and Katie,
You referenced Calero's paper on your poster; was his data experimentally found, and you are verifying that the computational models agree with experimental findings? I also was wondering how the CO2 absorption by zeolite compares to other compounds that absorb CO2.
James |
Joe Scanlon (Scanlonj)
Advanced Member
Username: Scanlonj
Post Number: 22
Registered: 2-2009
| | Posted on Wednesday, February 23, 2011 - 5:02 pm: | |
Diane and Katie,
Great poster, I enjoyed reading it. How much does the longer interaction cutoffs increase your simulation time? In future simulations, will you use the longer cutoffs or do you feel that they are not necessary. Was the purpose of looking at them in this project to merely figure out the disagreement with the published data? Thanks
Joe |
Shane Bernard (Smb8363)
Member
Username: Smb8363
Post Number: 9
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 8:08 pm: | |
Hi Diane and Katie,
I may be misunderstanding, but is the LTA4A able to incorporate more Al for Na substitutions than MFI? If so, why is this so, and if not, what determines the number of possible substitutions/sites of substitutions?
Thanks,
Shane |
Patrick Louden (Loudenpatrick)
Advanced Member
Username: Loudenpatrick
Post Number: 23
Registered: 2-2010
| | Posted on Wednesday, February 23, 2011 - 8:36 pm: | |
Hey Diane and Katie,
Great poster, it was an awesome read. I don't have a question but moreso just a comment. I did a presentation earlier this semester on monte carlo simulations of zeolites and in my readings I came across this paper, "Computer Simulations of the Energetics and Siting of n-Alkanes in Zeolites by Smit,Berend; Siepmann, J. Ilja. Journal of Physical Chemistry 1994, 98, 8442-8452" it introduces an interesting way to model long alkanes inside of a zeolite. I'm not sure if it would be that useful for a small molecule like CO2, but just thought I'd share : ]
Patrick
|
Diane Walters (Waltersd)
Member
Username: Waltersd
Post Number: 8
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:00 pm: | |
Hello Stacey,
We intend to look at mixtures of CO2 and N2 in future studies. One way that CO2 is unique from other small gases is that it has quadrupole. We hope to continue Dani’s previous work in investigating this characteristic by comparing it to N2, which has a very similar size and weight compared to CO2 but does not have a quadrupole.
Previous work in our research group has shown that CO2’s quadrupole and its interactions with the electrostatic field of the zeolite dictates its behavior, not its size or the amount of space in the zeolite. Still, size is still important because it can limit where molecules can adsorb, for example CO2 and N2 can only adsorb in the larger alpha cages of LTA4A and can’t access the smaller beta cages.
Thanks for your question!
-Diane and Katie |
Katie Deeg (Katie_d)
Member
Username: Katie_d
Post Number: 8
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:01 pm: | |
James C –
That’s a great question. The other main application of CO2 adsorption by zeolites is the separation of CO2 from methane. Zeolites can selectively adsorb CO2 over methane in a mixture of the gases, so this is useful for purifying natural gas to use as a fuel.
This is slightly related, but another interesting application related to CO2 is that zeolites can be used as an additive in the production of warm asphalt concrete - they decrease the temperature level during manufacture and laying of asphalt concrete, resulting in lower consumption of fossil fuels, thus releasing less carbon dioxide and other pollutants.
Hope that helps!
Diane and Katie |
Diane Walters (Waltersd)
Member
Username: Waltersd
Post Number: 9
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:01 pm: | |
Hello Erin,
Katie and I have been using VMD to look at some of our simulations with MFI. We’re trying to find patterns in the locations of the CO2 and sodium cations within the channels. So far we have found some interesting differences between where CO2 adsorbs in the sinusoidal channels (the channels running in the x direction) versus the straight channels (running in the y direction), but we’ll need to use a more quantitative approach using probability maps to really understand what we think we’ve been seeing. In short, that’s exactly what we’re working on now!
-Diane and Katie |
Katie Deeg (Katie_d)
Member
Username: Katie_d
Post Number: 9
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:03 pm: | |
James G,
In the paper that we referenced, Calero introduces parameters and shows that they allow her simulations to agree reasonably well experimental results. In our work, we’re comparing our simulation results with Calero’s simulation results. We want to use the parameters she gives, so we want to make sure that our code with her parameters gives the same results as her code with the same parameters; this indicates that our computational methods are accurate.
Although we do want to make sure that our simulation results agree reasonably well with experimental data so that we know our results are sensible, our focus is not reproducing experimental data as well as possible. Instead, we focus on achieving a model that satisfactorily produces realistic results, and then use this model to explore different adsorption behaviors.
In response to your second question, of the zeolites we’ve investigated, the ITQ-3 framework type (not discussed in this poster) exhibits the greatest level of adsorption of CO2 compared with MFI, LTA4A, and other zeolites. ITQ-3 adsorbs more CO2 than metal-organic frameworks, which are the type of CO2-adsorbing compounds other than zeolites that are most studied.
Thanks for your questions!
Diane and Katie |
Diane Walters (Waltersd)
Member
Username: Waltersd
Post Number: 10
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:03 pm: | |
Hello Joe,
We’re not sure exactly how much longer increasing the cutoff increases our simulation time because our run with a cutoff of 100 still hasn’t finished! Simulations take weeks longer with the longer cutoff. Clearly this is too long to be practical to use in our everyday work. For that reason we will use a cutoff of 25 in our future simulations since that gave us accurate enough results. The purpose of our work was to figure out the disagreement with Calero’s data, and by determining that the disagreement was caused only by too-small cutoffs we confirmed there was not a more fundamental problem in our methods. Now we believe our methods are accurate and will be reliable in exploring the behavior of CO2 in zeolites. Thank you for the questions!
-Diane and Katie |
Katie Deeg (Katie_d)
Member
Username: Katie_d
Post Number: 10
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:05 pm: | |
Hi Shane,
Thanks for your question. In reality, both LTA4A and MFI can have up to a 1:1 Al:Si ratio – half of the Si’s in the zeolite framework are substituted with Al. One mobile Na is then introduced for each aluminum substitution, to compensate for the change in charge. A higher ratio is not possible, since two silicons or two aluminums cannot be adjacent in the framework.
Right now, our code is able to model LTA4A with a 1:1 Al:Si ratio, where 96 of the 192 Si’s in a unit cell are substituted for Al. However, our code is able to model MFI with up to 4 Al substitutions per unit cell – out of 96 total Si’s in the unit cell. This is because our code originally dealt with all-silica zeolites (no Al substitutions), and we modified it to allow for Al substitutions. In MFI, the T12 and T12’ sites (marked in the figure in the Introduction section) have been shown to be the most likely sites for Al substitution to occur. There are 8 such sites in each unit cell, and up to 4 can be substituted (since aluminum atoms cannot be adjacent in the framework). With this known, it was easier to code for up to four Al substitutions in MFI. In the future we hope to modify our code to allow for more substitutions per unit cell in MFI; this will entail determining where the next most likely locations for Al substitutions are.
In general, substituting Si for Al can reduce the stability of the zeolite framework, and substitutions at some sites are more stable than substitutions at other sites. This is the case for MFI – the T12 and T12’ sites are known to be the most favored sites for substitution, since they’re the most stable. For LTA4A, by contrast, as far as we know, all sites are equally stable.
Diane and Katie |
Diane Walters (Waltersd)
Intermediate Member
Username: Waltersd
Post Number: 11
Registered: 2-2011
| | Posted on Wednesday, February 23, 2011 - 11:05 pm: | |
Hello Patrick,
Thanks, we hadn’t seen that paper before! You’re right that CO2 is too small for using a configuration bias to be practical. This method is very useful for long chains such as alkanes, but it isn’t necessary for a small molecule like CO2.
-Diane and Katie |
|