Limits...
In Silico Determination of Gas Permeabilities by Non-Equilibrium Molecular Dynamics: CO2 and He through PIM-1.

Frentrup H, Hart KE, Colina CM, Müller EA - Membranes (Basel) (2015)

Bottom Line: We study the permeation dynamics of helium and carbon dioxide through an atomistically detailed model of a polymer of intrinsic microporosity, PIM-1, via non-equilibrium molecular dynamics (NEMD) simulations.This work presents the first explicit molecular modeling of gas permeation through a high free-volume polymer sample, and it demonstrates how permeability and solubility can be obtained coherently from a single simulation.Solubilities in particular can be obtained to a very high degree of confidence and within experimental inaccuracies.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK. hendrik.frentrup10@imperial.ac.uk.

ABSTRACT
We study the permeation dynamics of helium and carbon dioxide through an atomistically detailed model of a polymer of intrinsic microporosity, PIM-1, via non-equilibrium molecular dynamics (NEMD) simulations. This work presents the first explicit molecular modeling of gas permeation through a high free-volume polymer sample, and it demonstrates how permeability and solubility can be obtained coherently from a single simulation. Solubilities in particular can be obtained to a very high degree of confidence and within experimental inaccuracies. Furthermore, the simulations make it possible to obtain very specific information on the diffusion dynamics of penetrant molecules and yield detailed maps of gas occupancy, which are akin to a digital tomographic scan of the polymer network. In addition to determining permeability and solubility directly from NEMD simulations, the results shed light on the permeation mechanism of the penetrant gases, suggesting that the relative openness of the microporous topology promotes the anomalous diffusion of penetrant gases, which entails a deviation from the pore hopping mechanism usually observed in gas diffusion in polymers.

No MeSH data available.


Related in: MedlinePlus

The pore size distribution of the two-dimensional PIM-1 sample used here (PIM-1 (2D)), in a solid blue line, compared with the molecular simulations of [26] (dashed black line). Also shown are available pore size distributions of experimental PIM-1 samples using positron annihilation lifetime spectroscopy (PALS) of PIM-1 in the powder form in red open circles [52] and using the Horvath–Kawazoe (H-K) method applied to N2 adsorption at 77 K; H-K in orange open squares [50]. The PALS data are reported in percent intensity according to the right axis.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4384093&req=5

f3-membranes-05-00099: The pore size distribution of the two-dimensional PIM-1 sample used here (PIM-1 (2D)), in a solid blue line, compared with the molecular simulations of [26] (dashed black line). Also shown are available pore size distributions of experimental PIM-1 samples using positron annihilation lifetime spectroscopy (PALS) of PIM-1 in the powder form in red open circles [52] and using the Horvath–Kawazoe (H-K) method applied to N2 adsorption at 77 K; H-K in orange open squares [50]. The PALS data are reported in percent intensity according to the right axis.

Mentions: The distribution of pore sizes within the framework is arguably the most influential polymer sample characteristic of gas permeation. The pore size distribution (PSD) profile of the polymer slice model, PIM-1 (2D), and 3D simulation samples compared to available experimental data are shown in Figure 3. As one can see, both the simulations and experiments show that a significant amount of the pore size distribution exists below 10 Å, although with slightly different distribution profiles. For the PIM-1 (2D) and the previous PIM-1 (sims.), the frequency and distribution of pore sizes are in excellent agreement. When comparing between simulated model PSDs with experimental PSDs, one must be aware of the method and interpretational models used to generate the experimental data. Shown in Figure 3 are two available experimental PSDs of PIM-1, which were generated using the Horvath-Kawazoe (H-K) method [50] and positron annihilation lifetime spectroscopy (PALS) [52]. The H-K method uses a slit pore model applied to the N2 adsorption isotherm at 77 K, and the distribution profile agrees remarkably well with the PIM-1 simulation models. Although, the H-K method of PSDs is limited by both the size of a N2 molecule and the low pressure limit of the equipment, the smaller pore sizes shown in the simulations were not observed. The existence of pores below 6 Å is confirmed by the PALS experiment. However, these data were interpreted with a bimodal distribution, which was not observed in either the simulation models or the H-K PSDs. Although the comparison between these different means of generating a PSD is not perfect, it may be concluded that the PIM-1 (2D) model has a reasonable PSD compared to both previous simulations and experimental data. As such, the PIM-1 (2D) model should exhibit similar gas permeability characteristics as a similarly-sized element of a bulk PIM-1 membrane as a result of the these similar pore structure characteristics.


In Silico Determination of Gas Permeabilities by Non-Equilibrium Molecular Dynamics: CO2 and He through PIM-1.

Frentrup H, Hart KE, Colina CM, Müller EA - Membranes (Basel) (2015)

The pore size distribution of the two-dimensional PIM-1 sample used here (PIM-1 (2D)), in a solid blue line, compared with the molecular simulations of [26] (dashed black line). Also shown are available pore size distributions of experimental PIM-1 samples using positron annihilation lifetime spectroscopy (PALS) of PIM-1 in the powder form in red open circles [52] and using the Horvath–Kawazoe (H-K) method applied to N2 adsorption at 77 K; H-K in orange open squares [50]. The PALS data are reported in percent intensity according to the right axis.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4384093&req=5

f3-membranes-05-00099: The pore size distribution of the two-dimensional PIM-1 sample used here (PIM-1 (2D)), in a solid blue line, compared with the molecular simulations of [26] (dashed black line). Also shown are available pore size distributions of experimental PIM-1 samples using positron annihilation lifetime spectroscopy (PALS) of PIM-1 in the powder form in red open circles [52] and using the Horvath–Kawazoe (H-K) method applied to N2 adsorption at 77 K; H-K in orange open squares [50]. The PALS data are reported in percent intensity according to the right axis.
Mentions: The distribution of pore sizes within the framework is arguably the most influential polymer sample characteristic of gas permeation. The pore size distribution (PSD) profile of the polymer slice model, PIM-1 (2D), and 3D simulation samples compared to available experimental data are shown in Figure 3. As one can see, both the simulations and experiments show that a significant amount of the pore size distribution exists below 10 Å, although with slightly different distribution profiles. For the PIM-1 (2D) and the previous PIM-1 (sims.), the frequency and distribution of pore sizes are in excellent agreement. When comparing between simulated model PSDs with experimental PSDs, one must be aware of the method and interpretational models used to generate the experimental data. Shown in Figure 3 are two available experimental PSDs of PIM-1, which were generated using the Horvath-Kawazoe (H-K) method [50] and positron annihilation lifetime spectroscopy (PALS) [52]. The H-K method uses a slit pore model applied to the N2 adsorption isotherm at 77 K, and the distribution profile agrees remarkably well with the PIM-1 simulation models. Although, the H-K method of PSDs is limited by both the size of a N2 molecule and the low pressure limit of the equipment, the smaller pore sizes shown in the simulations were not observed. The existence of pores below 6 Å is confirmed by the PALS experiment. However, these data were interpreted with a bimodal distribution, which was not observed in either the simulation models or the H-K PSDs. Although the comparison between these different means of generating a PSD is not perfect, it may be concluded that the PIM-1 (2D) model has a reasonable PSD compared to both previous simulations and experimental data. As such, the PIM-1 (2D) model should exhibit similar gas permeability characteristics as a similarly-sized element of a bulk PIM-1 membrane as a result of the these similar pore structure characteristics.

Bottom Line: We study the permeation dynamics of helium and carbon dioxide through an atomistically detailed model of a polymer of intrinsic microporosity, PIM-1, via non-equilibrium molecular dynamics (NEMD) simulations.This work presents the first explicit molecular modeling of gas permeation through a high free-volume polymer sample, and it demonstrates how permeability and solubility can be obtained coherently from a single simulation.Solubilities in particular can be obtained to a very high degree of confidence and within experimental inaccuracies.

View Article: PubMed Central - PubMed

Affiliation: Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ, UK. hendrik.frentrup10@imperial.ac.uk.

ABSTRACT
We study the permeation dynamics of helium and carbon dioxide through an atomistically detailed model of a polymer of intrinsic microporosity, PIM-1, via non-equilibrium molecular dynamics (NEMD) simulations. This work presents the first explicit molecular modeling of gas permeation through a high free-volume polymer sample, and it demonstrates how permeability and solubility can be obtained coherently from a single simulation. Solubilities in particular can be obtained to a very high degree of confidence and within experimental inaccuracies. Furthermore, the simulations make it possible to obtain very specific information on the diffusion dynamics of penetrant molecules and yield detailed maps of gas occupancy, which are akin to a digital tomographic scan of the polymer network. In addition to determining permeability and solubility directly from NEMD simulations, the results shed light on the permeation mechanism of the penetrant gases, suggesting that the relative openness of the microporous topology promotes the anomalous diffusion of penetrant gases, which entails a deviation from the pore hopping mechanism usually observed in gas diffusion in polymers.

No MeSH data available.


Related in: MedlinePlus