2023
DOI: 10.3389/fchem.2023.1100210
|View full text |Cite
|
Sign up to set email alerts
|

Storage and diffusion of CO2 in covalent organic frameworks—A neural network-based molecular dynamics simulation approach

Abstract: As a consequence of the accelerated climate change, solutions to capture, store and potentially activate carbon dioxide received increased interest in recent years. Herein, it is demonstrated, that the neural network potential ANI-2x is able to describe nanoporous organic materials at approx. density functional theory accuracy and force field cost, using the example of the recently published two- and three-dimensional covalent organic frameworks HEX-COF1 and 3D-HNU5 and their interaction with CO2 guest molecul… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
6
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
5

Relationship

3
2

Authors

Journals

citations
Cited by 6 publications
(8 citation statements)
references
References 44 publications
2
6
0
Order By: Relevance
“…For all temperatures, a notable increase of the diffusion is observed when increasing the amount of loading from one to four CO 2 molecules. A similar behavior was reported in a study of CO 2 diffusion in the covalent organic frameworks HEX-COF1 and 3D-HNU5 111 , as well as for other small guest molecules embedded in a host matrix, e.g. CH 4 and H 2 in zeolite ZK4 112 , 113 and CH 4 , CF 4 , He, Ne, Ar, Xe, and SF 6 in silicalite.…”
Section: Resultssupporting
confidence: 84%
See 1 more Smart Citation
“…For all temperatures, a notable increase of the diffusion is observed when increasing the amount of loading from one to four CO 2 molecules. A similar behavior was reported in a study of CO 2 diffusion in the covalent organic frameworks HEX-COF1 and 3D-HNU5 111 , as well as for other small guest molecules embedded in a host matrix, e.g. CH 4 and H 2 in zeolite ZK4 112 , 113 and CH 4 , CF 4 , He, Ne, Ar, Xe, and SF 6 in silicalite.…”
Section: Resultssupporting
confidence: 84%
“…However, these values deviate by several orders of magnitude when compared to experimentally measured values reported as 8.1 to 11.5 × 10 −9 cm 2 s −1 . 109 Nevertheless, a recent experimental study by Thissen et al 110 of fluorescein in the MOF HKUST-1 showed that the elimination of domain boundaries in the solid achieved by epitaxial growth of the MOF on a Si (111) substrate, increased the diffusivity of the organic guest molecule by more than one order of magnitude. The authors of this study concluded that macroscopic diffusion properties may be strongly affected by either boundaries between differently oriented grains or/and barriers associated with surface diffusion.…”
Section: Diffusion Analysismentioning
confidence: 99%
“…By analysis of the diffusion coefficients of the guest molecules as a function of loading, maxima in the associated activation energies could be identified for the different host materials. These maxima were found at loadings that coincide well with the experimentally determined maximum CO 2 adsorption capacities, again indicating an exceptional storage capacity for these compounds …”
supporting
confidence: 77%
“…These maxima were found at loadings that coincide well with the experimentally determined maximum CO 2 adsorption capacities, again indicating an exceptional storage capacity for these compounds. 57 The studied target systems presented in this work provide clear evidence that modern computational approaches combining the capabilities of semi-empirical DFTB methods and neural network potentials with established constrained molecular dynamics frameworks provide access to a broad range of structural, dynamical, and thermodynamical properties of complex guest@host systems. Although the computational demand of the discussed SCC DFTB/3ob/D3 MD simulations is orders of magnitude higher compared to the application of pairwise additive force fields, the execution times are well within the manageable range.…”
Section: The Journal Of Physical Chemistry Lettersmentioning
confidence: 88%
“…To execute the SCC DFTB MD simulations, the DFTB+ package has been interfaced to our in-house QM/MM MD simulation routines focused on the treatment of solid-state systems. The SHAKE/RATTLE algorithm , has been applied to satisfy holonomic bond constraints applied to all bonds involving hydrogen atoms, which enables the use of an increased time step of 2.0 fs in the time integration. The latter was realized using the velocity Verlet propagator .…”
Section: Methodsmentioning
confidence: 99%