TY  - THES
AU  - Selmert, Victor
TI  - Investigation of Polyacrylonitrile-based Carbon Nanofibers as Adsorbents for the Separation of Carbon Dioxide using Dynamic Gas Adsorption
PB  - RWTH Aachen University
VL  - Dissertation
M1  - FZJ-2025-04389
SP  - 215
PY  - 2025
N1  - Dissertation, RWTH Aachen University, 2025
AB  - Electrospun polyacrylonitrile-based carbon nanofibers (CNFs) exhibit a molecular sieving effect. The pore size can be controlled by the choice of the carbonization temperature, making such nanofibers a promising material for gas separation. In this work, these nanofibers are investigated for the application in the field of gas separation using dynamic gas adsorption and are tested against common reference materials. These experiments allow to analyze the adsorption behavior of a packed column of the fibers in terms of selectivity, stability, as well as adsorption and desorption kinetics under conditions similar to the application in an adsorption swing process. First, the procedure of the dynamic gas adsorption method is optimized for small, laboratory-scale amounts of sample. It is shown that an exact determination of the contribution of the void volume is paramount for accurate results. Furthermore, the measurement of desorption curves following the breakthrough curves increases the accuracy in the case of competitively adsorbing gases. In order to optimize the volumetric packing density and the pressure drop along the column, the pelletization of the fibers is also investigated. It is shown that performing the pelletization before carbonization of the fibers is advantageous, since the low mechanical stability of the non-carbonized fibers yields a greater compression of the fibrous material. Accordingly, a higher bulk density and, thus, a higher volumetric adsorption capacity is achieved. Subsequently, the separation capabilities of the pelletized fibers are evaluated on technically relevant separation problems such as CO2 separation from N2-rich mixtures like flue gases or CO2 and N2 separation from biogas-like or natural gas-like mixtures. The CO2/N2 separation on the fibers shows an exceptionally high selectivity, especially for a carbon-based material, and exceeds the values for most reference materials. The high selectivity of the CNFs is attributed to the high adsorption potential of CO2 associated with the narrow ultramicropores as well as the high heteroatom content present in CNFs carbonized at low temperature. In contrast, the separation of CO2 and N2 from CH4-richgas mixtures is kinetically driven and based on the size difference of CH4 to N2 and CO2 and on the narrow pores of the fibers. At higher carbonization temperature and thus, a narrower pore system, CH4 is excluded from the pores, leading to a strong kinetic limitation, while CO2 as well as N2 are adsorbed sufficiently fast. While the separation performance is attributed to the surface chemistry for the CO2/N2 mixture, the separation performance for CO2 and N2 from CH4 is based on the molecularsieve effect. Water adsorption experiments show that the CNFs adsorb water with high affinity and to a high adsorption capacity, which has significant influence on the CO2 adsorption. In cyclic adsorption/desorption experiments with moistened CO2, there is an accumulation of water on the CNFs, which reduces the CO2 capacity significantly. Therefore, drying of the gas mixture is necessary in order to apply the fibers for CO2 separation. However, the high water affinity also proves that the CNFs could be applied in the field gas drying as well. Overall, the electrospun polyacrylonitrile-based CNFs achieve a selective separation in the tested applications and offer the possibility to adjust selectivity as well as kinetics via the carbonization temperature to the needs of a given separation problem. Therefore, the material proves to be promising for the separation of CO2 from dry gas mixtures or for the drying of gases.
KW  - Hochschulschrift (Other)
KW  - carbon nanofibers ; gas separation ; adsorption ; carbon capture ; carbon molecular sieve ; breakthrough curve (Other)
LB  - PUB:(DE-HGF)11
DO  - DOI:10.18154/RWTH-2025-05853
UR  - https://juser.fz-juelich.de/record/1047563
ER  -