(589i) Unary and Binary CO2/N2 Dynamic Column Breakthrough Experiments Augmented By X-Ray Computed Tomography Imaging

Adsorption systems used for CO₂ capture and gas separation were identified as viable options to contribute to future CO₂ emissions mitigation efforts (Zanco et al., 2017). However, despite the numerous studies that have been conducted to characterise adsorption processes, many aspects are not fully understood because of the limitations of classical experimental methods like Dynamic Column Breakthrough experiments (DCB). Although DCB experiments can provide valuable information on the general properties of an adsorbent, such as adsorption rate and loading capacity, they are not capable of providing information about internal dynamics. Moreover, it has been shown that particularly competitive adsorption behaviours, for example of a CO₂ and N₂, remain difficult to describe with classical DCB experiments, an example being the necessity to divert to desorption experiments to measure the competitive loading of CO₂ and N₂ on Zeolite 13X (Wilkins et al. 2019).

With the limitations of classical DCB experiments in mind, the Digital adsorption approach has recently emerged as an experimental method that incorporates x-ray computed tomography (CT) imaging to determine the adsorption properties of microporous adsorbents (Joss et al., 2017). The approach exploits the gas-to-liquid phase change, i.e. the phase transition that a gaseous adsorbate undergoes as a result of adsorption, to generate space- and time-resolved information on the adsorption process. Applied to DCB experiments of CO₂ on activated carbon, it was demonstrated that this method can capture the internal dynamics of the adsorption and desorption process through the measurement of transient internal concentration profiles of the adsorbed phase (Pini et al., 2021). Therefore, Digital adsorption can not only provide an improved database to validate and calibrate computational models but potentially also improve our understanding of more advanced adsorber designs, such as layered beds or structured adsorbers.

However, against the great potential digital adsorption holds as a method, it has not yet been applied to study the adsorption on other adsorbent materials than activated carbon and of any gas other than CO₂. In this study, we, therefore, present the extension of the digital adsorption approach to the study of breakthrough experiments in a laboratory fixed-bed adsorption column involving unary N₂ and the binary gas mixture CO₂/N₂ on Zeolite 13X.

Methods

The experimental setup used for experiments is outlined in Figure 1. The core of the set-up and the only part exposed to the x-rays is the adsorption column itself. The latter has a length of 300mm, an inner diameter of 30mm, and is made of Polyether Ether Ketone (PEEK) – a material that is essentially transparent to x-rays. Ancillary equipment includes: pressure transmitters to monitor the pressure drop across the column; four ultra-fine thermocouples to record the internal temperature at distinct locations along the length of the column; mass flow controllers to maintain a fee of constant rate and composition; a mass flow meter and a mass spectrometer to measure rate and composition of the effluent stream. The x-ray CT scans are acquired with a Toshiba Aquilon 64 scanner, set to a current of 200 mA and peak tube voltage of 120 keV. The resolution of the reconstructed images is 0.07x0.07x1 mm, where the lowest resolution is in the axial direction of the column. The property measured by the x-ray CT instrument is the so-called ‘CT number’, which is linearly proportional to the local bulk density of the imaged object and can be used as a measure of gas adsorption.

All DCB experiments are conducted at ambient conditions (temperature, T ≈ 293.25 K and pressure, p ≈ 99 kPa). Before each individual DCB experiment is initiated, the column is purged with a pure helium stream, followed by the introduction of the single-component or gas mixture stream as a step-input. Two single-component DCB experiments, respectively one with CO₂ and one with N₂, at an inlet flow rate of 100ccm are conducted and x-ray scans are taken at subsequent time points until the column has been saturated with the feed gas. Competitive CO₂/N₂ DCB are conducted with the same total flow rate and CO₂ mole fractions, y=10%, 15%, and 50%, respectively.

Results & Discussion

The propagation of the adsorption front can be clearly identified from the single-component DCB experiments, both in the direct visualisation of the CT number at axial cross sections and in the internal profiles of the adsorbed amount at subsequent time steps. The adsorption front moves significantly faster in the experiment with pure N₂ than in the one with pure CO₂. Moreover, the observed change in x-ray attenuation is significantly lower in the experiment with pure N₂. Both observations can be directly related to the considerably smaller adsorption capacity of Zeolite 13X for N₂ relative to CO₂. It follows that uncertainty considerations become more important for experiments with weakly adsorbing gases, such as N₂, as the relative uncertainty for averaged voxel volumes of the size of one cross-sectional slice was found to be approximately 5-10% (Joss et al. 2017).

In the competitive CO₂/N₂ DCB experiments the internal profiles of the adsorbed amount cannot be calculated directly from the measured change in x-ray attenuation, without knowledge of the composition of the adsorbed phase and its density (or volume). Nevertheless, the dynamics can be observed in a qualitative manner from plots of the CT number at axial cross-sections as well as from the slice-averaged CT values along the column length. In these experiments, we observe two fronts moving simultaneously, the fast-traveling N₂ and the slow-traveling CO₂-rich front. In between the two fronts, the CT number stabilises to the value corresponding to pure N₂ adsorption. For the latter, we observe good agreement with the results from the pure-component experiments. Moreover, as expected, it is observed that the higher the CO₂ mole fraction in the feed, the faster the CO₂-rich front travels.

By applying the digital adsorption approach to competitive CO₂/N₂ DCB experiments on a Zeolite 13X bed, this study demonstrates the capability of the method to observe dynamics in competitive two-component DCB experiments. The ability to observe internal adsorption profiles may prove to be very beneficial for the study of competitive adsorption behaviour beyond classic DCB experiments. In fact, upon careful selection of feed and initial conditions, one can simultaneously probe adsorption and desorption fronts, demonstrating one practical advantage of x-ray CT image applied to the analysis of DCB experiments.

References

Zanco, S. E., Joss, L., Hefti, M., Gazzani, M., Mazzotti, M. (2017). Addressing the Criticalities for the Deployment of Adsorption-based CO₂ Capture Processes.

Wilkins, N. S., Rajendran, A. (2019). Measurement of competitive CO₂ and N₂ adsorption on Zeolite 13X for post-combustion CO₂ capture.

Joss, L., Pini, R. (2017). Digital Adsorption: 3D Imaging of Gas Adsorption Isotherms by X-ray Computed Tomography.

Pini, R., Joss, L., Hosseinzadeh Hejazi, S. A. (2021). Quantitative imaging of gas adsorption equilibrium and dynamics by X-ray computed tomography.