(583a) Digital Adsorption: 3D Imaging of Adsorption Equilibria and Dynamics in Technical Microporous Solids | AIChE

(583a) Digital Adsorption: 3D Imaging of Adsorption Equilibria and Dynamics in Technical Microporous Solids


Pini, R. - Presenter, Imperial College London
Joss, L., The University of Manchester
Hosseinzadeh Hejazi, S. A., University of Alberta
Gas adsorption is one of the key physical processes that underlie the numerous industrial uses of porous solids. Typical applications include separation processes, catalytic reactions, energy storage, and gas recovery from unconventional plays. Key to maximise the efficiency of these processes is our ability to understand the microsctructure of the material at the pores scale (1–10 Å) and its interaction with the operating fluids. Yet, to be useful for industrial implementation, the powders obtained in the laboratory need to be shaped into so-called technical solids, i.e. millimetre-sized bodies with sufficient porosity, with improved chemical and mechanical stability. However, these densified solids also reveal variations in terms of the structure and composition over length-scales much larger than the pore size, from a single pellet (µm–mm) up to the packed-bed (cm – dm)1. Understanding the scale and extent of heterogeneity in technical adsorbents by means of advanced experimentation is a precursor to the rational scale-up of tailor-made materials, from powders to the sub-kilogram scale, and to the robust design of separation processes. Multi-dimensional imaging techniques represent an essential tool to characterise these systems over the range of relevant length scales. However, the application of these methods to systems whose function is based on the presence of nanometre sized pores remains a challenge due to the trade-off between sample size and image resolution.

We show here how X-ray transmission Computed Tomography (XCT) can be applied to obtain a three-dimensional spatial characterisation of an adsorbent bed (50 cm3) in terms of its microstructural properties, non-invasively. We report the results of two case studies with commercially available samples of activated carbon and zeolite 13X, demonstrating the deployment of digital adsorption workflow2,3 to produce for the first time an integrated and quantitative three-dimensional visualisation of a packed-bed adsorber operando. Results are presented that include both static and dynamic (breakthrough) experiments using Helium and CO2as inert and adsorbing gas, respectively. From the static experiments, spatially distributed adsorption isotherms are measured at mm resolution and pressures 0.1–3 MPa; when combined with conventional analyses of physisorption data, we determine three-dimensional maps of the specific surface area, pore volume and other process performance metrics. We further use a machine learning approach to identify and locate different materials within the packed bed. The breakthrough experiments have been carried out at a total pressure of 0.1 MPa and at room temperature, while varying inlet composition, flow rates and tracking the adsorption fronts by XCT. All measured outputs (outlet andinternal profiles) are described by a detailed model of the column that solves the partial differential equations consisting of mass, energy and momentum balances coupled with the appropriate initial and boundary conditions. Notably, while 1D and average properties are well captured by the model, significant variability is observed in the local adsorbed amount. The latter is associated with packing variability at the mm-scale and, possibly, inter-pellet heterogeneity

This novel ability to measure the adsorbed amount in-situ with both spatial and temporal resolution paves the way towards the characterisation of adsorption processes over the continuum of relevant length scales. This is key towards reconciling fundamental studies on the scale of a single crystal or pellet with those performed at the column scale. The gained insight is expected to contribute in bridging the gap between materials research and process design.


[1] R. Pini, L. Joss, Curr. Opin. Chem. Eng. 24, 37–44, 2019; [2] L. Joss, R. Pini, J. Phys. Chem. C121, 26903–26915, 2017; [3] L. Joss, R. Pini ChemPhysChem20, 524–528, 2019