How Can (or Why Should) Process Engineering Aid the Screening and Discovery of Solid Sorbents for CO2 Capture?

ConspectusAdsorption using solid sorbents is emerging as a serious contender to amine-based liquid absorption for postcombustion CO2 capture. In the last 20+ years, significant efforts have been invested in developing adsorption processes for CO2 capture. In particular, significant efforts have been invested in developing new adsorbents for this application. These efforts have led to the generation of hundreds of thousands of (hypothetical and real) adsorbents, e.g., zeolites and metal-organic frameworks (MOFs). Identifying the right adsorbent for CO2 capture remains a challenging task. Most studies are focused on identifying adsorbents based on certain adsorption metrics. Recent studies have demonstrated that the performance of an adsorbent is intimately linked to the process in which it is deployed. Any meaningful screening should thus consider the complexity of the process. However, simulation and optimization of adsorption processes are computationally intensive, as they constitute the simultaneous propagation of heat and mass transfer fronts; the process is cyclic, and there are no straightforward design tools, thereby making large-scale process-informed screening of sorbents prohibitive.This Account discusses four papers that develop computational methods to incorporate process-based evaluation for both bottom-up (chemistry to engineering) screening problems and top-down (engineering to chemistry) inverse problems. We discuss the development of the machine-assisted adsorption process learning and emulation (MAPLE) framework, a surrogate model based on deep artificial neural networks (ANNs) that can predict process-level performance by considering both process and material inputs. The framework, which has been experimentally validated, allows for reliable, process-informed screening of large adsorbent databases. We then discuss how process engineering tools can be used beyond adsorbent screening, i.e., to estimate the practically achievable performance and cost limits of pressure vacuum swing adsorption (PVSA) processes should the ideal bespoke adsorbent be made. These studies show what conditions stand-alone PVSA processes are attractive and when they should not be considered. Finally, recent developments in physics-informed neural networks (PINNS) enable the rapid solution of complex partial differential equations, providing tools to potentially identify optimal cycle configurations. Ultimately, we provide areas where further developments are required and emphasize the need for strong collaborations between chemists and chemical engineers to move rapidly from discovery to field trials, as we do not have much time to fulfill commitments to net-zero targets.

Prediction of MOF Performance in Vacuum Swing Adsorption Systems for Postcombustion CO2 Capture Based on Integrated Molecular Simulations, Process Optimizations, and Machine Learning Models.

Postcombustion CO2 capture and storage (CCS) is a key technological approach to reducing greenhouse gas emission while we transition to carbon-free energy production. However, current solvent-based CO2 capture processes are considered too energetically expensive for widespread deployment. Vacuum swing adsorption (VSA) is a low-energy CCS that has the potential for industrial implementation if the right sorbents can be found. Metal-organic framework (MOF) materials are often promoted as sorbents for low-energy CCS by highlighting select adsorption properties without a clear understanding of how they perform in real-world VSA processes. In this work, atomistic simulations have been fully integrated with a detailed VSA simulator, validated at the pilot scale, to screen 1632 experimentally characterized MOFs. A total of 482 materials were found to meet the 95% CO2 purity and 90% CO2 recovery targets (95/90-PRTs)-365 of which have parasitic energies below that of solvent-based capture (∼290 kWhe/MT CO2) with a low value of 217 kWhe/MT CO2. Machine learning models were developed using common adsorption metrics to predict a material's ability to meet the 95/90-PRT with an overall prediction accuracy of 91%. It was found that accurate parasitic energy and productivity estimates of a VSA process require full process simulations.

The Challenge of Water Competition in Physical Adsorption of CO2 by Porous Solids for Carbon Capture Applications - A Short Perspective.

With ever-increasing efforts to design sorbent materials to capture carbon dioxide from flue gas and air, this perspective article is provided based on nearly a decade of collaboration across science, engineering, and industry partners. A key point learned is that a holistic view of the carbon capture problem is critical. While researchers can be inclined to value their own fields and associated metrics, often, key parameters are those that enable synergy between materials and processes. While the role of water in the chemisorption of CO2 is well-studied, in this perspective, it is hoped to highlight the often-overlooked but critical role of water in assessing the potential of a physical adsorbent for CO2 capture. This is a challenge that requires interdisciplinarity. As such, this document is written for a general audience rather than experts in any specific discipline.

Can a computer "learn" nonlinear chromatography?: Physics-based deep neural networks for simulation and optimization of chromatographic processes.

The design and optimization of chromatographic processes is essential for enabling efficient separations. To this end, hyperbolic partial differential equations (PDEs) along with nonlinear adsorption isotherms must be solved using computationally expensive numerical solvers to understand, simulate, and design the complex behavior of solute movement in chromatographic columns. In this study, physics-based artificial neural network framework for adsorption and chromatography emulation (PANACHE) is used to simulate and optimize chromatographic processes in a computationally faster and reliable manner. The proposed approach relies on learning the underlying PDEs in the form of a physics-constrained loss function to improve the accuracy of process simulations. The effectiveness of this approach is demonstrated by considering the complex dynamics of binary solute mixtures for generic pulse injections subjected to different isotherm systems, namely, the four cases of the generalized Langmuir isotherms. Unique neural network models were developed for each isotherm and the models accurately predicted the spatiotemporal concentrations of solute mixture in chromatographic columns for an arbitrary feed concentrations and injection volumes by facilitating up to 250 times computational speed-ups. Moreover, the neural network models were incorporated with process optimization routines to precisely determine the optimal injection volumes to enable baseline separation of solute components of the feed mixture.

A scalable metal-organic framework as a durable physisorbent for carbon dioxide capture.

Metal-organic frameworks (MOFs) as solid sorbents for carbon dioxide (CO2) capture face the challenge of merging efficient capture with economical regeneration in a durable, scalable material. Zinc-based Calgary Framework 20 (CALF-20) physisorbs CO2 with high capacity but is also selective over water. Competitive separations on structured CALF-20 show not just preferential CO2 physisorption below 40% relative humidity but also suppression of water sorption by CO2, which was corroborated by computational modeling. CALF-20 has a low enthalpic regeneration penalty and shows durability to steam (>450,000 cycles) and wet acid gases. It can be prepared in one step, formed as composite materials, and its synthesis can be scaled to multikilogram batches.

Three dimensional digital holographic profiling of micro-fibers.

A method to measure the size, orientation, and location of opaque micro-fibers using digital holography is presented. The method involves the recording of a digital hologram followed by reconstruction at different depths. A novel combination of automated image analysis and statistical techniques, applied on the intensity of reconstructed digital holograms is used to accurately determine the characteristics of the micro-fibers. The performance of the proposed method is verified with a single fiber of known length and orientation. The potential of the method for measurement of fiber length is further demonstrated through its application to a suspension of fibers in a liquid medium.

Simulated moving bed chromatography for the separation of enantiomers.

Simulated moving bed (SMB) chromatography, a continuous multi-column chromatographic process, has become one of the preferred techniques for the separation of the enantiomers of a chiral compound. Several active pharmaceutical ingredients, including blockbuster drugs, are manufactured using the SMB technology. Compared to single column preparative chromatography, SMB separations achieve higher productivity and purity, while reducing the solvent consumption. The SMB technology has found applications both at small and large scales. Design methods have been developed for robust operation and scale-up, using data obtained from analytical experiments. In the last few years, rapid developments have been made in the areas of design, improved process schemes, optimization and robust control. This review addresses these developments, as well as both the fundamentals of the SMB science and technology and some practical issues concerning the operation of SMB units. Particular emphasis is placed on the consolidation of the "triangle theory", a design tool that is used both in the academia and industry for the design of SMB processes.

Equilibrium theory-based design of simulated moving bed processes under reduced purity requirements linear isotherms.

The design of simulated moving bed processes under reduced purity requirements for systems whose isotherm is linear is considered. Based on the equilibrium theory of chromatography, explicit equations to uniquely identify the separation region that will ensure specified extract and raffinate purities are derived. The identification of the region requires only the knowledge of Henry constants of the solutes, the concentration of the solutes in the feed and the purity specifications. These results are validated using numerical simulations.

Effect of pressure drop on solute retention and column efficiency in supercritical fluid chromatography. Part 2: Modified carbon dioxide as mobile phase.

The effect of pressure drop on the performance of supercritical fluid chromatography systems using a modified mobile phase (carbon dioxide + ethanol) was studied. Experiments were performed on a Lichrospher-RP-18 column with phenanthrene as a solute. A wide range of back pressures (130 to 210 bar) and modifier concentrations (2 to 7% w/w) have been explored. Experiments yielding both small and large pressure drops were performed. From these experiments, parameters to describe pressure drop, retention, and column efficiency were extracted, and were used to simulate the dynamics of the chromatographic column. A good match between the experimentally measured and calculated values of pressure drop, retention times, and column efficiency was observed. At low back pressure and modifier composition, significant loss of column efficiency was observed.

Computational fluid dynamics study of viscous fingering in supercritical fluid chromatography.

Axi-symmetric numerical simulations are carried out to study the dynamics of a plug introduced through a mixed-stream injection in supercritical fluid chromatographic columns. The computational fluid dynamics model developed in this work takes into account both the hydrodynamics and adsorption equilibria to describe the phenomena of viscous fingering and plug effect that contribute to peak distortions in mixed-stream injections. The model was implemented into commercial computational fluid dynamics software using user-defined functions. The simulations describe the propagation of both the solute and modifier highlighting the interplay between the hydrodynamics and plug effect. The simulated peaks showed good agreement with experimental data published in the literature involving different injection volumes (5 µL, 50 µL, 1 mL and 2 mL) of flurbiprofen on Chiralpak AD-H column using a mobile phase of CO2 and methanol. The study demonstrates that while viscous fingering is the main source of peak distortions for large-volume injections (1 mL and 2 mL) it has negligible impact on small-volume injections (5 µL and 50 µL). Band broadening in small-volume injections arise mainly due to the plug effect.

Enantioseparation of 1-phenyl-1-propanol on cellulose-derived chiral stationary phase by supercritical fluid chromatography II. Non-linear isotherm.

The separation of the enantiomers of 1-phenyl-1-propanol by supercritical fluid chromatography on a chiral stationary phase, which consists of cellulose tris (3,5-dimethylphenylcarbamate) coated on a silica support (Chiralcel-OD), is studied under overloaded, non-linear chromatographic conditions. Pulse experiments are performed at a temperature of 30 degrees C using supercritical CO(2) modified with methanol as a mobile phase. The parameters of the binary Langmuir adsorption isotherm are determined by the inverse method, comparing experimental and simulated peak responses. Isotherm parameters are derived for modifier concentrations between 1 and 5% (w/w) and operating pressures between 125 and 185 bar, and the dependency of the isotherm parameters, namely the Henry constant and the saturation capacity, on operating conditions is investigated.

Local equilibrium theory analysis of chromatographic peak shapes in the presence of adsorbing modifiers.

The nature and shapes of solute and modifier peaks for a situation when the modifier and solute adsorb competitively through the Langmuir isotherm is studied using the local-equilibrium theory. The analysis yields 10 different cases depending on the relative adsorption strengths and the concentrations of the modifier and the solute. The origin of positive and negative Langmuirian/anti-Langmuirian-type peaks, even when the adsorption isotherms are strictly Langmuirian, and explicit conditions under which each combination is observed are derived. A fundamental explanation to the origin and propagation of rounded peaks reported in liquid and supercritical fluid chromatography [1-3] is offered. A rigorous theoretical basis for the rule of thumb proposed by Fornstedt and Guiochon [1,2] about the nature of peak shapes when an adsorbing modifier is present, is established.

Enantioseparation of 1-phenyl-1-propanol on Chiralcel OD by supercritical fluid chromatography. I. Linear isotherm.

The supercritical fluid chromatography (SFC) separation of the enantiomers of 1-phenyl-1-propanol on the chiral stationary phase Chiralcel OD under linear conditions is studied. Supercritical CO2 modified with methanol is used as a mobile phase. The effect of modifier concentration, pressure and temperature is studied. An empirical isotherm to account for the effect of density of the mobile phase and modifier concentration has been used to model the experimental results. It was observed that the selectivity and resolution were higher at 30 degrees C as compared to those at 40 degrees C.

Effect of pressure drop on solute retention and column efficiency in supercritical fluid chromatography.

The effect of pressure drop on the performance of supercritical fluid chromatographic systems is studied. Experiments have been conducted at different pressures and at 55 and 65 degrees C. Experiments at conditions leading to large and small pressure drops have been performed. Parameters to describe the pressure drop, retention time and efficiency have been extracted from these experiments. Using these parameters the dynamics of the chromatographic column have been modeled. Darcy's law was used to describe the pressure drop. The efficiency was modeled by considering the contributions from axial dispersion, resistance to mass transfer from the fluid film, and to diffusion in the pores. Good description of the pressure drop, retention time and mass transfer characteristics under normal operating conditions was obtained. The parameters extracted were used to predict the elution profile by numerical simulations. Considerable loss in column efficiency was observed when operating the column at lower values of the back pressure.

Enantioseparation of 1-phenyl-1-propanol by supercritical fluid-simulated moving bed chromatography.

The enantioseparation of 1-phenyl-1-propanol through the supercritical fluid-simulated moving bed (SF-SMB) process is studied. Non-linear isotherms were measured on an analytical column, and used together with the triangle theory for SMB design to select operating conditions for the SF-SMB. Experiments were carried out on a pilot-scale SF-SMB plant at conditions that corresponded to the non-linear range of the isotherm. Under conditions of low feed concentration, complete separation (extract purity = 99.5%; raffinate purity = 98.4%) was achieved. Under conditions of larger feed concentration, the best separation corresponded to an extract purity of 98.0% and a raffinate purity of 94.0%, and yielded a productivity of 110 g of racemate per kg stationary phase per day.

Peak distortions arising from large-volume injections in supercritical fluid chromatography.

Preparative separations in supercritical fluid chromatography (SFC) involve the injection of large volumes of the solute. In SFC, the mobile phase is typically high pressure CO2+modifier and the solute to be injected is usually dissolved in the modifier. Two-types of injection methods, modifier-stream and mixed-stream, are common in commercial preparative SFC systems. In modifier-stream injection, the injection is made in the modifier stream which is later mixed with the CO2 stream, while in the mixed-stream injection, the injection is made in a mixed CO2+modifier stream. In this work a systematic experimental and modelling study of the two techniques is reported using single-enantiomers of flurbiprofen on Chiralpak AD-H with CO2+methanol as the mobile phase. While modifier-stream injection shows non-distorted peaks, mixed-stream injection results in severe peak-distortion. By comparing the modelling and experimental results, it is shown that the modifier "plug" introduced in the mixed-stream injection is the primary cause of the peak distortions. The experimental results also point to the possible existence of viscous fingering which contributes to further peak distortion.

Absence of experimental evidence of a delta-shock in the system phenetole and 4-tert-butylphenol on Zorbax 300SB-C18.

We investigate the system consisting of phenetole (PNT) and 4-tert-butylphenol (TBP) in methanol-water (63:37 v:v) on a Zorbax 300SB-C18 column by characterising single component isotherms, by performing a large number of binary experiments of different types and by describing the experiments through simulations carried out using a novel, rather powerful competitive adsorption isotherm, that we call the generalized bi-Langmuir isotherm. This system is of great interest because it was previously reported to yield a new type of transition in nonlinear chromatography, the so-called delta-shock. Such transition had been discovered earlier through a theoretical analysis and confirmed by detailed simulations. The initial aim of this work was to reach a satisfactory agreement between delta-shock experiments and corresponding numerical simulations. In the course of this work however, a number of inconsistencies in the interpretation of the previous experimental results were highlighted and explained. This led to a new experimental campaign, which is reported here and has allowed to reach two important conclusions: (1) The binary system PNT-TBP mentioned above does not exhibit a delta-shock; the spike in the UV profile, which has previously been interpreted as an experimental evidence of the delta-shock, results from liquid-liquid phase separation within the chromatographic column. (2) The same system exhibits a rather peculiar behavior in breakthrough and displacement experiments, which could be well described using the generalized bi-Langmuir isotherm.

Equilibrium theory-based analysis of nonlinear waves in separation processes.

Different areas of engineering, particularly separation process technology, deal with one-dimensional, nonstationary processes that under reasonable assumptions, namely negligible dispersion effects and transport resistances, are described by mathematical models consisting of systems of first-order partial differential equations. Their behavior is characterized by continuous or discontinuous composition (or thermal) fronts that propagate along the separation unit. The equilibrium theory (i.e., the approach discussed here to determine the solution to these model equations) predicts this with remarkable accuracy, despite the simplifications and assumptions. Interesting applications are in adsorption, chromatography and ion-exchange, distillation, gas injection, heat storage, sedimentation, precipitation, and dissolution waves. We show how mathematics can enlighten the engineering aspects, and we guide the researcher not only to reach a synthetic understanding of properties of fundamental and applicative interest but also to discover new, unexpected, and fascinating phenomena. The tools presented here are useful to teachers, researchers, and practitioners alike.

Design of batch chromatography for separation of binary mixtures under reduced purity requirements.

Two methods are presented for designing separation of binary mixtures in a batch chromatography column under reduced purity requirements such that no waste or recycle fractions are generated. The first one is based on the equilibrium theory of chromatography and requires adsorption isotherm parameters. The second one is a shortcut method that uses a single experimental or simulated design chromatogram as input and is recommended under strongly non-ideal conditions with significant dispersive effects. Both approaches allow prediction of the injection volume and the cut position that lead to given target purities. In principle, they apply for all systems with convex or concave isotherms. The applicability of the design methods is evaluated by using numerical simulations. Both design methods work the better the higher the column efficiency and the lower the purity constraints are.