Ronny Pini

Direct air capture (DAC) of CO ₂ requires adsorbents that combine high uptake with fast adsorption kinetics, yet whether these properties can be co-optimized remains unresolved. Here we couple automated high-throughput synthesis and characterization of amine-grafted porous polymers with multi-objective Bayesian optimization in a closed loop. Jointly optimizing equilibrium CO ₂ uptake and adsorption rate at 400 ppm CO ₂ , we discover a kinetics-uptake trade-off described by a linear Pareto curve, uncovering adsorbents with order-of-magnitude higher adsorption kinetics than benchmark adsorbents at half the CO ₂ uptake. Three transferable synthesis-performance design rules are derived from the trained Gaussian process (GP) models and grafting with larger polyamines is established as a lever to extend the observed Pareto front. Process modeling shows that these adsorbents achieve twice the productivity of a realistic DAC process compared to the benchmark DAC adsorbent, Lewatit VP OC 1065. Beyond DAC, the framework provides a blueprint for other closed-loop multi-objective materials optimization studies.

The operation of a heat exchanger is a multiphysics problem, involving the simultaneous flow of two fluids and heat transfer between them. Velocity, pressure and temperature gradients within each fluid vary with location, operating conditions, flow arrangement and heat exchanger geometry. While robust design procedures exist, including analytical approaches, their solution can be hard to fully comprehend without computational tools that visualize the underlying physical processes within the heat exchanger. Here, we present a classroom exercise aimed at improving student’s understanding of double pipe heat exchangers using the computational fluid dynamics software COMSOL Multiphysics. This computer-based laboratory class is integrated within a ‘Process Heat Transfer’ module offered to undergraduate and postgraduate chemical engineering students. The exercise is fully documented, including the problem statement and solution, and guides students to apply the design workflow and discuss their solutions in class. A survey distributed at the end of the module indicates overall satisfaction with the hands-on approach, especially the opportunity to receive training on a computational software. A pre- and post-test evaluation demonstrates that the computer-based class helped eliminate persistent student misconceptions regarding the operation of heat exchangers. Ensuring proper software set-up, in-class support, and the provision of instructional material is crucial due to the learning curve associated with using the software. By developing a digital model of the heat exchanger, the exercise serves as a cognitive tool—encouraging students to explore more advanced problems, such as the completion of parametric studies, thereby reinforcing the understanding of theoretical concepts.

Structured sorbents can efficiently process large air volumes required for Direct Air Capture (DAC), but their broad implementation is limited by challenges in producing reproducible and homogeneous materials. This work presents a robust procedure to graft amines onto monolithic sorbents and systematically investigates how capture performance is influenced by three fundamental material building blocks: substrate type (mullite vs. cordierite), alumina loading, and amine type (APTMS vs. TRI). X‐ray Computed Tomography and gravimetric analyses confirm uniform amine distribution, with adsorption capacity scaling linearly with alumina mass. From a process perspective, lower density substrates achieve higher specific uptake, thus reducing the energy demand for regeneration. Breakthrough experiments demonstrate that humidity significantly enhances adsorption thermodynamics and kinetics: uptake increases by 30% for APTMS and 70% for TRI, and adsorption kinetics are accelerated. This study provides quantitative material performance relationships necessary for process design and contactor optimization.

Direct Air Capture (DAC) is a negative emissions technology whose performance is inherently linked to ambient conditions, which directly affect its primary feed stream (air). A common simplification in DAC model simulations is the use of fixed weather conditions, which can bias the predicted performance under weather variability. In response, this study quantifies the impact of local meteorological variability and temporal weather aggregation on the performance of DAC units. Building on a previously developed and validated 1D mechanistic model of a fixed-bed Steam-assisted Temperature Vacuum Swing Adsorption (S-TVSA) DAC process, we simulate its operation using weather data from the Met Office station at Buchan (UK), near the Saint Fergus terminal - a strategic hub for Carbon Capture and Storage (CCS) activities in Scotland. A two-branch methodological framework is developed combining optimization and forward simulations. Operating conditions are optimized using a multi-objective genetic algorithm (NSGA-II) to maximize productivity (Pr) and minimize specific equivalent work (Weq) at two temporal resolutions. Furthermore, daily weather inputs are aggregated on a monthly and yearly scale to assess the impact of data resolution on model predictions and real operational gains. Results show that temporal weather aggregation to yearly averages biases DAC key performance indicators, overestimating Pr by up to 5% while underestimating Weq by up to 31%, relative to performance based on daily weather variations. Moreover, optimization strategies that explicitly account for monthly weather variability present monthly gains, by increasing Pr by up to 10%. Yet, these monthly gains do not necessarily translate into significant operational performance benefits at the annual scale when daily weather data is propagated in the process model.

The design of cyclic adsorption processes is computationally expensive as it involves screening many process designs, each of which involve a time-consuming simulation to reach cyclic steady state. In this work, we demonstrate how differentiable programming can be used to accelerate both the simulation and attainment of cyclic steady state for a four-step pressure vacuum swing adsorption (PVSA) process to concentrate carbon dioxide from flue gas. A mechanistic one-dimensional dynamic adsorption model was implemented in JAX, enabling automatic differentiation and just-in-time compilation for efficient solution and accurate sensitivity evaluation. The latter was exploited to implement a Newton-based direct determination method for accelerated convergence to cyclic steady state, avoiding repeated cycle simulations. Across 4096 designs sampled in a six-dimensional design space, the direct determination method converged in an average of 4.6 iterations, compared to 145 cycles required by successive substitution. When combined with the computational gains from the JAX framework, this resulted in an overall speed-up of over 20 times relative to the conventional MATLAB-based implementation.

Designing large-scale adsorption-based separation units requires a multiscale understanding of material behavior, from molecular interactions to process-level performance. An underpinning challenge is the development of predictive multicomponent adsorption equilibrium models for systems departing from the classic type I isotherm for microporous adsorbents. Adsorbents that undergo adsorption-induced transitions (flexible adsorbents) belong to this category and are a class of framework materials that exhibit unique features such as sigmoidal equilibrium isotherms and intrinsic thermal management capabilities. Despite their promise for practical applications, flexible adsorbents remain underexplored at the process scale due to the lack of a suitable equilibrium isotherm model that mechanistically captures their adsorption-induced structural transitions. Here, we present a simplified statistical isotherm model for transition materials (SSI-T) as a generalized approach to parametrize─using a combination of sorbate-dependent and -independent physical parameters─the adsorption equilibria in flexible adsorbents that exhibit a broad range of structural and configurational transitions upon adsorption. We have validated the model using unary and binary equilibrium data for gate-opening, breathing, and configurational transitions in multiple adsorbent–adsorbate systems. Our formulation represents the first continuous, differentiable, and explicit isotherm model in the literature capable of accurately describing unary adsorption and desorption isotherms in flexible adsorbents, and predicting binary equilibria for multiple types of adsorption-induced transitions without any additional parameters. Because SSI-T is an explicit function of state variables, it can be seamlessly integrated into process-scale simulators, enabling the design and optimization of adsorption-based technologies that use flexible adsorbents.

This study extends the digital adsorption (DA) method─i.e., adsorption experiments augmented with X-ray computed tomography (XCT)─from unary systems with strongly adsorbing CO₂ to unary systems with weakly adsorbing N₂ and to binary systems involving both species. Using this method, equilibrium adsorption isotherms on commercial activated carbon and zeolite 13X were measured for both gases at 294.15 K between 0.1 and 9 bar. Our analysis shows that X-ray attenuation by the bulk phase must be considered for weak adsorbates, while it is relevant only at elevated pressures for strong adsorbates. Unary dynamic column breakthrough (DCB) experiments with N₂ showed that, with a correction factor, the DA method can describe the transient progression of the internal adsorbed phase under near-isothermal, equilibrium-controlled conditions quantitatively. In binary CO₂-N₂ DCB experiments, the direct determination of adsorbed amounts requires local composition data; however, when complemented by a one-dimensional DCB model, the experimental results are shown to capture dynamic adsorption behavior, including fast N₂ and slower CO₂ uptake, as well as a thermally enhanced roll-up. These results demonstrate that XCT provides valuable insight into dynamic adsorption processes and that, with care, the DA method can also be applied to weakly adsorbing systems.