Publications

Synergizing data-driven and knowledge-based hybrid models for ionic separations

Published in ACS ES&T Engineering, 2024

A hybrid modeling framework has been developed for electrodialysis (ED) and resin-wafer electrodeionization (EDI) in brackish water desalination, integrating compositional modeling with machine learning techniques. Initially, a physics-based compositional model is utilized to characterize the behavior of the unit. Synthetic data are then generated to train a machine learning-based surrogate model capable of handling multiple outputs. This model is further refined using a limited set of experimental data. The effectiveness of this approach is demonstrated by its ability to accurately predict experimental results, indicating an acceptable representation of the system’s behavior. Through an analysis of feature importance facilitated by the machine learning model, a nuanced understanding of the interaction between the chosen ion-exchange resin wafer type and ED/EDI operational parameters is obtained. Notably, it is found that the applied cell voltage has a predominant impact on both the separation efficiency and energy consumption. By employing multiobjective optimization techniques, experimental conditions that enable achieving 99% separation efficiency while keeping energy consumption below 1 kWh/kg are identified.

Recommended citation: Teslim Olayiwola, Luis A. Briceno-Mena, Christopher G. Arges, and Jose A. Romagnoli ACS ES&T Engineering 2024 4 (12), 3032-3044. https://pubs.acs.org/doi/abs/10.1021/acsestengg.4c00405

Deconvoluting charge-transfer, mass transfer, and ohmic resistances in phosphonic acid–sulfonic acid ionomer binders used in electrochemical hydrogen pumps

Published in Energy & Environmental Science, 2023

Ion-pair high-temperature polymer electrolyte membranes (HT-PEMs) paired with phosphonic acid ionomer electrode binders have substantially improved the performance of HT-PEM electrochemical hydrogen pumps (EHPs) and fuel cells. Blending poly(pentafluorstyrene-co-tetrafluorostyrene phosphonic acid) (PTFSPA) with Nafion™, and using this blend as an electrode binder, improved proton conductivity in the electrode layer resulting in a 2 W cm−2 peak power density of fuel cells at 240 °C (a HT-PEM fuel cell record). However, much is unknown about how phosphonic acid ionomers blended with perfluorosulfonic acid materials affect electrode kinetics and gas transport in porous electrodes. In this work, we studied the proton conductivity, electrode kinetics, and gas transport resistances of 3 types of phosphonic acid ionomers, poly(vinyl phosphonic acid), poly(vinyl benzyl phosphonic acid), and PTFSPA by themselves and when blended with Aquivion® (a perfluorosulfonic acid material). These studies were performed using EHP platforms. For all phosphoric acid ionomer types, the addition of Aquivion® promoted ionic conductivity, hydrogen oxidation/evolution reaction kinetics (HOR/HER), and hydrogen gas permeability. Solid-state 31P NMR revealed that the addition of Aquivion® eliminated or significantly reduced phosphate ester formation in phosphoric acid ionomers and this plays a vital role in enhancing ionomer blend conductivity. Using the best blend variant, PTFSPA-Aquivion®, an EHP performance of 5.1 A cm−2 at 0.4 V at T = 200 °C was attained. Density functional theory (DFT) calculations identified that phosphonic acids with electron-withdrawing moieties reduced the propensity of the phosphonic acid to adsorb on platinum electrocatalyst surfaces. The relative adsorption affinity of the various phosphonic acid ionomers from DFT is consistent with experimentally determined charge transfer resistance. A voltage loss breakdown model revealed that the addition of Aquivion® reduced activation and concentration overpotentials in EHPs. Overall, a systematic experimental and modeling approach provided further insight as to how perfluorosulfonic acid ionomers blended with phosphoric acid ionomers affect ionic conductivity, reaction kinetics, and gas permeability in EHP platforms.

Recommended citation: Arunagiri, Karthik, Wong, Andrew Jark-Wah, Briceno-Mena, Luis, Elsayed, Hania Gaber Hassan, Romagnoli, José A., Janik, Michael J., & Arges, Christopher G. (2023). Deconvoluting charge-transfer, mass transfer, and ohmic resistances in phosphonic acid–sulfonic acid ionomer binders used in electrochemical hydrogen pumps. Energy & Environmental Science, 16(12). https://doi.org/10.1039/d3ee01776a https://pubs.rsc.org/en/content/articlelanding/2023/ee/d3ee01776a/unauth

Published in , 2025

Optimization of multi-mode classification for process monitoring

Published in Frontiers in Chemical Engineering, 2022

Process monitoring seeks to identify anomalous plant operating states so that operators can take the appropriate actions for recovery. Instrumental to process monitoring is the labeling of known operating states in historical data, so that departures from these states can be identified. This task can be challenging and time consuming as plant data is typically high dimensional and extensive. Moreover, automation of this procedure is not trivial since ground truth labels are often unavailable. In this contribution, this problem is approached as a multi-mode classification one, and an automatic framework for labeling using unsupervised Machine Learning (ML) methods is presented. The implementation was tested using data from the Tennessee Eastman Process and an industrial pyrolysis process. A total of 9 ML ensembles were included. Hyperparameters were optimized using a multi-objective evolutionary optimization algorithm. Unsupervised clustering metrics (silhouette score, Davies-Bouldin index, and Calinski-Harabasz Index) were investigated as candidates for objective functions in the optimization implementation. Results show that ensembles and hyperparameter selection can be aided by multi-objective optimization. It was found that Silhouette score and Davies-Bouldin index are strong predictions of the ensemble’s performance and can then be used to obtain good initial results for subsequent fault detection and fault diagnosis procedures.

Recommended citation: Zachary Webb, Miriam Nnadili, Estelle Seghers, Luis Briceno-Mena, José Romagnoli (2022). Optimization of Multi-Modal Classification for Process Monitoring. Frontiers in Chemical Engineering, 4, 2673-2718 https://www.frontiersin.org/articles/10.3389/fceng.2022.900083/full

Data mining and knowledge discovery in chemical processes: Effect of alternative processing techniques

Published in Data-Centric Engineering, 2022

Data mining and knowledge discovery (DMKD) can be a powerful tool in the analysis of chemical plant data. However, method selection and implementation can be a difficult task. In this contribution, all stages of the DMKD process are considered in detail, examining the fundamentals of various methods as well as practical insights on how to select and tune these methods. Both simulated and real plant data are analyzed to showcase the properties of data cleaning, sampling, scaling, dimensionality reduction, clustering, and cluster analysis, and their applicability in real-life scenarios. Also, a friendly graphic user interface to implement DMKD is presented.

Recommended citation: Briceno-Mena, L., Nnadili, M., Benton, M., & Romagnoli, J. (2022). Data mining and knowledge discovery in chemical processes: Effect of alternative processing techniques. Data-Centric Engineering, 3, E18. doi:10.1017/dce.2022.21 https://www.cambridge.org/core/journals/data-centric-engineering/article/data-mining-and-knowledge-discovery-in-chemical-processes-effect-of-alternative-processing-techniques/6E58B1133A9390177130A292422F4786

Electrochemical Pumping for Challenging Hydrogen Separations

Published in ACS Energy Letters, 2022

Conventional hydrogen separations from reformed hydrocarbons often deploy a water gas shift (WGS) reactor to convert CO to CO2, followed by adsorption processes to achieve pure hydrogen. The purified hydrogen is then fed to a compressor to deliver hydrogen at high pressures. Electrochemical hydrogen pumps (EHPs) featuring proton-selective polymer electrolyte membranes (PEMs) represent an alternative separation platform with fewer unit operations because they can simultaneously separate and compress hydrogen continuously. In this work, a high-temperature PEM (HT-PEM) EHP purified hydrogen to 99.3%, with greater than 85% hydrogen recovery for feed mixtures containing 25–40% CO. The ion-pair HT-PEM and phosphonic acid ionomer binder enabled the EHP to be operated in the temperature range from 160 to 220 °C. The ability to operate the EHP at an elevated temperature allowed the EHP to purify hydrogen from gas feeds with large CO contents at 1 A cm–2. Finally, the EHP with the said materials displayed a small performance loss of 12 μV h–1 for purifying hydrogen from syngas for 100 h at 200 °C.

Recommended citation: TGokul Venugopalan, Deepra Bhattacharya, Evan Andrews, Luis Briceno-Mena, José Romagnoli, John Flake, and Christopher G. Arges, ACS Energy Letters 2022 7 (4), 1322-1329. https://pubs.acs.org/doi/full/10.1021/acsenergylett.1c02853

PemNet: A Transfer Learning-Based Modeling Approach of High-Temperature Polymer Electrolyte Membrane Electrochemical Systems

Published in Industrial & Engineering Chemistry Research, 2022

Widespread adoption of high-temperature electrochemical systems such as polymer electrolyte membrane fuel cells (HT-PEMFCs) requires models and computational tools for accurate optimization and guiding new materials for enhancing fuel cell performance and durability. While robust and better suited for extrapolation, knowledge-based modeling has limitations as it is time-consuming and requires information about the system that is not always available (e.g., material properties and interfacial behavior between different materials). Data-driven modeling, on the other hand, is easier to implement but often necessitates large datasets that could be difficult to obtain. In this contribution, knowledge-based modeling and data-driven modeling are combined by implementing a few-shot learning (FSL) approach. A knowledge-based model originally developed for a HT-PEMFCs was used to generate simulated data (887,735 points) and used to pretrain a neural network source model tuned via a genetic algorithm-based AutoML. Then, experimental datasets from HT-PEMFCs with different materials and operating conditions (∼50 points each) were used to train six target models via FSL. Models for the unseen data reached high accuracies in all cases (rRMSE < 10%).

Recommended citation: Luis A. Briceno-Mena, José A. Romagnoli, and Christopher G. Arges Industrial & Engineering Chemistry Research 2022 61 (9), 3350-3357 DOI: 10.1021/acs.iecr.1c04237. https://pubs.acs.org/doi/10.1021/acs.iecr.1c04237

Electrochemical Pumping for Challenging Hydrogen Separations

Published in ACS Energy Letters, 2022

Recommended citation: Gokul Venugopalan, Deepra Bhattacharya, Evan Andrews, Luis Briceno-Mena, José Romagnoli, John Flake, and Christopher G. Arges ACS Energy Lett. 2022, 7, XXX, 1322–1329 DOI: 10.1021/acsenergylett.1c02853. https://pubs.acs.org/doi/full/10.1021/acsenergylett.1c02853

Machine learning for guiding high-temperature PEM fuel cells with greater power density

Published in Patterns, 2021

A multilevel modeling and data analysis framework were constructed for high-temperature polymer electrolyte membrane fuel cells (HT-PEMFCs) and their electrolyte materials. The framework employed Machine Learning tools (e.g., support vector regression, dimension reduction, and clustering) that seamlessly linked materials characteristics with fuel cell device performance and design, allowing for the accelerated discovery of material property attributes and fuel cell operating parameters and configurations that achieve greater power density and efficiency while co-currently addressing costs.

Recommended citation: Luis A. Briceno-Mena, Gokul Venugopalan, José A. Romagnoli, Christopher G. Arges. (2021). "Machine learning for guiding high-temperature PEM fuel cells with greater power density, " Patterns. Volume 2, Issue 2, 2021, 100187, ISSN 2666-3899, https://doi.org/10.1016/j.patter.2020.100187. https://doi.org/10.1016/j.patter.2020.100187

Heterogeneous Catalytic Ozonation of Phenol over Iron-based Catalysts in a Trickle Bed Reactor

Published in Ingenieria, 2020

The use of continuous reactors for heterogeneous catalytic ozonation is yet to be investigated in order to develop a viable technology for industrial applications. This paper presents hydrodynamic and degradation studies on the use of a co-current down flow trickle bed reactor for heterogeneous catalytic ozonation of phenol (as model pollutant) over Fe-Diatomite pellets and Fe-coated glass beads. It was found that the reactor can operate under trickle or pulsing flow regimes, promoting mass transfer augmentation. Residence time distribution data, fitted with n-CSTR and axial dispersion (ADM) models, showed low axial dispersion and high flow distribution. Just the Fe-diatomite pellets showed important phenol adsorption (16 %). Degradation experiments demonstrated that phenol conversion was substantial when using both catalysts, up to 19,7 % pollutant conversion with liquid-phase space times of just 6 s. Compared to direct ozonation, the use of the Fe-diatomite pellets and Fe-coated glass beads enhanced the reactor performance by 48 % and 23 % respectively. It was confirmed that mass transfer is an important factor that restricts this reaction system performance; consequently, further improvement in mass transport rate is necessary for system optimization.

Recommended citation: Briceño, L., Durán, J. E. (2020). "Heterogeneous Catalytic Ozonation of Phenol over Iron-based Catalysts in a Trickle Bed Reactor." Ingenieria. 20(July-December 2020). https://revistas.ucr.ac.cr/index.php/ingenieria/article/view/39236/41886