Towards estimation of CO2 adsorption on highly porous MOF-based adsorbents using gaussian process regression approach

Machine learning-powered estimation of malachite green photocatalytic degradation with NML-BiFeO3 composites

This study explores the potential of photocatalytic degradation using novel NML-BiFeO3 (noble metal-incorporated bismuth ferrite) compounds for eliminating malachite green (MG) dye from wastewater. The effectiveness of various Gaussian process regression (GPR) models in predicting MG degradation is investigated. Four GPR models (Matern, Exponential, Squared Exponential, and Rational Quadratic) were employed to analyze a dataset of 1200 observations encompassing various experimental conditions. The models have considered ten input variables, including catalyst properties, solution characteristics, and operational parameters. The Exponential kernel-based GPR model achieved the best performance, with a near-perfect R2 value of 1.0, indicating exceptional accuracy in predicting MG degradation. Sensitivity analysis revealed process time as the most critical factor influencing MG degradation, followed by pore volume, catalyst loading, light intensity, catalyst type, pH, anion type, surface area, and humic acid concentration. This highlights the complex interplay between these factors in the degradation process. The reliability of the models was confirmed by outlier detection using William's plot, demonstrating a minimal number of outliers (66-71 data points depending on the model). This indicates the robustness of the data utilized for model development. This study suggests that NML-BiFeO3 composites hold promise for wastewater treatment and that GPR models, particularly Matern-GPR, offer a powerful tool for predicting MG degradation. Identifying fundamental catalyst properties can expedite the application of NML-BiFeO3, leading to optimized wastewater treatment processes. Overall, this study provides valuable insights into using NML-BiFeO3 compounds and machine learning for efficient MG removal from wastewater.

Integrative analysis of multi machine learning models for tetracycline photocatalytic degradation with MOFs in wastewater treatment

This study focuses on the utilization of connectionist models, specifically Independent Component Analysis (ICA), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Genetic Algorithm-Particle Swarm Optimization (GAPSO) integrated with a least-squares support vector machine (LSSVM) to forecast the degradation of tetracycline (TC) through photocatalysis using Metal-Organic Frameworks (MOFs). The primary objective of this study was to evaluate the viability and precision of these connectionist models in estimating the efficiency of TC degradation, particularly within the context of wastewater treatment. The input parameters for these models cover essential MOF characteristics, such as pore size and surface area, along with critical operational factors, such as pH, TC concentration, catalyst dosage, and illumination duration, all of which are linked to the photocatalytic performance of MOFs. Sensitivity analysis revealed that the illumination duration is the primary influencer of TC photodegradation with MOF photocatalysts, while the MOFs' surface area is the second crucial parameter shaping the efficiency and dynamics of the TC-MOF photocatalytic system. The developed LSSVM models display impressive predictive capabilities, effectively forecasting the experimental degradation of TC with high accuracy. Among these models, the GAPSO-LSSVM model excels as the top performer, achieving notable evaluation metrics, including STD, RMSE, MSE, MRE, and R2 at values of 3.09, 3.42, 11.71, 5.95, and 0.986, respectively. In comparison, the PSO-LSSVM, ICA-LSSVM, and GA-LSSVM models yield mean relative errors of 6.18%, 7.57%, and 11.37%, respectively. These outcomes highlight the exceptional predictive capabilities of the GAPSO-LSSVM model, solidifying its position as the most accurate and dependable model for predicting TC photodegradation in this study. This study contributes to advancing photocatalytic research and effectively reinforces the importance of leveraging machine learning methodologies for tackling environmental challenges.

Augmented Reality for Enhanced Visualization of MOF Adsorbents

Augmented reality (AR) is an emerging technique used to improve visualization and comprehension of complex 3D materials. This approach has been applied not only in the field of chemistry but also in real estate, physics, mechanical engineering, and many other areas. Here, we demonstrate the workflow for an app-free AR technique for visualization of metal-organic frameworks (MOFs) and other porous materials to investigate their crystal structures, topology, and gas adsorption sites. We think this workflow will serve as an additional tool for computational and experimental scientists working in the field for both research and educational purposes.

DFT-Quality Adsorption Simulations in Metal-Organic Frameworks Enabled by Machine Learning Potentials

Nanoporous materials such as metal-organic frameworks (MOFs) have been extensively studied for their potential for adsorption and separation applications. In this respect, grand canonical Monte Carlo (GCMC) simulations have become a well-established tool for computational screenings of the adsorption properties of large sets of MOFs. However, their reliance on empirical force field potentials has limited the accuracy with which this tool can be applied to MOFs with challenging chemical environments such as open-metal sites. On the other hand, density-functional theory (DFT) is too computationally demanding to be routinely employed in GCMC simulations due to the excessive number of required function evaluations. Therefore, we propose in this paper a protocol for training machine learning potentials (MLPs) on a limited set of DFT intermolecular interaction energies (and forces) of CO2 in ZIF-8 and the open-metal site containing Mg-MOF-74, and use the MLPs to derive adsorption isotherms from first principles. We make use of the equivariant NequIP model which has demonstrated excellent data efficiency, and as such an error on the interaction energies below 0.2 kJ mol-1 per adsorbate in ZIF-8 was attained. Its use in GCMC simulations results in highly accurate adsorption isotherms and heats of adsorption. For Mg-MOF-74, a large dependence of the obtained results on the used dispersion correction was observed, where PBE-MBD performs the best. Lastly, to test the transferability of the MLP trained on ZIF-8, it was applied to ZIF-3, ZIF-4, and ZIF-6, which resulted in large deviations in the predicted adsorption isotherms and heats of adsorption. Only when explicitly training on data for all ZIFs, accurate adsorption properties were obtained. As the proposed methodology is widely applicable to guest adsorption in nanoporous materials, it opens up the possibility for training general-purpose MLPs to perform highly accurate investigations of guest adsorption.

Theoretical investigation of CO2 capture in the MIL-88 series: effects of organic linker modification

CO₂ capture is a crucial strategy to mitigate global warming and protect a sustainable environment. Metal-organic frameworks with large surface area, high flexibility, and reversible adsorption and desorption of gases are good candidates for CO₂ capture. Among the synthesized metal-organic frameworks, the MIL-88 series has attracted our attention due to their excellent stability. However, a systematic investigation of CO₂ capture in the MIL-88 series with different organic linkers is not available. Therefore, we clarified the topic via two sections: (1) elucidate physical insights into the CO₂@MIL-88 interaction by van der Waals-dispersion correction density functional theory calculations, and (2) quantitatively study the CO₂ capture capacity by grand canonical Monte Carlo simulations. We found that the 1π_g, 2σ_u/1π_u, and 2σ_g peaks of the CO₂ molecule and the C and O p orbitals of the MIL-88 series are the predominant contributors to the CO₂@MIL-88 interaction. The MIL-88 series, i.e., MIL-88A, B, C, and D, has the same metal oxide node but different organic linkers: fumarate (MIL-88A), 1,4-benzene-dicarboxylate (MIL-88B), 2,6-naphthalene-dicarboxylate (MIL-88C), and 4,4'-biphenyl-dicarboxylate (MIL-88D). The results exhibited that fumarate should be the best replacement for both the gravimetric and volumetric CO₂ uptakes. We also pointed out a proportional relationship between the capture capacities with electronic properties and other parameters.

Estimating hydrogen absorption energy on different metal hydrides using Gaussian process regression approach

Hydrogen is a promising alternative energy source due to its significantly high energy density. Also, hydrogen can be transformed into electricity in energy systems such as fuel cells. The transition toward hydrogen-consuming applications requires a hydrogen storage method that comes with pack hydrogen with high density. Among diverse methods, absorbing hydrogen on host metal is applicable at room temperature and pressure, which does not provide any safety concerns. In this regard, AB₂ metal hydride with potentially high hydrogen density is selected as an appropriate host. Machine learning techniques have been applied to establish a relationship on the effect of the chemical composition of these hosts on hydrogen storage. For this purpose, a data bank of 314 data point pairs was used. In this assessment, the different A-site and B-site elements were used as the input variables, while the hydrogen absorption energy resulted in the output. A robust Gaussian process regression (GPR) approach with four kernel functions is proposed to predict the hydrogen absorption energy based on the inputs. All the GPR models' performance was quite excellent; notably, GPR with Exponential kernel function showed the highest preciseness with R², MRE, MSE, RMSE, and STD of 0.969, 2.291%, 3.909, 2.501, and 1.878, respectively. Additionally, the sensitivity of analysis indicated that ZR, Ti, and Cr are the most demining elements in this system.

Intelligent route to design efficient CO2 reduction electrocatalysts using ANFIS optimized by GA and PSO

Recently, electrochemical reduction of CO₂ into value-added fuels has been noticed as a promising process to decrease CO₂ emissions. The development of such technology is strongly depended upon tuning the surface properties of the applied electrocatalysts. Considering the high cost and time-consuming experimental investigations, computational methods, particularly machine learning algorithms, can be the appropriate approach for efficiently screening the metal alloys as the electrocatalysts. In doing so, to represent the surface properties of the electrocatalysts numerically, d-band theory-based electronic features and intrinsic properties obtained from density functional theory (DFT) calculations were used as descriptors. Accordingly, a dataset containg 258 data points was extracted from the DFT method to use in machine learning method. The primary purpose of this study is to establish a new model through machine learning methods; namely, adaptive neuro-fuzzy inference system (ANFIS) combined with particle swarm optimization (PSO) and genetic algorithm (GA) for the prediction of *CO (the key intermediate) adsorption energy as the efficiency metric. The developed ANFIS-PSO and ANFIS-GA showed excellent performance with RMSE of 0.0411 and 0.0383, respectively, the minimum errors reported so far in this field. Additionally, the sensitivity analysis showed that the center and the filling of the d-band are the most determining parameters for the electrocatalyst surface reactivity. The present study conveniently indicates the potential and value of machine learning in directing the experimental efforts in alloy system electrocatalysts for CO₂ reduction.

Dramatic Enhancement of Rare-Earth Metal-Organic Framework Stability Via Metal Cluster Fluorination

Rare-earth polynuclear metal-organic frameworks (RE-MOFs) have demonstrated high durability for caustic acid gas adsorption and separation based on gas adsorption to the metal clusters. The metal clusters in the RE-MOFs traditionally contain RE metals bound by μ₃-OH groups connected via organic linkers. Recent studies have suggested that these hydroxyl groups could be replaced by fluorine atoms during synthesis that includes a fluorine-containing modulator. Here, a combined modeling and experimental study was undertaken to elucidate the role of metal cluster fluorination on the thermodynamic stability, structure, and gas adsorption properties of RE-MOFs. Through systematic density-functional theory calculations, fluorinated clusters were found to be thermodynamically more stable than hydroxylated clusters by up to 8-16 kJ/mol per atom for 100% fluorination. The extent of fluorination in the metal clusters was validated through a ¹⁹F NMR characterization of 2,5-dihydroxyterepthalic acid (Y-DOBDC) MOF synthesized with a fluorine-containing modulator. ¹⁹F magic-angle spinning NMR identified two primary peaks in the isotropic chemical shift (δ_iso) spectra located at -64.2 and -69.6 ppm, matching calculated ¹⁹F NMR δ_iso peaks at -63.0 and -70.0 ppm for fluorinated systems. Calculations also indicate that fluorination of the Y-DOBDC MOF had negligible effects on the acid gas (SO₂, NO₂, H₂O) binding energies, which decreased by only ∼4 kJ/mol for the 100% fluorinated structure relative to the hydroxylated structure. Additionally, fluorination did not change the relative gas binding strengths (SO₂ > H₂O > NO₂). Therefore, for the first time the presence of fluorine in the metal clusters was found to significantly stabilize RE-MOFs without changing their acid-gas adsorption properties.

Stiffness estimation of planar spiral spring based on Gaussian process regression

Planar spiral spring is important for the dimensional miniaturisation of motor-based elastic actuators. However, when the stiffness calculation of the spring arm is based on simple beam bending theory, the results possess substantial errors compared with the stiffness obtained from finite-element analysis (FEA). It deems that the errors arise from the spiral length term in the calculation formula. Two Gaussian process regression models are trained to amend this term in the stiffness calculation of spring arm and complete spring. For the former, 216 spring arms' data sets, including different spiral radiuses, pitches, wrap angles and the stiffness from FEA, are employed for training. The latter engages 180 double-arm springs' data sets, including widths instead of wrap angles. The simulation of five spring arms and five planar spiral springs with arbitrary dimensional parameters verifies that the absolute values of errors between the predicted stiffness and the stiffness from FEA are reduced to be less than 0.5% and 2.8%, respectively. A planar spiral spring for a powered ankle-foot prosthesis is designed and manufactured to verify further, of which the predicted value possesses a 3.25% error compared with the measured stiffness. Therefore, the amendment based on the prediction of trained models is available.

An insight into tetracycline photocatalytic degradation by MOFs using the artificial intelligence technique

Tetracyclines (TCs) have been extensively used for humans and animal diseases treatment and livestock growth promotion. The consumption of such antibiotics has been ever-growing nowadays due to various bacterial infections and other pathologic conditions, resulting in more discharge into the aquatic environments. This brings threats to ecosystems and human bodies. Up to now, several attempts have been made to reduce TC amounts in the wastewater, among which photocatalysis, an advanced oxidation process, is known as an eco-friendly and efficient technology. In this regard, metal organic frameworks (MOFs) have been known as the promising materials as photocatalysts. Thus, studying TC photocatalytic degradation by MOFs would help scientists and engineers optimize the process in terms of effective parameters. Nevertheless, the costly and time-consuming experimental methods, having instrumental errors, encouraged the authors to use the computational method for a more comprehensive assessment. In doing so, a wide-ranging databank including 374 experimental data points was gathered from the literature. A powerful machine learning method of Gaussian process regression (GPR) model with four kernel functions was proposed to estimate the TC degradation in terms of MOFs features (surface area and pore volume) and operational parameters (illumination time, catalyst dosage, TC concentration, pH). The GPR models performed quite well, among which GPR-Matern model shows the most accurate performance with R², MRE, MSE, RMSE, and STD of 0.981, 12.29, 18.03, 4.25, and 3.33, respectively. In addition, an analysis of sensitivity was carried out to assess the effect of the inputs on the TC photodegradation by MOFs. It revealed that the illumination time and the surface area play a significant role in the decomposition activity.

On the Investigation of Effective Factors on Electronic Structure Properties of Transition Metal Complexes: Robust Modeling Using GPR Approach

Materials discovery is usually done using high-throughput computational screening. The use of costly and complex direct density functional theory (DFT) simulation methods has been commonly used to determine subtle trends in spin-state ordering and inorganic bonding of inorganic materials and, in general, to predict the electronic structure properties of transition metal complexes. A Gaussian process regression (GPR) framework consisting of four kernel functions is introduced for spin-state splitting estimation through inorganic chemistry-appropriate empirical inputs. To this end, the present study reviewed an extensive range of data values from earlier works. According to statistical analysis, the GPR model showed very good performance. The coefficients of determination were calculated to be 0.986 for the exponential and Matern kernel functions, suggesting the highest predictive power of these methods. Moreover, the sensitivity of output to inputs was measured. Artificial intelligence (AI) helped accurately predict the target values through various input ranges.

A novel composite of ionic liquid-containing polymer and metal-organic framework as an efficient catalyst for ultrasonic-assisted Knoevenagel condensation

1-Butyl-3-vinylimidazolium chloride was synthesized and polymerized with acrylamide to furnish an ionic liquid-containing polymer, which was then used for the formation of a composite with iron-based metal-organic framework. The resultant composite was characterized with XRD, TGA, FE-SEM, FTIR, EDS and elemental mapping analyses and its catalytic activity was appraised for ultrasonic-assisted Knoevenagel condensation. The results confirmed that the prepared composite could promote the reaction efficiently to furnish the corresponding products in high yields in very short reaction times. Moreover, the composite exhibited high recyclability up to six runs. It was also established that the activity of the composite was higher compared to pristine metal-organic framework or polymer.