Abstract
In the portfolio of technologies available for net zero-enabling solutions, such as carbon capture and low-carbon production of hydrogen, membrane-based gas separation is a sustainable alternative to energy-intensive processes, such as solvent-based absorption or cryogenic distillation. Detailed knowledge of membrane materials performance in wide operative ranges is a necessary prerequisite for the design of efficient membrane processes. With the increasing popularization of data-driven methods in natural sciences and engineering, the investigation of their potential to support materials and process design for gas separation with membranes has received increasing attention, as it can help compact the lab-to-market cycle. In this work we review several machine learning (ML) strategies for the estimation of the gas separation performance of polymer membranes. New hybrid modelling strategies, in which ML complements physics-based models and simulation methods, are also discussed. Such strategies can enable the fast screening of large databases of existing materials for a specific separation, as well as assist in de-novo materials design. We conclude by highlighting the challenges and future directions envisioned for the ML-assisted design and optimization of membrane materials and processes for traditional, as well as new, membrane separations.
Funding source: Royal Society of Edinburgh
Award Identifier / Grant number: 2915
About the authors
Eleonora Ricci is a Marie Skłodowska-Curie postdoctoral fellow at the National Centre for Scientific Research “Demokritos”, Athens, Greece, working on the application of artificial intelligence methods in multiscale molecular simulation frameworks. She obtained an MSc in Chemical Engineering from the University of Bologna, Italy, and earned a PhD in 2020 from the same university. Her research interests include the thermodynamic, molecular, and data-driven modelling of polymeric materials, for the study of multicomponent phase equilibria and mass transfer in polymeric membranes.
Maria Grazia De Angelis is a professor of Institute for Materials and Processes, School of Engineering, University of Edinburgh, EH9 3FB Edinburgh, ScotlandDepartment of Civil, Chemical, Environmental and Materials Engineering (DICAM), Alma Mater Studiorum – University of Bologna, Bologna, Italy. Her expertise is in the multiscale modelling of the physical phenomena associated to fluid transport in solid polymeric materials, with application in membrane separation, fluid purification and packaging design. She elucidated the mechanism of sorption of pure and mixed gases in glassy, composite and semicrystalline membranes, and is pursuing the integration of different theories to enhance the predictive ability of modelling tools.
-
Research ethics: Not applicable.
-
Author contributions: The authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: The authors declare no conflicts of interest regarding this article.
-
Research funding: The present work is partly supported by the RSE Research Workshop Autumn 2022 Award 2915 “A Machine Learning-Aided Modeling Platform for the design of Hydrogen-Ready materials”.
-
Data availability: Not applicable.
Appendix A: Short introduction to machine learning
In general terms, machine learning is a subset of artificial intelligence, and it involves the development of models that learn to make predictions based on the data used to train them. There is an overlap in the classification between ML methods and non-ML statistical modeling techniques, for example in the case of linear regression analysis. Sometimes the “scale” of the task, i.e. the size of the dataset considered, is used to discriminate between ML and non-ML applications.
ML methods can be divided into three main categories: i) supervised learning, ii) unsupervised learning, and iii) reinforcement learning.
Supervised learning involves training a model on a dataset where, for a set of given input properties (features) the values of the output properties (labels) are known. The model is trained by minimizing a penalty function (also called loss function) related to the accuracy of the predicted labels in comparison to the known values. Afterwards, the trained model is used to predict the labels of input features for which the labels were unknown.
Supervised learning methods are further split into a) classification, used for predicting categorical values, and b) regression, used for predicting continuous values. Several ML methods can be applied both for classification and regression tasks. Supervised learning regression methods are the most frequently applied for the prediction of gas transport properties in membrane materials and include neural networks, decision trees, logistic regression, Gaussian processes, support vector machines, and several variants of these methods, which are defined in the Table below.
Unsupervised learning involves training a model on an unlabeled dataset, where the desired output is not known. The aim is to identify patterns or structure in the data without a predefined categorization. Examples of unsupervised learning are clustering techniques and dimensionality reduction methods, such as principal component analysis.
Reinforcement learning also does not make use of labeled data, but rather it involves training a model by providing it with a reward signal based on its performance, and the model learns to make decisions that maximize the reward. An example of foreseeable application for membrane separations would be process control: the model would receive a reward signal based on the target set point of the process and it would learn to adjust the operating conditions to minimize deviations from the set point, thus maximizing the reward signal. To the best of our knowledge, no applications of reinforcement learning methods were reported concerning gas transport properties predictions.
A recent short communication by Tayyebi et al. (2022), provides some useful model selection guidelines in their supporting information.
Glossary and references for machine learning methods and terminology
Name | Acronym | Definition | References |
---|---|---|---|
Activation function | – | In artificial neural networks, activation functions are applied to the output of a neuron to modulate the response depending on the input values. E.g., different activation functions can be used to threshold the output of a neuron or introduce nonlinear relations between input and output. | Russell and Norvig (2020) |
Agglomerative clustering | AC | Clustering method in which initially each point forms a separate cluster and the clusters are progressively merged, based on the chosen similarity measure. | Maimon and Rokach (2006) |
(Artificial) neural network | (A)NN | A numeric-mathematical construction that can model complex non-linear relationships. Usually consisting of input, hidden (in the case of “deep” models), and output layers, each with a suitable number of nodes. The values assumed by each node are calculated from the connections with the nodes in the previous layer, expressed by weight and bias parameters that are tuned during the training of the model. The application of activation functions between each layers introduces nonlinearity into the model. | Russell and Norvig (2020) |
Adaptive neuro-fuzzy inference system | ANFIS | Model consisting of a series of fuzzy rules and appropriate membership functions, which, after initialization, are tuned to minimize the output error. | Jang (1993) |
Accelerated particle swarm optimization | APSO | Simplified variant of the PSO algorithm that does not use the “velocity” concept, thus speeding up convergence. | Yang et al. (2012) |
Bayesian linear regression | BLR | Modeling method that aims at determining the posterior probability distribution of the model parameters conditional upon the training inputs and outputs, which are typically sampled from a normal distribution. | MacKay (1992) |
Chaos-enhanced accelerated particle swarm optimization | CEAPSO | Improvement of the APSO algorithm that employs the Lorenze equations to generate a chaotic sequence for tuning the acceleration coefficients of the APSO method. | Ru-Ting and Xing-Yuan (2015) |
Committee machine intelligent system | CMIS | Model combining the responses of multiple ML models into a single response. | Russell and Norvig (2020) |
Cross validation | – | Test of the generalization capacity of an ML model, which consists of splitting the data into n portions and performing n tests in which every portion of the data is in turn left out of the training set and used for testing the obtained model. | Russell and Norvig (2020) |
Decision tree/Regression trees | DT/RT | Non-parametric supervised learning algorithm, which is utilized for both classification and regression tasks. It has a hierarchical, tree structure, consisting of a root node, branches and internal nodes, which are created to learn the paths between input and output. Decision trees where the target variable can take continuous values are called regression trees. | Russell and Norvig (2020) |
Double-population particle swarm optimization based on diffusion theory | DP-DT-PSO | Variation of the PSO algorithm, in which the iterative update function is inspired by diffusion theory. | Mengshan et al. (2017a) |
Extremely randomized trees | ERT | Variation of the random forest (RF) algorithm, which does not result in deterministic splits of the decision trees (DT). | Geurts et al. (2006) |
Fuzzy clustering | FC | Unsupervised clustering method in which each data point can belong to more than one cluster, with a membership grades value between 0 and 1 for each cluster. | Abiyev (2011) |
Fuzzy neural network | FNN | Neural network models that tune fuzzy rules and membership functions automatically from the input data. | Russell and Norvig (2020) |
Fuzzy logic | FL | A form of logic in which the value of “true” can correspond to any real number between 0 and 1, unlike Boolean logic, where only two integer values are considered, 0 or 1. | Zadeh (1988) |
Gaussian process regression | GPR | GPR is a probabilistic approach to regression analysis, in which the distribution of all possible functions relating input and output variables that could explain the observed data is considered. This distribution is a Gaussian process, which is defined by a mean function and a covariance function, which allows calculating the most likely values and associated uncertainties for the model predictions. | Rasmussen and Williams (2005) |
Genetic algorithms | GA | A type of optimization algorithm inspired by the process of natural selection and evolution: a population of candidate solutions is initially generated randomly and iteratively evolved over many generations, using a fitness function. The fittest individuals are then selected for reproduction, and combined to create new candidate solutions, iterating until a satisfactory solution is found or a maximum number of generations is reached. | Koza (1994) |
Genetic function approximation | GFA | Symbolic regression performed through genetic algorithm optimization. | Rogers (1999) |
Genetic programming | GP | A technique of creating programs fit for a specific task by “evolving” an initial random population through crossover and mutation of the population members to create new “generations” until the required accuracy criterion is satisfied. | Koza (1994) |
Hyperparameter | – | A parameter used to specify a characteristic of the ML model used (e.g., number of neurons in a neural network) or of the optimization algorithm (e.g., the learning rate). | Yang and Shami (2020) |
k-means clustering | KM | Unsupervised ML method for the partitioning of a dataset. Each data point is assigned to the cluster with the nearest mean distance. | Russell and Norvig (2020) |
k-harmonic means clustering | KHM | Center-based clustering algorithm considering the harmonic means of the distances from each data point to the centers as components to its object function. | Zhang et al. (2001) |
Kernel Ridge regression | KRR | Combination of Ridge regression (linear least squares with l2-norm regularization) with the kernel trick. The kernel trick is used to measure the similarity between two points in a higher-dimensional space, without actually computing the coordinates of the points in that space, thus implicitly mapping data from a low to a higher-dimensional space, without actually computing the transformed feature vectors. | Burges (1998) |
Least square SVM | LSSVM | While SVM separation hyperplanes are obtained through solving systems of inequality constraints, the least squares formulation involves equality constraints only. | Suykens and Vandewalle (1999) |
Multivariate imputation by chained equations | MICE | Multiple imputation method used to complete missing data values in a data set, under the assumption that they are missing at random. | Buuren and Groothuis-Oudshoorn (2011) |
Natural language processsing | NLP | AI techniques that focus on enabling computers to understand, interpret, and generate human language, leveraging elements of computer science and linguistics to analyze and process large amounts of natural language data. Applications of NLP include machine translation, speech recognition, and text mining. | |
Overfitting | – | Creation of a model that represents the training data with a very high accuracy but fails to generalize to unseen data, often because of the presence of too many model parameters with respect to training data set size. | Russell and Norvig (2020) |
Principal component analysis | PCA | Statistical method that applies an orthogonal transformation to convert the feature set into a set of linearly uncorrelated variables (principal components). Often applied for dimensionality reduction. | Russell and Norvig (2020) |
Particle swarm optimization | PSO | Group evolution optimization algorithm to solve problems that can be represented as a point or surface in a multi-dimensional space. In PSO, a “particle” is designed as a potential solution in the search space. A particle updates its “speed” and “position” iteratively, until converging to a solution. | Kennedy and Eberhart (1995) |
Polynomial regression | PR | Form of regression in which the relationship between the independent variable and the dependent variable is modelled as an nth degree polynomial function. | Kleinbaum et al. (2013) |
Random forest | RF | Ensemble model combining the prediction of multiple decision tree models. | Russell and Norvig (2020) |
Radial basis function neural network | RBFNN | NN model typically consisting of 3 layers: input, hidden, and output. In the hidden layer nodes, the input is transformed by the application of radial basis function (RBF). Training is performed in two steps. At first the parameters the RBF are tuned (e.g., centers and widths, if a Gaussian function is used), for example using unsupervised clustering algorithms, and at a second stage the connection weights between the hidden layer and the output layer are optimized. | Elanayar and Shin (1994) |
Stochastic gradient boosting | SGB | Improvement of the gradient boosting method, which consists of an ensemble of “weak” prediction models, typically decision trees, applied in sequence to minimize the residual error of the previous model. | Mason et al. (1999) |
Shapley additive explanations | SHAP | Calculation of the individual contribution of each feature to the predicted output of an ML model to increase the explainability of the results. | Lundberg and Lee (2017) |
Symbolic regression | SR | Search in the space of mathematical expressions to find the function that best fits a given dataset, both in terms of accuracy and simplicity. | Billard and Diday (2002) |
Support vector machines | SVM | Supervised learning models that construct hyperplanes in a multidimensional space to separate cases of different class labels. They can be applied to multiple continuous and categorical variables, thus allowing both classification and regression analysis. | Russell and Norvig (2020) |
Transfer learning | TL | Transfer learning consists of reusing parts of a previously trained model (usually trained on a large dataset) on a new model used for a different but similar problem, retraining only a smaller portion of the parameters on the dataset related to the new task. | Russell and Norvig (2020) |
Transformer | – | The transformer is a deep learning architecture used in natural language processing and other tasks, based on self-attention mechanisms. It is an embedding technique that converts input data, like words or sentences (or chemical formulas specifically), into dense numerical representations, capturing semantic meaning, and enabling efficient handling of long-range dependencies. | Vaswani et al. (2017) |
Glossary and references for gas separation with membranes terminology
Name | Definition | References |
---|---|---|
Diffusivity | The diffusion coefficient gives the rate at which dilute gas molecules can move through the membrane according to the diffusive mechanism and it is related to diffusive flux and transmembrane concentration gradient across the membrane of thickness
|
Matteucci et al. (2006) |
Permeability | The permeability coefficient (
|
Matteucci et al. (2006) |
Selectivity | Membrane selectivity to a gas pair (i, j) is defined as in other separation processes as the ratio between the enrichment of the most permeable component (i) from the feed to the permeate stream, versus the same value calculated for the less permeable component (j) (y
i
P
/y
i
F
)/(y
j
P
/y
j
F
). If the downstream side is at negligible pressure, this value coincides with the ratio between the permeability coefficient of i to that of j,
|
Matteucci et al. (2006) |
Solubility | Assuming that interphase equilibrium is valid at the membrane/gas contact surface, which is usually the case, solubility is given quantitatively by the equilibrium solubility coefficient of the gas in the polymer (
|
Matteucci et al. (2006) |
Solution-diffusion theory | If Fick’s law and interfacial equilibrium are valid, the definition of permeability together with the corresponding physical laws imply that that permeability coefficient (
|
Wijmans and Baker (2006) |
Swelling | Gas-induced swelling in gas separation membranes refers to the phenomenon where certain gases, when sorbed into the membrane material, cause it to expand or swell, leading to changes in its structure and properties, and affecting gas permeation and separation performance. It is quantified as relative polymer volume increase, ΔV/V0, at a given gas pressure or concentration. If mixing volume effects are negligible, the volume increase of the polymer phase is equal to the volume of pure liquid penetrant absorbed. | Matteucci et al. (2006) |
Acronyms
- CMS
-
carbon molecular sieve
- CPEs
-
carboxylated polyesters
- CPI
-
cardo-type poly-Imide
- DEA
-
diethanolamine
- HDPE
-
high-density polyethylene
- LDPE
-
low-density polyethylene
- NBR
-
nitrile butadiene rubber
- PB
-
polybutadiene
- PBS
-
polybutylene succinate
- PBSA
-
poly(butylene succinate-co-adipate)
- PC
-
polycarbonate
- PCP
-
polychloroprene
- PDMB
-
polydimethylbutadiene
- PDMS
-
polydimethylsiloxane
- PEM
-
polyethylmethacrylate
- PES
-
polyethersulfone
- PET
-
polyethylene terephthalate
- PIB
-
polyisobutylene
- PIP
-
cis-1,4-polyisoprene
- PLGA
-
poly(D,L-lactide-co-glycolide)
- PLLA
-
poly(L-lactide)
- PMP
-
poly (4-methyl-1-pentane)
- pNA
-
p-nitroaniline
- POM
-
polyoxymethylene
- POSS
-
octa-trimethylsiloxy polyhedral oligomeric silsesquioxane
- PP
-
polypropylene (PP)
- PPO
-
poly(2,6-dimethyl-1,4-phenylene ether) or poly(2,6-dimethyl-1,4-phenylene oxide)
- PS
-
polystyrene
- PSf
-
polysulfone
- PTFE
-
polytetrafluoroethylene
- PU
-
polyurethane
- PVAc
-
poly(vinyl acetate)
- PVC
-
polyvinylchloride
- PVLA
-
polyacetyl lactone
References
Abdollahi, S.A. and Ranjbar, S.F. (2023). Modeling the CO2 separation capability of poly(4-methyl-1-pentane) membrane modified with different nanoparticles by artificial neural networks. Sci. Rep. 13: 8812, https://doi.org/10.1038/s41598-023-36071-x.Search in Google Scholar PubMed PubMed Central
Abiyev, R.H. (2011). Fuzzy wavelet neural network based on fuzzy clustering and gradient techniques for time series prediction. Neural Comput. Appl. 20: 249–259, https://doi.org/10.1007/s00521-010-0414-4.Search in Google Scholar
Addis, B., Castel, C., Macali, A., Misener, R., and Piccialli, V. (2023). Data augmentation driven by optimization for membrane separation process synthesis. Comput. Chem. Eng. 177: 108342, https://doi.org/10.1016/j.compchemeng.2023.108342.Search in Google Scholar
Ahmad, A.L., Adewole, J.K., Leo, C.P., Ismail, S., Sultan, A.S., and Olatunji, S.O. (2015). Prediction of plasticization pressure of polymeric membranes for CO2 removal from natural gas. J. Membr. Sci. 480: 39–46, https://doi.org/10.1016/j.memsci.2015.01.039.Search in Google Scholar
Alesadi, A., Cao, Z., Li, Z., Zhang, S., Zhao, H., Gu, X., and Xia, W. (2022). Machine learning prediction of glass transition temperature of conjugated polymers from chemical structure. Cell Rep. Phys. Sci. 3: 100911, https://doi.org/10.1016/j.xcrp.2022.100911.Search in Google Scholar
Amamoto, Y. (2022). Data-driven approaches for structure-property relationships in polymer science for prediction and understanding. Polym. J. 54: 957–967, https://doi.org/10.1038/s41428-022-00648-6.Search in Google Scholar
Asghari, M., Dashti, A., Rezakazemi, M., Jokar, E., and Halakoei, H. (2020). Application of neural networks in membrane separation. Rev. Chem. Eng. 36: 265–310, https://doi.org/10.1515/revce-2018-0011.Search in Google Scholar
Barnett, J.W., Bilchak, C.R., Wang, Y., Benicewicz, B.C., Murdock, L.A., Bereau, T., and Kumar, S.K. (2020). Designing exceptional gas-separation polymer membranes using machine learning. Sci. Adv. 6: 1–8, https://doi.org/10.1126/sciadv.aaz4301.Search in Google Scholar PubMed PubMed Central
Basdogan, Y., Pollard, D.R., Shastry, T., Carbone, M.R., Kumar, S.K., and Wang, Z.-G. (2023). Machine learning-guided discovery of polymer membranes for CO2 separation. chemRxiv 1–23, https://doi.org/10.26434/chemrxiv-2023-5h4s7.Search in Google Scholar
Bejagam, K.K., Lalonde, J., Iverson, C.N., Marrone, B.L., and Pilania, G. (2022). Machine learning for melting temperature predictions and design in polyhydroxyalkanoate-based biopolymers. J. Phys. Chem. B 126: 934–945, https://doi.org/10.1021/acs.jpcb.1c08354.Search in Google Scholar PubMed
Billard, L. and Diday, E. (2002). Symbolic regression analysis. In: Jajuga, K., Sokołowski, A., and Bock, H.H. (Eds.). Classification, clustering, and data analysis. studies in classification, data analysis, and knowledge organization. Springer, Berlin, Heidelberg.10.1007/978-3-642-56181-8_31Search in Google Scholar
Burges, C.J. (1998). A tutorial on support vector machines for pattern recognition. Data Min. Knowl. Discov. 2: 121–167, https://doi.org/10.1023/A:1009715923555.10.1023/A:1009715923555Search in Google Scholar
Buuren, S.V. and Groothuis-Oudshoorn, K. (2011). Mice: multivariate imputation by chained equations in R. J. Stat. Softw. 45, https://doi.org/10.18637/jss.v045.i03.Search in Google Scholar
Carvalho, F.S. and Braga, J.P. (2022). Physics informed neural networks applied to liquid state theory. J. Mol. Liq. 367: 120504, https://doi.org/10.1016/j.molliq.2022.120504.Search in Google Scholar
Cencer, M.M., Moore, J.S., and Assary, R.S. (2022). Machine learning for polymeric materials: an introduction. Polym. Int. 71: 537–542, https://doi.org/10.1002/pi.6345.Search in Google Scholar
Chen, C., Ye, W., Zuo, Y., Zheng, C., and Ong, S.P. (2019). Graph networks as a universal machine learning framework for molecules and crystals. Chem. Mater. 31: 3564–3572, https://doi.org/10.1021/acs.chemmater.9b01294.Search in Google Scholar
Chen, H., Zeng, M., Zhang, H., Chen, B., Guan, L., and Li, M. (2022). Prediction of carbon dioxide solubility in polymers based on adaptive particle swarm optimization and least squares support vector machine. ChemistrySelect 7: e202104447, https://doi.org/10.1002/slct.202104447.Search in Google Scholar
Chen, L., Pilania, G., Batra, R., Huan, T.D., Kim, C., Kuenneth, C., and Ramprasad, R. (2021). Polymer informatics: current status and critical next steps. Mater. Sci. Eng. R Rep. 144: 100595, https://doi.org/10.1016/j.mser.2020.100595.Search in Google Scholar
Cheng, X., Liao, Y., Lei, Z., Li, J., Fan, X., and Xiao, X. (2023). Multi-scale design of MOF-based membrane separation for CO2/CH4 mixture via integration of molecular simulation, machine learning and process modeling and simulation. J. Membr. Sci. 672: 121430, https://doi.org/10.1016/j.memsci.2023.121430.Search in Google Scholar
Chi, M., Gargouri, R., Schrader, T., Damak, K., Maâlej, R., and Sierka, M. (2021). Atomistic descriptors for machine learning models of solubility parameters for small molecules and polymers. Polym. 14: 26, https://doi.org/10.3390/polym14010026.Search in Google Scholar PubMed PubMed Central
Coker, D.T., Freeman, B.D., and Fleming, G.K. (1998). Modeling multicomponent gas separation using hollow-fiber membrane contactors. AIChE J. 44: 1289–1302, https://doi.org/10.1002/aic.690440607.Search in Google Scholar
Coley, C.W., Barzilay, R., Jaakkola, T.S., Green, W.H., and Jensen, K.F. (2017). Prediction of organic reaction outcomes using machine learning. ACS Cent. Sci. 3: 434–443, https://doi.org/10.1021/acscentsci.7b00064.Search in Google Scholar PubMed PubMed Central
Creton, B., Veyrat, B., and Klopffer, M.-H. (2022). Fuel sorption into polymers: experimental and machine learning studies. Fluid Phase Equilib. 556: 113403, https://doi.org/10.1016/j.fluid.2022.113403.Search in Google Scholar
Dashti, A., Harami, H.R., and Rezakazemi, M. (2018). Accurate prediction of solubility of gases within H2 -selective nanocomposite membranes using committee machine intelligent system. Int. J. Hydrogen Energy 43: 6614–6624, https://doi.org/10.1016/j.ijhydene.2018.02.046.Search in Google Scholar
Dashti, A., Raji, M., Azarafza, A., Rezakazemi, M., and Shirazian, S. (2020). Computational simulation of CO2 sorption in polymeric membranes using genetic programming. Arab. J. Sci. Eng. 45: 7655–7666, https://doi.org/10.1007/s13369-020-04783-1.Search in Google Scholar
Doan Tran, H., Kim, C., Chen, L., Chandrasekaran, A., Batra, R., Venkatram, S., Kamal, D., Lightstone, J.P., Gurnani, R., Shetty, P., et al.. (2020). Machine-learning predictions of polymer properties with Polymer Genome. J. Appl. Phys. 128: 171104, https://doi.org/10.1063/5.0023759.Search in Google Scholar
Dobbelaere, M.R., Plehiers, P.P., Van de Vijver, R., Stevens, C.V., and Van Geem, K.M. (2021). Machine learning in chemical engineering: strengths, weaknesses, opportunities, and threats. Engin. 7: 1201–1211, https://doi.org/10.1016/j.eng.2021.03.019.Search in Google Scholar
Doghieri, F. and Sarti, G.C. (1996). Nonequilibrium lattice fluids: a predictive model for the solubility in glassy polymers. Macromol. 29: 7885–7896, https://doi.org/10.1021/ma951366c.Search in Google Scholar
Durant, J.L., Leland, B.A., Henry, D.R., and Nourse, J.G. (2002). Reoptimization of MDL keys for use in drug discovery. J. Chem. Inf. Comput. Sci. 42: 1273–1280, https://doi.org/10.1021/ci010132r.Search in Google Scholar PubMed
Ebrahimi, S., Mollaiy-Berneti, S., Asadi, H., Peydayesh, M., Akhlaghian, F., and Mohammadi, T. (2016). PVA/PES-amine-functional graphene oxide mixed matrix membranes for CO2/CH4 separation: experimental and modeling. Chem. Eng. Res. Des. 109: 647–656, https://doi.org/10.1016/j.cherd.2016.03.009.Search in Google Scholar
Elanayar, V.T.S. and Shin, Y.C. (1994). Radial basis function neural network for approximation and estimation of nonlinear stochastic dynamic systems. IEEE Trans. Neural Network. 5: 594–603, https://doi.org/10.1109/72.298229.Search in Google Scholar PubMed
Fakhroleslam, M., Samimi, A., and Rezaei, R. (2016). Prediction of the effect of polymer membrane composition in a dry air humidification process via neural network modeling. Iran. J. Chem. Eng. 13: 73–83.Search in Google Scholar
Farno, E., Ghadimi, A., Kasiri, N., and Mohammadi, T. (2011). Separation of heavy gases from light gases using synthesized PDMS nano-composite membranes: experimental and neural network modeling. Sep. Purif. Technol. 81: 400–410, https://doi.org/10.1016/j.seppur.2011.08.008.Search in Google Scholar
Farno, E., Rezakazemi, M., Mohammadi, T., and Kasiri, N. (2014). Ternary gas permeation through synthesized pdms membranes: experimental and CFD simulation basedon sorption-dependent system using neural network model. Polym. Eng. Sci. 54: 215–226, https://doi.org/10.1002/pen.23555.Search in Google Scholar
Felton, K.C., Ben-Safar, H., and Lapkin, A.A. (2022). DeepGamma: A deep learning model for activity coefficient prediction. In: 1st Annual AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE).Search in Google Scholar
Friess, K., Izák, P., Kárászová, M., Pasichnyk, M., Lanč, M., Nikolaeva, D., Luis, P., and Jansen, J.C. (2021). A review on ionic liquid gas separation membranes. Membranes 11: 97, https://doi.org/10.3390/membranes11020097.Search in Google Scholar PubMed PubMed Central
Galinha, C.F. and Crespo, J.G. (2021). From black box to machine learning: a journey through membrane process modelling. Membranes 11: 574, https://doi.org/10.3390/membranes11080574.Search in Google Scholar PubMed PubMed Central
Galizia, M., Chi, W.S., Smith, Z.P., Merkel, T.C., Baker, R.W., and Freeman, B.D. (2017). 50th anniversary perspective: polymers and mixed matrix membranes for gas and vapor separation: a review and prospective opportunities. Macromolecules 50: 7809–7843, https://doi.org/10.1021/acs.macromol.7b01718.Search in Google Scholar
Geurts, P., Ernst, D., and Wehenkel, L. (2006). Extremely randomized trees. Mach. Learn. 63: 3–42, https://doi.org/10.1007/s10994-006-6226-1.Search in Google Scholar
Ghadimi, A., Sadrzadeh, M., and Mohammadi, T. (2010). Prediction of ternary gas permeation through synthesized PDMS membranes by using Principal Component Analysis (PCA) and fuzzy logic (FL). J. Membr. Sci. 360: 509–521, https://doi.org/10.1016/j.memsci.2010.05.055.Search in Google Scholar
Ghanem, B.S., Swaidan, R., Litwiller, E., and Pinnau, I. (2014). Ultra-microporous triptycene-based polyimide membranes for high-performance gas separation. Adv. Mater. 26: 3688–3692, https://doi.org/10.1002/adma.201306229.Search in Google Scholar PubMed
Golzar, K., Amjad-Iranagh, S., and Modarress, H. (2013). QSPR prediction of the solubility of CO2 and N2 in common polymers. Measurement 46: 4206–4225, https://doi.org/10.1016/j.measurement.2013.08.012.Search in Google Scholar
González-Miquel, M. and Díaz, I. (2021). Green solvent screening using modeling and simulation. Curr. Opin. Green Sustain. Chem. 29: 100469, https://doi.org/10.1016/j.cogsc.2021.100469.Search in Google Scholar
González, M.P. and Helguera, A.M. (2003). TOPS-MODE versus DRAGON descriptors to predict permeability coefficients through low-density polyethylene. J. Comput. Aided. Mol. Des. 17: 665–672, https://doi.org/10.1023/B:JCAM.0000017373.50020.41.10.1023/B:JCAM.0000017373.50020.41Search in Google Scholar PubMed
González, M.P., Helguera, A.M., and Dı́az, H.G. (2004). A TOPS-MODE approach to predict permeability coefficients. Polymer 45: 2073–2079, https://doi.org/10.1016/j.polymer.2003.12.014.Search in Google Scholar
Goubko, M., Miloserdov, O., Yampolskii, Y., Alentiev, A., and Ryzhikh, V. (2017). A novel model to predict infinite dilution solubility coefficients in glassy polymers. J. Polym. Sci. Part B Polym. Phys. 55: 228–244, https://doi.org/10.1002/polb.24263.Search in Google Scholar
Goubko, M.V., Miloserdov, O.A., Yampolskii, Y.P., and Ryzhikh, V.Y. (2019). Prediction of solubility parameters of light gases in glassy polymers on the basis of simulation of a short segment of a polymer chain. Polym. Sci. 61: 718–732, https://doi.org/10.1134/S0965545X19050067.Search in Google Scholar
Gu, G.H., Noh, J., Kim, I., and Jung, Y. (2019). Machine learning for renewable energy materials. J. Mater. Chem. A 7: 17096–17117, https://doi.org/10.1039/C9TA02356A.Search in Google Scholar
Guan, J., Huang, T., Liu, W., Feng, F., Japip, S., Li, J., Wang, X., and Zhang, S. (2022). Design and prediction of metal organic framework-based mixed matrix membranes for CO2 capture via machine learning. Cell Reports Phys. Sci. 3: 100864, https://doi.org/10.1016/j.xcrp.2022.100864.Search in Google Scholar
Gupta, S. and Li, L. (2022). The potential of machine learning for enhancing CO2 sequestration, storage, transportation, and utilization-based processes: a brief perspective. JOM 74: 414–428, https://doi.org/10.1007/s11837-021-05079-x.Search in Google Scholar
Gurras, A. and Gergidis, L.N. (2021). Modeling sorption and diffusion of alkanes, alkenes, and their mixtures in silicalite: from MD and GCMC molecular simulations to artificial neural networks. Adv. Theory Simul. 4: 1–12, https://doi.org/10.1002/adts.202000210.Search in Google Scholar
Hasnaoui, H., Krea, M., and Roizard, D. (2017). Neural networks for the prediction of polymer permeability to gases. J. Membr. Sci. 541: 541–549, https://doi.org/10.1016/j.memsci.2017.07.031.Search in Google Scholar
Hatakeyama-Sato, K. and Oyaizu, K. (2021). Generative models for extrapolation prediction in materials informatics. ACS Omega 6: 14566–14574, https://doi.org/10.1021/acsomega.1c01716.Search in Google Scholar PubMed PubMed Central
He, P., Liu, X., Gao, J., and Chen, W. (2020). DeBERTa: decoding-enhanced BERT with disentangled attention. Available at: <https://github.com/jsunn-y/PolymerGasMembraneML> [WWW Document, n.d].Search in Google Scholar
Hu, P., Jiao, Z., Zhang, Z., and Wang, Q. (2021). Development of solubility prediction models with ensemble learning. Ind. Eng. Chem. Res. 60: 11627–11635, https://doi.org/10.1021/acs.iecr.1c02142.Search in Google Scholar
Ismaeel, H., Gibson, D., Ricci, E., and De Angelis, M.G. (2024). Estimating gas sorption in polymeric membranes from the molecular structure: a machine learning based group contribution method for the non-equilibrium lattice fluid model (ML-GC-NELF). J Memb. Sci. 691: 122220. https://doi.org/10.1016/j.memsci.2023.122220.Search in Google Scholar
Jackson, N.E., Webb, M.A., and de Pablo, J.J. (2019). Recent advances in machine learning towards multiscale soft materials design. Curr. Opin. Chem. Eng. 23: 106–114, https://doi.org/10.1016/j.coche.2019.03.005.Search in Google Scholar
Jang, J.-S.R. (1993). ANFIS: adaptive-network-based fuzzy inference system. IEEE Trans. Syst. Man. Cybern. 23: 665–685, https://doi.org/10.1109/21.256541.Search in Google Scholar
Jin, W., Barzilay, R., and Jaakkola, T. (2020). Hierarchical generation of molecular graphs using structural motifs. In Proceedings of the 37th International Conference on Machine Learning, PMLR, Vol. 119, 4839–4848.Search in Google Scholar
Jirasek, F., Alves, R.A.S., Damay, J., Vandermeulen, R.A., Bamler, R., Bortz, M., Mandt, S., Kloft, M., and Hasse, H. (2020). Machine learning in thermodynamics: prediction of activity coefficients by matrix completion. J. Phys. Chem. Lett. 11: 981–985, https://doi.org/10.1021/acs.jpclett.9b03657.Search in Google Scholar PubMed
Jomekian, A. and Poormohammadian, S.J. (2019). Improved prediction of solubility of gases in polymers using an innovative non-equilibrium lattice fluid/Flory–Huggins model. Fluid Phase Equilib. 500: 112261, https://doi.org/10.1016/j.fluid.2019.112261.Search in Google Scholar
Kalinin, S.V., Ziatdinov, M., Sumpter, B.G., and White, A.D. (2022). Physics is the new data. arXiv 1–8, https://doi.org/10.48550/arXiv.2204.05095.Search in Google Scholar
Kennedy, J. and Eberhart, R. (1995). Particle swarm optimization. In: Proceedings of ICNN’95 - international conference on neural networks. IEEE, Perth, WA, Australia.Search in Google Scholar
Keshavarz, M.H., Shafiee, M., and Jazi, B.N. (2022). Simple approach for reliable prediction of solubility of polymers in environmentally compatible solvents. Ind. Eng. Chem. Res. 61: 2425–2433, https://doi.org/10.1021/acs.iecr.1c04737.Search in Google Scholar
Khajeh, A. and Modarress, H. (2010). Prediction of solubility of gases in polystyrene by adaptive neuro-fuzzy inference system and radial basis function neural network. Expert Syst. Appl. 37: 3070–3074, https://doi.org/10.1016/j.eswa.2009.09.023.Search in Google Scholar
Khajeh, A., Modarress, H., and Mohsen, N.M. (2007). Solubility prediction for carbon dioxide in polymers by artificial neural network. Iran Polym. J. 16: 759–768.Search in Google Scholar
Khajeh, A., Modarress, H., and Rezaee, B. (2009). Application of adaptive neuro-fuzzy inference system for solubility prediction of carbon dioxide in polymers. Expert Syst. Appl. 36: 5728–5732, https://doi.org/10.1016/j.eswa.2008.06.051.Search in Google Scholar
Kim, C., Batra, R., Chen, L., Tran, H., and Ramprasad, R. (2021). Polymer design using genetic algorithm and machine learning. Comput. Mater. Sci. 186: 110067, https://doi.org/10.1016/j.commatsci.2020.110067.Search in Google Scholar
Kim, C., Chandrasekaran, A., Huan, T.D., Das, D., and Ramprasad, R. (2018). Polymer genome: a data-powered polymer informatics platform for property predictions. J. Phys. Chem. C 122: 17575–17585, https://doi.org/10.1021/acs.jpcc.8b02913.Search in Google Scholar
Kleinbaum, D.G., Kupper, L.L., Nizam, A., and Rosenberg, E.S. (2013). Applied regression analysis and other multivariable methods, 5th ed. Cengage Learning, Boston.Search in Google Scholar
Koza, J. (1994). Genetic programming as a means for programming computers by natural selection. Stat. Comput. 4: 87–112, https://doi.org/10.1007/BF00175355.Search in Google Scholar
Kuenneth, C. and Ramprasad, R. (2023). polyBERT: a chemical language model to enable fully machine-driven ultrafast polymer informatics. Nat. Commun. 14: 4099, https://doi.org/10.1038/s41467-023-39868-6.Search in Google Scholar PubMed PubMed Central
Kurotani, A., Kakiuchi, T., and Kikuchi, J. (2021). Solubility prediction from molecular properties and analytical data using an in-phase deep neural network (Ip-DNN). ACS Omega 6: 14278–14287, https://doi.org/10.1021/acsomega.1c01035.Search in Google Scholar PubMed PubMed Central
Landrum, G. (2006). RDKit: open-source cheminformatics. https://www.rdkit.org.Search in Google Scholar
Lee, S., Lee, M., Gyak, K.-W., Kim, S.D., Kim, M.-J., and Min, K. (2022). Novel solubility prediction models: molecular fingerprints and physicochemical features vs graph convolutional neural networks. ACS Omega 7: 12268–12277, https://doi.org/10.1021/acsomega.2c00697.Search in Google Scholar PubMed PubMed Central
Li, M., Huang, X., Liu, H., Liu, B., and Wu, Y. (2013a). Prediction of the gas solubility in polymers by a radial basis function neural network based on chaotic self-adaptive particle swarm optimization and a clustering method. J. Appl. Polym. Sci. 130: 3825–3832, https://doi.org/10.1002/app.39525.Search in Google Scholar
Li, M., Huang, X., Liu, H., Liu, B., Wu, Y., and Deng, X. (2013b). Solubility prediction of gases in polymers using fuzzy neural network based on particle swarm optimization algorithm and clustering method. J. Appl. Polym. Sci. 129: 3297–3303, https://doi.org/10.1002/app.39059.Search in Google Scholar
Li, M., Huang, X., Liu, H., Liu, B., Wu, Y., and Wang, L. (2015). Solubility prediction of supercritical carbon dioxide in 10 polymers using radial basis function artificial neural network based on chaotic self-adaptive particle swarm optimization and K-harmonic means. RSC Adv. 5: 45520–45527, https://doi.org/10.1039/C5RA07129A.Search in Google Scholar
Li, M., Huang, X., Liu, H., Liu, B., Wu, Y., Xiong, A., and Dong, T. (2013c). Prediction of gas solubility in polymers by back propagation artificial neural network based on self-adaptive particle swarm optimization algorithm and chaos theory. Fluid Phase Equilib. 356: 11–17, https://doi.org/10.1016/j.fluid.2013.07.017.Search in Google Scholar
Liu, Y., Esan, O.C., Pan, Z., and An, L. (2021). Machine learning for advanced energy materials. Energy AI 3: 100049, https://doi.org/10.1016/j.egyai.2021.100049.Search in Google Scholar
Luan, F., Zhang, X.Y., Zhang, H.X., Zhang, R.S., Liu, M.C., Hu, Z.D., and Fan, B.T. (2006). QSPR study of permeability coefficients through low-density polyethylene based on radial basis function neural networks and the heuristic method. Comput. Mater. Sci. 37: 454–461, https://doi.org/10.1016/j.commatsci.2005.11.003.Search in Google Scholar
Lundberg, S.M. and Lee, S.-I. (2017). A unified approach to interpreting model predictions. In: Advances in neural information processing systems 30 (NIPS 2017) , p. 10. Long Beach, CA.Search in Google Scholar
Ma, R. and Luo, T. (2020). PI1M: a benchmark database for polymer informatics. J. Chem. Inf. Model. 60: 4684–4690, https://doi.org/10.1021/acs.jcim.0c00726.Search in Google Scholar PubMed
MacKay, D.J.C. (1992). Bayesian interpolation. Neural Comput. 4: 415–447, https://doi.org/10.1162/neco.1992.4.3.415.Search in Google Scholar
Madaeni, S.S., Zahedi, G., and Aminnejad, M. (2008). Artificial neural network modeling of O2 separation from air in a hollow fiber membrane module. Asia-Pacific J. Chem. Eng. 3: 357–363, https://doi.org/10.1002/apj.155.Search in Google Scholar
Maimon, O. and Rokach, L. (2006). Clustering methods. In: Maimon, O. and Rokach, L. (Eds.). Data mining and knowledge discovery handbook, Vol. 2. Springer, New York, pp. 321–352.10.1007/b107408Search in Google Scholar
Mannodi-Kanakkithodi, A., Pilania, G., Huan, T.D., Lookman, T., and Ramprasad, R. (2016). Machine learning strategy for accelerated design of polymer dielectrics. Sci. Rep. 6: 20952, https://doi.org/10.1038/srep20952.Search in Google Scholar PubMed PubMed Central
Mason, Ll., Baxter, J., Barlett, P., and Frean, M. (1999). Boosting algorithms as gradient descent. Adv. Neural Inf. Process. Syst. 12: 512–518.Search in Google Scholar
Matei, I., de Kleer, J., Somarakis, C., Rai, R., and Baras, J.S. (2020). Interpretable machine learning models: a physics-based view, https://doi.org/2003.10025.Search in Google Scholar
Matteucci, S., Yampolskii, Y., Freeman, B.D., and Pinnau, I. (2006). Transport of gases and vapor in glassy and rubbery polymers. In: Yampolskii, Y., Pinnau, I., and Freeman, B.D. (Eds.). Materials science of membranes for gas and vapor separation. John Wiley & Sons, Ltd, Chicester, pp. 1–47.10.1002/047002903X.ch1Search in Google Scholar
Mengshan, L., Liang, L., Xingyuan, H., Hesheng, L., Bingsheng, C., Lixin, G., and Yan, W. (2017a). Prediction of supercritical carbon dioxide solubility in polymers based on hybrid artificial intelligence method integrated with the diffusion theory. RSC Adv. 7: 49817–49827, https://doi.org/10.1039/C7RA09531G.Search in Google Scholar
Mengshan, L., Wei, W., Bingsheng, C., Yan, W., and Xingyuan, H. (2017b). Solubility prediction of gases in polymers based on an artificial neural network: a review. RSC Adv. 7: 35274–35282, https://doi.org/10.1039/c7ra04200k.Search in Google Scholar
Miloserdov, O. (2020). Classifying amorphous polymers for membrane technology basing on accessible surface area of their conformations. Adv. Syst. Sci. Appl. 20: 91–104, https://doi.org/10.25728/assa.2020.20.3.897.Search in Google Scholar
Mizrahi Rodriguez, K., Wu, W.-N., Alebrahim, T., Cao, Y., Freeman, B.D., Harrigan, D., Jhalaria, M., Kratochvil, A., Kumar, S., Lee, W.H., et al.. (2022). Multi-lab study on the pure-gas permeation of commercial polysulfone (PSf) membranes: measurement standards and best practices. J. Membr. Sci. 659: 120746, https://doi.org/10.1016/j.memsci.2022.120746.Search in Google Scholar
Modarress, H., Mohsen-Nia, M., and Safamirzaei, M. (2008). Modelling the solubility of 1,1,1,2-tetrafluoroethane, 1-chloro-1,1-difluoroethane, butane and iso-butane in LDPE with artificial neural network. Iran. Polym. J. 17: 483–491.Search in Google Scholar
Moghaddam, A.H. and Alihosseini, A. (2020). Permeability and selectivity prediction of poly (4-methyl 1-pentane) membrane modified by nanoparticles in gas separation through artificial intelligent systems. Polyolefins J. 7: 91–98, https://doi.org/10.22063/poj.2020.2638.1150.Search in Google Scholar
Mousavi, S.-P., Nakhaei-Kohani, R., Atashrouz, S., Hadavimoghaddam, F., Abedi, A., Hemmati-Sarapardeh, A., and Mohaddespour, A. (2023). Modeling of H2S solubility in ionic liquids: comparison of white-box machine learning, deep learning and ensemble learning approaches. Sci. Rep. 13: 7946, https://doi.org/10.1038/s41598-023-34193-w.Search in Google Scholar PubMed PubMed Central
Nasir, R., Suleman, H., and Maqsood, K. (2022). Multiparameter neural network modeling of facilitated transport mixed matrix membranes for carbon dioxide removal. Membranes 12: 421, https://doi.org/10.3390/membranes12040421.Search in Google Scholar PubMed PubMed Central
Nguyen, D., Tao, L., and Li, Y. (2022). Integration of machine learning and coarse-grained molecular simulations for polymer materials: physical understandings and molecular design. Front. Chem. 9: 1–26, https://doi.org/10.3389/fchem.2021.820417.Search in Google Scholar PubMed PubMed Central
Nistane, J., Chen, L., Lee, Y., Lively, R., and Ramprasad, R. (2022). Estimation of the Flory-Huggins interaction parameter of polymer-solvent mixtures using machine learning. MRS Commun. 12: 1096–1102, https://doi.org/10.1557/s43579-022-00237-x.Search in Google Scholar
Noé, F., Tkatchenko, A., Müller, K.-R., and Clementi, C. (2020). Machine learning for molecular simulation. Annu. Rev. Phys. Chem. 71: 361–390, https://doi.org/10.1146/annurev-physchem-042018-052331.Search in Google Scholar PubMed
Ohno, H. (2022). Training data augmentation using generative models with statistical guarantees for materials informatics. Soft. Comput. 26: 1181–1196, https://doi.org/10.1007/s00500-021-06533-3.Search in Google Scholar
Otsuka, S., Kuwajima, I., Hosoya, J., Xu, Y., and Yamazaki, M. (2011). PoLyInfo: polymer database for polymeric materials design. In: 2011 international conference on emerging intelligent data and web technologies. IEEE, Tirana, Albania.10.1109/EIDWT.2011.13Search in Google Scholar
Patel, H.C., Tokarski, J.S., and Hopfinger, A.J. (1997). Molecular modeling of polymers 16. Gaseous diffusion in polymers: a quantitative structure-property relationship (QSPR) analysis. Pharm. Res. 14: 1349–1354, https://doi.org/10.1023/A:1012156318612.10.1023/A:1012156318612Search in Google Scholar PubMed
Patel, R.A., Borca, C.H., and Webb, M.A. (2022). Featurization strategies for polymer sequence or composition design by machine learning. Mol. Syst. Des. Eng. 7: 661–676, https://doi.org/10.1039/D1ME00160D.Search in Google Scholar
Peer, M., Mahdyarfar, M., and Mohammadi, T. (2008). Evaluation of a mathematical model using experimental data and artificial neural network for prediction of gas separation. J. Nat. Gas Chem. 17: 135–141, https://doi.org/10.1016/S1003-9953(08)60040-7.Search in Google Scholar
Pilania, G., Iverson, C.N., Lookman, T., and Marrone, B.L. (2019). Machine-learning-based predictive modeling of glass transition temperatures: a case of polyhydroxyalkanoate homopolymers and copolymers. J. Chem. Inf. Model. 59: 5013–5025, https://doi.org/10.1021/acs.jcim.9b00807.Search in Google Scholar PubMed
Rasmussen, C.E. and Williams, C.K.I. (2005). Regression. In: O. Bousquet O., von Luxburg, U., and Rätsch, G. (Eds.). Gaussian processes for machine learning. Springer, Berlin, pp. 7–31.10.7551/mitpress/3206.003.0005Search in Google Scholar
Rezakazemi, M., Azarafza, A., Dashti, A., and Shirazian, S. (2018). Development of hybrid models for prediction of gas permeation through FS/POSS/PDMS nanocomposite membranes. Int. J. Hydrogen Energy 43: 17283–17294, https://doi.org/10.1016/j.ijhydene.2018.07.124.Search in Google Scholar
Rezakazemi, M., Dashti, A., Asghari, M., and Shirazian, S. (2017). H2-selective mixed matrix membranes modeling using ANFIS, PSO-ANFIS, GA-ANFIS. Int. J. Hydrogen Energy 42: 15211–15225, https://doi.org/10.1016/j.ijhydene.2017.04.044.Search in Google Scholar
Rezakazemi, M. and Mohammadi, T. (2013). Gas sorption in H2-selective mixed matrix membranes: experimental and neural network modeling. Int. J. Hydrogen Energy 38: 14035–14041, https://doi.org/10.1016/j.ijhydene.2013.08.062.Search in Google Scholar
Riasat Harami, H., Dashti, A., Ghahramani Pirsalami, P., Bhatia, S.K., Ismail, A.F., and Goh, P.S. (2020). Molecular simulation and computational modeling of gas separation through polycarbonate/p-nitroaniline/zeolite 4A mixed matrix membranes. Ind. Eng. Chem. Res. 59: 16772–16785, https://doi.org/10.1021/acs.iecr.0c02827.Search in Google Scholar
Ricci, E., Minelli, M., and De Angelis, M.G. (2022). Modelling sorption and transport of gases in polymeric membranes across different scales: a review. Membranes 12, https://doi.org/10.3390/membranes12090857.Search in Google Scholar PubMed PubMed Central
Robeson, L.M. (2008). The upper bound revisited. J. Membr. Sci. 320: 390–400, https://doi.org/10.1016/j.memsci.2008.04.030.Search in Google Scholar
Rogers, D. (1999). Genetic function approximation: evolutionary construction of novel, interpretable, nonlinear models of experimental data. In: Truhlar, D.G., Howe, W.J., Hopfinger, A.J. Blaney, J., and Dammkoehler, R.A. (Eds.). Rational Drug Design. The IMA Volumes in Mathematics and its Application, Vol. 108. Springer, New York, pp. 163–189.10.1007/978-1-4612-1480-9_13Search in Google Scholar
Rogers, D. and Hahn, M. (2010). Extended-connectivity fingerprints. J. Chem. Inf. Model. 50: 742–754, https://doi.org/10.1021/ci100050t.Search in Google Scholar PubMed
Rostamizadeh, M., Rezakazemi, M., Shahidi, K., and Mohammadi, T. (2013). Gas permeation through H2-selective mixed matrix membranes: experimental and neural network modeling. Int. J. Hydrogen Energy 38: 1128–1135, https://doi.org/10.1016/j.ijhydene.2012.10.069.Search in Google Scholar
Ru-Ting, X. and Xing-Yuan, H. (2015). Predictive calculation of carbon dioxide solubility in polymers. RSC Adv. 5: 76979–76986, https://doi.org/10.1039/C5RA15109K.Search in Google Scholar
Russell, S. and Norvig, P. (2020). Artificial intelligence: a modern approach, 4th ed. Pearson, London.Search in Google Scholar
Safamirzaei, M. and Modarress, H. (2011). Hydrogen solubility in heavy n-alkanes; modeling and prediction by artificial neural network. Fluid Phase Equilib. 310: 150–155, https://doi.org/10.1016/j.fluid.2011.08.004.Search in Google Scholar
Sanchez-Lengeling, B., Roch, L.M., Perea, J.D., Langner, S., Brabec, C.J., and Aspuru-Guzik, A. (2019). A Bayesian approach to predict solubility parameters. Adv. Theory Simulations 2: 1800069, https://doi.org/10.1002/adts.201800069.Search in Google Scholar
Sanchez, I.C. and Lacombe, R.H. (1978). Statistical thermodynamics of polymer solutions. Macomolecules 11: 1145–1156, https://doi.org/10.1021/ma60066a017.Search in Google Scholar
Scheffler, M., Aeschlimann, M., Albrecht, M., Bereau, T., Bungartz, H.-J., Felser, C., Greiner, M., Groß, A., Koch, C.T., Kremer, K., et al.. (2022). FAIR data enabling new horizons for materials research. Nature 604: 635–642, https://doi.org/10.1038/s41586-022-04501-x.Search in Google Scholar PubMed
Schmid, T., Hildesheim, W., Holoyad, T., and Schumacher, K. (2021). The AI methods, capabilities and criticality grid: a three-dimensional classification scheme for artificial intelligence applications. Kunstl. Intell. 35: 425–440, https://doi.org/10.1007/s13218-021-00736-4.Search in Google Scholar
Schütt, K.T., Sauceda, H.E., Kindermans, P.-J., Tkatchenko, A., and Müller, K.-R. (2018). SchNet – a deep learning architecture for molecules and materials. J. Chem. Phys. 148: 241722, https://doi.org/10.1063/1.5019779.Search in Google Scholar PubMed
Segler, M.H.S., Preuss, M., and Waller, M.P. (2018). Planning chemical syntheses with deep neural networks and symbolic AI. Nature 555: 604–610, https://doi.org/10.1038/nature25978.Search in Google Scholar PubMed
Shahsavand, A. and Chenar, M.P. (2007). Neural networks modeling of hollow fiber membrane processes. J. Membr. Sci. 297: 59–73, https://doi.org/10.1016/j.memsci.2007.03.011.Search in Google Scholar
Shokrian, M., Sadrzadeh, M., and Mohammadi, T. (2010). C3H8 separation from CH4 and H2 using a synthesized PDMS membrane: experimental and neural network modeling. J. Membr. Sci. 346: 59–70, https://doi.org/10.1016/j.memsci.2009.09.015.Search in Google Scholar
Sholl, D.S. and Lively, R.P. (2016). Seven chemical separations to change the world. Nature 532: 435–437, https://doi.org/10.1038/532435a.Search in Google Scholar PubMed
Škerget, M., Mandžuka, Z., Aionicesei, E., Knez, Ž., Ješe, R., Znoj, B., and Venturini, P. (2010). Solubility and diffusivity of CO2 in carboxylated polyesters. J. Supercrit. Fluids. 51: 306–311, https://doi.org/10.1016/j.supflu.2009.10.013.Search in Google Scholar
Sodeifian, G., Raji, M., Asghari, M., Rezakazemi, M., and Dashti, A. (2019). Polyurethane-SAPO-34 mixed matrix membrane for CO2/CH4 and CO2/N2 separation. Chinese J. Chem. Eng. 27: 322–334, https://doi.org/10.1016/j.cjche.2018.03.012.Search in Google Scholar
Soleimani, R., Saeedi Dehaghani, A.H., Rezai-Yazdi, A., Hosseini, S.A., Hosseini, S.P., and Bahadori, A. (2020). Evolving an accurate decision tree-based model for predicting carbon dioxide solubility in polymers. Chem. Eng. Technol. 43: 514–522, https://doi.org/10.1002/ceat.201900096.Search in Google Scholar
Song, Z., Shi, H., Zhang, X., and Zhou, T. (2020). Prediction of CO2 solubility in ionic liquids using machine learning methods. Chem. Eng. Sci. 223: 115752, https://doi.org/10.1016/j.ces.2020.115752.Search in Google Scholar
Sun, J., Sato, Y., Sakai, Y., and Kansha, Y. (2023). A review of ionic liquids and deep eutectic solvents design for CO2 capture with machine learning. J. Clean. Prod. 414: 137695, https://doi.org/10.1016/j.jclepro.2023.137695.Search in Google Scholar
Suykens, J. and Vandewalle, J. (1999). Least squares support vector machine classifiers. Neural Process. Lett. 9: 293–300, https://doi.org/https://doi.org/10.1023/A:1018628609742.10.1023/A:1018628609742Search in Google Scholar
Swain, M.C. and Cole, J.M. (2016). ChemDataExtractor: a toolkit for automated extraction of chemical information from the scientific literature. J. Chem. Inf. Model. 56: 1894–1904, https://doi.org/10.1021/acs.jcim.6b00207.Search in Google Scholar PubMed
Talukder, M.J., Alshami, A.S., Tayyebi, A., Ismail, N., and Yu, X. (2023). Membrane science meets machine learning: future and potential use in assisting membrane material design and fabrication. Sep. Purif. Rev. 53: 1–14, https://doi.org/10.1080/15422119.2023.2212295.Search in Google Scholar
Tao, L., He, J., Arbaugh, T., McCutcheon, J.R., and Li, Y. (2023). Machine learning prediction on the fractional free volume of polymer membranes. J. Membr. Sci. 665: 121131, https://doi.org/10.1016/j.memsci.2022.121131.Search in Google Scholar
Tao, L., Varshney, V., and Li, Y. (2021). Benchmarking machine learning models for polymer informatics: an example of glass transition temperature. J. Chem. Inf. Model. 61: 5395–5413, https://doi.org/10.1021/acs.jcim.1c01031.Search in Google Scholar PubMed
Tayyebi, A., Alshami, A.S., Yu, X., and Kolodka, E. (2022). Can machine learning methods guide gas separation membranes fabrication? J. Membr. Sci. Lett. 2: 100033, https://doi.org/10.1016/j.memlet.2022.100033.Search in Google Scholar
Thornton, A.W., Freeman, B.D., and Robeson, L.M. (2012). Polymer gas separation membrane database. Available at: <https://membrane-australasia.org/polymer-gas-separation-membrane-database/> [WWW Document].Search in Google Scholar
Tian, Y., Wang, X., Liu, Y., and Hu, W. (2023). Prediction of CO2 and N2 solubility in ionic liquids using a combination of ionic fragments contribution and machine learning methods. J. Mol. Liq. 383: 122066, https://doi.org/10.1016/j.molliq.2023.122066.Search in Google Scholar
Tiwari, S.P., Shi, W., Budhathoki, S., Baker, J., Hopkinson, D., and Steckel, J. (2023). Creation of polymer datasets with targeted backbones for screening of gas permeability and selectivity. chemRxiv 1–19, https://doi.org/10.26434/chemrxiv-2023-qs4jw.Search in Google Scholar
Tokarski, J.S., Hopfinger, A.J., David Hobbs, J., Ford, D.M., and Faulon, J.-L.M. (1997). Molecular modelling of polymers 17. Simulation and QSPR analyses of transport behavior in amorphous polymeric materials. Comput. Theor. Polym. Sci. 7: 199–214, https://doi.org/10.1016/S1089-3156(98)00007-5.Search in Google Scholar
Toropov, A.A., Toropova, A.P., Begum, S., and Achary, P.G.R. (2016). Towards predicting the solubility of CO2 and N2 in different polymers using a quasi-SMILES based QSPR approach. SAR QSAR Environ. Res. 27: 293–301, https://doi.org/10.1080/1062936X.2016.1172666.Search in Google Scholar PubMed
Varnek, A., Fourches, D., Hoonakker, F., and Solov’ev, V.P. (2005). Substructural fragments: an universal language to encode reactions, molecular and supramolecular structures. J. Comput. Aided Mol. Des. 19: 693–703, https://doi.org/10.1007/s10822-005-9008-0.Search in Google Scholar PubMed
Vasudevan, R.K., Choudhary, K., Mehta, A., Smith, R., Kusne, G., Tavazza, F., Vlcek, L., Ziatdinov, M., Kalinin, S.V., and Hattrick-Simpers, J. (2019). Materials science in the artificial intelligence age: high-throughput library generation, machine learning, and a pathway from correlations to the underpinning physics. MRS Commun. 9: 821–838, https://doi.org/10.1557/mrc.2019.95.Search in Google Scholar PubMed PubMed Central
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., and Polosukhin, I. (2017). Attention is all you need. In: Advances in neural information processing systems 30 (NIPS 2017).Search in Google Scholar
Vieth, W.R., Tam, P.M.H.I.M., Michaels, A.S., Vieth, W.R., and Michaels, A.S. (1966). Dual sorption mechanisms in glassy polystyrene. J. Colloid Interface Sci. 22: 360–370, https://doi.org/10.1016/0021-9797(66)90016-6.Search in Google Scholar
Villanueva, N., Flaconnèche, B., and Creton, B. (2015). Prediction of alternative gasoline sorption in a semicrystalline poly(ethylene). ACS Comb. Sci. 17: 631–640, https://doi.org/10.1021/acscombsci.5b00094.Search in Google Scholar PubMed
Wang, J., Tian, K., Li, D., Chen, M., Feng, X., Zhang, Y., Wang, Y., and Van der Bruggen, B. (2023). Machine learning in gas separation membrane developing: ready for prime time. Sep. Purif. Technol. 313: 123493, https://doi.org/10.1016/j.seppur.2023.123493.Search in Google Scholar
Wang, L., Shao, C., Wang, H., and Wu, H. (2006). Radial basis function neural networks-based modeling of the membrane separation process: hydrogen recovery from refinery gases. J. Nat. Gas Chem. 15: 230–234, https://doi.org/10.1016/S1003-9953(06)60031-5.Search in Google Scholar
Webb, M.A., Jackson, N.E., Gil, P.S., and de Pablo, J.J. (2020). Targeted sequence design within the coarse-grained polymer genome. Sci. Adv. 6: eabc6216, https://doi.org/10.1126/sciadv.abc6216.Search in Google Scholar PubMed PubMed Central
Wen, C., Liu, B., Wolfgang, J., Long, T.E., Odle, R., and Cheng, S. (2020). Determination of glass transition temperature of polyimides from atomistic molecular dynamics simulations and machine-learning algorithms. J. Polym. Sci. 58: 1521–1534, https://doi.org/10.1002/pol.20200050.Search in Google Scholar
Wessling, M., Mulder, M.H.V., Bos, A., van der Linden, M., Bos, M., and van der Linden, W.E. (1994). Modelling the permeability of polymers: a neural network approach. J. Membr. Sci. 86: 193–198, https://doi.org/10.1016/0376-7388(93)E0168-J.Search in Google Scholar
Westermayr, J., Gastegger, M., Schütt, K.T., and Maurer, R.J. (2021). Perspective on integrating machine learning into computational chemistry and materials science. J. Chem. Phys. 154: 230903, https://doi.org/10.1063/5.0047760.Search in Google Scholar PubMed
Wijmans, J.G. and Baker, R.W. (1995). The solution-diffusion model: a review. J. Membr. Sci. 107: 1–21, https://doi.org/10.1016/S0166-4115(08)60038-2.Search in Google Scholar
Wijmans, J.G. and Baker, R.W. (2006). The Solution-Diffusion model: a unified approach to membrane permeation. In: Yampolskii, Y., Pinnau, I., and Freeman, B. (Eds.). Materials science of membranes for gas and vapor separation. John Wiley & Sons, Chichester, pp. 159–189.10.1002/047002903X.ch5Search in Google Scholar
Willard, J., Jia, X., Xu, S., Steinbach, M., and Kumar, V. (2020). Integrating physics-based modeling with machine learning: a survey. arXiv 1–34, https://doi.org/10.1145/1122445.1122456.Search in Google Scholar
Wu, Y., Liu, B., Li, M., Tang, K., and Wu, Y. (2013). Prediction of CO2 solubility in polymers by radial basis function artificial neural network based on chaotic self-adaptive particle swarm optimization and fuzzy clustering method. Chinese J. Chem. 31: 1564–1572, https://doi.org/10.1002/cjoc.201300550.Search in Google Scholar
Wu, Z., Ramsundar, B., Feinberg, E.N., Gomes, J., Geniesse, C., Pappu, A.S., Leswing, K., and Pande, V. (2018). MoleculeNet: a benchmark for molecular machine learning. Chem. Sci. 9: 513–530, https://doi.org/10.1039/c7sc02664a.Search in Google Scholar PubMed PubMed Central
Xu, Q. and Jiang, J. (2022). Recent development in machine learning of polymer membranes for liquid separation. Mol. Syst. Des. Eng. 7: 856–872, https://doi.org/10.1039/D2ME00023G.Search in Google Scholar
Xu, Q. and Jiang, J. (2020). Machine learning for polymer swelling in liquids. ACS Appl. Polym. Mater. 2: 3576–3586, https://doi.org/10.1021/acsapm.0c00586.Search in Google Scholar
Yampolskii, Y., Shishatskii, S., Alentiev, A., and Loza, K. (1998). Group contribution method for transport property predictions of glassy polymers: focus on polyimides and polynorbornenes. J. Membr. Sci. 149: 203–220, https://doi.org/10.1016/S0376-7388(98)00152-5.Search in Google Scholar
Yang, J., Tao, L., He, J., McCutcheon, J.R., and Li, Y. (2022). Machine learning enables interpretable discovery of innovative polymers for gas separation membranes. Sci. Adv. 8: eabn9545, https://doi.org/10.1126/sciadv.abn9545.Search in Google Scholar PubMed PubMed Central
Yang, L. and Shami, A. (2020). On hyperparameter optimization of machine learning algorithms: theory and practice. Neurocomputing 415: 295–316, https://doi.org/10.1016/j.neucom.2020.07.061.Search in Google Scholar
Yang, X.-S., Deb, S., and Fong, S. (2012). Accelerated particle swarm optimization and support vector machine for business optimization and applications, https://doi.org/10.48550/arXiv.1203.6577.Search in Google Scholar
Yap, C.W. (2011). PaDEL-descriptor: an open source software to calculate molecular descriptors and fingerprints. J. Comput. Chem. 32: 1466–1474, https://doi.org/10.1002/jcc.21707.Search in Google Scholar PubMed
Ye, H., Xian, W., and Li, Y. (2021). Machine learning of coarse-grained models for organic molecules and polymers: progress, opportunities, and challenges. ACS Omega 6: 1758–1772, https://doi.org/10.1021/acsomega.0c05321.Search in Google Scholar PubMed PubMed Central
Yuan, Q., Longo, M., Thornton, A.W., McKeown, N.B., Comesaña-Gándara, B., Jansen, J.C., and Jelfs, K.E. (2021). Imputation of missing gas permeability data for polymer membranes using machine learning. J. Membr. Sci. 627: 119207, https://doi.org/10.1016/j.memsci.2021.119207.Search in Google Scholar
Zadeh, L.A. (1988). Fuzzy logic. Computer 21: 83–93, https://doi.org/10.1109/2.53.Search in Google Scholar
Zeng, M., Kumar, J.N., Zeng, Z., Savitha, R., Chandrasekhar, V.R., and Hippalgaonkar, K. (2018). Graph convolutional neural networks for polymers property prediction. arXiv 1–7, https://doi.org/10.48550/arXiv.1811.06231.Search in Google Scholar
Zhang, B., Hsu, M., and Dayal, U. (2001). K-harmonic means - a spatial clustering algorithm with boosting. In: Roddick, J.F. and Hornsby, K. (eds.). Temporal, spatial, and spatio-temporal data mining. TSDM 2000. Lecture Notes in Computer Science(), Vol. 2007. Springer, Berlin, Heidelberg.10.1007/3-540-45244-3_4Search in Google Scholar
Zhang, K., Wu, J., Yoo, H., and Lee, Y. (2021). Machine learning-based approach for tailor-made design of ionic liquids: application to CO2 capture. Sep. Purif. Technol. 275: 119117, https://doi.org/10.1016/j.seppur.2021.119117.Search in Google Scholar
Zhang, Y. and Xu, X. (2020). Machine learning glass transition temperature of polymers. Heliyon 6: e05055, https://doi.org/10.1016/j.heliyon.2020.e05055.Search in Google Scholar PubMed PubMed Central
Zhao, H., Li, X., Zhang, Y., Schadler, L.S., Chen, W., and Brinson, L.C. (2016). Perspective: NanoMine: a material genome approach for polymer nanocomposites analysis and design. APL Mater. 4: 053204, https://doi.org/10.1063/1.4943679.Search in Google Scholar
Zhao, M., Zhang, C., and Weng, Y. (2023). Improved artificial neural networks (ANNs) for predicting the gas separation performance of polyimides. J. Membr. Sci. 681: 121765, https://doi.org/10.1016/j.memsci.2023.121765.Search in Google Scholar
Zhong, X., Gallagher, B., Liu, S., Kailkhura, B., Hiszpanski, A., and Han, T.Y.-J. (2022). Explainable machine learning in materials science. npj Comput. Mater. 8: 204, https://doi.org/10.1038/s41524-022-00884-7.Search in Google Scholar
Zhu, G., Kim, C., Chandrasekarn, A., Everett, J.D., Ramprasad, R., and Lively, R.P. (2020a). Polymer genome–based prediction of gas permeabilities in polymers. J. Polym. Eng. 40: 451–457, https://doi.org/10.1515/polyeng-2019-0329.Search in Google Scholar
Zhu, T., Jiang, Y., Cheng, H., Singh, R.P., and Yan, B. (2020b). Development of pp-LFER and QSPR models for predicting the diffusion coefficients of hydrophobic organic compounds in LDPE. Ecotoxicol. Environ. Saf. 190: 110179, https://doi.org/10.1016/j.ecoenv.2020.110179.Search in Google Scholar PubMed
Ziaee, H., Hosseini, S.M., Sharafpoor, A., Fazavi, M., Ghiasi, M.M., and Bahadori, A. (2015). Prediction of solubility of carbon dioxide in different polymers using support vector machine algorithm. J. Taiwan Inst. Chem. Eng. 46: 205–213, https://doi.org/10.1016/j.jtice.2014.09.015.Search in Google Scholar
© 2023 Walter de Gruyter GmbH, Berlin/Boston