Biomass is a valuable renewable resource to produce sustainable fuels and chemicals; its efficient utilization depends on accurate compositional characterization. This study presents a machine learning framework to predict the elemental composition of biomass carbon, hydrogen, oxygen, nitrogen, and sulfur from proximate analysis, net calorific value, and biomass clustering. A dataset of 426 samples from the Phyllis2 database was used, with outliers removed using the Local Outlier Factor (LOF) method. Biomass samples were first grouped using k-means clustering based on proximate analysis and net calorific value, identifying four distinct clusters with different compositional characteristics. These clusters were encoded and, together with the input variables, used as inputs to an artificial neural network (ANN) to predict elemental composition. The ANN achieved high predictive performance, with a mean squared error of 1.84 and R² of 0.994 on the test set. For validation, new samples were classified using a logistic regression model trained on the clustering results, and their elemental composition was subsequently predicted using the ANN. The proposed approach outperformed literature models for carbon, hydrogen, and oxygen. Lower accuracy for nitrogen and sulfur was attributed to their high variability and skewed distributions. The integration of biomass classification extends the applicability of the model to a wide range of carbonaceous materials, including organic residues and coal, providing a practical tool for biomass selection in thermochemical processes such as gasification.
This study presents an application of Physics-Informed Neural Networks (PINNs) and Galerkin Physics-Informed Neural Networks (G-PINNs) to the modeling of tracer dispersion in porous media using experimental data from sodium chloride transport in Berea sandstone cores. The proposed framework considers Dirichlet and Neumann boundary conditions in a 1D cylindrical coordinate system and employs artificial neural networks as basis functions in the G-PINN formulation to improve numerical accuracy. Both methodologies are trained to satisfy the underlying advection-diffusion equation while fitting the available experimental measurements. The study includes direct and inverse problem applications, with particular emphasis on estimating the effective tracer dispersion coefficient from outlet concentration data. To improve convergence in the inverse problem, a modified optimization strategy combining a scaled L-BFGS scheme and a reduced number of Adam iterations is introduced. The results show that both PINN and G-PINN can reproduce the concentration evolution with good agreement, while the proposed scaled L-BFGS + Adam strategy significantly reduces computational time compared with using Adam alone. These findings highlight the potential of PINN-based and Galerkin-based neural formulations for experimental-data-driven modeling and parameter estimation in porous media transport problems, with applications in reservoir engineering, contaminant transport, and related subsurface flow systems.
Green methods for nanoparticle synthesis are highly attractive due to their versatility, purity, cost-effectiveness, and environmental friendliness. Green Ag and Au nanoparticles (NPs) biosynthesized with fungi present varying sizes, shapes, colloidal stability, and outstanding biological activity, but their structural properties have not been widely researched. In this study, the structural characterization of green AuNPs synthesized with the extracellular extract of Epicoccum nigrum at pH conditions of 5, 6, and 8 was performed by transmission electron microscopy (TEM). TEM analysis revealed NPs with diverse geometric structures and a broad size distribution (1–380 nm), following a trimodal distribution. Electron diffraction and fast Fourier transform patterns evidenced stacking faults and dislocations within the NPs, generating single, multiple, and cyclic twinning. The stacking faults influenced nanocrystals’ anisotropic growth, which led to the formation of nanoplatelets. Lastly, the acidic medium resulted in more triangular nanoplatelets than the alkaline medium.
With the extraction of increasingly heavier crudes worldwide, transporting these crudes from wells to processing centers has become a complex and challenging task. This complexity has led to the development of various advanced technologies and the continuous improvement of existing processes to meet this growing demand.This review updates the state of the art in pipeline transportation of heavy and extra-heavy crude oil, building on a review by our working group published in 2011 (J. Pet. Sci. Eng, 75 (2011) 274-282). The methods used for transporting heavy crude oil from production sites to processing facilities are examined, including topics such as viscosity reducers, chemical additives, and friction reduction. Advances in those technologies are now presented, and other innovative technological solutions are also discussed, such as aquathermolysis and in situ upgrading, the use of microwaves and cavitation, magnetic nanoparticles (MNPs) and superhydrophobic and superamphiphobic materials for the transport of heavy and extra-heavy crudes, with special emphasis on recent more environmentally friendly product developments and more sustainable technologies; and, finally, some applications of artificial intelligence (AI) and machine learning (ML) to study various aspects of heavy crude oil transport are also discussed.
Managing excessive water production in oil fields during primary, secondary, or enhanced recovery remains challenging. It increases costs and reduces hydrocarbon recovery, particularly in reservoirs with high-conductivity pathways such as high-permeability zones and fractures. Hydrogels are commonly used for water blocking and retention; however, their effectiveness diminishes at higher flow rates due to mechanical weaknesses and structural limitations. These problems are intensified under harsh environmental conditions, including high temperatures, salinity, and hardness. In this study, we investigate how altering the molecular suprastructure of preformed particle gel (PPG) can improve its effectiveness in shear-responsive water-blockage treatments, particularly when traditional PPGs cannot control rising flow rates. We enhance the shear-responsive mechanical properties of a composite PPG by increasing the density and diversity of intermolecular interactions. We use two different strategies: first, incorporating cationic groups into the polymer backbone to form a polyampholyte network with stronger electrostatic interactions; second, adding a linear anionic polymer to generate a secondary interpenetrating network that can undergo a coil-stretch transition under thermal and shear stimuli, thereby enhancing its own solvation and whole-network expansion. Molecular simulations provide an interpretation of the experimentally observed shear-thickening response and enhanced disproportionate permeability reduction at high flow rates. The water residual resistance factor of the improved PPGs deviates from the typical shear-thinning power-law behavior (n < 1) observed in conventional PPG, showing shear-thickening (n > 1). Tests reveal a strong ability to preferentially reduce water flow over oil, with Disproportionate Permeability Reduction increasing from 8 to 117 in the high-flow-rate zone. The enhanced strength and thermal stability also improve resistance to washout under high-pressure gradients. This research provides a novel approach to tailoring the microscopic architecture of PPGs to achieve selective, robust water blockage, offering a high-efficiency solution for complex reservoir environments.
