Andreia Fonseca de Faria

The literature on graphene oxide (GO) comprises hundreds of thousands of publications spanning synthesis, characterization, and applications in diverse domains. Advancing GO materials requires a structured knowledge map derived from systematic mining of this extensive corpus. However, extracting meaningful insights from this vast, heterogeneous, sparse, and unstructured data remains a challenge. To address this challenge, Natural Language Processing techniques, including topic modeling, community discovery, and parameter extraction, are here employed to establish a framework. Latent Dirichlet Allocation (LDA) was applied to the full text of more than 120,000 articles, enabling the identification of eight macro-domains within the GO literature, and neural topic modeling was subsequently employed to characterize the granularity and heterogeneity of these domains. Parameter extraction was performed using Large Language Models (LLMs) on a subset of 8000 full-text articles, with a proto-ontology incorporated into the extraction prompt. Even with a broad prompt, an F1 score of 94% was achieved against manual validation. The extracted parameters were then clustered through community modeling and compared with the LDA-derived topics. At this intermediate resolution of the literature, we demonstrate the application of a recommendation system for relevant parameters according to the material family and application domain. The applicability of the framework at higher resolutions was further demonstrated through the construction of a data set of CO2 adsorption for graphene- and GO-based materials. This data set was subsequently employed in a downstream predictive case study using an Extremely Randomized Trees Regressor, with different training-test splitting strategies evaluated to assess in-domain and out-of-domain generalization at article-level resolution.

Natural language processing (NLP) pipelines can mine the nanotoxicology literature at a scale and resolution that cannot be achieved by manual curation. Here, we established a NLP pipeline that coupled topic modeling and end point-specific extraction LLM prompts to convert ∼10⁶ sentences extracted from abstracts of scientific articles into a structured knowledge base containing 13 nanotoxicology features. The pipeline is capable of analyzing and extracting nanomaterial descriptors such as size, ζ potential, and surface area, along with biological end points such as minimum inhibitory concentration (MIC), minimum bactericidal concentration (MBC) and lethal concentration 50% (LC₅₀). Statistical convergence across multiple quantitative end points - MIC, MBC, microbial log reduction, and biofilm killing efficiency - shows that Ag-based nanomaterials are the most potent antimicrobial agents, showing lower MIC and MBC than ZnO, TiO₂ and Au analogs. This trend was also observed for individual pathogens such as Escherichia coli and Staphylococcus aureus. Most nanomaterials are within 1 to 100 nm, with nanoparticles featured in >80% of the studies. Although nanomaterials 50, toxicity within a single size class spans orders of magnitude, underscoring the influence of surface chemistry, coatings, and colloidal behavior. In addition, adverse reproductive effects in Caenorhabditis elegans and Daphnia magna, and developmental abnormalities in Danio rerio, are predominantly related to Ag and TiO₂. In general, our automated data extraction and the curation strategy transforms disparate literature into a machine-readable knowledge base that paves the way for data-driven predictions of nanomaterial hazards.

Patients dealing with chronic wounds frequently experience persistent pain and a heightened risk of infection due to microbial contamination. To improve the quality of life for these patients while reducing treatment duration and costs, advanced wound dressings are being developed. This study aimed to develop a wound dressing from natural bacterial-derived cellulose fibers (BC) and calcium alginate (A), which were functionalized with silver-containing graphene oxide nanoparticles (GOAg). Carbon-based nanomaterials, such as GOAg, are recognized for their antimicrobial properties, noncytotoxic nature, and their strong ability to absorb exudate, which is crucial for controlling infections in dermal lesions. In vitro testing revealed that the BC-A-GOAg dressing possessed optimized characteristics, including a uniform distribution of GOAg nanoparticles within the polymeric blend of BC and A. When in direct contact with bacterial cells in suspension, the BC-A-GOAg dressing exhibited 74% antimicrobial activity against Staphylococcus aureus and 59% against Pseudomonas aeruginosa. Additionally, the antimicrobial BC-A-GOAg demonstrated no significant cytotoxicity to mouse fibroblast cells (L-929), maintaining 90.8% ± 5.2% cell viability after 48 h of exposure. Furthermore, the in vitro assessments showed that the BC-A-GOAg dressing could inhibit the activity of the myeloperoxidase enzyme, highlighting its effectiveness in reducing inflammation.

Water scarcity, contamination, and lack of sanitation are global issues that require innovations in chemistry, engineering, and materials science. To tackle the challenge of providing high-quality drinking water for a growing population, we need to develop high-performance and multifunctional materials to treat water on both small and large scales. As modern society and science prioritize more sustainable engineering practices, water treatment processes will need to use materials produced from sustainable resources via green chemical routes, combining multiple advanced properties such as high surface area and great affinity for contaminants. Lignin, one of the major components of plants and an abundant byproduct of the cellulose and bioethanol industries, offers a cost-effective and scalable platform for developing such materials, with a wide range of physicochemical properties that can be tailored to improve their performance for target water treatment applications. This review aims to bridge the current gap in the literature by exploring the use of lignin, both as solid bulk or solubilized macromolecules and nanolignin as multifunctional (nano)materials for sustainable water treatment processes. We address the application of lignin-based macro-, micro-, and nanostructured materials in adsorption, catalysis, flocculation, membrane filtration processes, and antimicrobial coatings and composites. Throughout the exploration of recent progress and trends in this field, we emphasize the importance of integrating principles of green chemistry and materials sustainability to advance sustainable water treatment technologies.

This study reports on the modification of bacterial cellulose (BC) membranes produced by static fermentation of Komagataeibacter xylinus bacterial strains with graphene oxide-silver nanoparticles (GO-Ag) to yield skin wound dressings with improved antibacterial properties. The GO-Ag sheets were synthesized through chemical reduction with sodium citrate and were utilized to functionalize the BC membranes (BC/GO-Ag). The BC/GO-Ag composites were characterized to determine their surface charge, morphology, exudate absorption, antimicrobial activity, and cytotoxicity by using fibroblast cells. The antimicrobial activity of the wound dressings was assessed against Staphylococcus aureus, Escherichia coli, and Pseudomonas aeruginosa. The results indicate that the BC/GO-Ag dressings can inhibit ∼70% of E. coli cells. Our findings also revealed that the porous BC/GO-Ag antimicrobial dressings can efficiently retain 94% of exudate absorption after exposure to simulated body fluid (SBF) for 24 h. These results suggest that the dressings could absorb excess exudate from the wound during clinical application, maintaining adequate moisture, and promoting the proliferation of epithelial cells. The BC/GO-Ag hybrid materials exhibited excellent mechanical flexibility and low cytotoxicity to fibroblast cells, making excellent wound dressings able to control bacterial infectious processes and promote the fast healing of dermal lesions.

In this study, we propose the production of a thin-film composite (TFC) membrane with efficient antimicrobial properties by modifying it with cellulose nanocrystals (CNC) and silver nanoparticles (AgNP). Two approaches (i.e., “one-pot” and “layer-by-layer”) were employed to incorporate CNC and AgNP onto the membrane surface and their impact on the membrane antimicrobial properties was examined. Firstly, the “one-pot” method involved reacting CNC with AgNO₃ and NaBH₄ to form a CNC/Ag hybrid material. This CNC/Ag hybrid was then applied to the TFC drawmembranes using polydopamine (PDA) chemistry, allowing the deposition of both CNC and AgNP in a single step. Secondly, the “layer-by-layer” method involved attaching CNC to the membrane surface using PDA chemistry, which was then followed by an in situ reaction of AgNO₃ and NaBH₄ to deposit the AgNP directly on top of the CNC layer. Membranes modified with a CNC/Ag hybrid using the one-pot method were able to deactivate 75.7 ± 15.6% of attached E. coli cells without detrimental impact to membrane separation performance. The membranes modified using the layer-by-layer assembly demonstrated improved antimicrobial activity when compared to those modified with the one-pot CNC/Ag hybrid, being capable of inactivating 90% of the attached E. coli cells. On the other hand, the layer-by-layer modification method led to a significant loss in the membrane's salt rejection capacity. These water-based, straightforward, and simple functionalization strategies can tailor the antimicrobial and anti-biofouling performance of membranes for water treatment applications.

In this study, we added value to lignocellulosic biomass-derived lignin by converting it into antimicrobial nanoparticles using a simple self-assembling method in solution. Transmission electron microscopy (TEM) and zeta potential analyses showed that the self-assembled lignin nanoparticles (SA-LNPs) had a spherical-like morphology, 80 nm average size, and a surface charge of −29 ± 4 mV. Previous studies have shown that LNPs are toxic to bacteria, though the potential mechanisms of action leading to antimicrobial properties of LNPs are lacking in the literature. Therefore, we conducted a thorough investigation of the antibacterial activity of SA-LNPs using four bacteria strains: Escherichia coli and Pseudomonas aeruginosa (Gram-negative) and Bacillus subtilis and Lactobacillus fermentum (Gram-positive). The antimicrobial assays performed in saline media revealed that SA-LNPs were selectively toxic to Gram-positive bacteria, and no significant antimicrobial effects were found against the Gram-negative strains. Time-kill experiments showed that 25 μg mL⁻¹ SA-LNPs were able to inactivate more than 90% of the Gram-positive bacteria after 30 min exposure. We conducted in vitro and in vivo assays to evaluate the production of reactive oxidative species (ROS), such as glutathione and 2′,7′-dichlorodihydrofluorescein diacetate (H₂DCFA). These assays indicated that oxidative stress was not the underlying mechanism involved in the antimicrobial activity of SA-LNPs. This finding corroborates that SA-LNPs could scavenge radicals of 2,2-diphenyl-1-(2,4,6-trinitrophenyl)hydrazin-1-yl (DPPH), confirming their strong antioxidant property. Although direct oxidative stress was ruled out as the probable mechanism of action, we still cannot dismiss an indirect pro-oxidant effect resulting from the SA-LNPs-containing adsorbed ROS coming into direct contact with the cell wall.