Efficient Removal of Uranyl Ions Using PAMAM Dendrimer: Simulation and Experiment.

In this work, using atomistic molecular dynamics (MD) simulations and polymer-assisted ultrafiltration experiments, we explore the adsorption and removal of uranyl ions from aqueous solutions using poly(amidoamine) (PAMAM) dendrimers. The effects of uranyl ion concentration and the pH of the solution were examined for PAMAM dendrimers of generations 3, 4, and 5. Our simulation results show that PAMAM has a high adsorption capacity for the uranyl ions. The adsorption capacity increases with increasing concentration of uranyl ions for all 3 generations of PAMAM in agreement with experimental findings. We find that the number of uranyl ions bound to PAMAM is significantly higher in acidic solutions (pH < 3) as compared to neutral solutions (pH ∼ 7) for all uranyl ion concentrations. Additionally, we find an increase in the number of adsorbed uranyl ions to PAMAM with the increase in the dendrimer generation. This increase is due to the greater number of binding sites present for higher-generation PAMAM dendrimers. Our simulation study shows that nitrate ions form a solvation shell around uranyl ions, which allows them to bind to PAMAM binding sites, including the amide, amine, and carbonyl groups. In polymer-assisted ultrafiltration (PAUF) experiments, the removal percentage of uranyl ions by G3 PAMAM dendrimer increased from 36.3% to 42.6% as the metal ion concentration increased from 2.1 × 10-5 M to 10.5 × 10-5 M at a pH of 2. Our combined experiment and simulation study suggests that PAMAM is an effective adsorbent for removing uranyl ions from aqueous solutions.

Data-Driven Approaches to Predict Dendrimer Cytotoxicity.

Dendrimers are employed as functional elements in contrast agents and are proposed as nontoxic vehicles for drug delivery. Toxicity is a property that is to be evaluated for this novel class of bionanomaterials for in vivo applications. The current research is hampered due to the lack of structured data sets for toxicity studies for dendrimers. In this work, we have built a data set by curating literature for toxicity data and augmented it with structural and physicochemical features. We present a comprehensive, feature-rich database of dendrimer toxicity measured across various cell lines for prediction, design, and optimization studies. We have also explored novel computational approaches for predicting dendrimer cytotoxicity. We demonstrate superior outcomes for toxicity prediction using essential regression in the space of small data sets.