Search

With growing environmental consciousness, biomaterials (BMs) have garnered attention as sustainable materials for the adsorption of hazardous water contaminants. These BMs are engineered using surface treatments or physical alterations to enhance their adsorptive properties. The lab-scale methods generally employ a One Variable at a Time (OVAT) approach to analyze the impact of biomaterial modifications, their characteristics and other process variables such as pH, temperature, dosage, etc., on the removal of metals via adsorption. Although implementing the adsorption procedure using BMs seems simple, the conjugate effects of adsorbent properties and process attributes implicate complex nonlinear interactions. As a result, artificial neural networks (ANN) have gained traction in the quest to understand the complex metal adsorption processes on biomaterials, with applications in environmental remediation and water reuse. This review discusses recent progress using ANN frameworks for metal adsorption using modified biomaterials. Subsequently, the paper comprehensively evaluates the development of a hybrid-ANN system to estimate isothermal, kinetic and thermodynamic parameters in multicomponent adsorption systems.

Per- and polyfluoroalkyl substances (PFAS) are potential human carcinogens that have been ubiquitously detected in drinking water sources. Ion exchange (IX) resins offer promising potential for the treatment of water sources impacted by high natural organic matter (NOM) and PFAS concentrations. Several kinetic models such as the pseudo-first-order (PFO), pseudo-second-order (PSO) and intraparticle diffusion (ID) have been investigated to examine the PFAS uptake kinetics on IX resins. However, the kinetic parameters are strongly impacted by the sorption test conditions, especially the sorbate-to-sorbent ratio, which were highly variable in past studies, resulting in several discrepancies when comparing published data. This study examined the use of the homogenous surface diffusion model (HSDM) coupled with the equivalent background concentration (EBC) model as an approach to describe competitive uptakes of regulated long- and short-chain PFAS, NOM, and inorganic ions (such as sulphate and nitrate) with IX resins. Kinetic studies confirmed surface diffusion as the rate-limiting step in NOM-rich waters (Biot number >30), while the multicomponent equilibrium model demonstrated that the initial NOM concentration of approximately 0.35–0.95 μmol L−1 (C0 = 5 mg C per L) competed with PFAS for active exchange sites. More importantly, the HSDM (R2 > 0.98)/EBC (R2 > 0.95) models provided an adequate fit to describe PFAS removal in a recycled wastewater.

Cyanobacterial blooms are a growing concern around the world. A feasible approach for small treatment plants fed by sources contaminated with cyanobacteria is vacuum UV (VUV). VUV is a promising advanced oxidation process used to treat water impacted by cyanobacterial blooms, with potential applicability in small and remote communities because of its simplicity. In this work, water samples from three Canadian lakes periodically affected by cyanobacteria were used to assess the impact of natural and algal organic matter (NOM/AOM) on treatment with VUV. NOM and AOM were characterized before and after VUV treatment by size exclusion chromatography (SEC) and fluorescence emission–excitation matrix (FEEM). FEEM spectra were analyzed with the parallel factor analysis (PARAFAC) tool. As a result, we found seven principal components describing the whole dataset. Disinfection by-product (DBP) formation after VUV treatment was analyzed and trihalomethanes (THM) yield was calculated. THM yield increased by 15–20% after VUV treatment. Regarding DBP formation and NOM/AOM fractions from SEC, we found that humic substances are the most important fraction causing the increase in DBP formation with at least 3 times higher yield than the other fractions: biopolymers, building blocks, low weight molecular acids and neutrals.

Perfluoroalkyl substances (PFAS) are drinking water contaminants of emerging concern due to their persistence in the environment and tendency to bio-accumulate. Although anionic ion exchange (IX) resins offer a cost-effective alternative for removing PFAS from natural waters, the studies on PFAS removal by IX under the influence of variable characteristics of the organic compounds present in the natural water sources, have been overlooked. This is critically important given that the organic matter (OM) characteristics are spatially variable and can change seasonally. In the present study, a strongly basic anion exchange resin was used to remove two of the most persistent PFAS, namely per-fluorooctanoic acid (PFOA) and per-fluorooctanesulfonic acid (PFOS). Factors influencing the uptake behavior included the PFOA and PFOS concentrations, resin dosage, and background OM characteristics, more specifically the charge density and molecular weight distribution of source water OM. The equivalent background concentration (EBC) was employed to evaluate the competitive uptake between OM and PFAS. Experimental data were fitted to different mathematical and physical models to evaluate the competitive interactions. Further, IX was able to achieve complete PFAS removal with simultaneous >60% dissolved organic carbon (DOC) removal. Evidence of size exclusion and pore blockage was also observed in the presence of humics and larger molecular weight organic fractions. Results of this study indicate that IX exhibits great potential for simultaneous OM and PFAS removal for drinking and potable water reuse applications.

The increasing frequency with which cyanobacterial blooms are affecting sources of drinking water is a growing concern worldwide. Such events are usually responsible for the presence of cyanotoxins as well as taste and odor (T&O) compounds. Vacuum UV (VUV) is a promising advanced oxidation process used to treat water impacted by cyanobacterial blooms, with potential applicability in small and remote communities because of its simplicity. Here, we present the performance of a VUV process, in both a collimated beam reactor (CBR) and a pilot scale flow-through reactor, using two cyanobacterial-laden Canadian source waters with different inorganic and natural organic matter contents. First, VUV performance was assessed by comparing the removal of microcystin-LR (MC-LR), 2-methylisoborneol (MIB), and geosmin (GSM). The average kinetic rates obtained in the CBR case were 2.9 × 10−3 cm2 mJ−1 for MIB and GSM and 6.6 × 10−3 cm2 mJ−1 for MC-LR. Under bloom conditions, removals of 40–60% for T&O compounds and MC-LR were achieved in the flow-through reactor. It was observed that although MC-LR, GSM, and MIB were impacted by VUV treatment, the removals achieved may not be sufficient to completely eliminate toxicity and T&O at the tested fluences (up to 400 mJ cm−2). In addition, we observed a 20% increase in disinfection by-products (DBPs), on average. Hence, achieving high MC-LR, MIB, and GSM removals with VUV may cause the generation of more DBPs.