Rapid prediction of the activated carbon adsorption ratio by a regression model

Identification of odor-causing compounds in six species of odor-producing microalgae separated from drinking water source with distinct fishy odor: Insight into microalgae growth and odor characteristics

Fishy odor, as an offensive and unpleasant odor, could occur in drinking water source with poor nutrition, it is usually considered to be associated with odor-producing microalgae. However, the specific relations among fishy odor, fishy odorants and responsible microalgae were not elucidated comprehensively. In this paper, the odor-causing compounds generated from six microalgae with fishy odor characteristic, isolated in drinking water source Tongyu River, were identified simultaneously. The sensory evaluation result indicated that Tongyu River was principally related to fishy odor (odor intensity 6.5-7.6). Correspondingly, seven, nine, seven, six, seven and seven olfactory detection peaks (ODP) were screened by combining data of GC/O/MS and GC/GC/TOFMS in Cyclotella, Cryptomonas ovate, Melosira, Dinobryon sp., Synedra and Ochromonas sp., which were isolated in Tongyu River and cultured in laboratory. Totally twenty odor-causing compounds, including hexanal, 2-hexenal, 3-hexen-1-ol, heptanal, 1-octen-3-one, 2,4-heptadienal, 2-tetradecanone, 3,5-octadien-2-one, octanal, 1-octen-3-ol, 2-octenal, nonanal, 2,4-octadienal, 2-nonenal, decanal, 2,6-nonadienal, 2-decenal, undecanal, 2,4-decadienal and dodecanal, were screened and identified in all isolated microalgae. Additionally, fishy odor intensity for all identified microalgae increased obviously as microalgae cell number increased and microalgae cell ruptured in cultivation cycles through pearson and spearman correlation analysis. For the first time, twenty odor-causing compounds associating with fishy odor were recognized from six isolated microalgae, which would provide more scientific basis and theoretical support for preventing and treating fishy odor episode of drinking water source.

Comparison of Organic Matter Properties and Disinfection By-Product Formation between the Typical Groundwater and Surface Water

The disinfection by-product (DBP) formation was affected by the dissolved organic matter (DOM). Therefore, the DOM properties and DBP formation potential of the two most widely used source waters: groundwater (GW) and surface water (SW), were comparatively studied in this work. The results suggested that the GW mainly consisted of protein-like organics with smaller molecular weight (Mw) less than 3000 Da, while the SW contained the humic- and fulvic-like substances with larger Mw. The tap water DBP concentration of GW as source water was lower than that of SW as well as the cytotoxic index (CI). The total DBP formation potential of the SW under chlorine and chloramine disinfection was higher than that of GW, especially the trihalomethanes (THMs) and haloacetic acid (HAAs). The higher THM and HAA formation potential of the SW was mainly attributed to the relatively hydrophobic and aromatic humic and fulvic substances. The halonitromethanes (HANs) formation was mainly due to the less hydrophobic protein-like components with smaller Mw. In addition, the total CI of the GW was lower than the SW under both chlorine and chloramine disinfection. Therefore, for the DBPs control, using the GW as source water was more beneficial to human health.

Mechanisms and application of the IAST-EBC model for predicting 2-MIB adsorption by PAC in authentic raw waters: Correlation between NOM competitiveness and water quality parameters

Natural organic matter (NOM) exerts negative impacts on 2-methylisoborneol (2-MIB) removal by powdered activated carbon (PAC), thus adding to the difficulty in accurate PAC dose prediction. Our study investigated the application of the ideal adsorbed solution theory-equivalent background compound (IAST-EBC) model and its simplified version for PAC dose prediction. Four raw water samples were employed, and the corresponding C_0,EBC values, indicating NOM competitiveness, were calculated. The results showed that the IAST-EBC model presented ideal predictive performance in 2-MIB adsorption under both equilibrium and nonequilibrium conditions and the C_0,EBC values of the Huangpu River (8800 ng/L) and Qiantang River (10300 ng/L) were high, representing the higher NOM competitiveness in these two rivers, which may be caused by municipal effluent and industrial wastewater discharge. In contrast, Tai Lake water showed a lower C_0,EBC value (6400 ng/L), which was likely associated with algae and other microbial activities. The fluorescence index (FI, the ratio of E_x/E_m = 370/470 nm to E_x/E_m = 370/520 nm) can be applied to estimate C_0,EBC, thus facilitating prediction. Our study also showed that the IAST-EBC model can be further simplified under lower initial 2-MIB concentrations or longer contact times, which is particularly useful for practical applications.

Degradation of odorous sulfide compounds by different oxidation processes in drinking water: Performance, reaction kinetics and mechanism

Swampy/septic odor caused by various sulfides is one of the most frequently encountered odor problems in drinking water. However, even though it is much more offensive, few studies have specifically focused on swampy/septic odor compared to the extensively studied musty/earthy problems. In this work, four sulfide odorants, diamyl sulfide (DAS), dipropyl sulfide (DPS), dimethyl disulfide (DMDS) and diethyl disulfide (DEDS), were selected to evaluate the treatment performance of different oxidation processes in drinking water. The results demonstrated that DMDS, DEDS, DPS and DAS could be oxidized effectively by KMnO_4, NaClO and ClO₂. The oxidation processes could be well described by the second-order kinetic model, in which k values of selected sulfides followed the order DMDS≈DEDS ≪ DPS≈DAS. As for the three oxidants, the order of reactivity was KMnO₄ ≪ ClO₂ < NaClO, which was also verified in raw water. The results of oxidation treatability, reaction kinetics and mechanisms confirmed that the characteristics of the central sulfur atom rather than the side chain is the decisive factor in controlling the oxidation rate and transformation pathway of sulfides. The transformation products and pathways were significantly different for the three oxidants. Sulfones (DPSO, DASO) were always formed by cycloaddition reactions during KMnO₄ oxidation, yet recombination reactions proceeded during ClO₂ oxidation and formed more products, such as MADS, DADS and EADS. Density functional theory (DFT) calculations confirmed that the differences in transformation pathways were caused by the variations in the activity of the oxidants and sulfides. Finally, NaClO was certified as the most effective oxidant for controlling sulfide odorants in drinking water treatment.

Predicting Aqueous Adsorption of Organic Compounds onto Biochars, Carbon Nanotubes, Granular Activated Carbons, and Resins with Machine Learning

Predictive models are useful tools for aqueous adsorption research; existing models such as multilinear regression (MLR), however, can only predict adsorption under specific equilibrium concentrations or for certain adsorption isotherm models. Also, few studies have discussed data processing beyond applying different modeling algorithms to improve the prediction accuracy. In this research, we employed a cosine similarity approach that focused on mining the available data before developing models; this approach can mine the most relevant data concerning the prediction target to build models and was found to considerably improve the prediction accuracy. We then built a machine-learning modeling process based on neural networks (NN), a group-selection data-splitting strategy for grouped adsorption data for adsorbent-adsorbate pairs under different equilibrium concentrations, and polyparameter linear free energy relationships (pp-LFERs) for aqueous adsorption of 165 organic compounds onto 50 biochars, 34 carbon nanotubes, 35 GACs, and 30 polymeric resins. The final NN-LFER models were successfully applied to various equilibrium concentrations regardless of the adsorption isotherm models and showed less prediction deviations than the published models with the root-mean-square errors 0.23-0.31 versus 0.23-0.97 log unit, and the predictions were improved by adding two key descriptors (BET surface area and pore volume) for the adsorbents. Finally, interpreting the NN-LFER models based on the Shapley values suggested that not considering equilibrium concentration and properties of the adsorbents in the existing MLR models is a possible reason for their higher prediction deviations.