Machine learning modeling of predictive external corrosion rates of spent nuclear fuel carbon steel canister in soil.

Soil corrosion is always a critical concern to corrosion engineering because of the economic influence of soil infrastructures as has been and has recently been the focus of spent nuclear fuel canisters. Besides corrosion protection, the corrosion prediction of the canister is also important. Advanced knowledge of the corrosion rate of spent nuclear fuel canister material in a particular environment can be extremely helpful in choosing the best protection method. Applying machine learning (ML) to corrosion rate prediction solves all the challenges because of the number of variables affecting soil corrosion. In this study, several algorithms of ML, including series individual, boosting, bagging artificial neural network (ANN), series individual, boosting, bagging Chi-squared automatic interaction detection (CHAID) tree decision, linear regression (LR) and an ensemble learning (EL) merge the best option that collects from 3 algorithm methods above. From the performance of each model to find the model with the highest accuracy is the ensemble stacking method. Mean absolute error performance matrices are shown in Fig. 15. Besides applying ML, the significance of the input variables was also determined through sensitivity analysis using the feature importance criterion, and the carbon steel corrosion rate is the most sensitive to temperature and chloride.

Adsorptive removal of five heavy metals from water using blast furnace slag and fly ash.

Heavy metals can be serious pollutants of natural water bodies causing health risks to humans and aquatic organisms. The purpose of this study was to investigate the removal of five heavy metals from water by adsorption onto an iron industry blast furnace slag waste (point of zero charge (PZC) pH 6.0; main constituents, Ca and Fe) and a coal industry fly ash waste (PZC 3.0; main constituents, Si and Al). Batch study revealed that rising pH increased the adsorption of all metals with an abrupt increase at pH 4.0-7.0. The Langmuir adsorption maximum for fly ash at pH 6.5 was 3.4-5.1 mg/g with the adsorption capacity for the metals being in the order Pb > Cu > Cd, Zn, Cr. The corresponding values for furnace slag were 4.3 to 5.2 mg/g, and the order of adsorption capacities was Pb, Cu, Cd > Cr > Zn. Fixed-bed column study on furnace slag/sand mixture (1:1 w/w) revealed that the adsorption capacities were generally less in the mixed metal system (1.1-2.1 mg/g) than in the single metal system (3.4-3.5 mg/g). The data for both systems fitted well to the Thomas model, with the adsorption capacity being the highest for Pb and Cu in the single metal system and Pb and Cd in the mixed metal system. Our study showed that fly ash and blast furnace slag are effective low-cost adsorbents for the simultaneous removal of Pb, Cu, Cd, Cr and Zn from water.

Polycyclic aromatic hydrocarbons in road-deposited sediments, water sediments, and soils in Sydney, Australia: Comparisons of concentration distribution, sources and potential toxicity.

Sixteen polycyclic aromatic hydrocarbons (PAHs) considered as priority environmental pollutants were analysed in surface natural soils (NS), road-deposited sediments (RDS), and water sediments (WS) at Kogarah in Sydney, Australia. Comparisons were made of their concentration distributions, likely sources and potential toxicities. The concentrations (mg/kg) in NS, RDS, and WS ranged from 0.40 to 7.49 (mean 2.80), 1.65 to 4.00 (mean 2.91), and 0.49 to 5.19 (mean 1.76), respectively. PAHs were dominated by relatively high molecular weight compounds with more than three fused benzene rings, indicating that high temperature combustion processes were their predominant sources. The proportions of high molecular weight PAHs with five or six fused benzene rings were higher in NS than in RDS, whereas the low molecular weight PAHs were higher in RDS. Concentrations of all PAHs compounds were observed to be the lowest in WS. The concentrations of most of the high molecular weight PAHs significantly correlated with each other in RDS and WS. All PAHs (except naphthalene) were significantly correlated in NS suggesting a common PAH source. Ratios for individual diagnostic PAHs demonstrated that the primary source of PAHs in WS and NS was of pyrogenic origin (combustion of petroleum (vehicle exhaust), grass, and wood) while in RDS it was petrogenic (i.e. unburned or leaked fuel and oil, road asphalt, and tyre particles) as well as pyrogenic. The potential toxicities of PAHs calculated using a toxicity equivalent quotient (TEQ) were all low but higher for NS compared to WS and RDS.