Search

Water research 195, 116940 (2021). doi:10.1016/j.watres.2021.116940, Published by Elsevier Science, Amsterdam [u.a.]

Groundwater resources are under growing pressures in water scarce countries. Membrane distillation (MD) is an interesting option for drinking water production from groundwater with elevated salinity and fluoride (F) contamination. A direct contact MD (DCMD) at a moderate feed temperature of 55 °C was able to concentrate a synthetic solution representing F contaminated groundwater. An average initial flux of 13.3 L/m 2 /h was achieved at 75% water recovery, resulting in only 15–17% flux decline, while producing high quality permeate (96–99% F rejection). Membrane autopsy showed presence of Ca, Na, S and Mg on the used membrane. Particularly, 51–53% F precipitation was estimated in line with the saturation index (SI) model. The used MD membrane with groundwater showed only 10–12% reduction in membrane hydrophobicity, which was largely recovered with simple cleaning. Meanwhile, synthetic groundwater solution spiked with humic substances resulted in brownish deposition on MD membrane, reducing the membrane hydrophobicity significantly by 37–40%. Additionally, DCMD operation with vacuum at the permeate side (vacuum enhanced DCMD; VEDCMD) was beneficial in increasing the permeate flux by 42%. Continuous VEDCMD operation with intermediate membrane cleaning showed positive results in treating F contained groundwater while producing good quality permeate at 67% water recovery.

The removal of five selected pesticide compounds in a brackish model groundwater solution was examined using a bench scale direct contact membrane distillation (DCMD) system. It was found that the rejection rate of the pesticides in DCMD is mainly influenced by its properties. Compounds with low hydrophobic characteristics and low vapour pressure showed a high rejection rate (70–99%), whereas compounds with a high vapour pressure or high hydrophobicity (LogD) showed a reduced rejection (30–50%) at a water recovery of 75%. The influence of groundwater feed solution contents such as the presence of organics (humic acid) and inorganic ions (Na+, Ca2+, Mg2+, Cl− and SO4 2−) as well as feed temperature (40, 55 and 70 °C) on the rejection of the pesticides in DCMD operation was also evaluated. The results showed that the presence of inorganic ions and organics in the feed solution influences the pesticides rejection in DCMD operation to a minor degree. In contrast, reduced rejection of pesticides with high vapour pressure was observed. A rapid small-scale column test (RSSCT) was carried out to study the removal of any remaining substances in the permeate by adsorption onto granular activated carbon (GAC). RSSCT showed promising performance of GAC as a post-treatment option.

Because organic micropollutants (MP) are frequently detected in river waters that are used as drinking water sources, combining a relatively cost-efficient natural treatment with upstream advanced oxidation processes (AOP) appears promising for their efficient abatement. Such a multi-barrier system can be integrated in drinking water production schemes to minimize risks from potentially hazardous MPs. This study investigates the impact of an UV/H2O2 AOP before soil aquifer treatment (SAT) on the abatement of selected MPs (EDTA, acesulfame, iopamidol, iomeprol, metformin, 1H-benzotriazole, iopromide), dissolved organic matter (DOM) (apparent molecular size distribution, specific UV absorbance at 254 nm – SUVA) and microbial parameters (intact cell count, cell-bound ATP). A pilot plant consisting of an AOP (0.5 m3 h−1, 4 mg L−1 H2O2, 6000 J m−2) and two parallel soil columns (filtration velocity: 1 m d−1, column height: 1 m) was continuously operated over a period of 15 months with Rhine river water pre-treated with rapid sand filtration. The investigations revealed a shift towards longer retention times of the humic substances peak in LC analysis of DOM, lower SUVA and higher biodegradability of DOM after UV/H2O2 treatment. In addition, an overall higher abatement of all investigated MPs by the combined treatment was observed (AOP with subsequent SAT) compared to either process alone. This observation could be explained by an addition of the single treatment effects. The strong primary disinfection effect of the AOP was detectable along the first meter of infiltration, but did not lead to any change in the column performance (i.e., similar abatement of dissolved organic matter).