UTFacultiesTNWClustersResearch groupsMTEWTResearchCompleted ProjectsAssessing biological stability of drinking water after reverse osmosis and remineralization

Assessing biological stability of drinking water after reverse osmosis and remineralization

mohaned sousi (UNESCO-IHE/UT)

Promotors: Maria Kennedy (UNESCO-IHE), Walter van der Meer (UT). Co-promotor: Sergio Salinas (UNESCO-IHE).

Additional supervisors: Gang Liu (Key Laboratory of Drinking Water Science and Technology), Antoine Kemperman (UT)

Funded by Water Company Oasen

Defense date: December 9th, 2021.

The Ph.D. thesis can be downloaded here.

Summary of the Ph.D. thesis

Biological stability definition and controlling factors

Water utilities aim at providing drinking water that remains biologically safe and stable during distribution. The term biological stability refers to maintaining the microbiological quality of drinking water without extensive bacterial growth, either quantitative or qualitative, during distribution. Growth of planktonic bacteria or biofilms in water distribution networks leads to several problems, such as health threats posed by (opportunistic) pathogens, operational problems including bio-corrosion of pipes and fittings, and adverse effects on the aesthetic characteristics of drinking water (i.e., taste, colour, and odour). Therefore, controlling bacterial growth in distribution systems is an important factor in complying with biological stability requirements, which can be achieved by the addition of residual disinfectant. However, drinking water in the Netherlands is distributed without residual disinfectant because of the health threats associated with the formation of disinfection by-products. Alternatively, multi-barrier treatment strategies are applied in the Netherlands to remove bacterial growth-promoting nutrients from source water.

Membrane technology, including reverse osmosis (RO) filtration, is increasingly applied worldwide in water purification processes, providing an effective solution for retaining substances that promote bacterial growth. Water after RO filtration (i.e., RO permeate) lacks the essential minerals for human consumption (e.g., calcium and magnesium). Therefore, post-treatment including remineralisation of RO permeate is required before distribution. However, limited knowledge is available on the effect of post-treatment (remineralisation) on the biological stability of ultra-low nutrient drinking water produced by RO filtration.

This study was conducted at a drinking water production site in the Netherlands where anaerobic groundwater is currently treated by conventional means comprising dry sand filtration (rapid sand filters fed with spray aerated water), pellet softening, rapid sand filtration, activated carbon filtration, and UV disinfection. At the same site, anaerobic groundwater is also treated by a pilot-scale unit comprising reverse osmosis and post- treatment (including remineralisation and aeration). Water samples were collected from both conventional and RO-based treatment line and assessed for biological stability using the bacterial growth potential (BGP) method.

Method development

The main aim of this dissertation was to further develop the current BGP method, using flow cytometry (FCM), for the assessment of biological stability of ultra-low nutrient water produced by RO and remineralisation. The results showed that cell count by FCM is reproducible for this type of water where changes in cell count and fluorescence fingerprints could be detected after on-site treatment or adjustments at the laboratory (Chapter 2, 3, and 4). Furthermore, the limit of detection of the BGP method could be lowered by using ultrapure blank prepared from RO permeate that is remineralised at the laboratory. The lower limit of detection ensures obtaining reliable BGP results for water with very low BGP such as remineralised RO permeate (Chapter 2). The effect of sample pre-treatment prior to BGP measurement was assessed, where pre-treatment by pasteurisation or membrane filtration are commonly applied. It was found that membrane filtration has a pronounced negative influence on BGP of ultra-low nutrient water due to the leaching of organic filter material in water, whereas no significant effect of pasteurisation on the nutrient content of water was observed (Chapter 3).

Pre-treated water samples were inoculated using a natural bacterial consortium originating from conventionally treated drinking water, which was accompanied with the addition of nutrients to ultra-low nutrient water when a high inoculum concentration was added (Chapter 2). Alternatively, using bacteria originating from water produced by RO and remineralisation as an inoculum was evaluated, where it was observed that these bacteria have limited ability to utilise some types of organic compounds, more specifically organic carbon with complex molecular properties (Chapter 3). Therefore, an inoculum concentration of 10 × 103 intact cells/mL originating from conventionally treated drinking water was considered throughout this research since the associated nutrient addition was found to be insignificant (Chapter 2). The developed BGP method was useful for the identification of bacterial growth-limiting nutrient in ultra-low nutrient water (Chapter 4). Moreover, the developed BGP method was combined with nutrient analysis, adenosine triphosphate (ATP) measurement, and 16S rRNA gene sequencing to obtain an in-depth understanding of why growth occurs (limiting nutrients) and which bacteria grow (dominant bacterial genera) in the types of water studied (Chapter 4 and 5).

Method application

The application of the developed BGP method showed that the type of water treatment plays an important role in determining the biological stability of drinking water, where long term monitoring of BGP revealed that RO-based treatment (RO and remineralisation) resulted in >75% reduction in BGP compared with conventional treatment (Chapter 3), due to the high removal of organic and inorganic nutrients achieved by RO filtration (>99% removal). Moreover, it was observed that the BGP of RO permeate was negatively influenced by post-treatment, where the BGP increased by 100% when RO permeate underwent post-treatment processes at the production site. Post-treatment, especially remineralisation by calcite contactors, led to the addition of organic and inorganic nutrients in RO permeate (Chapter 4). For both water types, i.e., conventional drinking water and ultra-low nutrient water, organic carbon was the bacterial growth-limiting nutrient, even at very low concentrations of phosphate (<1 μg/L PO4-P) (Chapter 4). The type of water treatment also shaped the bacterial community of the finished treated water, where the bacterial genera identified in RO-treated water were mainly introduced during the post-treatment processes (Chapter 4). Furthermore, it was found that the assessment of BGP using both FCM and ATP is especially useful as additional complementary information can be obtained from the combined tests, such as the effect of phosphate limitation on bacterial growth, where cellyield from phosphate was estimated at 0.70 ± 0.05 × 109 cells/μg PO4-P (Chapter 5).


This study clearly demonstrated that RO filtration is a promising technology to considerably limit bacterial growth in drinking water. However, measures are recommended to mitigate the negative influences of post-treatment on BGP, such as using high quality calcite grains for remineralisation to prevent the introduction of organic and inorganic nutrients, and more frequent maintenance of the aeration towers, or even investigating alternative aeration methodologies (e.g., membrane aeration). Further research is needed to investigate the effect of RO-treated water on bacterial growth and ecology in the current distribution system, with a special focus on the transition period when switching from the current conventionally treated drinking water to RO-treated water. In the future, research should also focus on improving analytical techniques to elucidate the specific organic compounds that promote growth in drinking water as well as ensuring that these specific compounds are adequately removed during treatment.


Sousi M., Salinas-Rodriguez S.G., Liu G., Dusseldorp J., Kemperman A.J.B., Schippers J.C., Van der Meer W.G.J., Kennedy M.D., Comparing the bacterial growth potential of ultra-low nutrient drinking water assessed by growth tests based on flow cytometric intact cell count versus adenosine triphosphate, Water Research, 203 (15 September 2021), article number 117506, https://doi.org/10.1016/j.watres.2021.117506

Sousi, M., Liu, G., Salinas-Rodriguez, S.G., Chen, L, Dusseldorp, J., Wessels, P., Schippers, J.C., Kennedy, M.D., van der Meer, W., Multi-parametric assessment of biological stability of drinking water produced from groundwater: Reverse osmosis vs. conventional treatment, Water Research, 186 (1 November 2020), article number 116317, https://doi.org/10.1016/j.watres.2020.116317

Mohaned Sousi, Sergio G. Salinas-Rodriguez, Gang Liu, Jan C. Schippers, Maria D. Kennedy, Walter van der Meer, Measuring Bacterial Growth Potential of Ultra-Low Nutrient Drinking Water Produced by Reverse Osmosis: Effect of Sample Pre-treatment and Bacterial Inoculum, Frontiers in Microbiology, 11 (2020), article 791, https://doi.org/10.3389/fmicb.2020.00791

Mohaned Sousi, Gang Liu, Sergio G. Salinas-Rodriguez, Aleksandra Knezev, Bastiaan Blankert, Jan C. Schippers, Walter van der Meer, Maria D. Kennedy, Further developing the bacterial growth potential method for ultra-pure drinking water produced by remineralization of reverse osmosis permeate, Water Research, 145 (2018), 687-696, https://doi.org/10.1016/j.watres.2018.09.002