Natural viruses to monitor the integrity of membrane installations

Urbanisation, industrialisation, climate change and other phenomena are placing pressure on the sources used in the production of impeccable (drinking) water. For a variety of treatment objectives, membrane processes can supplement and/or replace several – sometimes traditional – water treatment processes. This can make a positive contribution to the provision of bacteriologically safe water and the removal of harmful substances, including iron, manganese, arsenic, pesticides, endocrine disruptors, pharmaceutical products and personal care products. This project is investigating how viruses can be used to monitor the integrity of membranes.


Membrane technology is being applied with greater frequency and at an increased scale in the production of high-quality (drinking) water from a variety of sources, such as groundwater, surface water, waste water and seawater. One of its key functions is the removal of pathogens, particularly viruses. The water sources contain a number of natural viruses at concentrations of 1×108 to 1×1010 viruses per litre. This is 10 to 100 times more than the bacteria present. Since the size of the viruses means than they cannot pass through the membrane pores, it is technically possible to use membrane processes to produce water that is completely free of viruses. Intact membranes have a natural virus log reduction value (LRV) of 8-10. This project will study whether this LRV can be drawn on to use viruses to detect whether leaks occur in membrane installations.


Current methods can only determine a maximum LRV of 3, and the capacity of membranes to remove viruses completely cannot be verified. At present, then, adequate monitoring of membrane integrity with the LRV is not possible. This is nevertheless required if there is damage to the membrane or the membrane module.

Damage to membranes reduces the efficiency with which they remove viruses (as well as other contaminants in the water sources). The seals and O-rings in the membrane modules can, for instance, become defective. Fibre breakage can occur in hollow fibres on the membrane surface or glue lines can become damaged. Leakage can considerably reduce the virus reduction values of membrane modules. In principle, the virus LRV is therefore a potential candidate to test for leakage.


KWR has developed a new method based on natural viruses (NV) for monitoring the integrity of membranes. It uses viruses that are naturally present in high concentrations in the water. Using low sample volumes, and without having to add chemical or biological virus surrogates beforehand, NV makes it possible to monitor virus removal very sensitively (LRV of 7 or more), and therefore to keep a finger on the pulse of the membrane integrity. Moreover, the high virus concentration in the source water makes detection possible without requiring laborious sample concentration procedures.

Thanks to the simple Natural Virus method, the integrity of full-scale membrane installations can be routinely monitored with a much higher level of sensitivity than usual. This means that an important obstacle to the application of high-pressure membranes has been eliminated.


The TKI research into the monitoring of membrane integrity with natural viruses consists of five subject areas:

  1. The development of the current test – based on qPCR – for NV in surface waters that monitors the LRV of membrane installations. On the basis of, among other things, new metagenomic approaches and innovative microscopic analyses, the natural viruses are being characterised for the first time, for example to determine how big they are, the load and which types of virus are involved.
  2. The testing and further development of the NV method under different conditions, and the development of new NV methods with qPCR for the determination of natural viruses in other water types: groundwater, waste water and seawater
  3. The determination of the integrity of intact and damaged ultrafiltration membranes based on virus removal on the laboratory, pilot and practical scales. Tests have been conducted for this purpose on the laboratory and pilot scales. The impact of fibre breakage on UF membrane integrity has been studied using the NV method. It has emerged from the tests that a single fibre breakage leads to a significant reduction in the LRV value. Measurements on the laboratory, pilot and practical scales show that using the NV method can determine an LRV of 5 for UF membranes.
  4. The determination of the integrity of intact and damaged reverse osmosis membranes based on virus removal on the laboratory, pilot and practical scales. Tests have been conducted for this purpose on the laboratory and pilot scales. The impact of physical (e.g. O-rings) and chemical (e.g. chlorine) damage to membranes on RO membrane integrity has been examined using the NV method. The different types of damage inflicted led to a reduction in LRV values. The NV method would seem to be very suitable for determining the impact of damage on UF and RO membrane integrity. In addition, measurements were conducted on the practical scale during different periods.
  5. Research into the development of a biosensor for target viruses in water systems for rapid on-site measurement. The research will show – or it has already made methods available for this purpose – whether the NV method can be use on-site (in time) for the fast monitoring of membrane integrity without calling on the services of specialised laboratories. The developments still required in this respect are being identified and described.