“To date, we at Vitens have applied almost exclusively groundwater extraction and bank filtration for the production of drinking water. We have extraction permits that specify the volume of groundwater we may extract per month and per year. In recent hot and dry summers, however, we already witnessed that our customers’ demand for water in some areas exceeded these levels. One of the methods we apply in our endeavour to tackle this problem is to encourage our customers to use water both sparingly and consciously. Despite this policy, we expect to reach the limits of our extraction permits more often in the coming years. Given that the development of new groundwater extraction sites can easily take a decade or longer, we are to launch two projects that investigate the possibilities of using surface water as a raw material.”

Ultrapure water
“The first of the projects is to examine whether we might produce ultrapure water by applying reverse osmosis membranes to remove all unwanted substances from surface water. We subsequently intend to blend this ultrapure water with purified groundwater. Should this prove feasible, we will proceed to construct a membrane system that is both modular and can be conveniently assembled and disassembled. This would enable us to swiftly deploy the system in areas where we cannot – temporarily or otherwise – extract sufficient groundwater and where we have insufficient control of the quality of the surface water used.”

“If one could succeed in removing only the problematic substances, it would yield a far more efficient solution.”

Abel continues: “One of the disadvantages of reverse osmosis is that it involves high energy consumption. Moreover, the water produced is not immediately suitable for human consumption, as all the minerals have been removed. We are therefore keen to investigate whether membrane filtration might prove suitable for the direct production of reliable drinking water. After all, the ability to remove solely problematic substances would yield a far more efficient solution. To carry out trials with this purification technique, we intend to construct a pilot plant alongside the River IJssel, which will have to run for at least a year. This should enable us to establish whether membranes serve as a sufficiently effective barrier, how quickly they become clogged and how they behave in the long term. It is also an opportunity for Vitens to simultaneously acquire vital experience in the purification of new sources and all that this might entail.”

Test bed
Abel goes on to explain that another research project involves the construction of a test bed in Friesland. “Our ultimate aim is to develop a completely new purification system there, which applies all the innovative techniques we have researched at other production sites in recent years. We trust that this will enable us to design the most sustainable, innovative and energy-efficient purification system. Trials are scheduled to commence next year. We intend to build three parallel purification lines – one that applies conventional techniques, while the other two feature a combination of innovative technologies – to establish which operates most favourably over the course of the year. We will then equip the new treatment plant with them.

“Another project again involves research into the removal of PFAS. These toxic substances deteriorate very gradually, if at all, and are becoming highly abundant in the environment. This implies that water companies such as ourselves need to devote increasing attention to these harmful substances. Our existing treatment plants have so far succeeded in removing PFAS by means of activated carbon and membrane filtration. In anticipation of the fact that surface water will increasingly serve as the basis for our drinking water production, we intend to carry out model-based research into the removal of PFAS by means of carbon filtration.”

“We are keen to avoid releasing a potent greenhouse gas like methane into the air, and are therefore looking for useful applications that reduce the environmental impact.”

Methane valorisation
Vitens increasingly extracts dissolved methane from the groundwater at its various production sites. Abel: “Methane is a potent greenhouse gas. We are keen to avoid simply releasing extracted methane into the air, and are therefore looking for useful applications that reduce the environmental impact. We are collaborating closely with the team from the Circular Economy and Society theme in this regard. Given that the amount of gas extracted differs per site, as do possible applications in the immediate vicinity – e.g., using the methane to heat properties or mixing it with the biogas produced by a manure digester – we need to map the most favourable applications at each particular site. Should the search fail to yield a feasible application, however, we could simply opt to flare the gas, as the environmental impact of the resulting CO₂ emission is significantly lower than that of methane itself.

The final innovation project Abel mentions is aimed at the development of a pellet reactor for the deferrisation of groundwater: “In collaboration with Delft University of Technology, we are to examine the possibilities of developing a continuous process for the removal of iron, thus avoiding the need to backwash sand filters in future. We are currently considering the use of a compact pellet reactor, which promotes the adsorption of a layer of iron to the granular filter material, which then need only be replenished now and then. We are due to launch a small scale pilot programme at our production company in Wierden in the spring of 2022.”

Contribution?

Would you like to contribute to the efforts Vitens’ innovation programme is making within this theme? In that case, please contact us through: abel.heinsbroek@vitens.nl

*This is an article from the magazine Vitens Innovates – Purification theme.

“Reverse osmosis is a highly suitable means of desalinating our raw water,” says process technologist and membrane expert Bas Reitman. “This technique consists of us pumping water at a pressure of seven to eight bar through membranes mounted in tubes that retain over 98 percent of dissolved salts. This produces clean permeate – water from which the salts have been removed – and a concentrate flow of brine. During the process, the concentration of salts on the feed side increases to such an extent that they begin to precipitate on the membrane surface. This causes such a decline in permeability that it is no longer efficient to continue with the production of permeate. Although we can briefly prevent precipitation of salts by adding a kind of liquid Calgon to the feed water, we do eventually have to dispose of the membrane concentrate.”

“We began to consider the issue of minimising water loss, as some twenty per cent of the feed water is lost in discharging the concentrate”

Minimising water loss
“Some twenty per cent of the feed water is lost in discharging the concentrate,” process technologist Nico Wolthek explains. “Not only is this a waste of the water itself, but also the energy required to extract and pressurise it, and the chemicals we use to treat the feed water. We therefore began to consider the issue of minimising water loss. This has led to the launch of two research projects. The first – which we have since completed – investigated whether we might reduce water loss by improving the entire membrane filtration process. The second research project is examining the possibility of our reducing water loss by post-treating the concentrate stream.”

Nico continues: “During the first project, we investigated the possibilities of feed flow reversal. The principle of this technique is that you reverse the direction of flow of the feed water shortly before signs of salt scaling appear on the surface of the membrane. This scaling always starts in the end section of the membrane tubes (see ROTEC diagram). The salt concentration is highest there, as it is furthest away from the point where the feed water enters. Reversing the direction of flow and directing the feed water through the pipe where the concentrate flow originally exited causes the salt concentration in this section of the pipe to decline once more.”

“We built a pilot set-up next to the existing membrane system at our Het Engelse Werk production site, which is fed with the same raw water. Trials carried out with this system show that we can reduce the water loss to seven per cent. This is a highly favourable result. We are nevertheless also investigating another means of reducing water loss. After all, we would have to install many additional valves and complex control technology in our existing systems in order to apply feed flow reversal. That would call for not only substantial investment, but also increased maintenance.”

“The advantage of this technique is that one can continue to use the existing systems, which implies the investment costs are much lower”

Concentrating concentrate flow

Bas: “In the second project, we are examining the possibilities of further concentrating the concentrate flow. In particular, we are considering a technique called closed circuit reverse osmosis. It involves the additional extraction of water from the concentrate flow, while regularly expelling the thick concentrate, as it were. We hope to complete this study in the course of 2022.”

Nico adds: “The advantage of this technique is that one can continue to use the existing systems, and need only install a small additional unit for post-treatment of the concentrate flow – which has a relatively small volume. The investment costs are therefore considerably lower than those that would be incurred in rendering the membrane systems suitable for feed flow reversal.”

Inspired?
For further information or a partnership, please contact us through bas.rietman@vitens.nl or nico.wolthek@vitens.nl.

A proven technique, sand filtration is the oldest and most widely used purification process at Vitens,” explains process technologist Frank Schoonenberg. “A sand filter removes iron, manganese and ammonium dissolved in the water. Contrary to popular belief, a sand filter is not a simple sieve but a kind of reactor in which physical, chemical and biological processes take place that also influence one another. However, we are neither really sure how this influence is brought to bear, nor how the various processes actually take place. Despite the long-term application of this form of filtration, sand filters are still largely a black box, and it remains more or less a mystery why one sand filter works well while the other does not.”

“Despite the long-term application of this form of filtration, sand filters are still largely a black box”

Cross-pollination
“With a view to improving our understanding of the way sand filters work, we embarked on the Sand Filtration research programme together with drinking water company Dunea, thanks to financial support from the Netherlands Organisation for Scientific Research (NWO). Launched in 2020, this programme comprises five research projects, in which no fewer than five Dutch universities and three research institutes participate. Two of these projects are focused on rapid sand filtration, and are therefore highly relevant to Vitens. The two are being carried out by Delft University of Technology, Utrecht University and Radboud University Nijmegen. Each project is executed by two PhD students from different departments. This approach was adopted because it allows both different insights to be combined and cross-fertilisation to take place. In performing their research, the PhD students make use of filter material and water supplied by Vitens and Dunea.”

“The studies are all pretty fundamental and focus on various aspects of sand filtration. For example, the researchers from Utrecht and Nijmegen endeavour to gain an insight into the individual microbiological and geochemical processes that take place in sand filters. To this end, they combine modern DNA analyses with geochemical measurements. The DNA analyses enable them to make an inventory of all the bacteria that are present and active in the sand filters. The geochemical measurements – based on microscopic photographs taken by the researchers, among other things – map out highly precisely where in a sand filter the iron and manganese are adsorbed. Our own laboratory also contributes to the research projects. For example, our colleagues measure the amount of iron-oxidising bacteria in the sand filters using a qPCR method.”

Process solutions
“Delft’s research projects are executed by PhD students from its biotechnology and water management departments. Among other things, they are investigating which adjustments to the process conditions might lead to improved operation of sand filters. They are also investigating the possibility of applying more efficient process solutions. For example, should it transpire that various processes have a detrimental effect on one another or operate more favourably individually, it may prove beneficial to ‘separate’ them. And we have done just that in the pilot project being run at our Hammerflier production site. There, we first remove the iron using a sand filter under relatively low-oxygen conditions. We then increase the oxygen content by means of aeration, to remove manganese and ammonium from the water in another filter.”

“A better understanding of the way processes work should enable us to both further optimise and design them more efficiently”

Frank is enthusiastic about the research programme: “This kind of in-depth long-term research is highly suitable in gaining a better understanding of purification processes that we have been using for quite some time, yet still do not fully comprehend. Of course, it is not simply a matter of comprehension. After all, a better understanding of the way processes work should enable us to both further optimise and design them more efficiently”

Inspired?
For further information or a partnership, please contact us through frank.schoonenberg@vitens.nl

“If one considers our current drinking water infrastructure, I would not hesitate to say that it will not be adequate beyond the year 2050,” says Rian. “Its flexibility is insufficient to cope with the consequences of climate change, for example. In recent hot and dry summers, for example, we already witnessed that our customers’ demand for water in some areas exceeded the quantity that our permits entitle us to extract. This was the case, for instance, in the high, sandy territory on the eastern side of our supply area. There, we extract groundwater from vulnerable water systems, where the risk of drought damage is relatively high. Given that this type of negative environmental impact is exacerbated by climate change, we may have to close those extraction sites in due course.”

“This should render us less dependent on local conditions, while ensuring our infrastructure becomes more resilient.”

“In other parts of our region, we do not face these problems. There, we extract groundwater from deep aquifers, which has little effect on the environment. The plan is to extract more water from these ‘strategic cores’ in the future. We also want to establish a mains network that will enable us to quickly transport this water to other areas. This should render us less dependent on local conditions, while ensuring our infrastructure becomes more resilient.”

Rian continues: “Developing additional extraction sites and establishing a mains network is a long-term process. That is the reason we are currently examining the possibilities of using membrane filtration to purify surface water, so that we can anticipate water shortages in the coming years. At the same time, we are acquiring knowledge through these research projects on possible means of producing suitable drinking water using other water qualities in the future.”

“Soft sensors in combination with the dashboard are fantastic tools to support operators in their work”

Controlling quality
In addition to establishing a resilient infrastructure, we also endeavour to gain the greatest possible control of the quality of the water we produce, and in real time,” adds Doeke. “This is the basic objective of the SLIMM project. As part of the initial pilot, we fitted as many sensors as possible at a treatment plant so as to continuously monitor all the process conditions. It soon became apparent that this was not the solution, as sensors require considerable maintenance. Moreover, colossal investment would be required to purchase sensors for all our treatment plants. We therefore began to consider other solutions. This led to the development of what are known as soft sensors, which calculate the required data on the basis of a limited number of measured values. “These in combination with the dashboard are fantastic tools to support operators in their work. And the digital twins developed as part of SLIMM offer great opportunities to improve treatment processes.”

“The digital twins developed as part of SLIMM offer great opportunities to improve treatment processes.”

“Another long-term objective – stemming from our ambition to make our operations as sustainable as possible – is to use everything we extract from the ground as effectively as possible,” Doeke continues. “Our research into reducing water loss using reverse osmosis is a good example, as is the pilot at Hammerflier with vacuum degasification. There, we use membranes to remove 99% of the greenhouse gas methane from the groundwater, leaving us with almost pure methane that can be put to good use. We also recover substances such as calcium carbonate and humic acid, using selective separation techniques in such a way that we can market them as useful raw materials. Another example is the research into deferrisation using a pellet reactor. If we manage to achieve this type of deferrisation, we will probably be left with a residual flow that is far more readily reusable. And even if that does not prove the case, the cost of disposing of the residual flow will at least be greatly reduced. After all, the pellets largely consist of iron, whereas of the iron sludge that remains after deferrisation using sand filters comprises 92% water.”   

Safe and healthy
Rian: “It goes without saying that our main objective remains the production and supply of safe and healthy drinking water in the long term. Given the increasing number of undesirable man-made substances in the environment – such as medicine residues, pesticides and resistant substances like PFAS – this is a major challenge. Particularly if future climate change and greater peak demands increasingly oblige us to rely on sources other than groundwater, which are of far more variable quality. Research into new purification processes with which to remove these types of substances as effectively as possible therefore remains essential.”

Contribution?

Would you like to contribute to the efforts Vitens’ innovation programme is making within this theme? In that case, please contact us through: rian.kloosterman@vitens.nl or doeke.schippers@vitens.nl

“When SLIMM was originally launched, there were various reasons to digitise and automate the treatment plants,” Nico explains. “We at Vitens have over ninety production sites that are highly diverse. In the past, each site was run by an operator who often had decades of experience, and could, as it were, hear, smell and taste whether the processes were running smoothly therefore. Many of these experienced operators have since retired, or are due to do so in the foreseeable future. We therefore had to come up with something clever to keep the knowledge in-house.”

Rein points out another reason: “While our experienced process operators did know how to set up and operate the purification processes, they focused largely on optimising the quantity of drinking water produced. The only means of monitoring water quality was to take water samples now and then, which were then analysed by our laboratory in Leeuwarden. This implied that we ran the risk of not detecting undesirable anomalies in time. This way of working was at odds with our ambition to invariably supply high quality drinking water. If we are to achieve that ambition, we need to be able to manage the water quality in real time. Such an approach calls for constant awareness of both the quality of the incoming water and how favourably the various stages of treatment are proceeding. While this could be achieved by equipping all treatment plants with a large number of sensors, that would require huge investments and a great deal of maintenance. We therefore opted for a different approach.”

Digital twins
“Our approach boils down to making a digital twin of each treatment plant,” Nico explains. “This is a simulation model that accurately reflects the course that the water takes at a production site – from the extraction wells to the purification tank – as well as the various purification processes. Important components in these digital twins are software sensors, also known as soft sensors. These are mathematical models that allow us to calculate the most important process stages. For example, we have developed a sensor with which we can continuously calculate the quality of the raw water. Each production site is fed by several extraction wells, each with its own particular water quality. Constant switching takes place between the wells, partly to prevent well blockages. As a result, the quality the incoming water constantly varies, which has an effect on the various purification stages.”

“We will continue to develop new sensors in the years ahead, so that we can simulate the most complex processes”

“We similarly created a soft sensor that allows us to calculate the hardness of the water coming out of a pellet softening reactor. Furthermore, we developed sensors that can predict when a plate aerator needs to be cleaned, how much scaling is occurring or what dosage of a particular substance is required. We will continue to develop new sensors in the years ahead, so that we can simulate the most complex processes. To validate each soft sensor, we compare the calculated values with those measured in the laboratory. We carry out such comparisons regularly, with a view to further improving the predictive power of the sensors.”

“We have developed dashboards that display the properties of the water produced at each site in real time”

Central database
Rein: “A central database forms the core of our digitisation operation. We store the findings of laboratory research, the readings taken by physical sensors, and all data from the assets at the various production sites – such as the treatment processes in place, the dimensions of pipes and the capacity of pumps – in this database. We use all this data as input for our soft sensors. The results of the calculations are also stored in the database. We then use the stored data to support our operators. For instance, we have developed dashboards that display the properties of the water produced at each site in real time, including hardness, the pH and turbidity level.”

“Around fifty of our production sites currently have digital twins, and digitisation is embedded in daily production. Initially, we used the data from the SLIMM database mainly to inform our operators. We now also use it to advise them on optimum settings for water quality, energy consumption and the dosage of chemicals alike. We trust that we will eventually be able to use the data to actually control dynamic processes – such as soda dosing – so that we are no longer entirely dependent on operators.”

“The simulations reduce unwanted quality deviations and can also be used to train new employees.”

Calculating scenarios
Nico emphasises that digitisation is not only improving the daily operations of the production sites – there are far fewer undesired quality deviations, for instance – but also has a number of other advantages: “The digital twins are highly suitable for training new employees. Furthermore, unlike the production locations themselves, they offer us the opportunity to experiment with settings. For example, you can calculate all kinds of scenarios to establish how to produce the required water quality at the lowest possible cost and with minimum consumption of energy and chemicals. This is entirely in keeping with the Vitens-wide ambition of ‘Every single drop sustainable by 2030’.”

In-house development
The SLIMM innovation project commenced with a pilot project at the Oldeholtpade treatment plant. Vitens cooperated with national and international parties in its execution. Once it became apparent that the results were favourable, Vitens opted to continue independently. Rein: “Since then, we have done almost everything ourselves, without external assistance. We mainly use open source software to develop the soft sensors and digital twins. There are a number of advantages to ‘in-house’ development. Firstly, external costs are low, as we only have to invest in our own time. Another benefit is that we can fine-tune everything to our own business processes. And if a user has specific requirements, we can readily anticipate them.” 

Inspired?
For further information or a partnership, please contact us through nico.wolthek@vitens.nl or rein.wuestman@vitens.nl