2021

‘Reduce incidents by implementing a source approach’

An abnormal discharge can always happen at a factory. But if you have your operations properly under control, you can prevent it from developing into an incident in the surface water. This arises from Hans Geijselaers' tale on technical service provider Sitech Services, which is responsible for the wastewater management at the Chemelot industrial complex.

Chemelot is an industrial estate covering 800 hectares in Zuid-Limburg. It is home to 54 factories and over 150 different businesses. Sitech arranges that the wastewater from the factories is transported via a 290 km sewerage system to the central biological wastewater treatment plant, where it is purified before being discharged into the Grens Meuse (Border Meuse) and Meuse.

What is special about the discharge situation at Chemelot is that a little further on at Roosteren, drinking water is made from the Meuse water. The river water there must meet the strict abstraction standards to produce drinking water. The discharge from Chemelot must take this into account. This means that the effluent that leaves the discharge pipe from the industrial estate is monitored 24/7.

"But at the I-WWTP (industrial wastewater treatment plant), you’re sometimes already too late," admits Hans Geijselaers, who has been manager of wastewater treatment at Sitech for 3½ years. He also pleads for an approach at the source to reduce discharges. On request, Hans answers 11 questions about how Sitech goes to work at Chemelot.

1. Fifty-four factories discharge into your industrial wastewater treatment plant (I-WWTP). How do you get a grip on this complex process?

"Agreements with the companies are largely laid down in contracts. Besides this, we operate within the discharge permit, which is one permit for the entire site that includes all the factories. Our contracts with the companies are therefore essentially an extension of our discharge permit. Our I-WWTP team coordinates all the circumstances concerning the permit, and how we can meet the requirements.

Our discharge permit is relatively new. For several years, the discharge has been viewed in a completely different light, namely at the level of individual substances. Before this, we controlled the quality of the effluent based on group parameters and only on a few individual substances. To get a view of any more individual substances in the effluent, we invested significantly in new monitoring and analysis techniques.

All this also fits in with the 'greening vision' of the entire Chemelot site, aimed at enhancing and reaching sustainability and circularity. To this end, specific discharges are going to be significantly reduced into the future. At this point, there is a survey underway of everything we can do to reach that objective (circularity)."

2. You supervise the process and know what is happening in it. What role does monitoring have in this?

"Monitoring is crucial and it happens at various places on the site. It starts in the factories themselves. Next, the wastewater goes into the sewerage pipes where we have analysis equipment present in several places. Finally, we measure the wastewater just before and after it is purified in the I-WWTP. In this case, we conduct targeted analyses and a broad screening of the effluent. This moment is your last line of defence: if something has gone wrong in a factory, you would naturally prefer to detect it as early as possible. If you notice it in time, you can then act accordingly to prevent it leading to an improper discharge into surface water bodies."

3. In summary: you monitor the substances in the permit at different places, and as a safety net, you have extra screening of the effluent in case incidents occur, is this right?

"Yes, we’ve been doing the measurement of effluents for years. But in recent years, the measurement package has indeed been significantly expanded. We use five different analytical techniques to see whether any deviations can be detected in the effluent.

For example, for one year we have been working with a bio-monitor to measure toxicity; this is done with mussels. They are highly sensitive to toxic substances. These cause the mussels to close up. We can record this movement using electrodes, and then we check whether something is the matter.

The importance of good screening and monitoring is that you can act quickly in the case of an irregular situation. You preferably want to prevent breaches of the standard taking place. But if this should happen, then you want to restrict the breach as far as possible."

4. What do you do in case of deviations in the discharge?

"Firstly, we want to know the concentration of the substance. To be clear: if we see something in the screening, we're by definition talking about very low concentrations. Then fortunately we’re still far from the phase where there is a problem. For surface water, we screen substances at concentrations of 0.1 micrograms per litre, while the warning level for abstraction for the drinking water companies is 1 microgram per litre. You’ve already built in a safety factor with this factor of 10.

Follow-up? This can go into two directions. If it's a known substance, we can respond very quickly. If it's an unknown substance, we must start looking. This involves real ‘detective’ work: identification and tracking. We do this together with Aqualab Zuid's laboratory. By applying different analysis techniques, we define a molecular formula and then we search for this molecule further back in the stream."

5. How do you proceed if the problem is unknown substances at very low concentrations?

"We work with five different screening methods, of which a couple are also used by the drinking water companies themselves. This is useful, because then we’re speaking the same language. If we detect a peak in an effluent sample, that single peak can also be caused by three different substances together. For this reason, the laboratory possesses techniques with which they can determine the molecular mass. Once you know this, your next steps can be highly focused. Then you can also do an actual analysis on your water streams to trace the substance. This way of working is really intended to put the dots on the i’s, and to ensure that you don't see any alien substances. Because when you screen, you really see everything, in spite of concentrations at hand being often very low."

6. Some substances, such as PFAS, are already problematic for drinking water production even at extremely low concentrations. How do you determine whether a new, unknown substance is a problem?

"The laboratory then gets to work to determine the molecular structure and molecular mass. If you're in luck you know which substance it could be. Then you can search in a directed way for information about the substance to find out the degree to which it is 'water-problematic'. You can use resources including the ECHA/REACH files for this, but these are unfortunately often incomplete.

So, we research further in literature, or in the databases of suppliers of chemicals. The laboratories themselves also have information. But if everything still yields insufficient information, we have a comprehensive study done. Then you have to have toxicity tests conducted."

7. When do you determine how problematic a substance is: what procedure do you use to set priorities for all these substances?

"We cooperate closely with the drinking water companies and we seek advice from KWR and Aqualab Zuid. You must realise that we abstract Meuse water and that we also discharge it again. Whatever is in the Meuse, we draw from it. This is why we also monitor the background level of the substances in the water that we abstract. Recently, we found four PFAS components in our effluent, while these substances do not arise in our usual discharge. These substances proved to be already present in the Meuse water that we abstract. They have no priority for us to search any further."

8. Do you also use screening to manage abnormal discharges?

"Be aware: investigating emerging substances and managing incidents are two different things. The screening discussed above is intended as a safety net. Large streams are never involved here, because the substances in the large streams are known and you don't need any screening for this. We know all the individual substances that we discharge; these are listed in the permit and are measured.

Things proceed differently when it comes to an unusual discharge. If something unforeseen happens in a factory, we have monitoring points at various places in the process to spot an unintended discharge coming along. In this case, we can switch the stream to the storage basin and enter discussion with the factory to stop the discharge. In this way, we prevent the discharge ending up in the I-WWTP and then in the surface water.

But I must confess that this doesn't always work. The wastewater goes quickly through the sewer. We have an average of four hours before it arrives at the I-WWTP. Sometimes therefore we exceed our discharge standard. Thanks to all the monitors we have installed, we can ensure that discharge is noticed quickly. We can prevent the discharge being long-term, with detrimental consequences for the surface water or at any event minimise the potential consequences.

Usually, the impact of an abnormal discharge proves to have less detrimental consequences. Often the abnormal discharge is not observable at the drinking water abstraction point. Recently, we had to deal with a short-term standards breach. To prevent the drinking water companies being jeopardized by this, we immediately phoned the WML drinking water company. But they didn't see any peak in their screening."

9. How can it happen that a standards breach is not measured further down the Meuse?

"This has to do with extra safety margins that are built in. If I talk about a breach in standards, I mean a standard that applies to our I-WWTP permit. The standard for surface water is translated back into a standard for our effluent in the permit.

Because we discharge into the Grensmaas, this standard is adapted to the receiving surface water, and moreover calculated for a very low Meuse water level. The standard is therefore set for situations of low water flow. In other words: if we exceed our standard in the effluent, whilst the water discharge in the Meuse is not low, it will usually cause no problem. It remains the case that we don't accept any standards breaches."

10. If we are talking about the impact of substances, what about indicative target values?

"This is quite complicated. If you identify a known substance, you can find toxicological information about it. Once you have this data you can derive a standard based on it and then you know what your limit is with regards to the discharge. The more data you have, the more accurately you can derive the standard. If you have too little data for a standard, you then must work with safety factors and thereby the standard becomes an indicative one. You can then compare a substance with another that closely resembles it. Based on this you can still derive an indicative standard.

You don't do this just like that; all kinds of strict rules apply to it. The RIVM has instructions on how you should do this. This costs time. Mainly because the standard application must be tested in a scientific sounding board group that only meets a couple of times a year and they can only handle a limited number of applications each time.

Briefly: if you must apply for many standards, it can take quite a while. This is a stumbling block for innovation.” We often want to introduce new substances into the production process, which are better or less harmful than existing substances. You want to implement this as quickly as possible."

11. Your way of working has proved successful because drinking water companies see far fewer incidents than previously. Do you agree?

"We’ve had a better grip on the process for quite a while, ever since the incident with pyrazole in 2015. This was a national wake-up call at the time; afterwards there was attention to individual substances. We went to work immediately, and I now dare to say that we have the situation well under control. That's not to say that nothing ever happens. But if something goes wrong, we're on it in a flash. We also always communicate with the Limburg Water Board, Rijkswaterstaat and the drinking water companies. In this way, we can prevent a repeat of an incident like that one of 2015.

Our vision of the future? The dot on the horizon is finally zero discharges. Until that time, we will need to manage and mitigate the current residual risk even better. Therefore, we work more and more at the source, in the factories themselves. The source approach is extremely important, especially for abnormal discharges."