Guardians of the Green and Blue Planet

The UN’s 2019 Global Chemical Outlook report tells us that the current chemical production capacity of 2.3 billion tons is set to double by 2030 (1). That’s less than 10 years from now – and what about the decade that follows? There is more pressure on the analytical community than ever before to monitor compounds, identify and quantify the harmful ones and, ultimately, ensure they are properly regulated. 

We spoke to Jacob de Boer, Derek Muir, Valeria Dulio, and Diana Aga about some of the key issues around the regulation of chemicals in the environment, and asked them – where do we go from here?

Meet the Gurus

Jacob de Boer

Jacob is a professor of environmental chemistry and toxicology. He started out at the Netherlands Institute for Fisheries Research, where he used fish to study the identification, quantification, and behavior of contaminants – in particular, persistent organic pollutants and bioaccumulating contaminants. Nowadays, his scope has widened to study these compounds across the whole environment – from the sea and rivers to the soil and air. And even in furniture. Jacob’s background is in chemistry, and so, although he’s often referred to as a toxicologist because of his body of work – he says he’s not a “true” toxicologist.

Derek Muir

Derek is a research scientist with Environment and Climate Change Canada – the government department responsible for coordinating environmental policies and programs. His focus is mainly on persistent organic chemicals and, more recently, emerging contaminants. Derek has also done a lot of work in the Arctic; for example, recently he has been assessing the links between temporal trends in POPs in the biota in relation to climate change. Indeed, he is the co-chair of the POP’s Expert Group of the Arctic Monitoring and Assessment Program and has been their resident POP expert since the 1990s. He also leads the review of perfluoroalkyl substances (PFAS) in water for the Global Monitoring Program.

Valeria Dulio

Valeria – an industrial chemist from the University of Torino, Italy, is senior program manager on emerging contaminants at INERIS. Since its creation in 2005, she has coordinated the NORMAN network on contaminants of emerging concern, a former EU-funded project – and now a permanent network with over 80 members in 20 countries. She is a technical expert in European programs and ensures the animation of various national and international working groups. Her early career focused on pollution prevention strategies for industrial installations, eventually at EU Commission level.

Diana Aga

Diana is a Chemistry professor at the University at Buffalo (UB), the State University of New York, and the director of the UB RENEW Institute. In addition to supervising and mentoring PhD students in Analytical Chemistry, she is also coordinating interdisciplinary research groups to tackle problems related to energy, environment, and water. Recently, she and her team have been investigating the fate, effects, transport, and treatment of emerging contaminants in the environment, ranging from antimicrobials to perfluoroalkyl substances (PFAS). Diana is an active member of the Philippine-American Academy of Science and Engineering, and has been involved in promoting collaborations among scientists and engineers of Philippine descent, writing white papers and preparing proposals for the Philippine government on resolving environmental problems in an effort to advance science and technology in the Philippines. 

Has the environmental monitoring of chemicals changed over the years?

Jacob: An interesting question – and I think everyone will have a different perspective. In Europe, and the Netherlands in particular, there have certainly been fluctuations – mostly tied to funding cuts. I remember the 1990s poultry scare in Belgium – often referred to as the Dioxin Affair (even though it was actually PCBs). It was a big deal at the time, and I ended up being called in to give some data on dioxins and PCBs in fish. The problem: I could only give data from 1992. Someone in the ministry told me: “That’s not good enough, we need the latest numbers.” And I said: “Well, you cut the monitoring program in 1990 so we don’t have them!” Within a year of that case, the funding was back and monitoring was increased again. Such stories vary from country to country. But even in individual countries, there are some amazing examples of a less than strategic approach – like companies requiring a permit for the PFAS compounds they emit into water but not into the air! 

In Canada, there appear to have been a lot of closures of research facilities in recent years. And in the US, while Trump was in power? Well, basically nothing happened. If you look to developing countries, it is completely different because very little monitoring is going on. In  China, the trend is almost the opposite. I’ve worked there as an expert for years, and it’s interesting to see how their attitude to the environment has changed. During their booming years in the 1990s, I’d say almost no attention was given to the environment. But this changed after the Beijing Olympics and the amount of smog kicked them into actually doing something with environmental analysis. Now, they are focusing on cleaning the water and soil – but it’s one hell of a job and it will probably be another 100 years or so to get there. It’s nice to see so many papers and research coming out of China – but the damage is already done in some ways. 

Derek: It’s interesting to hear Jacob say that, because I don’t think I have the same perspective – as he predicted! I would say there’s more interest now than ever in screening chemicals for persistence or bioaccumulation. However, I’ve published several articles on chemical inventories globally – across Canada, the US, China, and Europe – and they are massive. And that’s why, when you ban one substance, industry is quick to simply find a substitute that we know less about – this issue has been tightened up significantly in Europe under REACH – but not so much in other countries. 

I will say there’s probably been less of an effort made with persistent organic pollutants (POPs) in comparison to all the other organic chemicals being monitored. But overall, I don’t think there’s been a reduction in monitoring effort – it’s just spread out over more chemicals. I’d go as far as saying that there are now more environmental chemists involved in monitoring trace organics than ever. 

Diana: I think it definitely depends on where you are in the world. And also on the pressure from the public – like in the case of PFAS. These compounds are being regulated now because of increasing evidence showing they are toxic and ubiquitous in the environment. Recent advances in analytical technologies have enabled us to detect mixtures of PFAS in complex environmental samples even at very low levels. I wouldn’t say that there’s been a reduction in the monitoring efforts in the US. On the contrary, the US EPA recently set a timeline for regulating PFAS in drinking water and controls on industrial discharges of PFAS-containing wastewater – but it’s certainly slow to get regulations enforced.  

In developing countries, there are no regulations on many of these so-called “emerging contaminants” because there are different (and arguably more challenging) problems that the government is trying to tackle – like poverty and hunger. We know that emerging contaminants, such as pharmaceuticals and personal care products, are important to monitor, but developing countries simply don’t have the capacity to do so. That said, some progress is being made – mostly because of the wider education around issues like the increased emergence of antimicrobial resistance. For example, in the Philippines they have recognized the problem of the widespread occurrence of antibiotic residues, antimicrobial resistance genes, and resistant bacteria in wastewater. Hence, several researchers are now involved in understanding how to control the spread and proliferation of antimicrobial resistance in aquatic environments.

Valeria: I’d have to disagree in terms of Europe. There are several recent examples of large-scale monitoring projects supported by environmental authorities, like the Joint Danube Survey – now the largest multinational monitoring campaign in the world, connected with the implementation of the EU Water Framework Directive (WFD), involving 51 sites in 13 countries with more than 2700 chemical substances in different matrices. Also, in France, every 3–5 years large national exploratory monitoring campaigns (more than 100 sites) are organized to acquire information on CECs in the aquatic compartment in support of the implementation of the Water Framework Directive.

The issue is that, as we are now able to measure almost any chemical present in the environment, the authorities must recognize the need for a more integrated strategy to deal with these extensive lists of substances and the associated mixture effects. And that’s where it might seem that the monitoring effort is reduced – because the task has got so much larger! The authorities also need to be able to communicate to their citizens the actual progress made in improving the environment as a result of the mitigation measures that have already been implemented.

Has too much responsibility for monitoring and data collection been given to industry?

Jacob: In some ways, the answer ties into the previous point related to funding. There were further cuts in Europe in the 2000s, particularly after the 2008 financial crash, but the key change here was not just the money – it was the idea that industry should start to take on the burden for monitoring their outputs themselves. And, of course, industry was pretty happy to do that because it meant they only had to report the data needed for permits. 

I’ll demonstrate this with an example. In the 1990s, we had the Quality Program of Agricultural Products in the Netherlands, which involved monitoring pesticide levels in fruits and vegetables. Eventually, the responsibility for assessing this was given to the trading companies, so they will now self-report things like the level of nitrates in green vegetables. We know that levels are higher in certain periods of the year, and there’s nothing to stop industry reporting their results around these periods to suit their needs. So, I think it’s fine to give industry more responsibility, but there should be something else in place as well – like unscheduled checks. 

Derek: The whole organic chemical industry is a trillion-dollar operation, and it’s true that, to an extent, they are self-regulating. Because these are businesses, there is a lot of proprietary information, which makes life difficult for environmental chemists as it prevents the dissemination of some of this information. Oftentimes, we are very much in the dark (except for some legacy chemicals like PCBs). We have access to a lot more data on things like pharmaceuticals and pesticides. But with industrial chemicals, that’s not the case – there is a lot of information that industry simply isn’t prepared to make available. 

Valeria: What has been overlooked in the past year is the crucial role of monitoring data in support of the data collected by industry. The risk assessment protocols applied under the Plant Protection Products and Biocidal Products Regulations, the EMEA Guidelines for risk assessment of veterinary and human pharmaceuticals, and the REACH Regulation are mainly based on a prospective assessment of the risks derived from models looking at consumption, use, and hazard properties data. 

This approach should be corrected in the future. Post-registration monitoring data should be required in a common platform and monitoring data should systematically be used for re-authorization decisions. 

Diana: It’s a complicated issue. We know that industry is required to do standard testing before anything is released into the environment (in most cases), but a lot of this standard testing might not be enough. For example, when EPA requires a company to show a certain pesticide degrades in the environment, they don’t require the company to identify what is formed from this degradation when the by-products are less than 5%, whether these products are toxic or not. So I think there’s a great deal more to be done in terms of regulating what happens to products after they enter into the environment. Oftentimes, when EPA goes through the process to institute new regulations, EPA will establish a voluntary stewardship program that challenges the companies to reduce chemical releases into the environment.

How does the complex nature of many pollutants affect their regulation?

Jacob: It makes regulation difficult because as chemists, we like to focus on a specific compound – or, in the case of PFAS, an individual congener that is more toxic than another. But this causes problems, because time and time again you find new congeners – it’s tiring just keeping on top of them! And by the time you get one regulated, you’ve found another that’s even more toxic. I’ve worked for years on brominated flame retardants, of which there are around 75 different compounds identified. After years of research, arguments and discussion, we’ve managed to get three of them banned. Okay, this is important, but the industry just continues with the other 72 in the meantime. So it’s great to see more focus being given to a whole group of compounds, like PFAS, in the Green New Deal and I think that’s a massive improvement. 

Valeria: Yes, I’d say grouping compounds is a management solution for regulators to be able to restrict the use of problematic compounds with common modes of action. PFAS are a great example of compounds that need to be treated as a group. In the case of PFAS, the regulatory approach is even more complex because this group encompasses more than 4000 different compounds. In the Drinking Water Directive (DWD) there are already lists of PFAS that will be monitored and regulated; in the WFD there are also actions underway to integrate about 20 PFAS on the list. But it is very difficult to prioritize the most hazardous compounds to measure or the appropriate list of specific compounds to be monitored with associated limit values because toxicological data are still limited. 

But what is most important is that PFAS are going to be banned in Europe as a whole class of compounds (except for those applied for essential uses). So the crucial step is also to define (chemical or effect-based) indicators to assess the effectiveness of PFAS management measures.

Diana: We obviously want to be inclusive of all the chemicals when we analyze them for regulatory purposes. However, environmental contaminants vary from small to large, and polar to nonpolar molecules. It makes it impossible to analyze all of them in one method, hence analysis of a wide range of chemical contaminants can be cost-prohibitive. It takes a lot of work and resources to extract them and analyze them using different techniques. Some people have tried freeze-drying and concentrating all the compounds, but this means you just concentrate the background matrix as well. It’s something we’re still struggling with as chemists, but I hope we see improvements in the future. The highly complex nature of pollutants make it difficult to regulate them, because if you cannot detect them, obviously you cannot regulate them.

One challenge with environmental analysis is that new toxicological insights are being revealed all the time. How does this affect the regulatory landscape and the max tolerance limits set?

Jacob: A striking example is from 2019, when the European Food Safety Authority came up with new advisory tolerance levels for PFAS. As in many cases, the chemists first detect a specific contaminant, but there is little toxicological information. Due to this lack of information, the initial tolerance level was relatively relaxed. Then two or three years later the toxicologists came back with their data and said actually, the limit needs to be 1000 fold stricter. That is a massive difference! The reason for this change was that the initial studies were mostly focused on cancer occurrence, whereas the toxicologists found an effect related to immunotoxicity that occurred at a much lower level.  

This is just one striking example, but we do see this sort of thing happening often.

Derek: Though I don’t really work in this area, I can share one good example; recently, an antioxidant used in tires was found to degrade and then find its way into rivers after a storm, where it was causing sudden death in salmon (2). I think this example is particularly interesting (and worrying) because it was a standard chemical being used in thousands of kilograms of tires each year. How many more cases like this could be occurring that we don’t yet know about? The toxicological tools are there, but there are too many chemicals and degradates to monitor.

Diana: Let’s take the case of the herbicide metolachlor as an example. This herbicide has two enantiomers, S-metolachlor and R-metolachlor, which can only be separated using a chiral column. The R isomer only has 50 percent herbicidal activity relative to the S isomer. So when applied as a mixture, most of the R-isomer only contaminate the environment without providing any herbicidal activity! Now, the new herbicide formulation is enriched with S-isomer to reduce unnecessary pollution in the environment. This is a positive example of how new analytical techniques that provide efficient separation of chemical mixtures to reveal components that have different toxicities can drive changes in environmental regulations.

Valeria: New knowledge means that the threshold values are regularly subject to review – and they often become more stringent; it can also have implications on the analytical techniques that must be routinely employed. However, it is important to distinguish between the regulated substances for which the applied threshold values are more stable (warranting  periodic reviews), and emerging contaminants for which conservative or predicted threshold values are precautionarily applied during investigative monitoring studies before they are progressively refined in view of integration of these substances in the regulatory monitoring context (for example, isothiazolinones, fipronil). 

What are the key differences between labs in so-called developed and developing countries?

Jacob: A lot of the work I do has a global reach, and I’ve spent a great deal of time in developing countries. In fact, part of the UN Environment Program is devoted to building capacity in these countries. But in many cases, environmental monitoring is just not a priority for the government. They’ve got more important things to contend with; for example, supplying everyone with food and water – and the prioritization is reflected in the laboratories, which simply don’t get the resources. On occasion, I’ve visited labs and found high-tech mass specs sitting there unused – perhaps a gift from Japan or Sweden! It may be still in its original packaging because they didn’t know where to begin – or it’s fallen into a state of disrepair because there’s no one to service or fix it. When advanced training is provided, it’s impossible to properly remember methods that you’re not using routinely – and the expertise is lost. Finally, it can take months to order reference materials and standards – and when they finally arrive they can be stuck in customs for another few months. All these issues can add up to an almost hopeless situation for these labs. 

That said, if you are able to obtain the right investment and ensure the right resources are available, it can be done. In fact, I’ve seen it work – but sadly it’s the exception rather than the rule, with dedicated people working very hard to make it happen.

Derek: Jacob has been directly involved in these capacity building efforts and, as he mentioned, the expertise may exist or can be developed, but the other challenges are harder to deal with. A further challenge is that, with sufficient training (or even a PhD obtained elsewhere) opportunities outside of the country in need may tempt people away – the brain drain. 

Looking at the problem from a global perspective, it means we rarely get the full picture of chemical contamination.

So, where do we go from here?

Jacob: I think we’ve seen a lot of improvement over the years in environmental analysis – obviously I am now talking more generally and more so from my perspective in the Western world. But we’ve seen MS become much more sensitive, much faster, and much more reliable. And it’s our job as analytical chemists to apply that pressure for better methods. We’ve even developed direct probes to analyze compounds without the need for chromatography – these would be fantastic tools for developing countries. We’ve also made significant gains in terms of the speed of analysis and reduced sample preparation. 

When I started working in this area in 1974, I was determining the concentration of DDT in fish – and we struggled to see 0.1 mg/kg levels! Now, we can find picograms. And every time the detection limits get lower, I think, “This is really it this time.” And every time I’m wrong! It’s amazing how far the field has come. 

The big challenge now is data collection and data analysis. We can now run so many more samples per day than we ever could before, so it’s about how to best handle all that data. 

Derek: Absolutely. We need to look towards more artificial intelligence-based approaches for analyzing and collecting data. I’m actually co-author on a recent paper (3); my Chinese colleagues applied a sort of deep neural network to testing chemicals on the Chinese and European inventories – just to see if tools could be used to rapidly screen these massive lists. And it worked quite well – though we need to improve the reliability. I’d like to see more research into such approaches in the future. I’m not sure about other countries, but the topic certainly isn’t getting enough attention or funding in Canada. 

Importantly, such advances would give us the capacity to screen more widely, before chemicals become commercial. We also need to start looking at how to get rid of contaminants that have been in use for decades, but aren’t being investigated because they were registered before many of our current toxicity concerns.

On top of this, there’s also room for improved analytical methods; we always need better tools, but we also need to apply the ones we have more widely. For example, high-resolution (Orbitrap or ToF) MS should be adopted fully into the monitoring sphere – not just in research labs. But again, this is why AI is important; we don’t want routine labs drowning in seas of data. 

Diana: The ideal situation for water analysis, for me at least, is that the regulations in every country are the same. And I don’t think this needs to be at the extreme of saying, “No chemicals below a certain level,” as there are some that we know are less toxic than others (like pharmaceuticals). So perhaps it would depend on the purpose of the water – whether it’s drinking water or being reused for irrigation in agriculture. 

If I could wave a magic wand, I would present the world with a cheap probe that can be dipped into water to detect every chemical present, before sending the results to a smartphone. Such technology would be game changing for everyone. For now, we can only hope that HRMS becomes increasingly accessible – in terms of both the expertise required and the cost to acquire and maintain the instrument.

Valeria: I’d have to agree on the data front – this should be a key focus in the coming years. And a common platform that provides comprehensive information (in terms of spatial and temporal coverage across a range of matrices) on the exposure and effects of chemicals during the entire life cycle of products would be a game changer.

Such a platform would allow more efficient and systematic harmonization of all data required for the assessment of chemical compounds across different regulatory regimes. It would also ensure more consistent links between prospective risk assessment at the moment of authorization (to allow chemicals into the market) and retrospective risk assessment using monitoring data. It would also facilitate grouping of chemicals and identification of common profiles based on criteria such as chemical structure, mode of action, and sector of use.

Jacob: In an ideal world, we wouldn’t need regulations or permits. Industry would either i) no longer need to use harmful chemicals or ii) would have filters or other technology in place to prevent their waste from entering the environment at all. Almost as unlikely as Diana’s magic wand. In the meantime, we must hope that the combination of increasingly sensitive but user-friendly analytical techniques and software tools will support continued progress.