How to avoid pharmaceuticals in the aquatic environment

Journal of Biotechnology 113 (2004) 295–304 How to avoid pharmaceuticals in the aquatic environment Tove A. Larsen∗ , Judit Lienert, Adriano Joss, Hansruedi Siegrist EAWAG (Swiss Federal Institute For Environmental Science And Technology), Ueberlandstrasse 133, P.O. Box 611, CH-8600 Duebendorf, Switzerland Received 2 September 2003; accepted 4 March 2004 Abstract Pharmaceuticals and other micropollutants in wastewater pose a new challenge to wastewater professionals as well as to the pharmaceutical industry. Although there is a great deal of uncertainty concerning the possible detrimental effects on the aquatic ecosystems, the precautionary principle – or possibly new scientific evidence – may give rise to more stringent demands on wastewater treatment in the future. In conventional wastewater treatment plants, a combination of biological treatment with high sludge residence times and ozonation of the effluent seems to be the most promising technology. Ozonation, however, is an energy-intensive technology. Moreover, in conventional end-of-pipe systems a large part of the pollutants will always be lost to the environment due to leaking, primarily during rain. In the long term, source separation offers the more sustainable solution to the entire wastewater problem, including organic micropollutants. Urine source separation is an elegant solution to the problems of nutrients and pharmaceuticals alike and losses of untreated pollutants to the environment can be minimized. Although few technologies for the separate treatment of urine have been developed to date, the 100–500 times higher concentrations of micropollutants promise more efficient conditions for all removal technologies known from conventional wastewater treatment. © 2004 Elsevier B.V. All rights reserved. Keywords: Cleaner production; Micropollutants; Precautionary principle; Wastewater treatment; Sustainability; Urine source separation 1. Introduction The existence of micropollutants (e.g. pharmaceuticals and hormonally active substances) in the aquatic environment and their possible effects on living organisms are giving rise to growing concern (Heberer, 2002). Substances such as natural estrogenic hormones ∗ Corresponding author. Tel.: +41-1-823-50-39; fax: +41-1-823-53-89. E-mail address: [email protected] (T.A. Larsen). 0168-1656/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2004.03.033 and especially synthetic female steroid hormones are biologically very potent compounds. They are only partly eliminated during conventional wastewater treatment and have been measured in the effluent of various European wastewater treatment plants (Desbrow et al., 1998; Pickering and Sumpter, 2003; Eggen et al., 2003). For instance, 27 of 32 pharmaceutical substances and four of five metabolites were detected in municipal wastewater treatment plant effluents, and in surface waters peak values of over 1 ␮g l−1 were measured (e.g. lipid regulating agents, antiphlogistics, 296 T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 beta-blockers and antiepileptic drugs; Ternes, 1998). To date, over 80 compounds have been found in sewage effluents, surface waters, and even in ground waters (Heberer, 2002). The potential impact on the environment is largely unknown (Länge and Dietrich, 2002) and a causal relationship between e.g. estrogenic compounds and adverse effects on populations of water organisms has not yet been established. However, in many rivers and streams in Switzerland, a 50% decline in fish catch has been observed over the last 15 years, and there is some evidence that organic micropollutants could contribute to this effect (Jobling et al., 1998; Routledge et al., 1998; Burkhardt-Holm et al., 2000, 2002). Traditionally, measures of action were only undertaken, once a sound relationship between cause and effect were established. A combination of rising ethical concern for the environment (Harremoës, 2003) and an increasing awareness of ignorance and uncertainty in science (Kriebel et al., 2001; Harremoës, 2003; Rogers, 2003a) slowly changes this approach. Where scientific causeeffect relationships have not (yet) been established, but discharge of a substance could have potentially harmful effects on the environment, the precautionary principle comes into action (Kriebel et al., 2001; deFur and Kaszuba, 2002; Harremoës, 2003; Rogers, 2003a). In its most extreme interpretation, where the burden of proof has to be carried by the risk generator, the precautionary principle can be considered as a paradigm change regarding the introduction of chemical substances to the environment. The current concern of micropollutants in the receiving waters may also call for new approaches in wastewater treatment. Wastewater treatment plants are designed to deal with bulk substances that arrive regularly and in large quantities (primarily organic matter and the nutrients nitrogen and phosphorus). Pharmaceuticals are entirely different. They are single compounds with an individual behaviour in the treatment plant, and they represent only a minor part of the wastewater organic load. So far, the majority of the compounds and their metabolites cannot even be detected, since the chemical analytical methods have not yet been developed. And at least in some cases (e.g. the antiphlogistic drug diclofenac and the antiepileptic drug carbamazepine), significant removal in conventional wastewater treatment plants has not yet been demonstrated (see Heberer, 2002 for references). Wastewater professionals are considering various options to deal with these new substances. The traditional end-of-pipe processes can be optimized or new treatment steps introduced. A third approach that is receiving increasing interest by wastewater professionals is waste design and source separation (Henze, 1997; Larsen and Gujer, 2001). The idea is to produce a waste with an optimal composition for further treatment and disposal (waste design), primarily based on the separation of different household wastewaters (source separation) and when possible in close cooperation with the producing industry (source control). With the rise of initiatives such as “Responsible Care”, “Green Chemistry” or “Cleaner Production” the pharmaceutical industry itself is giving increasing attention to environmental and sustainability issues (Brandt, 2002). Source control and optimization of the production process have been successfully applied by industry (e.g. Kahn et al., 2001; Overcash, 2002). However, to achieve the goal of a “minimal waste society”, ongoing technological innovation to generate extremely low levels of waste as well as low toxicity and persistence of wastes is needed (Norberg-Bohm, 1999). Including not only waste at the production site but also the waste products occurring after the pharmaceuticals have passed through the human metabolism into this process is an additional challenge that calls for cooperation between industry and wastewater professionals. In this paper, we discuss the different technical possibilities for dealing with this waste and the possible motivation of the pharmaceutical industry to cooperate with wastewater professionals on this task. 2. Technical possibilities of removing pharmaceuticals from wastewater Basically, four different approaches for removal of micropollutants from wastewater are distinguished: optimizing existing treatment technology, upgrading existing treatment plants with new end-of-pipe technology, source separation methods, and source control measures. Improving wastewater treatment plants and applying source control measures are the traditional weapons to improve the quality of wastewater emissions to receiving waters. Source separation is a recent approach, which receives increasing acceptance in the wastewater treatment community (see e.g. Henze, 1997; Larsen and Gujer, 2001). T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 297 Fig. 1. Biological degradation resp. transformation of a micropollutant depends on the aerobic solids retention time, SRT (Siegrist et al., 2003). 2.1. Elimination processes in wastewater treatment plants In a typical European wastewater treatment plant, biological degradation is the only real elimination process of micropollutants. Where sewage sludge is incinerated, however, sorption will also contribute to elimination. In special cases, physical processes produce a concentrated solution of micropollutants that can also be incinerated. Photochemical processes, which can contribute to the elimination of micropollutants in the aquatic ecosystems (Jürgens et al., 2002), only play a minor role in wastewater treatment plants. Chemical oxidation in the effluent is a new treatment technology, still mainly in the experimental phase. Sorption is an important process for the final distribution of micropollutants, but actual elimination depends on the fate of the treatment sludge. Stripping shifts the micropollutants from the aquatic environment to the atmosphere. 2.1.1. Biological degradation or transformation in wastewater treatment plants In wastewater treatment plants, organic micropollutants occur in concentrations of 10−5 to 10−9 g l−1 (Golet et al., 2002). Their degradation is only partial: some compounds are not removed, some partly, and some below the detection limit (Golet et al., 2002). Parameters influencing the degradation efficiency are not yet fully understood; in the focus of current research are sludge age (solids retention time, Fig. 1), substrate availability (substrate inhibition), redox conditions (aerobic, anoxic or anaerobic), sorption (as competitive reaction), and reactor configuration (number of cascaded compartments, biofilm growth surface, sand filtration). Due to the extremely small concentrations of organic micropollutants in wastewater, the mechanisms of biological transformation and degradation are not fully understood. In the case of NTA (a synthetic metalchelating agent), Egli (2001) showed that enzyme induction depended on concentration and the availability of a primary substrate. 2.1.2. Chemical oxidation in wastewater treatment plants A new end-of-pipe technology discussed in connection with the problem of organic micropollutants is the ozonation of wastewater effluents. Ternes et al. (2003) showed that ozone doses of 10 and 15 mg l−1 were capable of reducing the concentrations of all target pharmaceuticals (five antibiotics, five betablockers, four antiphlogistics, two lipid regulator metabolites, and the antiepileptic carbamazepine), of the natural estrogen estrone, and of two polycyclic musk fragrances below the detection limit in the effluent of a biological wastewater treatment plant. However, information on transformation products is still lacking and iodinated X-ray contrast media were detected in appreciable concentrations. These results were supported also by kinetic experiments (Huber et al., 2002). Ozonation of wastewater effluent is not expensive, but rather energy intensive. With approximately 0.1 kWh m−3 , ozonation would cause a 40–50% increase in the energy demand of a normal treatment plant. 2.1.3. Sorption in wastewater treatment plants Sorption of organic micropollutants to the sludge in treatment plants (Fig. 2) depends on two main mechanisms: absorption and adsorption. Absorption is the 298 T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 tion of heavy metals in the soil, and potentially dangerous effects of micropollutants (Seyman, 2003). In most other countries, a significant part of the sewage sludge is still disposed of on agricultural land, a far cheaper solution with the additional advantage of being able to recycle phosphorus from human waste. Fig. 2. Absorption and adsorption of micropollutants to particulate matter. As observed for norfloxacin (Golet et al., 2002). hydrophobic interactions of the aliphatic and aromatic groups of a compound with the lipophilic cell membrane of the microorganisms and with the lipid fraction of the sludge. Adsorption is the electrostatic interactions of positively charged groups of the chemicals with the negatively charged surfaces of the biomass (Schwarzenbach et al., 2003, p. 275ff). With the concentrations of micropollutants encountered in wastewater, an approximate linear correlation between particulate and solute concentration of a given compound can be assumed: Xi = Kd,i · Xss · Si (1) Xi is the particulate concentration of the micropollutant i; Xss the sludge concentration; Si the soluble concentration of the micropollutant i; Kd,i the sorption constant. With mainly hydrophobic interactions, the sorption constant Kd,i (1 gss −1 ) can be estimated from the octanol–water partition coefficient Kow (or even better from the partitioning coefficient to particulate organic matter Koc ), whereas by electrostatic interactions, Kd,i is found empirically. Since sorption only involves a phase shift of the pollutants and no degradation or even transformation, the fate of the sewage sludge is central to the environmental evaluation of this process. In Switzerland, the spreading of sewage sludge on agricultural land is banned with immediate effect for market gardens and forage crops, and from 2006 on for other crops (Seyman, 2003). Incineration of sludge is to become standard practice. The main reason for the ban was the application of the precautionary principle, as consequence of increasing doubts about food safety, mad cow disease, accumula- 2.1.4. Stripping in wastewater treatment plants Stripping takes place in the aerobic part of a wastewater treatment plant due to the intensive aeration of the mixed liquor. Stripping depends on the aeration intensity and the Henry coefficient of a given compound. Stripping removes about 90% of perchlorethylene (H = 0.77), but only 5% of the musk compound tonalide (H ≈ 0.005). Since the majority of pharmaceuticals have a molecular mass above 250 g mol−1 and are hydrophobic with a Henry coefficient below 0.005 (Schwarzenbach et al., 2003, Appendix C), stripping is not of much practical concern. 2.1.5. Physical removal processes in wastewater treatment plants In membrane bioreactors (MBR), the secondary clarifier (characteristic for conventional activated sludge wastewater treatment) is substituted by membranes. Since the micro- and ultrafiltration membranes used for this purpose have a pore size between 100 and 1000 times bigger than the physical size of compounds qualifying as micropollutants (molecular weight between 100 and 1000 g mol−1 ), no direct physical retention of the compounds by membrane bioreactors can be expected, if not in connection to the previously discussed sorption or biological degradation. One important feature of MBR is that the solids retention time can be increased significantly above the levels that can be obtained with secondary clarifiers. Its impact on the degradation of micropollutant is currently being studied. Generally MBR do have a rather high energy demand (1 kWh m−3 ) and high costs: membrane bioreactors are economically competitive where sensitive surface water require advanced treatment or where the space availability is very limited (Walther, 2001). The removal of organic micropollutants in the effluent from wastewater treatment plants by nanofiltration, reverse osmosis or activated carbon are mainly discussed in view of reuse of wastewater (Schafer et al., 2003; Wintgens et al., 2002), because of the high energy (ca. 1 kWh m−3 only for this membrane filtration T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 step) and material demand of these technologies (Côté et al., 1997; Nghiem et al., 2002; Wintgens et al., 2002). Hartig et al. (2001) suggest a combination of ultrafiltration to remove the bulk organic matter, followed by treatment with activated carbon. 2.2. The source separation approach Traditionally, there have been two ways of reducing impact from anthropogenic substances in the aquatic ecosystems: removal of the substance in wastewater treatment plants or giving up (voluntarily or following a ban) production of the substance. Industry, however, invented a third approach: more intelligent production processes to reduce the emission of pollutants. In the last 15–20 years there was a dramatic increase in pollution prevention programmes within advanced industrialized countries, which is often referred to as a paradigm shift (Overcash, 2002). Recently, these ideas have been taken up for pollutants from households as well. The reason for this was not the occurrence of micropollutants, but the wish to find more sustainable solutions to the complex task of protecting the environment from the hazards of water-borne pollution (see Henze, 1997; Larsen and Gujer, 1997). Synergies between this general wish and the wish to tackle more efficiently the problem of organic micropollutants rapidly became obvious (Larsen et al., 2001). Here, we will shortly review the discussion of waste design in general and the issue of urine separation in particular, the latter being of special interest for the question of pharmaceuticals in wastewater. 2.2.1. Waste design Henze (1997) introduced the notion of waste design, defined as measures taken in households (and industry) with the goal of producing a waste with an optimal composition for further treatment and disposal. Flexibility and adaptability are important aspects of this approach. Possible actions in households would be storage (e.g. of urine, Larsen and Gujer, 1996), source separation of mixed toilet wastewater followed by anaerobic treatment (Otterpohl, 2002), or integration of treatment technology in different wastewater producing devices like the washing machine (Larsen and Gujer, 2001). In the case of pharmaceuticals, a pre-treatment of highly contaminated wastewater from hospitals (Giger et al., 2003) would be an obvious example of waste design. 299 2.2.2. Urine source separation Source separation of urine is one of the main research topics related to waste design. Since anthropogenic organic chemicals are in general metabolized to a polar water-soluble form to allow excretion by the kidney (Sheldon et al., 1986), this technology is of special interest for the question of pharmaceuticals in wastewater. Different varieties of urine-separation toilets exist (see http://www.novaquatis.eawag.ch) and a number of larger European wastewater authorities run pilot projects to test the technology, especially the functionality of the urine-separating toilets (Johansson, 2001; Bastian et al., 2002; Kühni et al., 2002; PeterFröhlich et al., 2003). Major reasons for introducing urine source separation are the possibility of better water pollution control with respect to nutrients and micropollutants and the possibility of closing the nutrient cycles. Although urine constitutes less than 1% of the wastewater volume, it contains most of the nutrients ending up in wastewater and many micropollutants from the human metabolism (pharmaceuticals, hormones). Besides the therapeutic application, waste disposal of unused pharmaceuticals via the toilet in private households presumably also contributes to the contamination of wastewater with micropollutants, but this seems to be of minor importance (Heberer, 2002). In most cases, efficient urine source separation would render nutrient removal at treatment plants obsolete; to obtain more stringent threshold values for phosphorus, merely a small technical effort would be necessary (Larsen and Gujer, 1996). Additional to water pollution control, urine source separation offers an elegant solution to nutrient recycling, a sustainability issue especially for phosphorus (Driver et al., 1999; Lienert et al., 2003; Maurer et al., 2003). Perhaps the most important aspect of urine separation is the possibility of flexible adaptation of the present wastewater system without losing capital bound in existing infrastructure (sewers and treatment plants). Larsen and Gujer (1996) suggested transition scenarios with storage of urine in households and subsequent release when nitrogen is required at the treatment plant (a typical example of waste design). Furthermore, storage capacity could be chosen such that urine in combined sewer overflow (CSO, release of untreated wastewater to receiving waters during rain) could be avoided. In a typical Swiss wastewater treatment plant, Rauch et al. (2003) showed that with a very moderate 300 T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 storage capacity (10 l per toilet), which could possibly be integrated into the toilet itself, and a simple control strategy, a 30% increase of nitrification capacity and a 50% reduction of urine in CSOs can be achieved. For the question of pharmaceuticals in wastewater, the transition scenarios are primarily of interest due to the possibility of reducing the amount of urine in CSOs. Further reaching scenarios with separate treatment of urine promise better chances of removing pharmaceuticals and their metabolites quantitatively at an acceptable price. The perhaps most important advantage of tackling the problem of pharmaceuticals as close to the ‘source’ as possible, is the inherent problem of leaking of pollutants from the wastewater system. Due to loss of wastewater from leaky pipes, households not connected to the sewer system, failures of the treatment plants and CSOs, it is hardly possible to obtain more than 80% treatment efficiency from prevailing wastewater systems (Larsen and Gujer, 2001). mental work it is known that about 85% of the organic fraction in urine is biologically degradable (Udert, 2002). Specific results on biological degradation of micropollutants in source separated urine are still too scarce to be conclusive. First results indicate that the half-life of natural estrogens in a biological reactor treating urine is less than 15 min (Maurer, personal communication). Compared to wastewater treatment plants, we would expect a higher degree of transformation/degradation of pharmaceuticals due to substantially higher concentrations (100–500 times larger than in wastewater) and the possibility of obtaining a substantially higher solids retention time at very low costs (because of the small organic loading from urine). Substrate inhibition, which may possibly occur in wastewater treatment plants due to peak organic loadings, can be avoided more easily in urine-treating systems, but this will demand some storage capacity. 2.3. Removal mechanisms in source separated urine 2.3.2. Chemical oxidation in source-separated urine (ozonation) The efficiency of the ozonation process is expected to depend mainly on the concentration ratio between the target compound and the soluble background organic carbon (von Gunten et al., 2003). Since biologically treated urine contains about 10 times less soluble organic matter per person than the effluent from a typical wastewater treatment plant, we would expect ozonation in biologically treated urine to be substantially more energy-efficient. However, this still has to be proven experimentally. Research on removal mechanisms in source separated urine considerably lags behind research in conventional wastewater treatment plants. The most prominent reasons for this are the pioneering character of the projects and the requirement for building up experimental routines for dealing with urine. The following comparison with removal possibilities in conventional wastewater treatment plants is therefore to a large part based on theoretical considerations and preliminary experience with different treatment technologies. 2.3.1. Biological degradation or transformation in source separated urine Urine is a concentrated mixture containing a number of water-soluble waste products from the human metabolism. Due to rapid hydrolysis of urea once urine has left the urinary tract, the concentration of ammonia/ammonium and pH rise rapidly in source separated urine. Research on biological treatment of urine has mainly concentrated on partial nitrification, either to stabilize urine for further treatment (decreasing the pH below 7 and thereby preventing stripping of ammonia) or as a first step in nitrogen removal by autotrophic denitrification (Udert et al., 2003). From this experi- 2.3.3. Sorption in source separated urine To our knowledge, no experience with sorption of micropollutants to organic or inorganic material in urine has been published, although there is some experience with sorption of nutrients to different inorganic materials (Lind et al., 2000). In biological systems, the most important aspect of sorption is the fate of the sludge produced in the process. Due to the very low organic load of urine (only 5% of the wastewater organic load, Larsen and Gujer, 1996) and the possibilities of obtaining very large solids retention times, sludge production in biological systems treating urine is extremely small. Accordingly, incineration will be the logical route for this sludge, hereby eliminating any possible problem with micropollutants in T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 agriculture. If for some reason phosphorus precipitates are mixed with the sludge, a technology for recovery of this valuable element must be developed. The sorption process will be influenced by various factors, including the concentration of micropollutants and the amount of sludge (Eq. (1)). 2.3.4. Stripping in source separated urine As for sorption, no results of stripping of micropollutants from urine have been reported in the literature. Under identical aeration conditions, however, stripping would be increased in comparison to wastewater treatment plants due to the high concentrations of micropollutants. As in the case of sorption, we expect that possible detrimental effects of stripping can be prevented more easily in small biological urine-treating reactors than in large wastewater treatment plants (e.g. treatment of the exhaust air). 2.3.5. Physical processes in source separated urine In conventional wastewater treatment plants, membrane processes are extremely energy-intensive due to the large amount of water to be treated. For the removal of micropollutants from urine, first promising results have been obtained with nanofiltration. The energy demand of reverse osmosis of wastewater effluent and the energy demand of nanofiltration of source separated urine is in the same order of size (1 kWh m−3 for reverse osmosis of wastewater effluent; 0.5–5 kWh m−3 for nanofiltration of urine, depending on the conditions, Pronk, personal communication). With the very small production rate of urine (less than 1% of the wastewater production rate), the energy demand for this latter process is considerably smaller. The rapid progress in membrane technology in other areas (Matsuura, 2001) is very promising for the development of suitable membrane technology for urine treatment. 3. Role of the industry It is clearly in the interest of the involved industry to support the efforts of wastewater professionals to solve the problems of organic micropollutants in wastewater. There is a long tradition for such cooperation; a good example is the increased biodegradability of washingpowders and even pharmaceutical products. However, 301 recognizing that there might still be a problem of micropollutants in domestic wastewater may call for a more conceptual discussion. The chemical industry – including the pharmaceutical industry – is becoming increasingly aware of environmental problems and many companies have succeeded in strongly improving their ecological performance during production processes (see e.g. Kahn et al., 2001; Overcash, 2002). Reasons can be ethical ones, legislative restrictions or socio-economic pressure, i.e. because “Green Technology” products sell better. Also environmental catastrophes such as Bhopal (1984) or the pollution of the Rhine after the fire in a storehouse of Sandoz in 1996 lead to a re-orientation in the environmental policy of the chemical industry (see Brandt, 2002 for references). One major initiative was launched by the chemical industry itself: the “Responsible Care Initiative”. It was initiated in Canada in 1984 and adapted to European conditions by the European Chemical Industry Council, CEFIC a few years later (Brandt, 2002). Its goal is to contribute to an ecologically, economically and societal sustainable future by implementing individual programs in chemical companies. Other programs have been launched by national and international organizations (“Green Chemistry” by US Environmental Protection Agency (EPA), 1991; “Eco-Efficiency” by World Business Council for Sustainable Development (WBCSD), 1992; “Cleaner Production” by UNEP, 1998). All these initiatives strongly focus on the optimization of the production process within an industry and of waste disposal connected to the production of pharmaceuticals. Until lately, ecological considerations and “waste disposal” of pharmaceuticals once they had left the industry was hardly an issue; one major reason presumably being that human health will always rank higher than ecological safety. Hence, pharmaceuticals are designed to have the desired effect within the human body, without too much consideration for effect on ecosystems after having passed through the human metabolism. However, in Sweden, an environmental classification system for pharmaceuticals has recently been developed in cooperation with industry. This labelling would allow physicians and patients to choose the pharmaceutical with the least negative impacts on the environment (Wennmalm, 2003). The European Commission’s present policy for regulating chemicals in the EU does include risk assessment targeted at the 302 T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 chemicals of greatest concern or with an indication of unacceptable risk (Rogers, 2003b). So far data regarding adverse effects of pharmaceuticals in the environment are very scarce, the methodologies have not yet been fully established, and possible cause-effect relationships were not made, so that at the moment pharmaceuticals will be subject to closer scrutiny only as an exception. voked (Rogers, 2003a). Currently, the European Commission’s position seems to lie somewhere between version 1 and 2 (Rogers, 2003a,b). However, it is thinkable that this could change and that a stronger policy could be adopted, at least in special cases such as endocrine disruptive substances in the environment. The European Council has for instance already stressed the need to develop criteria for allergenic and endocrinedisrupting substances (Rogers, 2003b). 3.1. The precautionary principle There are indications that the approach towards pharmaceuticals in the environment may change in the future, primarily due to the implementation of the precautionary principle. The precautionary principle is articulated in different international treaties such as the Rio Declaration and became part of the European law in 1992—it is directly related to uncertainty due to lack of scientific knowledge, and may be invoked to decide on appropriate risk management actions (deFur and Kaszuba, 2002; Harremoës, 2003; Rogers, 2003a). Wastewater treatment was one of the first areas to experience the paradigm change caused by the precautionary principle. When in the 1980s the necessity of quick action to prevent further pollution and degradation of the North Sea was recognized, the precautionary principle was applied to all point source discharges, even if there was no hard scientific evidence that specific substances were harmful to the environment (deFur and Kaszuba, 2002). The precautionary principle is open to different interpretations ranging from “lack of full certainty is not a justification for preventing an action that might be harmful” to “take no action unless you are certain that it will do no harm” (Rogers, 2003a). According to Wiener and Rogers (2002) the many versions can be divided into three categories: (1) lack of full scientific certainty about a risk shall not justify postponing action to prevent it, (2) uncertainty about a risk justifies action to prevent it, and (3) the proponent of an activity posing uncertain risk bears the burden of proving that the activity poses “no” or an “acceptable” risk before the activity can go forward. In the last case, the proof that the risk level is acceptable has to be provided by the generator of the risk; i.e. the producer of a pharmaceutical (reversal of the burden of proof). It is never stated, which action should be undertaken, if any of the versions of the precautionary principle are in- 3.2. Applying the precautionary principle to pharmaceuticals Even if the strongest version of the precautionary principle were applied, i.e. the burden of proof be carried by the producer of a substance, and even if it could be shown with relative large certainty that the substance does indeed adversely affect the environment, it may still be difficult to prevent the substance from reaching the receiving waters because a ban might not outweigh the societal benefits in the case of pharmaceuticals. Rogers (2003a) gives three examples of difficult decisions regarding uncertain risks as opposed to socioeconomic benefits (of the herbicide atrazine, low-dose cadmium, and hydrogen fluoride). In our case, it is clearly in the interest of all parties to develop a technical solution that keeps pharmaceuticals (and other organic micropollutants) out of the aquatic ecosystems with a high degree of certainty. From the technical point of view there is a choice between end-of-pipe technologies and more fundamental changes in wastewater management. End-of-pipe technologies may be developed within a relatively short time, but they will never be able to solve the problem entirely, and it is likely that a large degree of uncertainty will remain (see above). More fundamental changes based on source separation measures will take significantly longer to implement, but are – from the conceptual point of view – more suitable for the complex task of dealing with micropollutants. 4. Conclusion Pharmaceutical and their metabolites are found in the effluent of wastewater treatment plants and in the aquatic environment. Although there is still little scientific evidence as to the detrimental effects of these com- T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 pounds on aquatic organisms, the precautionary principle may give rise to more stringent requirements in the future. Previous experience shows that the precautionary principle is well applied to wastewater treatment technologies, and consequently technologies are under development that will alleviate the situation in the aquatic environment with respect to organic micropollutants. End-of-pipe wastewater treatment concentrates on improved biological treatment and ozonation of the effluent. In the long term, source separation of urine that contains many of the pharmaceuticals and their transformation products from the human metabolism may offer the more effective solution to the problem of pharmaceuticals in the environment. Due to the higher concentrations of micropollutants, biological as well as physical processes are expected to be more efficient in urine than in diluted wastewater. Chemical oxidation (ozonation) may profit from the higher micropollutant to soluble organic matter ratio in biologically treated urine in comparison to the effluent from a conventional wastewater treatment plant. References Bastian, A., Schirmer, G., Londong, J., 2002. Zukunftsfähiges Abwassermanagement, Workshop in der Lambertsmühle. KA Wasserwirtschaft Abwasser Abfall 49 (10), 1339–1342 (in German). Brandt, Ch., 2002. Sustainable development and responsible care—Die chemische Industrie auf dem Weg in eine grüne Zukunft? Chemie in unserer Zeit 36 (4), 214–224 (in German). Burkhardt-Holm, P., Wahli, T., Meier, W., 2000. Nonylphenol affects the granulation pattern of epidermal mucous cells in rainbow trout, Oncorhynchus mykiss. Ecotox. Environ. Safety 46, 34–40. Burkhardt-Holm, P., Peter, A., Segner, H., 2002. Decline of fish catch in Switzerland—project fishnet: a balance between analysis and synthesis. Aquat. Sci. 64, 36–54 (also see www.fischnetz.ch). Côté, P., Buisson, H., Pound, C., Arakaki, G., 1997. Immersed membrane activated sludge for the reuse of municipal wastewater. Desalination 113 (2–3), 189–196. deFur, P.L., Kaszuba, M., 2002. Implementing the precautionary principle. Sci. Total Environ. 288, 155–165. Desbrow, C., Routledge, E.J., Brighty, G.C., Sumpter, J.P., Waldock, M., 1998. Identification of estrogenic chemicals in STW effluent. 1. Chemical fractionation and in vitro biological screening. Environ. Sci. Technol. 32 (11), 1549–1558. Driver, J., Lijmbach, D., Steen, I., 1999. Why recover phosphorus for recycling, and how? Environ. Technol. 20, 651–662. Eggen, R.I.L., Bengtsson, B.-E., Bowmer, C.T., Gerritsen, A.A.M., Gibert, M., Hylland, K., Johnson, A.C., Leonards, P., Nakari, T., Norrgren, L., Sumpter, J.P., Suter, M.J.-F., Svenson, A., Pickering, A.D., 2003. Search for the evidence of endocrine disruption in the aquatic environment: lessons to be learned 303 from joint biological and chemical monitoring in the European Project COMPREHEND. Pure Appl. Chem. 75 (11–12), 2445– 2450. Egli, T., 2001. Biodegradation of metal-complexing aminopolycarboxylic acids. J. Biosci. Bioeng. 92 (2), 89–97. EPA—US Environmental Protection Agency, 1991. Green Chemistry. http://www.epa.gov/greenchemistry. Last update: 24 June 2003, last visit to website: 31 July 2003. Giger, W., Alder, A., Golet, E., Kohler, H., McArdell, C, Molnar, E., Pham Thi, N., Siegrist, H., 2003. Antibiotikaspuren auf dem Weg von Spital- und Gemeindeabwasser in die Fliessgewässer: Umweltanalytische Untersuchungen über Einträge und Verhalten. Tutzing Symposium, March 2003, D (in German). Golet, E.M., Alder, A.C., Giger, W., 2002. Environmental exposure and risk assessment of fluoroquinolone antibacterial agents in wastewater and river water of the Glatt Valley Watershed, Switzerland. Environ. Sci. Technol. 36 (17), 3645–3651. Harremoës, P., 2003. Ethical aspects of scientific incertitude in environmental analysis and decision making. J. Clean. Prod. 11, 705–712. Hartig, C., Ernst, M., Jekel, M., 2001. Membrane filtration of two sulphonamides in tertiary effluents and subsequent adsorption on activated carbon. Water Res. 35 (16), 3998–4003. Heberer, T., 2002. Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment: a review of recent research data. Toxicol. Lett. 131, 5–17. Henze, M., 1997. Waste design for households with respect to water, organics and nutrients. Water Sci. Technol. 35 (9), 113–120. Huber, M., Canonica, S., Park, G., von Gunten, U., 2002. Oxidation of pharmaceuticals during ozonation and advanced oxidation processes. Environ. Sci. Technol. 37, 1016–1024. Jobling, S., Nolan, M., Tyler, C.R., Brighty, G., Sumpter, J.P., 1998. Widespread sexual disruption in wild fish. Environ. Sci. Technol. 32 (17), 2498–2506. Johansson, M., 2001. Urine separation—closing the nutrient cycle. Final report. Stockholm Water Company, Stockholm, S. Jürgens, M.D., Holthaus, K.I.E., Johnson, A.C., Smith, J.J.L., Hetheridge, M., Williams, R.J., 2002. The potential for estradiol and ethinylestradiol degradation in English rivers. Environ. Toxicol. Chem. 21, 480–488. Kahn, F.I., Natrajan, B.R., Revathi, P., 2001. GreenPro: a new methodology for cleaner and greener process design. J. Loss Prevent. Proc. 14, 307–328. Kriebel, D., Tickner, J., Epstein, P., Lemons, J., Levins, R., Loechler, E.L., Quinn, M., Rudel, R., Schettler, T., Stoto, M., 2001. The precautionary principle in environmental science. Environ. Health Perspect. 109 (9), 871–876. Kühni, M., Koch, G., Ott, E., 2002. Zukunftsweisende Sanitarund Abwassertechnik—erstes Pilotprojekt der Schweiz fur Urinseparierung, -speicherung und -steuerung im technischen Massstab. gwa (Gas Wasser Abwasser) 11, 827–835 (in German). Länge, R., Dietrich, D., 2002. Environmental risk assessment of pharmaceutical drug substances—conceptual considerations. Toxicol. Lett. 131, 97–104. Larsen, T.A., Gujer, W., 1996. Separate management of anthropogenic nutrient solutions (human urine). Water Sci. Technol. 34 (3–4), 87–94. 304 T.A. Larsen et al. / Journal of Biotechnology 113 (2004) 295–304 Larsen, T.A., Gujer, W., 1997. The concept of sustainable urban water management. Water Sci. Technol. 35 (9), 3–10. Larsen, T.A., Gujer, W., 2001. Waste design and source control lead to flexibility in wastewater management. Water Sci. Technol. 43 (5), 309–318. Larsen, T.A., Peters, I., Alder, A., Eggen, R., Maurer, M., Muncke, J., 2001. Re-engineering the toilet for sustainable wastewater management. Environ. Sci. Technol. 35, 192A–197A. Lienert, J., Haller, M., Berner, A., Stauffacher, M., Larsen, T.A., 2003. How farmers in Switzerland perceive fertilizers from recycled anthropogenic nutrients (urine). Water Sci. Technol. 48 (1), 47–56. Lind, B., Zsófia, B., Byden, S., 2000. Nutrient recovery from human urine by struvite crystallization with ammonia adsorption on zeolite and wollastonite. Bioresour. Technol. 73, 169–174. Matsuura, T., 2001. Progress in membrane science and technology for seawater desalination—a review. Desalination 134 (1–3), 47–54. Maurer, M., personal communication. EAWAG, P.O. 611, CH-8600 Dübendorf, Switzerland, [email protected]. Maurer, M., Schwegler, P., Larsen, T.A., 2003. Nutrients in urine: energetical aspects of removal and recovery. Water Sci. Technol. 48 (1), 37–46. Nghiem, L.D., Schafer, A.I., Waite, T.D., 2002. Adsorptive interactions between membranes and trace contaminants. Desalination 147 (1–3), 269–274. Norberg-Bohm, V., 1999. Stimulating “green” technological innovation: an analysis of alternative policy mechanisms. Policy Sci. 32, 13–38. Otterpohl, R., 2002. Options for alternative types of sewerage and treatment systems directed to improvement of the overall performance. Water. Sci. Technol 45 (3), 149–158. Overcash, M., 2002. The evolution of US pollution prevention, 1976–2001: a unique chemical engineering contribution to the environment—a review. J. Chem. Technol. Biot. 77, 1197–1205. Peter-Fröhlich, A., Kraume, I., Lesouëf, A., Oldenburg, M., 2003. Separate discharge and treatment of urine, faeces and greywater—pilot project. Hydroplus 135, 82–86. Pickering, A.D., Sumpter, J.P., 2003. COMPREHENDing endocrine disrupters in aquatic environments. Environ. Sci. Technol. 37 (17), 331A–336A. Pronk, W., personal communication. EAWAG, P.O. 611, CH-8600 Dübendorf, Switzerland. [email protected]. Rauch, W., Brockmann, D., Peters, I., Larsen, T.A., Gujer, W., 2003. Combining urine separation with waste design: an analysis using a stochastic model for urine production. Water Res. 37, 681–689. Rogers, M.D., 2003a. Risk analysis under uncertainty, the precautionary principle, and the new EU chemicals strategy. Regul. Toxicol. Pharm. 37, 370–381. Rogers, M.D., 2003b. The European Commission’s White Paper “strategy for a future chemicals policy”: a review. Risk Anal. 23 (2), 381–388. Routledge, E.J., Sheahan, D., Desbrox, C., Brighty, G.C., Waldock, M., Sumpter, J.P., 1998. Identification of estrogenic chemicals in sewage treatment works effluent. 2. In vivo responses in trout and roach. Environ. Sci. Technol. 32, 1559–1565. Schafer, A.I., Nghiem, L.D., Waite, T.D., 2003. Removal of the natural hormone estrone from aqueous solutions using nanofiltration and reverse osmosis. Environ. Sci. Technol. 37 (1), 182– 188. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003. Environmental Organic Chemistry, second ed. Wiley-Interscience. ISBN 0-471-35750-2. Seyman, I., 2003. Sewage sludge—no more land treatment in Switzerland. Hydroplus 135, 21. Sheldon, L., Umana, M., Bursey, J., Gutknecht, W., Handy, R., Hyldburg, P., Michael, L., Moseley, A., Raymer, J., Smith, D., Sparacino, C., Warner, M., 1986. Biological Monitoring Techniques for Human Exposure to Industrial Chemicals. Noyes Publications, USA. Siegrist, H., Joss, A., Alder, A., Gobel, A., Keller, E., McArdell, C., Ternes, T.A., 2003. Mikroverunreinigungen— Abwasserentsorgung vor neuen Anforderungen. EAWAG News 57, 7–10 (in German). Ternes, T.A., 1998. Occurrence of drugs in German sewage treatment plants and rivers. Water Res. 32 (11), 3245–3260. Ternes, T.A., Stüber, J., Herrmann, N., McDowell, D., Ried, A., Kampmann, M., Teiser, B., 2003. Ozonation: a tool for removal of pharmaceuticals, contrast media and musk fragrances from wastewater? Water Res. 37, 1976–1982. Udert, K.M., 2002. The fate of nitrogen and phosphorus in sourceseparated urine. Dissertation ETH, No. 14847. ETH-Zürich, CH. Udert, K.M., Fux, C., Munster, M., Larsen, T.A., Siegrist, H., Gujer, W., 2003. Nitrification and autotrophic denitrification of sourceseparated urine. Water Sci. Technol. 48 (1), 119–130. UNEP—United Nations Environment Programme, 1998. International declaration on cleaner production, http://www.uneptie.org/ pc/cp/declaration. [email protected]. Last visit to website: 31 July 2003. von Gunten, U., Huber, M., Ternes, T.A., Ried, A., 2003. Oxidation of pharmaceuticals during ozonation of drinking water and wastewater (abstract), Proceedings of the IWA Leading Edge Conference on Drinking Water and Wastewater, 26–28 May 2003, Noordwijk/Amsterdam, NL, pp. 82–83. Walther, H., 2001. Berücksichtigung der Membrantechnik im regionalen Abwasserentsorgungskozept. Kommunale Abwasserbehandlung 48 (8), 1092–1097 (in German). Wennmalm, A., 2003. A proposed environmental classification system for medicinal products. Envirpharma Conference, Lyon, F. Wiener, J.B., Rogers, M.D., 2002. Comparing precaution in the US and Europe. J. Risk Res. 5, 317–349. Wintgens, T., Gallenkemper, M., Melin, T., 2002. Endocrine disrupter removal from wastewater using membrane bioreactor and nanofiltration technology. Desalination 146 (1–3), 387–391. WBCSD—World Business Council for Sustainable Development, 1992. Eco-Efficiency. http://[email protected]. Last visit to website: 31 July 2003.