Keeping drugs out of drinking water
KELLYN S. BETTS / Environmental Science & Technology v.36, n.1, 1oct02
The first systematic investigation of how effectively drinking water treatment technologies remove pharmaceutical products has found that the technologies being used in Germany appear to do a good job, according to new ES&T research (Environ. Sci. Technol. 2002, 36, 3855–3863). However, the paper’s lead author, Thomas Ternes of Germany’s Institute for Water Research and Water Technology (ESWE), says that some of the technologies used elsewhere in the world—particularly in the United States—may be letting pharmaceuticals through.
Although research shows that pharmaceutically active products are found in surface waters throughout the United States (Environ. Sci. Technol. 2002, 36, 1202–1211) and Europe, there is as yet very little information on how effectively different drinking water treatment technologies remove these pharmaceutical residues.
Ternes’s work, which represents the most comprehensive assessment published to date, was funded in part by the European Union’s (EU’s) ongoing Poseidon program for assessing technologies for removing pharmaceuticals and personal care products in sewage and drinking water treatment facilities. The program is also aimed at evaluating the ability of advanced wastewater treatment technologies like membranes and source separation to reduce the volume of pharmaceuticals being discharged into EU waters. Ternes and his fellow researchers at ESWE and two other organizations associated with the German Technical and Scientific Association on Gas and Water (DVGW), the Technologiezentrum Wasser and the Institute for Water Research, investigated how effectively treatment technologies were able to eliminate five pharmaceutical products often found in German waters: two epilepsy medications, carbamazepine and prim done; one lipid regulator, bezafibrate, and a metabolite of the lipid regulator clofibrate, clofibric acid; and diclofenac, an arthritis drug. The paper is notable for evaluating the removal efficiency of a number of popular technologies— flocculation, ozonation, granular activated carbon (GAC), bank filtration, and slow sand filtration—both in the laboratory and in waterworks treatment facilities.
Both GAC and ozonation were very effective at removing carbamazepine and diclofenac at both laboratory and waterworks scales, Ternes found. Laboratory-scale ozonation treatments showed that 0.5 milligrams/liter (mg/L) of ozone reduced the two pharmaceuticals by more than 90%.
Although GAC was very successful in removing bezafibrate during waterworks testing, laboratory-scale ozonation treatments with much higher concentrations of ozone (3.0 mg/L) were required to reduce it by 90%. The laboratory scale also showed that 3.0 mg/L of ozone cut primidone levels by nearly 90%. Although neither GAC nor ozonation at any concentration was fully successful with clofibric acid separately, the two in combination cut levels of the drug to below the limits of detection during waterworks testing.
GAC is effective because carbon can bind a broad category of compounds, and many pharmaceuticals have components like benzene rings or amine groups that enhance their ability to be taken up by the activated carbon, Ternes explains. The powdered form of activated carbon can be even more effective, removing virtually every possible pharmaceutical and personal care product, adds Shane Snyder, the project manager for research and development at the Southern Nevada Water Authority. But he stresses that GAC—like all technologies—is only effective if used properly. Ozone’s effectiveness is tied to its ability to chemically attack the pharmaceutical molecules, Snyder says. For example, it can oxidize different kinds of functional aromatic bonds in pharmaceuticals, he says. Ternes adds that his research has shown ozone capable of easily oxidizing 90% of 51 different pharmaceuticals. It is particularly effective with compounds with amine groups and phenolic hydroxyl groups, he says.
On the basis of the testing he has conducted thus far, Ternes concludes that contamination of German drinking water by the chemicals he investigated is “rather unlikely”, given that the waterworks treating surface water, which is most likely to contain pharmaceutical residues, use either ozonation or activated carbon, or perhaps both. He believes that other advanced techniques, such as membrane filtration and advanced oxidative processes using a combination of ozone and UV or ozone and hydrogen peroxide, should also remove pharmaceuticals. Most pharmaceutical products are not removed by conventional flocculation technology because their polar- ity makes them unlikely to adsorb to organic matter, he explains.
The paper does not address the multitude of degradation products that can be produced by oxidative treatments such as ozone, points out Christian Daughton, chief of the environmental chemistry branch of the U.S. EPA’s National Exposure Research Laboratory’s Environmental Sciences division. For example, recent research conducted by Roberto Andreozzi of the University of Naples in Italy (Water Res. 2002, 36, 2869– 2877) shows that numerous metabolites can persist after carbamazepine has been treated with ozonation, Daughton says.
Ternes says that he is already investigating such oxidation products—as well as other drinking water treatment technologies, including membrane filtration—through the Poseidon program. Between the Poseidon program and a project begun this year by the American Water Works Association’s Research Foundation (AWWARF), virtually all treatment technologies in current use are being tested, says Snyder, who is the AWWARF project’s principal investigator. The AWWARF project’s goal is to evaluate how well different drinking water treatments remove 42 compounds representative of the pharmaceutical products found in U.S.
waters. It is notable for its goal of developing quantitative structure– activity relationship (QSAR) computer models that predict how pharmaceuticals will respond to different treatments based on their chemical structure. The study therefore includes a wide array of different types of compounds—such as acidic, basic, large, and small molecules—chosen to represent the pharmaceuticals and personal care products that can be found in the environment.
The AWWARF project will have succeeded in its goal if “when we read a paper in ES&T with the newest endocrine disrupter of the month, we can go back to our data set . . . and see how these treatment processes will remove [it],” says Snyder, who is the project’s principal investigator. However, Snyder and his colleagues have not yet published any results. Although more study is clearly needed, Ternes’s findings hint that conventional water treatment plants, such as those that use coagulation and flocculation followed by filtration and disinfection with a chlorine product, cannot efficiently remove trace organic pollutants such as low levels of pharmaceuticals and personal care products, says Jörg Drewes, an assistant professor in the Colorado School of Mines’ Environmental Science and Engineering division. In such systems, “it may be possible that these compounds add up to the contamination of drinking water,” Ternes adds.
Most countries outside central Europe—including the United States—use either chlorine, chloramines, or ozone to disinfect drinking water, Drewes says. To date, the AWWARF studies have shown that chlorination can add another barrier to pharmaceutically active compounds, but it is not nearly as effective as ozone, Snyder says, stressing that the results are as yet unpublished. He and his fellow researchers haven’t yet tackled chloramines, but he doesn’t expect them to be as effective as chorine.
Both ozonation and GAC are increasingly popular in the United States, but they are still only used by a relatively small number of larger drinking water treatment facilities, says Steve Via, a regulation engineer at AWWA. According to a survey of how U.S. facilities treat surface water by the U.S. EPA in 1997, the latest year for which figures are available, 5.4% of U.S. facilities serving 50,000– 100,000 people used ozone and 7.7% used GAC. Of treatment providers for over 100,000 people, 5.8% used ozone and 6.4% used GAC.
Although the pharmaceutical compounds that Ternes studied have been reported in wastewater and surface water outside Germany, Drewes says that clofibrate is no longer commonly used as a lipid regulator in the United States, and studies by his lab and others have found it only in some U.S. wastewater effluents. Drewes says he believes that Ternes’s carbamazepine results are most relevant in the United States; he notes that his studies of recharging groundwater have shown that the compound can survive intact after traveling 8–10 years through the subsurface.
Because of the increasing popularity of groundwater recharging in places like the United States, all of the scientists interviewed for this article concurred that groundwater may become contaminated with pharmaceuticals. Ternes adds that the risk that pharmaceutical compounds will be found in groundwater is “relatively high” for locations where surface water intrudes into the groundwater. Although the risk associated with consuming the low levels of pharmaceuticals found in U.S. and European waters is as yet unknown Environ. Sci. Technol. 2002, 36, 140A–145A), Ternes notes that any antibiotics that find their way into surface waters and other natural water bodies could promote antibiotic resistance. “The major concern associated with the presence of [pharmaceutically active compounds] in drinking water is not acute effects on human health (since environmental concentrations are orders of magnitude below the therapeutic dosages), but rather the manifestation of imperceptible effects that can accumulate over time to yield truly profound changes,” Drewes says. “And the latter is extremely difficult to determine.”
On the positive side, the new technologies that are likely to be put in place to reduce the probability of pharmaceuticals making their way into drinking water, such as activated carbon, could have the additional benefit of removing “all sorts of other chemicals we don’t even know about at the present,” Daughton says. “The analytical chemists have gotten way ahead of us,” in terms of their ability to devise new ways to detect chemicals that may be found in the environment, he explains. Because environmental chemists regularly discover new and unexpected compounds in the nation’s waters, he predicts that greater use of advanced water treatment technologies could represent a “much, much higher reduction of overall risk” from such compounds.
If you have come to this page from an outside location click here to get back to mindfully.org