A survey of disinfection byproduct (DBP) occurrence in the United States was conducted at 12 drinking water treatment plants. In addition to currently regulated DBPs, more than 50 DBPs that rated a high priority for potential toxicity were studied. These priority DBPs included iodinated trihalomethanes (THMs), other halomethanes, a nonregulated haloacid, haloacetonitriles, haloketones, halonitromethanes, haloaldehydes, halogenated furanones, haloamides, and nonhalogenated carbonyls. The purpose of this study was to obtain quantitative occurrence information for new DBPs (beyond those currently regulated and/or studied) for prioritizing future health effects studies. An effort was made to select plants treating water that was high in total organic carbon and/or bromide to enable the detection of priority DBPs that contained bromine and/or iodine. THMs and haloacetic acids (HAAs) represented the two major classes of halogenated DBPs formed on a weight basis. Haloacetaldehydes represented the third major class formed in many of the waters. In addition to obtaining quantitative occurrence data, important new information was discovered or confirmed at full-scale plants on the formation and control of DBPs with alternative disinfectants to chlorine. Although the use of alternative disinfectants (ozone, chlorine dioxide, and chloramines) minimized the formation of the four regulated THMs, trihalogenated HAAs, and total organic halogen (TOX), several priority DBPs were formed at higher levels with the alternative disinfectants as compared with chlorine. For example, the highest levels of iodinated THMs-which are not part of the four regulated THMs-were found at a plant that used chloramination with no prechlorination. The highest concentration of dichloroacetaldehyde was at a plant that used chloramines and ozone; however, this disinfection scheme reduced the formation of trichloroacetaldehyde. Preozonation was found to increase the formation of trihalonitromethanes. In addition to the chlorinated furanones that have been measured previously, brominated furanones-which have seldom been analyzed-were detected, especially in high-bromide waters. The presence of bromide resulted in a shift to the formation of other bromine-containing DBPs not normally measured (e.g., brominated ketones, acetaldehydes, nitromethanes, acetamides). Collectively, -30 and 39% of the TOX and total organic bromine, respectively, were accounted for (on a median basis) bythe sum of the measured halogenated DBPs. In addition, 28 new, previously unidentified DBPs were detected. These included brominated and iodinated haloacids, a brominated ketone, and chlorinated and iodinated aldehydes.

Iodoacid drinking water disinfection byproducts (DBPs) were recently uncovered in drinking water samples from source water with a high bromide/iodide concentration that was disinfected with chloramines. The purpose of this paper is to report the analytical chemical identification of iodoacetic acid (IA) and other iodoacids in drinking water samples, to address the cytotoxicity and genotoxicity of IA in Salmonella typhimurium and mammalian cells, and to report a structure-function analysis of IA with its chlorinated and brominated monohalogenated analogues. The iodoacid DBPs were identified as iodoacetic acid, bromoiodoacetic acid, (Z)- and (E)-3-bromo-3-iodopropenoic acid, and (E)-2-iodo-3-methylbutenedioic acid. IA represents a new class (iodoacid DBPs) of highly toxic drinking water contaminants. The cytotoxicity of IA in S. typhimurium was 2.9x and 53.5x higher than bromoacetic acid (BA) and chloroacetic acid (CA), respectively. A similar trend was found with cytotoxicity in Chinese hamster ovary (CHO) cells; IA was 3.2x and 287.5x more potent than BA and CA, respectively. This rank order was also expressed in its genotoxicity with IA being 2.6x and 523.3x more mutagenic in S. typhimurium strain TA100 than BA and CA, respectively. IA was 2.0x more genotoxic than BA and 47.2x more genotoxic than CA in CHO cells. The rank order of the toxicity of these monohalogenated acetic acids is correlated with the electrophilic reactivity of the DBPs. IA is the most toxic and genotoxic DBP in mammalian cells reported in the literature. These data suggest that chloraminated drinking waters that have high bromide and iodide source waters may contain these iodoacids and most likely other iodo-DBPs. Ultimately, it will be important to know the levels at which these iodoacids occur in drinking water in order to assess the potential for adverse environmental and human health risks.

Using gas chromatography/mass spectrometry (GC/MS), we investigated the formation of disinfection byproducts (DBPs) from high bromide waters (2 mg/L) treated with chlorine or chlorine dioxide used in combination with chlorine and chloramines. This study represents the first comprehensive investigation of DBPs formed by chlorine dioxide under high bromide conditions. Drinking water from full-scale treatment plants in Israel was studied, along with source water (Sea of Galilee) treated under carefully controlled laboratory conditions. Select DBPs (trihalomethanes, haloacetic acids, aldehydes, chlorite, chlorate, and bromate) were quantified. Many of the DBPs identified have not been previously reported, and several of the identifications were confirmed through the analysis of authentic standards. Elevated bromide levels in the source water caused a significant shift in speciation to bromine-containing DBPs; bromoform and dibromoacetic acid were the dominant DBPs observed, with very few chlorine-containing compounds found. Iodo-trihalomethanes were also identified, as well as a number of new brominated carboxylic acids and 2,3,5-tribromopyrrole, which represents the first time a halogenated pyrrole has been reported as a DBP. Most of the bromine-containing DBPs were formed during pre-chlorination at the initial reservoir, and were not formed by chlorine dioxide itself. An exception wasthe iodo-THMs, which appeared to be formed by a combination of chlorine dioxide with chloramines or chlorine (either added deliberately or as an impurity in the chlorine dioxide). A separate laboratory study was also conducted to quantitatively determine the contribution of fulvic acids and humic acids (from isolated natural organic matter in the Sea of Galilee) as precursor material to several of the DBPs identified. Results showed that fulvic acid plays a greater role in the formation of THMs, haloacetic acids, and aldehydes, but 2,3,5-tribromopyrrole was produced primarily from humic acid. Because this was the first time a halopyrrole has been identified as a DBP, 2,3,5-tribromopyrrole was tested for mammalian cell cytotoxicity and genotoxicity. In comparison to other DBPs, 2,3,5-tribromopyrrole was 8x, 4.5x, and 16x more cytotoxic than dibromoacetic acid, 3-chloro-4-(dichloromethyl)-5-hydroxy-2-[5H]-furanone [MX], and potassium bromate, respectively. 2,3,5-Tribromopyrrole also induced acute genomic damage, with a genotoxic potency (299 microM) similar to that of MX.

Using a combination of spectral identification techniquesgas chromatography coupled with low- and high-resolution electron-impact mass spectrometry (GC/EI-MS), low- and high-resolution chemical ionization mass spectrometry (GC/CI-MS), and infrared spectroscopy (GC/IR)we identified many drinking water disinfection byproducts (DBPs) formed by ozone and combinations of ozone with chlorine and chloramine. Many of these DBPs have not been previously reported. In addition to conventional XAD resin extraction, both pentafluorobenzylhydroxylamine (PFBHA) and methylation derivatizations were used to aid in identifying some of the more polar DBPs. Many of the byproducts identified were not present in spectral library databases. The vast majority of the ozone DBPs identified contained oxygen in their structures, with no halogenated DBPs observed except when chlorine or chloramine was applied as a secondary disinfectant. In comparing byproducts formed by secondary treatment of chlorine or chloramine, chloramine appeared to form the same types of halogenated DBPs as chlorine, but they were generally fewer in number and lower in concentration. Most of the halogenated DBPs that were formed by ozone−chlorine and ozone−chloramine treatments were also observed in samples treated with chlorine or chloramine only. A few DBPs, however, were formed at higher levels in the ozone−chlorine and ozone−chloramine samples, indicating that the combination of ozone and chlorine or chloramine is important in their formation. These DBPs included dichloroacetaldehyde and 1,1-dichloropropanone.

Using a combination of mass spectrometry and infrared spectroscopy, disinfection byproducts were identified in ozonated drinking water containing elevated bromide levels and in ozonated water treated with secondary chlorine or chloramine. Only one brominated byproductdibromoacetonitrilewas found in the water treated with only ozone. This compound was found only in one of the three treatment rounds and was also present in the untreated, raw water but at levels 20 times lower than in the ozonated water. Many more byproducts were identified when secondary chlorine or chloramine was applied after ozonation. A number of these byproducts have not been reported previously. When comparing low-bromide water to water with elevated bromide, a tremendous shift in speciation was observed for samples treated with secondary chlorine or chloramine. Without high bromide levels, chlorinated species dominate (e.g., chloroform, trichloroacetaldehyde, tetrachloropropanone, dichloroacetonitrile, trichloronitromethane); with elevated bromide levels (1 mg/L), these shift to brominated species (e.g., bromoform, tribromoacetaldehyde, tetrabromopropanone, dibromoacetonitrile, tribromonitromethane). An entire family of bromo- and mixed chlorobromopropanones was identified that was not present in library databases and has not been reported previously. They were observed mainly in the ozone−chloramine samples but were also present in ozone−chlorine-treated water. These brominated byproducts were also observed in water treated with only chloramine or chlorine.