Biocides used in Cooling Water Disinfection

Entry to the marine environment

Recorded levels in the marine environment

Fate and behaviour in the marine environment

Effects on the marine environment

Bioaccumulation

Potential effects on interest features of European marine sites

Entry into the marine environment

Biofouling by algae, fungi and bacteria can occur in cooling water systems as the systems offer a warm, moist environment, ideal for the promotion of biological growth. The growth of these organisms, if left unchecked, can rapidly lead to the formation and accumulation of slimes and biofilms which can cause obstructions in the cooling water systems. This can increase pumping costs and inhibit the effectiveness of heat transfer processes. Fouling can also lead to the proliferation of sulphate reducing bacteria, which can ultimately lead to production of hydrogen sulphide which can cause metallic corrosion problems.

Biocides are routinely employed in order to control the growth and development of such organisms. Within cooling water systems, the life of a biocide will depend on environmental factors as well as the amount added and the physical and chemical fate of the individual chemical. However, where cooling water systems discharge directly to estuaries or marine waters, there is the potential for residual quantities of biocides and their degradation/transformation products to be present in the effluents.

There are many biocides available to control biofouling in cooling water systems. These can be divided into two main groups: the oxidising biocides and the non-oxidising biocides. The classification is based on the mode of biocide action against biological material.

Oxidising biocides include chlorine and bromine-based compounds and are non selective with respect to the organisms they kill. Non-oxidising biocides are more selective, in that they may be more effective against one type of micro-organisms than another. A large variety of active ingredients are used as non-oxidising biocides, including quaternary ammonium compounds, isothiazolones, halogenated bisphenols, thiocarbamates as well as others. In view of the wide range of potentially different fate and behaviour and toxicity, consideration of the environmental fate and saltwater toxicity of biocides has been limited here to the effects of chlorine and bromine based biocides and their principal transformation products (chloroform and bromoform).

Recorded levels in the marine environment

Chloride concentrations in cooling water discharges are unlikely to be significant relative to the background levels present in the receiving waters where mean annual average concentrations have been reported in the range < 10 - >100 mg/l (Gardiner and Smith 1990).

Monitoring data for chloroform from the National Rivers Authority and the National Monitoring Programme Survey of the Quality of UK Coastal Waters are presented in Appendix D. One water column concentration was found to exceed the EQS value (see Appendix D). Monitoring data were not available for sediments or biota.

The available data suggest that concentrations of chloroform may be elevated in some UK coastal and estuarine waters but in general, are unlikely to exceed relevant quality standards derived for the protection of saltwater life.

Fate and behaviour in the marine environment

Chlorine and bromine

The chemistry of chlorine in water has been reviewed extensively elsewhere (White 1986). Consequently, only a summary of the main points is included here.

When chlorine gas is dissolved in water, it hydrolyses rapidly according to the following equation to yield hypochlorous acid:

Cl2 + H2O HOCl + H+ + Cl- (1)

Hypochlorous acid is also formed when sodium hypochlorite is used as the source of chlorine:

NaOCl + H2O HOCl + Na+ + OH- (2)

Hypochlorous acid is a weak acid, and will undergo partial dissociation as follows:

HOCl H+ + OCl- (3)

In waters of pH between 6 and 9, both hypochlorous acid and hypochlorite ion will be present; the proportion of each species depending on pH and temperature. Hypochlorous acid is significantly more effective as a biocide than the hypochlorite ion.

The reaction of hypochlorous acid with ammonia results in the formation of chloramines as follows:

NH3 + HOCl NH2Cl + H2O (4)

NH2Cl + HOCl NHCl2 + H2O (5)

NHCl2 + HOCl NCl3 + H2O (6)

These reactions are all dependent on pH, temperature, contact time and the relative concentrations of chlorine and ammonia. Essentially, any free chlorine will be converted to monochloramine at pH 7 to 8 when the ratio of chlorine to ammonia is equimolar (5:1 by weight) or less. At higher chlorine to ammonia ratios, or lower pH values, dichloramine and trichloramine will be formed. The interaction of the various competing reactions is complex and will not be considered here. A detailed discussion of chlorine-ammonia chemistry can be found in White (1986).

Chlorine will also oxidise bromide to form hypobromous acid:

HOCl + Br- HOBr + Cl-- (7)

Hypobromous acid is an effective biocide. It is worth noting that, for a given pH value, the proportion of hypobromous acid relative to hypobromite is significantly greater than the corresponding values for the hypochlorous acid - hypochlorite system. Thus, for example, at pH 8 and 20oC, hypobromous acid represents 83% of the bromine species present, compared with hypochlorous acid at 28%. When ammonia is also present, the competing reactions of chlorine with bromide and ammonia are likely to result in the rapid formation of both monochloramine and hypobromous acid. A number of other reactions can then occur:

NH2Cl + Br- + 2H2O HOBr + NH4OH + Cl- (8)

HOBr + NH4OH NH2Br + 2H2O (9)

NH2Br + HOBr NHBr2 + H2O (10)

Chlorine can also react with nitrogen-containing organic compounds, such as amino acids to form organic chloramines. Little is known about the biocidal properties of these compounds.

In natural waters, chlorine can undergo a range of reactions in addition to those discussed above. It will react with inorganic constituents of water such as iron (II), manganese (II), nitrite and sulphide.

The reaction of chlorine with organic constituents in aqueous solution can be grouped into several types:

(a) Oxidation, where chlorine is reduced to chloride ion, e.g.

RCHO + HOCl RCOOH + H+ + Cl- (11)

(b) Addition, to unsaturated double bonds, e.g.

RC = CR' + HOCl RCOHCClR' (12)

(c) Substitution to form N-chlorinated compounds, e.g.

RNH2 + HOCl RNHCl + H2O (13)

or C-chlorinated compounds, e.g.

RCOCH3 + 3HOCl RCOOH + CHCl3 + 2H2O (14)

Chlorine substitution reactions can lead to the formation of halogenated compounds, such as chloroform (e.g. reaction 14), and, where HOBr is present, mixed halogenated and brominated organic compounds. Although such reactions are significant in terms of the resultant halogenated by-products, it has been estimated that only a few percent of the applied chlorine ends up as halogenated organic products. Chlorine is a powerful oxidant, and a significant proportion of the applied chlorine is likely to be consumed in reactions such as 11, leading to the formation of non-halogenated organic products, with chlorine being reduced to chloride.

A number of other source water characteristics is likely to have an impact on the concentrations of organic by-products present in cooling water discharges:

Natural organic matter in water is the major precursor of halogenated organic by-products, and hence the organic content of the source water (often measured as total organic carbon, TOC) may affect the concentration of by-products formed. In general, the higher the organic content of the source water, the higher the potential for by-product formation. Whether this potential is realised will depend primarily on the applied chlorine dose, as well as the extent of competing reactions that lead to the consumption of chlorine. Freshwaters typically contain higher amounts of TOC than marine or estuarine, and hence have the potential to produce higher levels of halogenated organic compounds during chlorination. Any pre-treatment (e.g. settling) of the cooling water is likely to reduce by-product formation through the removal of organic precursors.

The ammonia concentration is likely to affect the extent of by-product formation, through reaction with chlorine to form chloramines. Although seawater generally contains low concentrations of ammonia than freshwater, under certain conditions (dependent on chlorine dose:ammonia nitrogen concentration) it can compete with bromide for the available chlorine to form monochloramine. In addition, hypobromous acid can react with ammonia to form bromamines. Although the sequence of reactions is complex, it is likely that the reaction of either hypochlorous or hypobromous acid with ammonia to form halamines will reduce organic by-product formation during the chlorination of seawater.

The pH of the incoming cooling water could also affect the nature of the by-products formed. Trihalomethanes (THMs) formation has been shown to rise with increasing pH, probably due to base-catalysed hydrolysis of intermediates in the haloform reaction. TCA formation on the other hand has been shown to decrease significantly at pH values above 7, possibly due to the reduced oxidising power of hypochlorite ion compared with hypochlorous acid (Miller and Uden 1983). In general, while variations in pH are likely to affect the concentrations of individual by-products, the overall quantity formed (i.e. AOX content) is likely to remain relatively constant. At sites where the pH is adjusted to reduce problems with scaling, a reduction in THM formation could occur, possibly at the expense of an increase in haloacetic acid production.

The presence of certain pollutants in source waters could lead to an increase in the levels of certain halogenated organics. The presence of phenol, for example, can lead to the formation of chlorophenols (see Section B42). THMs and halogenated phenols have been identified in a number of chlorinated cooling waters. Chloroform is the major THM formed at sites using freshwater sources, whereas bromoform predominates at estuarine and marine sites.

In estuarine and marine sites, where chlorination or bromination has been used in the cooling water, brominated products will predominate due to the influence of bromide in saline waters.

Chloroform and bromoform

Chloroform (trichloromethane) and bromoform (tribromomethane) are highly volatile and only moderately soluble in water (IARC 1979, Merck Index 1989). Chloroform is a commonly encountered chemical, having many industrial uses, hence, a lot of information is available on the environmental fate of this compound. Based on the data available in the literature, chloroform and bromoform do not adsorb onto sediments and soils to any great extent. Consequently, this process is not considered to be an important means of removing chloroform or bromoform from the aquatic environment (CCME 1992). Volatilisation (followed by oxidation) is the major fate process for removing chloroform (and most likely bromoform) from the aquatic environment. Biodegradation is slow but has a significant effect on the removal of chloroform or bromoform from soils and sediments. Hydrolysis, adsorption, photo-oxidation, photolysis, hydraulic processes, and bioaccumulation do not appear to reduce chloroform concentrations substantially in the environment.

Effects in the marine environment

Toxicity to marine organisms

An exhaustive literature review on the toxicity of chlorine, bromine, chloroform and bromoform to marine organisms has not been carried out for the purposes of this profile. The information provided in this section is taken from existing review documents (Lewis et al. 1994, 1997 ). The most sensitive groups of organisms have been identified.

Chlorine and bromine

A review of data on the toxicity of chlorine and bromine to saltwater organisms indicates that far more information is available on the toxicity of chlorine and chloramines to saltwater organisms than for bromine and bromamines with invertebrates (especially crustaceans) exhibiting greatest sensitivity (Lewis et al 1994 and Lewis et al 1997). However, Lewis et al (1997) has presented a comparison the toxicity of bromine and chlorine. Investigating the difference in sensitivity for two saltwater organisms, the silverside Menidia beryllina and the mysid Mysidopsis bahia only a slight increase in toxicity (by a factor of 2) was noted, the chlorine-induced oxidants appearing slightly more toxic than bromine-induced oxidants.

Liden et al (1980) used continuous flow bioassays to compare the effects of bromochlorinated and chlorinated condenser cooling effluent on several estuarine food-chain organisms. Two fish species, Atlantic menhaden Brevoortia tyrannus and spot Leiostomus xanthurus, two bivalve species, American oyster Crassostrea virginica and brackish water clam Rangia cuneata have been investigated. Similar total survival of menhaden and spot as well as oysters and clams exposed to BrCl and Cl2 treated effluents indicated that the toxicities of the residual oxidants were similar for both halogens.

Roberts and Gleeson (1978) determined the acute toxicity of bromochlorinated estuarine seawater (ca 20 l) for several estuarine organisms. When the BrCl toxicity data were compared with Cl2 toxicity data for the same species and LC50s are expressed as equivalents per litre, BrCl was found to be two to four times less toxic than Cl2. The ranking of species in terms of sensitivity was found to be the same for both disinfectants.

Bradley (1977) reported a calculated 24 hour static LC50 for Acartia tonsa of 362 26 &micro;g l-1 bromide chloride (applied in the form of sodium hypochlorite and sodium hypobromite). Toxicity was found to be similar to chlorine (applied as sodium hypochlorite) with a 24 hour static LC50 for Acartia tonsa of 403 46 &micro;g l-1.

Chloroform and bromoform

Studies on the toxicity of chloroform and bromoform to saltwater organisms are outlined below. These indicate that chloroform is of moderate to high toxicity to aquatic organisms. Fewer data are available on the toxicity of bromoform than for chloroform. However, as for chloroform, the data indicate that bromoform is of moderate to high toxicity. The saltwater mollusc Crassostrea virginica appears to be particularly sensitive to bromoform, with lethal and sub-lethal effects being reported at concentrations of 0.05 mg l-1 and less.

Experiments to determine the effect of chloroform on marine algal species Glenodinium halli, Isochrysis galbana, Skeletonema costatum and Thalassiosira pseudonana have been carried out by Erikson and Freeman (1978). After 7 days exposure at 20 C, none of the species showed any inhibition of cell division at a nominal chloroform concentration of 32 mg l-1. Stimulation of cell division was observed in the species at varying concentrations - G. halli at 32 mg l-1, I. galbana at 0.5 mg l-1, S. costatum at 8 mg l-1, and T. pseudonana at 32 mg l-1 nominal dose. However, due to the test conditions, it is likely that, because of evaporation, the actual concentration of chloroform at the end of the 7 day test period would be much lower than the initial 32 mg l-1.

Okubo and Okubo (1962) carried out bioassays with the brine shrimp Artemia salina and the amphibious crab Sesarma haematocheir. The LC100 of the brine shrimp was found to lie between 464 and 800 mg l-1, and concentrations of 480 mg l-1 resulted in 100 % mortality of the crab species. Bentley et al (1975) reported an LC50 of 81.5 mg l-1 for the pink shrimp Pennies duorarum.

Tests have been performed with two species of molluscs, the eastern oyster Crassostrea virginica and the common mussel Mytilus edulis. Fertilised eggs of C. virginica were exposed to chloroform and the survival of the larvae after 48 hour exposure was determined (Stewart et al 1979). Even though the test beakers were covered with aluminium foil to minimise evaporative losses, a loss of 85 % was found from one exposure concentration (100 &micro;g l-1). A 48 hour LC50 of 1 mg l-1 was found. However, if the same losses due to volatilisation can be assumed, then an 48 hour LC50 of 0.15 mg l-1 is probably more realistic. Okubo and Okubo (1962) reported that embryonic development of M. edulis was unaffected by 800 mg l-1.

Okubo and Okubo (1962) also studied the effects of chloroform on the fertilised eggs of the sea urchin Hemicentrotus pulcherrimus. Embryonic development was unaffected by concentrations of up to 800 mg l-1 which indicates that this life stage is very tolerant to exposure to chloroform.

Data for marine fish are limited to one test. Madeley (1973) carried out experiments with the dab Limanda limanda and reported LC0 and LC50 values of 23 and 28 mg l-1 respectively. The figures are similar to those recorded for freshwater fish. No reports of chronic studies have been found.

Sung et al (1978) state that the current practice of chlorination of seawater for power station cooling systems will produce acute toxic effects on marine organisms exposed for periods of 1-25 minutes. It has been known for some time that reproductive tissues, especially sperm, and the immature stages of the organisms are sensitive to very low concentrations of organohalogens, such as bromoform (Davis and Middaugh 1978, cited in Ali and Riley 1986).

Erikson and Freeman (1978) reported that a concentration of >32 mg l-1 bromoform was needed to cause a 50 % reduction in the cell division (EC50) to four species of marine phytoplankton, the marine diatom Skeletonema costatum; Thalassiosira pseudonana, a rapid-growing unicellular diatom; Glenodinium halli, a dinoflagellate; and Isochrysis galbana a microflagellate.

Gibson et al (1979a) studied the toxicity and effects of bromoform on five marine species (3 bivalve molluscs, 1 penaeid shrimp and 1 fish). Considerable difficulty was experienced in maintaining experimental concentrations, due to the volatility of bromoform. Protothaca staminea (littleneck clam), at concentrations of 300-400 mg l-1 were seen to close up their shells and retract their siphons, and thus able to avoid exposure to bromoform. At concentrations of 800 mg l-1, the clams died. The other two bivalve molluscs, Crassostrea virginica and Mercenaria mercenaria, ceased filter-feeding and closed their shells at <10 mg l-1 bromoform. Although there were no mortalities during exposure to 27 mg l-1, some test organisms died immediately after exposure. The LC50 was estimated to lie between 40 and 150 mg l-1 for both species (Gibson et al 1979a).

The species of crustacean tested by Gibson et al (1979a) was Pennies aztecus (a shrimp). This species was found to be more sensitive to bromoform than the mollusc species, with a 96-hour LC50 of 26 mg l-1. Bromoform was pumped in via an air supply at one end of the tank, and within 60 seconds of exposure, the animals had moved as far away from the source as possible. Sub-lethal effects occurred before death including the animals lying on their sides at the bottom of the tank undulating their abdominal appendages.

Stewart et al (1979) found that the by-products formed during chlorination of a power plant cooling water may have adverse effects on the growth of marine invertebrates during their larval stages. They found that concentrations as low as 0.05 mg l-1 were significantly toxic to the larval stages of the marine oyster Crassostrea virginica (approximately 20 % mortality), with a 48 hour LC50 of 1 mg l-1. Scott et al (1982 cited in Ali and Riley 1986) reported that adult oysters of the species Crassostrea virginica which had been exposed to seawater containing 25 &micro;g l-1 of bromoform had an increased rate of respiration while the rate of feeding and the size of gonads had been reduced. There are no data to confirm if this test was acute or chronic exposure. Rapid uptake of the compound occurred, but on removal to clean water, depuration was complete within 96 hours. Although the feeding rate then returned to normal, the damage to the gonads was irreversible.

The menhaden Brevoortia tyrannus was found to be the most sensitive species tested by Gibson et al (1979a), with a LC50 of 12 mg l-1. As individuals approached death, they experienced a loss of equilibrium and lay on the bottom of the tank. Opercular movement gradually decreased until it eventually stopped. Ward and Parrish (1980) conducted a 28 day (chronic) early-life stage test with the sheepshead minnow Cyprinodon variegata to determine the toxic effect of bromoform to embryo and juvenile growth and mortality. The results show that juvenile mortality was a more sensitive indicator of toxicity than the hatching success of embryos, whereas growth appeared to be comparatively non-sensitive. The lowest concentration resulting in juvenile mortality was found to be < 24 mg l-1.

Bioaccumulation

Chlorine and bromine

There are no available data on the bioaccumulation of bromine or bromamines in saltwater organisms. However, CCREM (1987) concluded that, for freshwater organisms, since chlorine and chloramines do not appear to have any potential for bioaccumulation or bioconcentration, it is reasonable to assume that this is probably the same for bromine and bromamines. Additional data are needed to confirm this. In addition, the reaction of residual oxidants with organic substances may yield brominated organic compounds which may well bioaccumulate.

Chloroform and bromoform

Bioaccumulation data for chloroform are contradictory and it appears that slight to moderate bioaccumulation may occur in some aquatic organisms. Based on the available bioaccumulation studies and estimated BCFs, bromoform appears to have a low potential for bioaccumulation.

Chloroform is highly volatile and has a relatively low octanol/water partition coefficient (log Kow = 1.97) Consequently, the bioaccumulation potential is expected to be relatively low and this is illustrated by the low BCFs. An estimated BCF of 18 was derived by Veith et al (1979) for fish and invertebrates. However, Mailhot (1987) calculated a bioconcentration factor (BCF) of 690 for the green alga Selenastrum capricornutum. For bluegill sunfish, channel catfish, largemouth bass and rainbow trout, BCFs ranged from 1.6 to 10 after a 1 day exposure (Anderson and Lusty 1980).

Concentrations of up to 180 &micro;g/kg wet weight have been reported (Pearson and McConnell 1975, cited in Oakley 1988). For Cerastoderma edule (cockle) from Liverpool Bay, concentrations as high as 150 &micro;g/kg (wet weight) have been obtained, compared with a maximum measured water concentration of 1 &micro;g l-1 in the bay. This would indicate a BCF of around 150.

Bioaccumulation potential of the trihalomethanes appears to be low, compared to many chlorinated organic compounds. Gibson et al (1979b) conducted 28 day uptake and depuration tests on five commercially and recreationally important species. These included three species of Penaeid shrimp and one species of fish. The authors found that both uptake and depuration were rapid, with an equilibrium reached after 24 hours. Bioconcentration factors were relatively low (between <1 and 10 times the water concentration). For example, experiments using the marine oyster Crassostrea virginica have shown that exposure to 90 &micro;g l-1 bromoform for 24 hours resulted in a bioconcentration factor of 7.6. Uptake and depuration of bromoform were both rapid, both being essentially complete in 24-48 hours. However, it was concluded that the rate of uptake and depuration was dependent on the individual, the species and the concentration of bromoform in the water.

Potential effects on the interest features of European marine sites

Potential effects include:

  • toxicity of chloroform to invertebrates (in particular molluscs) at concentrations above the EQS of 12 &micro;g l-1;
  • where chloroform is likely to form in the water column, confirmation that bioaccumulation in invertebrates, fish, birds and Annex II sea mammals should be sought.

Next Section                         References