Toxicity of Fire Retardant Chemicals to Aquatic Organisms: Progress Report
Results and Discussion
| All data are tentative and may change with further evaluation and review. This manuscript is a modification of the 1993 annual progress report submitted to the U.S. Department of Interior fire coordination committee. |
Fish
In general, the egg life stage of both species was the least sensitive to the five fire retardants tested and the swim-up stage was the most sensitive (Tables 4 and 5). The 60- and 90-day-old rainbow trout and 30- and 60-day-old fathead minnow were only slightly less sensitive than their respective swim-up life stage.
The five fire retardants were more toxic to several life stages of rainbow trout, especially for Fire-Trol GTS-R and Silv-Ex, and fathead minnow, especially Fire-Trol GTS-R and Phos-Chek D-75-F, in hard water than in soft water, which is unusual (Tables 4 and 5). Typically, the toxicity of toxicants, especially inorganics, is greater in soft water than in hard water (Rand and Petrocelli 1985).
The rank order from most toxic to least toxic of the chemicals tested for rainbow trout was: Phos-Chek WD-881 > Silv-Ex > Phos-Chek D-75-F > Fire-Trol GTS-R > Fire-Trol LCG-R. The rank order from most toxic to least toxic of the chemicals tested for fathead minnow was: Phos-Chek WD-881 > Silv-Ex > Fire-Trol GTS-R > Phos-Chek D-75-F > Fire-Trol LCG-R. The two foams were clearly much more toxic to both fish than were the three non-foam chemicals.
Ammonia
Ammonia concentrations in the low, medium, and high test concentrations of each fire-fighting chemical were measured and used in regression analysis to determine the total ammonia concentration as nitrogen that would have been present at the 96-hour LC50 concentration and are reported elsewhere (Gaikowski 1994). The concentrations of ammonia and unionized ammonia in tests with the swim-up life stage of rainbow trout and fathead minnow are given in Tables 6 and 7. The three non-foam chemicals (Fire-Trol LCG-R, Fire-Trol GTS-R, and Phos-Chek D-75-F) had substantially more ammonia than the two foam chemicals (Phos-Chek WD-881 and Silv-Ex). Fire-Trol LCG-R had the highest total ammonia concentration as nitrogen of the three non-foam chemicals.
Unionized ammonia concentrations were estimated by determining the NH3-N concentration from regression equation coefficients and the 96-hour LC50 concentration. The percentage of unionized ammonia was estimated by using the high and low measured pH recorded at test initiation.
The unionized ammonia predicted at the 96-hour LC50 for the three non-foam retardants for both fish species was very close to those reported in toxicity tests with NH3 alone. Thurston and Russo (1983) reported a 96-hour LC50 of 0.23-0.47 mg NH3/L for 0.12-0.15 g rainbow trout, which is nearly identical to our results with 0.09 g fry in soft water (0.32-0.50 mg NH3/L) and in hard water (0.24-0.56 mg NH3/L). Thurston and Russo (1983) also reported acute toxicity NH3 concentrations for other sizes of rainbow trout that are similar to concentrations in the present study. Thurston et al. (1983) reported a 96-hour LC50 for fathead minnow of 1.1-1.5 mg NH3/L for 0.09 g fish and 0.75 mg/L for 0.13 g fish. These values are nearly identical to our results with 0.12 g fish in soft water (1.22-1.28 mg NH3/L) and in hard water (0.95-2.77 mg NH3/L). Based on these results, it is most likely that the toxicity of the non-foam fire-fighting chemicals is due to unionized ammonia.
Toxicity of the fire-suppressant foams Phos-Chek WD-881 and Silv-Ex may be due to the surfactant portion of their formulation. Various authors have reported on the toxicity of surfactants, with results comparable to the 96-hour LC50s determined in this study. Muller (1980) reported a 24-hour LC50 of 8.5 mg/L for a commercial, non-ionic surfactant using 8-g rainbow trout as the test organism. Muller determined that surfactant toxicity was related to the surface tension reduction caused by the surfactant. The greater the reduction in surface tension, the greater the toxicity of the surfactant. In Muller's study, surface tension was reduced to approximately 45-60 dynes/cm at the 24-hour LT50 (LT50 is the concentration at which 50% of the population survives exposure for the specified time period). In comparison, the 0.6% Silv-Ex field application mixture has a surface tension of 22.92 dynes/cm (Ansul 1991), about half the surface tension reduction that caused mortality as reported by Muller (1980). Reduction in surface tension has also been shown to have adverse effects on gill epithelia, ranging from epithelial swelling to complete destruction of the gill epithelia (Bock 1967, as cited in Muller 1980). Holman and Macek (1980) determined the 96-hour LC50s for three different chain length linear alkylbenzene sulfonate (LAS) surfactants with 2-3 month-old fathead minnow juveniles tested in soft water (40 mg/L as CaCO3). The 96-hour LC50s ranged from 0.86 to 12.3 mg/L, with increasing chain length directly increasing toxicity. Although the exact surfactants used in Phos-Chek WD-881 and Silv-Ex were not known, the 96-hour LC50 of the C11.7 chain length LAS surfactant (12.3 mg/L) was extremely close to the 96-hour LC50s determined in this study for Silv-Ex and Phos-Chek WD-881.
The surfactants used in the fire-suppressant foams pose another threat to aquatic organisms besides their acute toxicity. Surfactants have been shown to alter the permeability of biological membranes (Helenius and Simons 1975). This change in permeability may be detrimental in situations in which multiple stressors are being placed upon an aquatic organism. LAS surfactants increased the uptake of cadmium across the perfused rainbow trout gill above that of gills exposed to cadmium without LAS (Part et al. 1985).
LAS can modify the toxicity of various substances, as well as change their uptake. Solon and Nair (1970) reported an increase in the toxicity of various phosphorothionate pesticides, such as parathion, by as much as 49% when fathead minnows were exposed to the pesticide in the presence of a sublethal (1 mg/L) LAS concentration. Thus, in aquatic ecosystems which are degraded by certain inorganic or organic pollutants, fire-suppressant foam toxicity may be altered, or may alter the uptake and toxicity of the additional pollutants.
Aquatic Invertebrates
Daphnia magna
The toxicity of the two foam chemicals to Daphnia was 10-200 times greater than that of the three non-foams (Table 8). This difference in toxicity of the five chemicals was similar to that observed in fish tests. There was no consistent effect of water quality on the toxicity of the five fire retardant chemicals: no difference for Fire-Trol LCG-R, Fire-Trol GTS-R, or Silv-Ex; the toxicity of Phos-Chek D-75-F was increased in soft water; the toxicity of Phos-Chek WD-881 was decreased in soft water (Table 8). From most toxic to least toxic the rank order of the five chemicals was: Silv-Ex = Phos-Chek WD-881 > Phos-Chek D-75-F > Fire-Trol GTS-R > Fire-Trol LCG-R. This rank order was similar to that for fish.
Hyalella azteca
The toxicity of the two foam chemicals to Hyalella was 5-50 times greater than that of the three non-foams (Table 9). Although this pattern was similar to that observed with Daphnia and fish, the magnitude of difference was not as great, especially for soft water tests. All five fire retardant chemicals were consistently more toxic in soft water than in hard water (Table 9). For the three non-foam chemicals the increase in toxicity in soft water was substantial. The rank order from most toxic to least toxic in soft water was: Phos-Chek WD-881 > Silv-Ex > Phos-Chek D-75-F = Fire-Trol LCG-R = Fire-Trol GTS-R. In hard water the rank order was: Phos-Chek WD-881 = Silv-Ex > Fire-Trol GTS-R = Phos-Chek D-75-F = Fire-Trol LCG-R.
Ammonia
The concentration of total ammonia in tests with aquatic invertebrates were measured in the low, medium, and high test chemical concentrations and used in regression analysis to determine the total ammonia concentration as nitrogen that would have been present at the 96-hour LC50 concentration (Tables 10 and 11). The three non-foam chemicals (Fire-Trol LCG-R, Fire-Trol GTS-R, and Phos-Chek D-75-F) had substantially more ammonia than the two foam chemicals (Phos-Chek WD-881 and Silv-Ex).
Unionized ammonia concentrations were estimated by determining the NH3-N concentration from regression equation coefficients and the 96-hour LC50 concentration. The percentage of unionized ammonia was estimated by using the high and low measured pH recorded at test initiation.
The unionized ammonia predicted at the 96-hour LC50 for the three non-foam retardants for both aquatic invertebrates was close to those reported in toxicity tests with NH3 alone. Studies conducted by Williams et al. (1986) reported 96-hour LC50s ranging from 0.71 to 2.95 mg NH3/L for 11 aquatic invertebrates. USEPA (1985) reported 48-hour LC50s for daphnids of 2.4-2.8 mg NH3/L in hard water and 0.53-0.90 mg NH3/L in soft water. These values are close to our results with daphnids in soft water (0.56-2.7 mg NH3/L for two Fire-Trol compounds) and in hard water (0.57-4.86 mg NH3/L for two Fire-Trol compounds). In our tests with Phos-Chek D-75-F, the amounts of unionized ammonia were lower but still relatively close to toxic concentrations (0.15-0.32 mg NH3/L in soft water; 0.48-0.89 mg NH3/L in hard water). Williams et al. (1986) reported a 96-hour LC50 of 2.05 mg NH3/L to Gammarus pulex (related to Hyalella azteca used in our tests) in hard water, which is substantially higher than unionized ammonia concentrations in tests with the three non-foam retardants, except Fire-Trol GTS-R tested in hard water. The toxicity of these compounds to Hyalella was probably due to other constituents in the retardant formulations.
Algae
The 96-hour IC50s for algae ranged from 10 mg/L for Fire-Trol LCG-R to 79 mg/L for Phos-Chek D-75-F (Table 12). Fire-Trol compounds were substantially more toxic to algae than aquatic invertebrates. The difference in toxicity ranged from 7 times more toxic to algae than Hyalella to 80 times more toxic to algae than Daphnia.
Critical nutrients for algal productivity are phosphorus and nitrogen (Shiroyama et al. 1975). The optimum ratio of nitrogen to phosphorus is about 11:1 to support optimal algal growth. Fire retardant chemicals contain nitrogen and phosphorus in the form of ammonium and diammonium compounds. These chemicals are not considered to be a threat to the environment because their constituents contain fertilizer elements. Without proper nutrient balance, these chemicals can stress aquatic plant species such as algae by limiting their growth and maturity, forcing them to become nitrogen or phosphorus limited. Inadequate nitrogen or phosphorus accounts for the majority of nutrient limitations experienced by algae. Addition of both nitrogen and phosphorus will support growth relative to the phosphorus content in water. Based on the ammonia analysis conducted on algal test concentrations at 0 and 96 hours (Table 13), algae could have been nutrient limited, either by phosphorus or other essential nutrients such as carbon when there was still nitrogen available. Carbon limitation is indicated by increased pH values (Fitzgerald 1975). Because pH values of lower test concentrations were frequently above pH 8.0, carbon limitation could be a strong possibility.
An alternative explanation might be that there were some toxins present in the compounds to adversely affect algae (Miller et al. 1978). Because phosphorus concentrations were not measured, inhibition due to nutrient limitation can not be determined.
Table 12. Acute toxicity (95-hour IC50; mg/L; 95% confidence interval in parentheses) of five fire retardant chemicals to Selenastrum capricornutum exposed in ASTM algal assay medium. Values with different letters are significantly different (p=0.05).
| Chemical | |
| Fire-Trol GTS-R | |
| Fire-Trol LCG-R | |
| Phos-Chek D-75-F | |
| Phos-Chek WD-881 | |
| Silv-Ex | |
Because ASTM algal assay medium is phosphorus limited, stimulation in response to the addition of ammonium and phosphorus compounds would be expected. Some stimulation was evident in four of five test chemicals. Phos-Chek WD-881 produced a stimulation in lower chemical concentrations even though the chemical did not contain ammonium or phosphorus, based on available information. Stimulation suggests the presence of additional phosphorus or a response to CO2 evolving from biodegradation of the chemical. Addition of phosphorus will increase algal biomass when in the presence of excess nitrogen (Miller et al. 1978), whereas addition of CO2 did not change algal productivity or decrease pH substantially in other tests (Fitzgerald 1975).
The greatest stimulations occurred in long-term retardant chemicals. Fire-Trol GTS-R stimulated twice as much growth in the two lowest concentrations as Fire-Trol LCG-R in similar concentrations, whereas the lowest Phos-Chek D-75-F treatment produced the greatest stimulation of all five test chemicals. The ammonium and phosphorus constituents are essentially identical in Phos-Chek D-75-F and Fire-Trol GTS-R, therefore a smaller stimulation in Fire-Trol GTS-R could be due in part to toxicants present in the test chemical as well as a lesser concentration of these constituents in the chemicals.
Comparison to Published/Manufacturer Data
There is very limited information on the toxicity of the five fire fighting compounds tested except for studies conducted by the manufacturers or their contract testing facilities (Table 14). Most of the results of these tests are within a factor of four of our values, which is within the typical range of interlaboratory variation for acute toxicity data (Schimmel 1981). However, our acute toxicity data tended to be lower than those reported by the manufacturers or their contract laboratories, and one of our results was greater than 4-fold different. Our test with Phos-Chek D-75-F tested in soft water with 0.6-g rainbow trout resulted in a 96-hour LC50 of 234 mg/L, which was greater than 4 times lower than the manufacturer's value of >1000 mg/L.
Relation to Environmental Conditions
Foam chemicals are applied at about 1% foam, which is equivalent to 1 g/100 ml. This field application rate can be related to our toxicity values by converting 96-hour LC50 values to the same units. For example, the 96-hour LC50 of Silv-Ex to swim-up rainbow trout is 20 mg/L, which is equivalent to 0.002 g/100 ml. Thus, the 96-hour LC50 value is equivalent to a 0.002% foam solution. This acute toxicity value is 500 times less than the 1% foam field application rate. Consequently, if a 1% foam solution came in contact with an aquatic environment it would have to be diluted 500 fold to reach the 96-hour LC50 concentration -- a concentration which would cause a substantial amount of mortality in an aquatic environment.
A safety factor may be applied to toxicity data to estimate a safe concentration for aquatic organisms. A safety factor for acute toxicity data is usually 100. Applying a safety factor of 100 to the above toxicity information would require a 50,000 dilution (100 × 500) of the 1% foam field application solution to approach a safe concentration. Similar approaches could be used with toxicity values for other fire-fighting chemicals.
Summary
Overall, the toxicity of the five fire retardant chemicals to these four species is remarkably similar. The two foams, Silv-Ex and Phos-Chek WD-881, have very similar toxicity and are substantially more toxic than the three non-foams. Of the non-foams, Fire-Trol GTS-R and Phos-Chek D-75-F have similar toxicity, which was substantially higher than Fire-Trol LCG-R except for the amphipod Hyalella azteca
Water quality did not seem to modify the toxicity of the five fire retardant chemicals in a consistent manner, except for Hyalella which were consistently more sensitive in soft water.
For the three non-foam chemicals, Hyalella was the most sensitive species in soft water, whereas fathead minnow was the most sensitive species in hard water (Table 15). For the two foam chemicals, Daphnia in three tests and Hyalella in one soft water test were the most sensitive species.
In 8 out of 10 tests Daphnia were more sensitive than the swim-up life stage of rainbow trout. The greatest difference in sensitivity associated with water quality was shown by Hyalella. In 4 out of 5 soft water tests Hyalella was the most sensitive species, but in 4 out of 5 hard water tests it was the least sensitive species.
Previous Section -- Methods and Materials
Return to Contents
Next Section -- References
