Document RpoKvQZN949pVp5BJb01V99qE

BACK TO MAIN POTENTIAL FOR ENVIRONMENTAL IMPACT OF AFA-6 SURFACTANT A Report for: PANARCTIC OILS LTD. CALGARY, ALBERTA - .- .. Submiitted by: BEAK CONSULTANTS LIMITED MISSISSAUGA, ONTARIO u.L Robertson Project Manager 13.c"2.. C. Craig f Director, Toxicological Services 7 January 1986 1.0 SUMMARY Panarctic Oils Ltd. of Calgary has applied to t h e Environmental Protection Service (EPS) of Environment Canada to use Ceotech AFA-6 Accelerated Freezing Additive in t h e construction of an ice island for exploration in t h e Arctic ocean. Although preliminary approval had been granted for t h e 1984-85 season, EPS remained concerned a b o u t potential environmental impacts. Beak Consu1tant:j Limited was retained by P a n a r c t i c to investigate the potential impacts and prepare this report addressing EPS concerns. Panarctic plans to apply AFA-6 at a concentration of 10 mg/L down to -21OC. AFA-6 is a perfluorinated alkyl ethoxylate surface active agent manufactured from a perfluorinated alkyl sulphonate (product number FC-95) and ethylene oxide. It will biodegrade slowly, producing the original perfluorinated alkyl sulphonate, plus C 0 2 and water as its final degradation products. The application concentration of AFA-6 is similar to t h e LC50 f o r representative salt w a t e r organisms. For Artemia (brine shrimp) t h e LC50 wa= !Lm&L , while for sea-water acclimatized trout the LC50 was 4 mg/L. The degradation product (FC-95) was slightly less toxic, having LC5O's of 9 mg/L and 14 mg/L, respectively. Octanol/water partition coefficients were measured for both AFA-6 and FC-95 and w e r e used to estimate bioconcentration potential. Estimated bioconcentration factors for AFA-6 and FC-95 were 3 and 1, respectively, indicating essentially no significant bioconcentration potential. Beause of t h e limited use of AFA-6, it is estimated t h a t a maximum of 9% of t h e total volume of the ice island will contain AFA-6. Consequently, it is probable t h a t dilution during melting will reduce the under ice concentration to less than t h e 1-2 mg/L nonlethal concentration observed in the bioassays. On the basis of the information available, application of AFA-6 in t h e concentrations and quantities proposed by Panarctic does not appear t o represent an environmental hazard. 9328.1 1.1 2 0 INTRODUCTION 2 1 Backgromd -- Each drilling season, P a n a r c t i c Oils Ltd. builds ice! islands in t h e Beaufort Sea to serve as exploration platforms. The rate at which thesie islands can be constructed depends largely upon the temperature during construction. During the 1984-85 drilling season Panarctic used a chemical product known as AFA-6 supplied by t h e Geotech Corporation to enhance the rate of freezing, through surface tension reduction. Use of t h e A-FA-6 was limited to a concentration of 10 mg/L in ii specific temperature range (down t o -21OC) below which i t had no effect. T h e Regional Office of Environment Canada had approved t h e use of t h e AFA-6 compound for t h e 1984-85 season, but had several concerns regarding potential environmental impacts. Because SEAK was conducting similar work on other surf a c t a n t s , EPS Industrial P r o g r a m s Branch Regional O f f i c e approached BEAK a b o u t inclusion of t h e AFA-6 in a parallel study to address some of the environmental issues. BEAK was subsequently retained by Panarctic to carry out an assessment of AFA-6 and its breakdown products. 2 2 Project Objectives and Scope EPS has expressed concern about t h e following issues: 0 -w - :?, .*. I - 0 persistence of t h e surfactant and i t s breakdown products, biodegradability, and measurement of the surfactant and breakdown products, as well as identity of t h e breakdown products, 0 distribution in various environmental compartments (aqueous phase, sediments, food chain, air), 0 potential for bioconcentration of t h e surfactant or its breakdown products, 0 acute toxicity of the surfactant and breakdown products. 9328.1 2.1 T h e objective of this project was to acquire sufficient d a t a to answer these concerns and establish t h e acceptability of t h e AFA-6 surfactant using these concerns as criteria. The following tasks were identified to achieve this objective: 0 Physical/Chemical characterization including chemical composition and structure, solubility, melting and boiling points, vapour pressure, octanol water partition coefficients. Data sources would include manufacturer's safety data s h e e t (MSDS), chemical l i t e r a t u r e , in-house testing. 0 Validation of appropriate analytical procedures including wet chemical and instrumental methods such as gas chromatography (GC)or high pressure liquid chromatography (HPLC). 0 Determination of biodegradability and identification of breakdown products. Data sources would include t h e chemical literature and in-house testing. 0 Determination of a c u t e toxicity to a variety of test organisms and in particular t h e salt water species Artemia (brine shrimp), representative of the branchipoda invertebrate group which is resident in the Arctic Ocean. Organisms to be tested included Daphnia (fresh water), Artemia (salt water), rainbow trout (fresh and salt water). 0 Estimation of t h e potential for accumulation in sediments and biota based on the 4 i octanol/water partition coefficient (KO,) Submit a final report detailing the test results and making a recommendation about the continued use of AFA-6. This report presents and summarizes all the information gathered, together with a recommendation.regarding t h e continued use of AFA-6. 9328.1 2.2 . !as?k 3.0 ENVlRONMENTAL CHEMISTRY OF AFA--6 3.1 General The designation AFA-6 is the Geotech produc:t number for a perfluorinated alkyl ethoxylate manufactured by 3-M (Minnesota Mining and Manufacturing). T h e 3-M product numbers for this type of compound are FC-171 or FC-760. The compound can be classified generally as a nonionic surface a c t i v e agent (surfactant) manufactured from _- a potassium perfluoroalkylsulphonate and ethylene oxide. The perfluoro-alkylsulphortate starting material is an anionic s u r f a c t a n t also manufactured by 344, and designated as FC-95. -. All organic surfactants contain both a hydrophilic and a hydrophobic moiety in the same molecule. With modem chemical processes, surfactants can essentially be "tailor-made" to exhibit specific properties. In general, t h e hydrophobic portion of a s u r f a c t a n t molecule is derived from a C8 to C20 hydrocarbon, selected on the basis of t h e particular properties desired. Hydrophilic groups can be divided into two groups: those t h a t ionize in aqueous solution (anionic and cationic) and those t h a t do not (non-ionic). FC-95, t h e starting m a t e r i a l for AFA-6, is an anionic surfactant compound containing a fluorinated alkyl hydrophobe and a hydrophilic sulphonate group as shown below. perf 1uorinated alkyl hydrophobe anionic hydrophile - To manufacture AFA-6, a polyoxyethylene (polyether) chain is added on to the sulphonate group using ethylene oxide. The polyether chain is non-ionic and confers somewhat different properties on t h e molecule as compared to the anionic starting material. T h e polyether chain length in most non-ionic surfactants can be varied simply by continuing t h e introduction of ethylene oxide into t h e reaction mixture (Swisher, 1970). T h e rate- controlling s t e p is the formation of t h e "mono-adduct", an intermediate product containing only one ethylene oxide unit. Thereafter, polymerization occurs rapidly, independent of chain length (Kirk-Othmer). Consequently, the polyether chain is built up before all the starting material is reacted and the final product is polydisperse with 9328.1 3.1 respect to chain length. This is illustrated by t h e structure of t h e AFA-6 shown below, where t h e mean polyether chain length is 7.2 ethoxylate units. The chain length probably varies from 5 to 9 units with t h e predominant length being 7 with a mean of 7.2. perfluoroalkyl sulphonate ethoxyl a t e (polyether) chain Table 3.1 summarizes t h e physical/chemical properties of AFA-6 and its precursor compound FC-95. Water solubility for AFA-6 is reported by Ceotech as 'fnegligibleff. In the toxicity tests reported in Chapter 4.0, solutions of up to 100 mg/L were prepared. cs- However, the AFA-6 would not readily dissolve at room temperature and had to be heated to 4O-5O0C with vigorous stirring to produc:e a stable solution. 3.2 Analytical Protocols Two analytical techniques were evaluated for analysis of the AFA-6 surfactant. The Cobalt Thiocyanate Active Substances (CTAS) test for non-ionic surfactants was evaluated and refined to allow detection of 0.2 mg/L. This is adequate to allow detection of the compound at concentrations lower than t h e lowest reported LC50 (0.4 mg/L to Daphnia). Although the method is not specific to AFA-6, i t is unlikely t h a t any other non-ionic surfactants will b e present in Arctic waters or in test solutions. The method will also d e t e c t nonionic surfactant brealkdown products containing more than 3 ethoxylate units. a Analysis of the surfactant was a t t e m p t e d by GC/EC (gas chromatography with electron - capture detection) to provide a highly sensitive, specific analysis of t h e AFA-6. Initially, direct analysis under various column conditions was assessed but none was unsuccessful. Subsequently, an HBr cleavage technique used for analysis of linear alchol ethoxylate nonionic surfactants was tested. While the technique showed typical ethoxylate unit cleavage and bromination response, the balance of the molecule did not chromatograph well. As the AFA-6 compound is an ethoxylated perfluoroalkyl sulphonate, t h e chromatographic problems encountered in method development can be attributed to t h e highly polar sulphonate group in the cleavage product and its associated high boiling point. 9328. I 3.2 . TABLE 3.1: PHYSICAL/CHEMICAL PROPERTIES OF AFA-6 AND FC-95 FLUOROCHEMICAL SURFACTANTS Property AFA-6 FC-95 Form -.h ? Composition 3 Ionic Type pH (0.1% Aqueous Solution) Melting Point (OC) Boiling Point (OC) Water Solubility (at 25OC) Specific Gravity Clean, yellowish liquid 100?6 a c t i v e Non-ionic - L 3OO0C Negligible* 1.4 Free-flowing white powder 100% active Anionic 7-8 Decomposes a t 39OoC Decomposes at 39OoC 2,000 mg/L approx. 0.5 (bulk density) ~~ 4 - -23.- * Solutions of up to 100 mg/L were prepared for bioasay tests, indicating significant solubility in environmental terms. L = less than The FC-95 compound could not be analysed by t h e CTAS method, b u t was found to be responsive to t h e Methylene Blue Active Substance (MBAS) tests used for anionic surfactants. As with t h e AFA-6 in the CTAS itest, FC-95 was less responsive in t h e MBAS test than other anionic surfactants. However, t h e response was sufficient to allow detection in 0.2-0.5 mg/L range as well. 3.3 Octanol/Water Partition Coefficients The partition coefficient (P) m a y b e determined empirically or calculated (Leo, 1971)-and represents t h e partitioning of a compound into the n-octanol phase of a n octanol/water mixture. The octanol/water interface models biological membrane systems and the P value can be related to lipid solubility. High P values are representative of lipid soluble persistent chlorinated hydrocarbons (e.g., PCB, DDT, mirex) and low ? values are associated with non-lipid-soluble non-persistent compounds (e+, phenols). Octanol/water partition coefficients can be used to predict bio-magnification potential and potential for accumulation in sediments (Veith et al., (1979); Neely et al. (1974); Mackay (1982); Karickhoff et al. (1979)). No d a t a on octanol/water partitioning for either AFA-6 or FC-95 was available from 3-M or t h e chemical literature. Consequently, o c t m o l / w a t e r partition coefficients were measured in t h e laboratory. The method used is essentially t h a t presented by Leo et al. (1971). Analytical methods were t h e CTAS and FIA'BAS methods indicated in Section 3.2. I Table 3.2 summarizes the partition coefficients [:PIand their base 10 logarithms (log P) 1':...-J for both AFA-6 and FC-95. Bioconcentration factors (BCF) have been calculated for 2 both using t h e equation developed by Veith et al. (1979). ' L-2 3.4 Biodegradation and Breakdown Products T h e study of s u r f a c t a n t biodegradability first beg,an in England in t h e 1950's, when many of t h e general principles of surfactant biodegradation were established during research on anionic surfactants. These principles were l a t e r extended to other s u r f a c t a n t classes. In t h e case of non-ionic surfactants, additional work was required on t h e effect of t h e hydrophilic group because, unlike t h e anionics, i t is potentially biodegradable as well. 9328.1 3.3 i TABLE 3.2: OCTANOL/ WATER PARTITION COEFFICIENTS AND DERIVED DATA FOR AFA-6 AND FC-95 Compound P log P BCF AFA-6 FC-95 25 11.4 6.0 0.8 NOTES: P= log P = BCF = octanol/water partition coefficient; also known as Kow base 10 logarithm of P or KO,, bioconcentration factor (from equation in Veith, 1979) fc +. F I (-sIg 7j 3(0.53) - 5t6-7 The following generalizations relating struc1:ure and biodegradation of nonionic surfactants have been reasonably well documented. i) Hydrophobe Structure: biodegradation is enhanced by hydrophobe linearity and deterred by branching or chemical substitution (e+, replacement of H by F); ii) Nature of Hydrophile generally has only a minor effect; in polyethoxylate non- & .ionics shorter chain length promotes complete conversion to carbon dioxide and water (ultimate biodegradability); and iii) Distance Principle generally, t h e g r e a t e r t h e distance between hydrophile and t h e far end of t h e hydrophobe inlcreases t h e speed of the primary biodegradation. This is particularly t r u e for ABS. The structure and cornpositon of the AFA-6 surfactant permits only partial biodegradation, as discussed below. i) Hydrophobe Structure: This portion of the molecule is completely fluorinated. consequently, i t will be essentially non-biodegradable, but will also be chemically and biologically inert, as i t has a structure and composition much like teflon. ii) Hydrophile: the polyether chain which solubilizes the AFA-6 molecule, will be biodegradable. Birch (1982) used standard OECQ biodegradation procedures to demonstrate t h a t alcohol ethoxylates with up to 20 ethoxylate units can b e readily and. completely degraded. With only about 7 ethoxylate units biodegradation of t h e polyether chain in AFA-6 will occur. In general, biodegradation of ethoxylated non-ionic surfactants can occur at t h e hydrophobic or hydrophilic group. A non-ionic s u r f a c t a n t may undergo "complete primary biodegradation" as measured by change:; in surface tension, foaming capacity, CTAS or other chemical analysis based on the presence of the ethyoxylate group. However, ltultimate biodegradation" Le., conversion to C 0 2 and water may be only partially achieved, or may be achieved at a much slower rate. This implies the formation of intermediate products of potential environmental significance. That is, by-products may be formed which also have surface a c t i v e properties, which may e x e r t toxicity, which may bioaccumulate, or which may persist in the environment for some time. 9328.1 3.4 Using a n HDr cleavage/GC analysis technique developed for the determination of t h e hydrophobe/hydrophile ratios in commercial non-ionic surfactants, Tobin et al. (1976a) showed t h a t t h e alkyl portion of a readily degradable alcohol ethoxylate was degraded rapidly, whereas the ethoxylate portion was degraded much more' slowly. The polyethylene glycol (PEG) like by-products from t h e ethoxylate portion were only 45?6 degraded in 530 hours, by which time, analysis :Ear t h e surfactant itself and t h e alkyl portion of t h e s u r f a c t a n t showed 100% degradation. Testing a similar surfactant at t h e s a m e starting concentration (20 mg/L) K r a v e t z (1982) showed t h a t t h e fish toxicity of t h e shake flask solution was reduced to less t h a n t h e surfactant LC50 in less than 15 days (360 hours) and to zero in a b o u t 600 hours. This suggests t h a t t h e PEG-like by-products are less toxic t h a n t h e original surfactant. However, no specific dose-response d a t a w e r e presented for t h e PEG-like products. In a subsequent paper, Tobin et al. (1976b) showed a similar pattern in lake w a t e r degradation studies conducted in-situ on the same surf actant. The previous experiments w e r e further expanded to include C 0 2 production as a comparative parameter. Anthony and Tobin (1977) subsequently used a sequential extraction scheme to demonstrate t h e rapid formation of polyglycol which is then slowly degraded. Cook (1979) used both HBr cleavage and trirnethylsilylation with GC analysis to demonstrate t h a t similar reactions ocurred in the activated sludge degradation of a similar linear alcohol ethoxylate. < K r a v e t z et al. (1982) used radio-labelling techniques t o demonstrate t h a t both alcohol s ethoxylates and alkylphenol ethoxylates produce soluble organic intermediate by- .? . products from the hydrophilic- polyethoxylate goup, with the alkylphenol producing - substantially more. Similarly, the hydrophobic alkylphenol portion produced metabolites as well. In contrast, t h e alcohol produced very little hydrophobe metabolite, as expected from t h e work of Tobin et al. (1976a,b). K r a v e t z et al. (1982) proposed t h e following mechanisms for the degradation of alcohol and alkylphenol ethoxylates. 9328.1 3.5 Later in 1982, Stephanou and Ciger confirmed this mechanism with t h e identification of nonylphenol mono - and di-ethoxylates in sewage treatment plant effluents. In AFA-6, only t h e ethoxylate portion of t h e molecule should be biodegradable, because of the complete fluorination of t h e alkyl hydrophobe. As y e t unpublished work by 3-M using radio-carbon tracer techniques in mice has demonstrated t h a t t h e ultimate degradation product of AFA-6 is a perfluoralkyl sulphonate similar to FC-95. Data from 3 4 show t h a t t h e 20 day BOD of FC-760 (AFA-6) is 170,000 mg/L while t h e COD is 700,000 mg/L indicating only 25% decomposition. Assuming t h e chemical composition and structure shown in Section 3.1, and assuming t h a t t h e fluorinated alkyl portion of the molecule is inert, t h e theoretical oxygen demand of AFA-6 is 1,050,000 mg/L. This indicates only partial biodegradation and a relatively low rate of u degradation. No inhibitory effect on activated sludge respiration rate was noted at 1,000 mg/L, indicating no bacterial toxicity. Biodegradation tests undertaken in brine solution in BEAK'S laboratories showed a similar slow degradation. Using a low initial concentration of 14.5 mg/L as dissolved organic carbon (DOC), no significant degradation had occurred a f t e r 9 days. A t 57 days, t h e DOC had been reduced to 6.0 mg/L. Based on t h e relative carbon contents of t h e fluorinated alkyl sulphonate and the average ethoxylate chain length, this indicates that biodegradation of the ethoxylate chain was approaching completion. T h a t is, 59% DOC removal had been achieved as compared to 64% for complete ethoxylate degradation. The DOC removal suggests an average of one ethoxylate group remaining undegraded. 9328.1 3.6 Test conditions were set up similar to t h e I S 0 standard aerobic shake flask test with nutrient salts in a brine (sea water) solution. Although IS0 recommends use of low concentrations of various organic growth factors, t h e APA-6 was used as t h e sole organic substrate to simplify DOC measurements, and to simulate pristine Arctic w a t e r conditions. This may have contributed to the nine-day lag observed. I t is clear from t h e data presented above t h a t AFA-6 does biodegrade and t h a t it does so relatively slowly. With an applied concentration of 10 mg/L concentration, no measureable e f f e c t on dissolved oxygen will b e observed. - 9328.1 3.7 4.0 TOXICOLOGY 4.1 Exposure Protocols Rainbow trout (Salmo gairdneri), water flea (Daphnia magna) and brine shrimp (Artemia salina) were used as t h e bioassay organisms to determine the toxicity of t h e AFA-6 and FC-95 compounds. Exposure protocols followed the guidelines outlined in "Standard Procedures f o r Testing Acute Lethality of Liquid Effluents" (Environment Canada, 1980). for rainbow trout, the International Standards Organization (1982) protocol for daphnia and t h e International Standards Organization (Vanhaecke and Persoone, 1981) d r a f t protocol for artemia. Bo.t.h rainbow t r o u t and daphnia a r e widely recognized as standard test organisms for fresh w a t e r toxicity assessments and theref o r e provide an excellent basis of comparison to establish relative toxicity to other compounds. Acclimation of rainbow t r o u t to saline conditions is also an accepted means to estimate toxicity of compounds in marine environments. Artemia are native to inland saltwater lakes and are physiologically adapted to a salt environment. The species has been adopted as a representative marine organism for toxicity screening tests by European member countries of t h e International Standards Organization (ISO). The use of Artemia provides a practical and standardized means of toxicological assessment and meets all of t h e requirements of biological testing. Culture and maintenance are well documented, standardized methods have been developed, and ----i tests are reproducible and comparable among laboratories. The strength of incorporating -"it artemia data is t h a t marine invertebrate information is provided. The genus is within t h e .- .e' 3 - crustacean family which includes crabs, shrimp and lobsters. The artemia data can therefore be considered to b e representative of marine crustacean sensitivities in t h e absence of specific species test data. All protocols were similar in t h a t they required test organisms to be exposed to a - logarithmic concentration series of test product for a set period (fish - 4 days, invertebrates 2 days). Mortality, dissolved oxygen and pH were recorded daily and t h e cumulative mortality-concentration plot was used to estimate t h e concentration producing 50% mortality (LC50). 9328.1 4.1 TABLE 4.1: DILUTION VATER CHARACTERISTICS (Dechlorinated Vississauga Tap Water) Chemical.Parameter* PH Conductivity (umhos/cm Alkalinity Hardness (as CaCo3) Ammonia4 Chlorine Anion Scan Chloride Fluoride Sulphate 0-Phospha te Nitrate N it rite DCP Scan Berylium Molybdenum Calcium Vanadium Aluminum Magnesium Borium Potassium Strontium Sodium Zinc. Cadmium Manganese Cobalt - Copper Silver Iron . Lead Chromium Nickel ~ ~~~ ~ ~~ ~~~ * Results in mg/L unless otherwise indicated. m g/L 7.94 377 88 137 < 0.06 < 0.05 28.0 1.1 0.74 0. I 0.46 0.1 < 0.005 0.015 40 <0.005 0.15 8.6 0.02 1.7 0.19 12.6 $::;<0.005 <0.005 <0.005 <o.o 1 <0.005 <O.Ol <0.005 <0.005 ' Tests were static, continuously a e r a t e d and at a constant temperature (fish - 15OC; invertebrates - 22OC). Due to t h e size of t h e fish t h e exposure solution was replaced a f t e r 48 hours to maintain an a c c e p t a b l e loading rate. Photoperiod was 12 hours light, 12 hours dark. Rainbow trout were purchased from a certified disease free hatchery (Rainbow Springs, Thamesford) and acclimated for 28 days in Mississauga dechlorinated tap water (Table 4.1). Daphnia were originally obtained from a culture maintained at the Canadian Centre for Inland Waters, Burlington and have been successfully reproducing f o r 1 year. Daphnia first instars (less than 24 hours old) were collected from adults and used in t h e tests. Artemia eggs (Salt Lake Brine Shrimp Inc.) were incubated at 22OC in 30 parts per thousand (ppth) NaCl solution and hatched naupuli (less than 24 hours old) were collected for exposure. The same 30 ppth NaCI solution was used to dilute t h e AFA-6 and FC-95. Fish and daphnia tests were conducted in duplicate, while artemia tests were completed in triplicate. Specific test conditions appear in Table 4.2. 4.2 Saltwater Acclimation of Rainbow Trout 23 Rainbow trout (4.2 g f 0.9 sd) w e r e transferred into a 350 L holding tank containing a 15 .3 ppth salt solution of dechlorinated t a p water. Each day for t h e following four days t h e c..d; , - s a l t concentration was i n c r e a s e d - 3 ppth by t h e addition of a stock s a l t solution. The ionic composition of the final s a l t solution appears in Table 4.3. The salinity was further increased by 2 ppth/day until a final salinity of 30 ppth was reached. The fish were held at this salinity for 8 days before testing. The e n t i r e reservoir of saline acclimation water was changed every 2 days. A 30 ppth solution which was considered representative of a r c t i c marine conditions was used as dilution water in t h e tests. 9328.1 4.2 TABLE 4.2: ACUTE STATIC TEST CONDITIONS USING RAINBOW TROUT, DAPHNIA AND ARTEMIA Test Organism Acclimation Period (days) Organism Size (g) Number Org/Vessel Exposure Test Volume (L) Loading R a t e @)d$ Test Temperature (OC) Exposure Period (Hrs) Replicates Rainbow Trout Fresh water Salt water 28 4.2 f 0.9 (sd) 6 35 0.69" 15 96 * 2 8 d at 30 ppth 4.4 f 1.9 (sd) 6 35 0.66" . 15 96 * 2 * - 96 hr s t a t i c test with 48 hr renewal Daphnia 24 hr 0.01 7 10 0.2 1.0 22 48 2 Artemia 24 hr 0.001 10 0.01 0.5 22 48 3 TABLE 4.3: COMPOSITION OF SYNTHETIC SEAWATER (30 ppth) USED FOR ACCLIMATION AND TESTING OF RAINBOW TROUT -_ Ion Concentration (g/L) Calcium Magnesium Sodium Ch1otide 0.4 . 1.3 10.8 19.6 ~~ ~~~ ~ Dilution water was Mississauga dechlorinated tap water (see Table 3.1). 4.3 Acute Toxicity Mortality d a t a for rainbow trout, daphnia, and artemia appears in Appendix 1 and is summarized in Table 4.4 f o r both AFA-6 and FC-95. Rainbow trout appeared to be slightly more sensitive to the AFA-6 in saltwater compared to freshwater while the degradation product FC-95 appeared slightly more toxic to fish in freshwater. The difference in toxicity between fresh and salt water !,. conditions f o r t h e two products is not biologically significant. The average LC50 value of 5.2 mg/L for AFA-6 and 11.3 mg/L for FC-95 suggests t h a t t h e parent product AFA-6 is about t w i c e t h e toxicity of t h e degradation product. Artemia, the representative marine invertebrate, displays twice the tolerance to AFA-6 as rainbow t r o u t while sensitivity to t h e FC-95 was comparable to rainbow trout. Daphnia, a freshwater invertebrate, is twenty fold more sensitive to AFA-6 than rainbow trout, but is a b o u t five fold more tolerant to t h e FC-95 than rainbow tqout. T h e reduction in toxicity to daphnia between the parent and degradation product is about 150 times, which represents a major response difference compared to the other two organisms. This serves to underline t h e importance of using test results for representative marine organisms to estimate marine toxicity rather than develop projections entirely from fresh water data. Recognizing t h a t t h e AFA-6 will be released into seawater during melting of t h e ice it would be important to know what concentrations would not be lethal during short-term (24 hr) exposure. Table 4.5 summarizes t h e range of highest concentrations t h a t were ~ observed to be non-lethal to rainbow trout and artemia. Since the estimated degradation half-life is about 30 days a c u t e exposure (24 hrs) of AFA-6 t o fish will represent t h e most critical potential impact. Once diffusion of t h e AFA-6 decreases t h e concentration below 2 mg/L, acute effects should not occur. 9328.1 4.3 . w& TABLE 4.4: ACUTE LC50 VALUES (mg/L) AND 95% CONFIDENCE LIMITS ( ) FOR RAINBOW TROUT, DAPHNIA AND ARTEMIA EXPOSED TO AFA-6 AND FC-95 UNDER FRESH'S'ATER AND SALTWATER CONDITIONS Species LC5Os (95% Confidence Limits) Freshwater . (mg/L) Saltwater (mg/L) - AFA-6 Rainbow Trout Artemia Daphnia Magna FC-95 ...;, .- Rainbow Trout Artemia Daphnia N/A - Not Applicable 6.4 (4.9 - 8.4) 6.1 (4.8 - 7.8) N/A N/A N/A . 0.29 (0.23 - 0.35) 0.27 (0.22 - 0.33) - 7.8 (6.2 9.8) - 9.9 (7.5 13.4) N/A N/A ' N/A 58 (46 - 72) 67 (48 - 92) 4.6 (3.6 - 5.9) '3.7 (3.2 - 4.3) 9.7 (7.5 - 12.7) 10.3 (7.9 - 12.5) 9.1 (7.2 - 11-41 N/A N/A - 13.7 (10.7 - 13.7 (10.7 17.7) 17.8) 9.4 (7.4 - 12.1) 9.4 (7.3 - 12.2) 8.9 (6.7 - 11.9) N/A N/A TABLE 4.5: HIGHEST NON-LETHAL CONCENTRATIONS OF AFA-6 6r FC-95 * (mg/L) Rainbow Trout Artemia AFA-6 2-3 3-5 FC-95 5 - 30 1-2 * Exposure Period = 24 hours 4.4 Sublethal Effects In addition to direct t o x i c effects, chemical compounds can also impair growth, reproduction, or behaviour of organisms. This affects t h e viability of t h e population without. causing mortality. These sublethal effects generally occur at lower exposure concentrations over longer periods of t i m e than acute toxic (lethal) effects. The potential chronic sublethal e f f e c t s of chemical compounds can be estimated using a c u t e lethality data. - Various safety factors have been adopted for application to acute toxicity data when information on chronic effects was unavailable. The ratio between a c u t e and chronic toxicity d a t a for similar compounds has been used to provide a safety factor. The application of an appropriate acute/chronic ratio to a c u t e data then provides an e s t i m a t e of chronic effect concentrations. In a review of non-ionic s u r f a c t a n t toxicity (BEAK,1985) the LC50 values for non-ionic surfactants containing 5-10 ethoxylate groups ranged between 1 to 10 mg/L bracketing t h e LC50 values reported in this study. However, no sublethal effect levels for t h e former compounds were found in t h e literature. Drawing on data generated for lauryl alkyl sulfonate, albeit an anionic surfactant but having a similar level of fish toxicity (LCSO about 4 mg/L) and probably a similar mode of action, i t would not be unreasonable to use the chronic/acute ratio developed for LAS (0.2) to estimate a chronic e f f e c t concentration for AFA-6 and FC-95. I t would appear t h e n that a n a m b i e n t value 0.2 of t h e a c u t e AFA-6 and FC-95 LC50 concentration would protect against chronic e f f e c t s in aquatic biota. A > - 4.5 Bioconcentration Factors The level to which a compound accumulates in aquatic organisms compared to t h e ambient concentration is ref erred to as t h e bioconcentration factor (BGF). Considerable research has been devoted to estimating BCF's based on the octanol water partition coefficient (P). The more lipid soluble a compound (high P value) t h e g r e a t e r t h e propensity for t h e compound to be sequestered in fat tissues and therefore be isolated from metabolic breakdown and excretion. During periods of organism stress (migration, reproduction, starvation! compounds in fat tissue can be mobilized and exert an effect. 9328.1 4.4 The bioconcentration f a c t o r can be expressed in t e r m s of the P value of a compound according to t h e relationship developed by Veith et al. (19791, log BCF = 0.85 log P - 0.70 T h e importance of t h e BCF derivation is t h a t it can be used to e s t i m a t e t h e exposed organism's toxicant dose from ambient exposure concentrations. Knowing that ambient concentrations produce a toxic response, t h e equivalent dose per unit body weight can be determined. To provide sufficient protection - t o aquatic organisms so they a r e not exposed to critical (effect) concentrations, ambient Concentrations can be limited so t h a t they do not exceed predicted equivalent organism dose concentrations and thereby ensure biological e f f e c t s are not produced. Limiting t h e exposure concentration to t h e inverse of t h e BCF value ensures t h a t t h e dose concentration (body burden Concentration) does not exceed the ambient effect concentration equivalent. Incorporating the chronic/acute ratio into the exposure limit ensures that the organism concentration does not exceed the ambient chronic effect concentration, thereby providing additional safety in developing a chronic exposure limit. The BCF values and inverse functions for AFA-6 and FC-95 which would be used to develop exposure limits are as follows: - AFA-6 FC-95 --II log P 1.4 0.8 - BCF 3 1 1/BCF 0.32 1.01 The above values indicate t h a t both AFA-6 and FC-95 would have very limited levels of accumulation in biota compared to recognized persistent highly accumulative compounds (e&., BCF f o r PCB = 1.5 x 105;DDT = 1.5 x 104 4.6 Exposure Limits T h e a c u t e toxicity concentrations provide the basis of exposure limit derivation and t h e mean LC50 values will be used to select the most sensitive representative trophic group from fish and invertebrates. 9328.1 4.5 A B Mean LC50 (mg/L) Chronic/Acute Ratio C 1/BCF Exposure Limit (mg/L) AxSxC Rainow Trout AFA-6 FC-95 4.15 0.2 13.70 0.2 0.32 0.27 1 .o 2.74 Artemia AFA-6 9.69 0.2 0.32 0.62 FC-95 9.23 0.2 1 .o 1.85 Based on application of il,e clIronicacute ratio anc, lBCF t o the acute effect level t h e maximum exposure limit of AFA-6 to protect against chronic sublethal effects in fish would be 0.27 mg/L. Inclusion of additional s a f e t y and rounding down to single decimal place accuracy would suggest a maximum level of 0.2 mg/L of AFA-6 should not be exceeded. This would afford threefold greater protection to invertebrates due to their greater tolerance and would provide tenfold protection against degradation product FC-95 effects due to its lower toxicity. 9328. I 4.6 5.0 FATE IN THE ARCTIC OCEAN ENVIRONMENT 5.1 Ice Island Construction P a n a r c t i c plans to use a maximum of 500 L of AFA-6 in the coming winter exploration season. Use last year amounted to only 120 L due to t e m p e r a t u r e conditions. Typical ice island dimensions are 7 m thick at t h e c e n t r e by 400 m in d i a m e t e r and about 2-3 m thick at t h e edges. This includes about 1 m of natural ice. An additional 0.5 m-of snow would typically accumulate on t h e surface. Total volume of t h e completed island is 690,000,000 L, of which 565,000,000 L is ice laid down in construction. A t an application rate of 10 mglL, the maximum volume of water (ice) containing AFA-6 will b e 50,000,000 L or about 9 % of t h e total volume. This is likely to be applied relatively e a r l y in the season and therefore will be interlayered into the ice below the surface. 5.2 Dilution, Dispersion, and Potential Impacts As indicated in Section 4.3, t h e observed level of no lethal effect for AFA-6 in t h e toxicity tests was 1-2 mg/L. Assuming a 10 mg/L application rate, dilution by a factor of 5-10 will reduce t h e AFA-6 concentration to the non-lethal level. As indicated in Section 5.1, only 9% or less of t h e ice in t h e island will contain AFA-6. Therefore, dilution of 5- to IO-fold (or greater) may occur within t h e ice-island itself as a.result of mixing of clean and "contaminated" melt-waters as they flow through t h e cracks and fissures or pond on the surface. Consequently, little or no lethal effect may be observed under t h e ice. 31 -c I In the summer months, Panarctic has observed t h a t up to 2 m of t h e 7 to 8 m ice thickness melts, and t h e island may drift up to 60 miles depending on wind and currents. As a rule more island movement is observed when melting is extensive. The melt-waters are likely to be less dense than t h e underlying seawater because of higher t e m p e r a t u r e , and the slightly lower salt content from snowmelt and "old" ice. Consequently, horizontal dispersion immediately under the ice will likely be considerably more prevalent than vertical dispersion. Further dilution should occur rapidly as a result of currents and ice island movement. 9328.1 5.1 Based on t h e above, only those a q u a t i c species which have t h e t o p of t h e water column and t h e undersurface of t h e ice as their natural habitat could b e affected. This would include such relatively abundant species as t h e branchipoda, of which Arternia is a member. Free-swimming species such as fish would not be affected, as their natural avoidance responses would tend to keep them o u t of t h e area. .. 9328.1 5.2 6.0 CONCLUSIONS Investigations on the chemical nature and potential biological effects of AFA-6 and i t s starting degradation product FC-95 have led to t h e following conclusions. 0 AFA-6 is slowly biodegradable to its slightly less toxic starting material FC-95. 0 FC-95 is chemically essentially inert, having a hydrocarbon structure similar to te f Ion. 0 Bioassay tests on two s a l t water species (the brine shrimp Artemia, and rainbow . trout) show a c u t e (24hr) toxicity in t h e 5-10 mg/L range. 0 The non-lethal concentration (no mortality of test species) was in t h e range 1-2 mg/L. 0 Dilution by melt-water in t h e ice island itself, may prevent under ice concentrations from exceeding t h e toxic threshold of 1-2 mg/L. 0 The recommended chronic no-effect concentration of AFA-6 was 0.2 mg/L based octanol/water partition coefficients and chronic/acute effect ratios for similar surfactant compounds. 0 Only those aquatic species which have the top of t h e water column and t h e underside of the ice as their natural habitat a r e likely to be affected. 0 Because of the low octanol/water partition coefficients, neither AFA-6 nor FC- 95 will biomagnify significantly. 0 Based on current knowledge t h e r e do not appear to be any other long-term biological effects (e.g., teratogenicity), t; On the basis of the information presented in this report, application of AFA-6 in t h e concentrations and quantities proposed by Panarctic does not appear to represent an - environmental hazard. 9328.1 6.1 REFERENCES - Birch, R.R. 61:Z;pp. (1984). 118iodegradation of 340-343 (February 1984). Non-Ionic SurfactantsP J. Amer. Oil chem. SOC. Cook, K.A. (1979). V e g r a d a t i o n of the Non-Ionic Surfactant Dobanol 45-7 by Activated Sludge." Water Research & pp. 259-266. Karickhoff, S.W., D.S. Brown and T.A. Scott. (1979). "Sorption of Hydrophobic Pollutants on Natural Sediments." Water Res. 13:241-248. Kravetz, L., K.F. Guin, W.T. Shebs, L.S. 5mith, H. Stupel. (1982). YJltimate Biodegradation of a n Alcohol Ethoxylates and a Nonylphenol Ethoxylates Under Realistic Conditions." Soap/Cosmetics/Chemical Specialites from April, 1982, pp. 34-43. Leo, A., Hans&, C., Elkins, 9.(1971). "Partition Coefficients and Their Uses." chem. Rev. _ 71.:(6), p. 525. Mackay, D. (1982). 'Correlation of Bioconcentration Factors." 163274-278. Environ. Sci. Technol. Neely, V.B., D.R. Branson and G.E. Blau (1974). 'Tattition Coefficient to Measure Bioconcentration Potential of Organic Chemicals in Fish." Environ. Sic. Technol. 8: 1113-1 115. Stephanou, E. and W. Giger (1982). llPersistent Organic Chemicals in Sewage Effluents." 2. Quantitative determinations of nonylphenols and nonylphenol ethoxylates by glass capillary gas chromatography. Environ. Sci. Technol. 16:800-805. Swisher, R.D. (1970). 3 u r f a c t a n t Biodegradation" Vol. 3 in Surfactant Science Series, . Marcel Dekker, 1970. Tobin, R.S., F.I. Onuska, D.H.J. Anthony, M.E. Comba (1976a). Won-Ionic Surfactants: - Conventional Biodegradation Products." Ambio 5 pp. 3-31, Test 1976. Methods Do Not Detect Persistent Polyglycol Tobin, R.S., F.I. Onuska, R.C..Rrownlee, D.H.J. Anthony and M.E. Comba (1976b). T h e Application of an Ether Cleavage Technique to a STudy of t h e Biodegradation of a Linear Alcohol Ethoxylate Non-Ionic Surfactant." Water Research 10,pp. 529-535 (1 976). Vanhaecke, P. and C. Persoone. (1981). W e p o r t on an Intercalibration Exercise on a Short-Term Standard Toxicity Test with Artemia Yauplii (Arc-Test).'l Inserm, 106:369-376. Veith, G.D., D.L. *Foe and B.V. Sergstedt. (1979). llMeasuring and Estimating t h e Bioconcentration Factor of Chemicals in Fish." J. Fish. Res. Board Can. 36:1040- 1048. GLOSSARY CTAS: EC: EPS: GC: HPLC: LC-50: MBAS: surf actant: - cobalt thiocyanate a c t i v e substances a non-specific test for all non-ionic surfactants of the polyethxylate type. . electron c a p t u r e d e t e c t o r for gas chromatography Enviromental P r o t e c t i o n Service of Environment Canada. gas chromatography; also called GLC or gas-liquid chromatography high pressure liquid chromatography Yethal concentration" at which 50% of an aquatic species dies when exposed f o r a specified t i m e period (e.g., 24 hours, 96 hours). methylene blue active substances - a non-specific test for anionic surf a t a n t s of t h e alkyl sulphonate type. a cpntractiori of t h e phrase ''surface active agent".