Document oMbJdqybgRJgX6vnK1ypqZe5X

' baTmly iquota. . comparable thion, van in .lant to 2.5 a a non- diffuse at from an aampiaa. 50 pg rathion and xtracta. ctad or ar to thoaa nd did not a peaticidea El mining zinphos- t elaborate f acreen irbophono- J24 (1964). . 1 (1966). 105 (1966). 17 (1966). >red Prod v. i and , and . , and hem. i | i Polychlorinated Biphenyls (PCB) Solubilized in /ater by Nonionic Surfactan for Studies of Toxity to Aquatic Animals by V. Zrrxo Fuharim Rateanh Board of Canada, Biological Station Si. Andrew, N.B. PCS have baan detected in Marine anlaali In Sweden. England, USA, and The Netherlands (1,2,3). Knowledge about the toxicity of PCS to wildlife It Halted. PCS are Inducers of hepatic enzymes that hydroxylate steroids In birds (1) and a latent toxicity of PCB to grasshoppers (Chorthippua brunnaut) at ecdysls has been suggested'(4). No data on the toxicity of PCB to aquatic animals have been published. The solubility of PCB In water Is very low and difficulties are encountered In dosing PCB In experiments with aquatic animals. PCB-ln-water emulsions, prepared by dilution of PCB, dissolved In an organic solvent, with water or by mechanical dis persing of PCB In water, coagulate on standing and contain PCB particles varying greatly In size. The results of toxicity tests using such PCB-ln-water emulsions are difficult to reproduce and may not Indicate the true toxicity of PCB. A PCB-ln-water emulsion may display different degrees of toxicity or appear nontoxic depending on the size of PCB particles In the emulsion. A more stable and homogeneous PCB-lnwater system was therefore sought. This paper describes the solubilization of PCB In water by nonionic surfac tants, the determination of concentration of the solubilized PCB by UV spectrophotometry and fluorescence, some properties of aqueous PCB solutions, and preliminary results on the toxicity of PCB to Atlantic salmon (Salmo aalar) parr*. EXPERIMENTAL Commercial PCB preparations containing 21 and 54% chlorine (Aroclor 1221 and Aroclor 1254) were used. Ethylene oxide adduct (9 ethylene oxide units) of laurlc acid, ethylene oxide adduct (8 ethylene oxide units) of 2,6 ,8-trimethyl-4-nonanol (both preparations from the Surfactant Kit, Chem Service, Media, Pa.), and a com mercial emulsifier preparation Corexlt 7664 (Enjay Chemical Co.) were used to solubilize PCB In water. Typically, Aroclor (500 mg) was dissolved In Corexlt 7664 (9500 mg) and the volume was adjusted to 50 ml with *A more detailed presentation of all results Is given In (5). Toxicity of PCB, solubilized by Corexlt 7664, to Gammarua oaaanieua has been recently described (6). Bulletin *f Environmental Contamination ft Tuiieolojry, Yoi. &, No. 3, )970, publiaiied b) ^pringer-VerUg New York Inc. mew^rm^ngmmwmram n wwtiemngrma^amiP--WW DSW 027905 STLCOPCB4011867 distilled water. St. Andrews tap or sea water was used for further dilutions. UV and fluorescence spectra were recorded on a Beckman DK-2A and on a Perkin-Elmer MPF-2A instrument respectively. Gas chromatography was carried out on a Varian GOOD instrument equipped with a 250 mC tritium electron capture detector, using a 5 ft, 1/8 in. glass column, operated at 1 300C and containing 42 SE-30 on Chromosorb W, 100-120 mesh. To determine the solubility of PCB in water, Aroclor (1.5 ml) was added to fresh or sea water (50 ml) In a Waring blendor and the mixture was homogenized for 10 min at room temperature. The resulting emulsion was centrifuged for 30 min at 30,000 g (5C) and the con centration of Aroclor in the clear supernatant was determined by spectrophotometry. A part of the super natant was extracted with hexane and the extract was analyzed by gas chromatography. Toxicity tests with Atlantic salmon parr (average weight 4.3 g, average length 7.2 cm) were carried out at 15C in 3 liter Erlenmeyer flasks con taining 2 liters of Aroclor-Corex1t solutions. Each flask was aerated by bubbling air through a glass tube immersed in the solution. Two flasks were used at each Aroclor concentration: one containing 2 fish, the other without fish, serving as a blank. Tests were terminated after 192 hours. At this time fish were still alive in solutions of Aroclor 1254 and Aroclor 1221, initial concentration 0.9 and 2.5 and 0.9 mg/1, respectively (Fig. 1,2). Samples (10 ml) were withdrawn from the flasks at different times and the concentration of Aroclor was determined by spectro photometry or by fluorescence. RESULTS Solubility of PCB in water. A fractionation according to the degree of chlorination takes place on breaking Aroclor-in-water emulsions (no surfactant added) by centrifugation. The dissolved fraction is richer In lower chlorinated biphenyls than the original pre paration. The quantitative composition of Aroclor 1254 and of the mixture of biphenyls recovered from the supernatant are given in Table 1. The difference between the original Aroclor 1254 and the mixture of biphenyls In the soluble fraction causes some uncer tainty in the determination of the solubility of Aroclors in water. The solubility of Aroclor 1254 Is 2-3 mg/1 In fresh and 1-1.5 mg/1 in sea water. This batch of Aroclor 1254 contains a relatively high amount of lower chlorinated biphenyls. The solubility of 280 T** DSW 027906 J another ba of lower cl both fresh 1221 Is 5.1 respective 1 Quantita and of t i Aroclor 12 Sol.fracti Aroclor 12 Sol.fracti Solubiliza oxide addu ta1n1ng is with water be used in dimethyl s In ethylen (weight ra (0.6 g/gPC solution o 1:20) y1e1 when d 11ut 1sopropano (weight ra with water Determina t of Aroclor maximum at ethanol : steadily t Aroclor 12 3.97 (in e Beer's law tration ra of Aroclor c - 35.1(Ai cell at 247 excitation fluorescent for Aroclo f1uorescen tration up limits are Aroclor 12 STLCOPCB4011868 was used irded on istrument iut on a ;ritlurn i. glass 30 on i water, r (50 ml) ized for sion was e conwas i supert was iarr >ere s con- Each ss tube at each the were were ocl or mg/1 , d the tro- cordlng eaking ) by er In re1 or from erence re of ncerf 254 Is This h amount of ano\...er batch of Aroclor 1 254 , containing only traces of lower chlorinated biphenyls, is 0,3-0.5 mg/1 ln both fresh and sea water. The solubility of Aroclor 1221 Is 5.0 and 3.8 mg/1 in fresh and sea water, respectively. TABLE 1 Quantitative composition of a batch of Aroclor 1254 and of its water-soluble fraction. Relative peak area in order of increasing retention time Aroclor 1254 0 0.2 0.1 0.1 3 8 1.2 1 .3 10.3 Sol.fraction 0.2 0.4 0.2 0.2 1 3 0.6 0.7 8.6 Aroclor 1254 16.0 22.0 15.7 13.6 11 1 2.7 1 .9 Sol.fraction 14.4 17.5 18.7 16.5 15 7 2.7 2.3 Solubilization of PCB. A solution of PCB in ethylene oxide adduct of 1auric acid (weight ratio 1:20), con taining isopropanol (0.6 g/gPCB), yields on dilution with water a clear solution. Acetone (1.5 g/gPC8) may be used Instead of isopropanol but the same amount of dimethyl sulfoxide Is not effective. A solution of PCB In ethylene oxide adduct of 2 ,6,8-trimethyl-4-nonanol (weight ratio 1:10) similarly requires Isopropanol (0.6 g/gPCB) to yield a clear aqueous solution. A solution of PCB In the same surfactant (weight ratio 1:20) yields, even without isopropanol, a clear solution when diluted with water. Corexlt 7664 contains Isopropanol and a solution of PCB in Corexlt 7664 (weight ratio 1:19) gives a clear solution on dilution with water. Determination of PCB solubilized in water. UV spectrum of Aroclor 1221 , solubilized by Corexit / 664 , has a maximum at 247 nm, log Z 4.18 (Aroclor 1221 in ethanol: 247 nm, log I * 4.23). Absorbance increases steadily towards shorter wavelengths in solutions of Aroclor 1254 containing Corexlt 7664; at 250 nm log Z 3.97 (in ethanol log Z = 4.00 at 250 nm). LambertBeer's law Is obeyed by both Aroclors in the concen tration range examined (0.5 -- 10 mg/11; concentration of Aroclor 1221, c (mg/1) * 12.6(A247); of Aroclor 1254, c 35.1(A250); (A247), (A250) absorbance in 1 cm cell at 247 and 250 nm, respectively. Fluorescence excitation maximum of both Aroclors is at 270 nm, fluorescence emission maximum Is at 316 and 320 nm for Aroclor 1221 and 1254, respectively. Relative fluorescence at 320 nm is a linear function of concen tration up to 0.5 and 4.0 mg/1 and the detection limits are approximately 0.025 and 0.25 mg/1 for Aroclor 1221 and 1254, respectively. 281 DSW 027907 STLCOPCB4011869 Variations occur between different batches of Aroclors with the sane declared chlorine content; both spectrophotometry and fluorescence require calibration using the particular batch of Aroclor which Is to be determined. ** Adsorption of PCB. In water solubilized PCS are strongly adsorbed by plastic surfaces such as poly ethylene sheets and tubes. Toxicity of PCB to Atlantic salmon parr. Aroclors solubilized by Corexlt 7664 were used because of the known low toxicity of the latter*. The experiments were carried out primarily to determine the behaviour of solubilized PCB In toxicity tests with aquatic * animals. Two fish were tested at each concentration, the toxicity data are therefore only preliminary. ( merits Is p flasks wit Oecrease 1 by fish Is concentrat salmon par concentrat the experl by fish Is tratlons t cond1tlon: concentrat strated 1r Aroclor II i According ; volatile. cold room I Figure 1. Concentration of Aroclor 1254 In the bioassay. *No mortality of Atlantic salmon parr exposed to 500 mg/1 of Corexlt 76(4 was observed In 96 hours (7). DSU 027908 r water wet In water procedun STLCOPCB4011870 satches of snt; both l1bration i to be ir poly- :1 ors of the ments haviour tic ration, ry. Concentration of Aroclor 1,^4 in the experi ments Is presented in Figure 1. The concentration in flasks without fish remained practically constant. Decrease in concentration caused by uptake of Aroclor by fish is noticeable with the exception of the lowest concentration. Aroclor 1254 is lethal to Atlantic salmon parr at concentrations higher than 2 mg/1. The concentration of Aroclor 1221 decreased steadily during the experiment (Fig. 2). The uptake of Aroclor 1221 by fish is pronounced at the two Intermediate concen trations tested. Aroclor 1221 is, under the described conditions, lethal to Atlantic salmon parr at initial concentrations higher than 2 mg/1. it has been demon strated in separate experiments that the losses of Aroclor 1221 are caused by its volatility on aeration. According to Figure 1 Aroclor 1254 appears to be non volatile. However, PCB were detected in the air of a cold room (5C) used for tests with Aroclor 1254. 0 500 (7). Figure 2. Concentration of Aroclor 1221 in the bioassay. DISCUSSION The determined levels of solubility of PCB in water were reached by very vigorous dispersing of PCB In water and it would be difficult to scale up this procedure to prepare a large volume of the PCB solution. j 283 DSW 027909 STLCOPCB4011871 As Indicated by the toxicity data, lethal levels Mould be reached only In the case of Aroclor 1221. Because of the solubility differences between higher and lower chlorinated biphenyls. Mainly the toxicity of the latter would be tested even In solutions prepared by dispersing Aroclors with higher chlorine content, In water. The solubilization of PCB In water by relatively nontoxic nonionic surfactants Is a reproducible and convenient Method for dosing PCB in experiments with aquatic animals. The whole PCB preparations are solubilized and the stock solution of the solubi lized PCB May be used In running-water toxicity tests. Concentration of the solubilized PCB can be readily monitored by spectrophotonetry or by fluorescence at a rate of 20-30 analyses per hour; a continuous _ monitoring Is also possible. From the point of view of optical properties there Is no difference between PCB solubilized In water by nonionic surfactants and true solutions of PCB In ethanol. The mechanism of the solubilization Is as yet unknown. The presence of Corexlt 7664 may enhance the toxicity of PCB by providing better contact between fish and PCB than a mechanical dispersion of PCB In water or a dilution of PCB dissolved In an organic solvent. It Is not known whether Corexlt 7664, at the concentrations used, affects the permeability of fish body surface to PCB. i/. S(G The preliminary tests Indicate that PCB may be less toxic to Atlantic salmon parr than chlorinated hydrocarbon pesticides. Inhalation of PCB represents a health n hazard (8). 8ecause of the volatility of PCB, experl- ' ments must be carried out In well ventilated facilities. REFERENCES 1. R. W. RISEBROUGH, P. RIECHE, 0. B. PEAKALL, S. G. HERMAN and M. N. KIRVEN, Nature 220, 1098 (1968). 2. J. H. K0LMAN, M. C. TEN N0EVER 0E BRAUH and R. H. 0E VOS, Nature 221, 1126 (1969). 3. S. JENSEN, A. 6. JOHNELS, M. 0LSS0N and G. 0TTERLIND, Nature 224. 247 (1969). 4. F. MORIARTY, Entomol. Exp. Appl. 12, 206 (1969); CA 71. 890S1 (1969). i 1 I l f 5. V. 21 MS Re deter and s resul 6. 0. J. Canad bl phe oatart 7. V. ZI Board oil. Xzlt 8. HAN0B Edlto Clave m STLCOPCB4011872 els would Se cause nd lower the red by nt, In I rel*11vely le and its with pre fib 1 ty tests, tad! ly nee at is `opertles I In ins of zation ance t Ion of an xit PCB experl acllitles. 1 969 ); 5. V. ZITKO, Fisheries Research Board of Canada MS Report No. 1083, Polychlorinated biphenyls: determination by optical methods, solubility and solubilization in water, preliminary results on toxicity to salmon (April 1970). 6. D. J. WILDISH, Fisheries Research Board of Canada MS Report No. 1084, Polychlorinated biphenyls in sea water: Bioassay of Gammarua oeeanicua (May 1970). 7. V. ZITKO and W. G. CARSON, Fisheries Research Board of Canada MS Report No. 1043, Bunker C oil. Dispersibility in water by Corexit and Xzlt at different temperatures (October 1969). 8. HANDBOOK OF ANALYTICAL TOXICOLOGY, I. Sunshine, Editor, p 614 (1969), The Chemical Rubber Co., Cleveland. 235 DSW 027911 STLCOPCB4011873