Document 4Jbd43jNaOQ4KeYwM5wNYwdLx
Varying PimvIsUmk'C of Polvrbloriualetl
in Six California Soils
tmclci* Laboratory Conditions
by Yl'TAKA IwATA, W'lU.IAM F-. WtSTLAKE. nd Francis A. Gi'nthkr
Citrm Kcacarch Center onti .1 yriruhtirnt Experiment Stntion
Vnivrrn'ry of Cnlifornia Hivertute, Cnti[. 02502
Widespread reports of polychlorinated biphenyls
(PCBs) as an apparently very stable environmental
contam) nant subject to biological magnification
(RISEBROUGH ct a_l. 1968; HOLMAN et ad. 1969, and
JENSEN et al_. 1969) prompted a study of the relative
behaviors ot Aroclor - 1254 vs.
'-DPT, both at
ten ppm, in several distinct soil types.
Each fortified and control sample, moistened to 40% saturation, was inoculated and kept in an enameled tray loosely covered with a glass plate to retard water evaporation, substrate volatilization, and photodecomposition. Thus, the substrates would be subject to possible microbial degradation, chemical degradation, and adsorption to soil colloids, but not losses due to leaching or soil transport. Properties of the soils used were determined by HEP. MANS ON and BIBLE (1967) (See Table I).
Experimental
Fortification. A hexane solution (one mg/ml) of either 30 mg of d,J2'~DDT or Aroclor 1254 was added to 300 g of air-dried soil, which was then mixed and stirred until the solvent evaporated. This fortified subsample was then added to sufficiert unfortified soil to yield a total weight equivalent to three kg of oven-dry soil. Since the soils bad been stored for about four years in an air-dried condition, three g of 2-mm sieved inoculant soil collected from the local mountains was added to insure the presence of microorganisms. The combined sample was mixed for one hour in a Twin Shell Dry Blender, placed in a 10-1/8 in x 16-1/8 in x 2-3/8 in enameled tray, distilled water was added to adjust the moisture content to 40% of saturation, and the tray was covered loosely with a glass plate.
Registered Trademark, Monsanto Company, St. Louis, Missouri, USA.
204
RattfHr* Wl, , N. 4
& T*lfo1ocv. I97J kj 54rrf>(tr-Vrrl^c Sew York Inc.
DSW 032724
yls
i ve
tion, Id
)
dded and f ied d kcr d' hroe c Dt
Condjtjons. Trnyn wore kept on shelves in a plant growth chamber at 30.) 1C ant about '10% relative humidity and continuously illuminated with "daylight" fluorescent tubes. Dvnrorative Usres were replaced witn distilled water three times a month to maintain a constant soil moisture level. Each sample was handmixed at this, time, and tray positions in the chamber were interchanged.
Extraction. At each sampling interval, three subsainples oi 20 a of oven-dry soil each were extracted without, {trying (KAMA ejt aj_ 1969) . To each subsainple in a 4-oz screw-cap bottle was added 40 ml of a 1:1 hexane-acctone mixture. The bottle, closed with aluminum foil and a teflon-lined cap, was iihaV.cn mechanically for one hour. The shaking was repeated twj.ee, using 40 ml of fresh solvent and ten min each time. Subsample supernatants vjere combined and passed through 25 g of anhydrous Na2S04- Soil, bottle, and Nai>S04 were rinsed with 40 ml more of hexane.
This total extract was reduced to about two ml using initially a Kuderna-Danish apparatus and finally a gentle stream of air. The extractives were transferred to a 160 mm x 90 mm ID column containing about 1.6 g of deactivated Florisil (15 g water/100 g) using three successive 5-ml portions of hexane to rinse the tube and elute the column. The volume of the eluate was adjusted to 10.0 ml, and the solution was analyzed by miccocoulometric glc.
Each Laveen loamy sand extract was further cleaned up for analysis by electron capture by placing it on a 30 cm x 2.5 cm OD glass column packed with a ten cm height of 60-100 mesh Florisil and prewashed with 50 ml of hexane, and then eluting with 200 ml of naDegrade hexane (REYNOLDS, 1969). The eluate was reduced to 50 ml and an aliquot analyzed.
Analsfsis. A Dohrmann Instruments Model S-200 furnace unit and Model C-200 microcoulometer with a T-300S titration cell were used for the microcoul.omotr ic determinations; operation was in Mode IT at 200 ohir.s . Gas flow rates were 100 ml/min for nitrogen (carrier) and 100 ml./min for oxygen. The boros i lien to column via.1; 6 ft x 4 mm ID packed with a 1 :1 mixture of 1.074 DC-200 and 15'/ QF-1, each coated on Gas-Chroin Q, 60-HO mesh. DDT quantitation was by peak area using internal, standards, and Aroclor composition was estimated from peak heights.
20S
osw 032725
The Aroclor 1254 chroma log rants shown in Figures lB~lG each represent:; the1 extractives from 24 mg of noil. No correction for background, as determined from rnfortified samples, was necessary for either substrate.
Data using an electron capture detector (tritium source) were obtained v/ith a Vnrian model 1700 gas chromatograph fitted with a 5-j ft x 2 mm ID borosilicate column packed with a 1:1 mixture of 10% DC-200 and 15% OF-1 . The carrier gar. was helium, 30 ml/min; injection, column, and detectortemperatures were 230, 100, and 205C, respectively. A standard curve (nonlinear) was used for quantitation. No corrections for background, as determined by an unfortified sample, was required.
Re sul 11' and Discussion
Figure 1A is the chromatogram obtained with a microcouloinetric detector for 240 ng cf Aroclor J 254 . The major peaks, disregarding shoulders due to poor resolution, are numbered 1-7 for discussion purposes. Figures 1B-1G are micrccoulomctric chromatograms of extracts of soil moisture content of 40-45% saturation for 4-12 months. Each chromatogram represents the extractives obtained from 24 mg of soil. Thus, Fig. 1A represents a 100?(- recovery of Aroclor 12 54 from 24 mg of soil fortified at ten ppm.
About 95% of the Aroclor 1254 added to Windy loam soil was recovered after one year had elapsed (Fig. IB); the relative peak heights of the various compounds present remained unchanged demonstrating no preferential losses. Results for the Santa Lucia silt loam were identical with these. Both soils have high organic matter (Table I).
Figure 1C represents extractives from the Linne clay; after one year it still retained all the peaks but pref.erent ia 1 looses of material were evident. While approximately 90/, of the material comprising peak 7 was recovered, only about 40% was recovered from peak 1. Losses were most apparent for the first two major peaks. Results for Madera sandy loam were similar to those for J.inno clay, although the soil characteristics were quite dissimilar (Table I). Figure ID represent a .``ocho silt, loam soil which differs from the clay and sandy loam in that, except for peaks 0 and 7. 2i>-3u% losses, of material were evident. The result was corrected for an 80% recovery of Aroclor 1254 from the initial sample.
206
DSW 032726
STLCOPCB4016688
(3 1/7
Xr <n Q
< o O
m
rt- n n v
(-> 31
1
o
D
\
KJ
o s: o M si M
TABLE I Chemical and physical data of some California soils*.
Soil type
Organic matter (%)
pH
Soil texture % sand % silt % clay
Laveen loamy sand San Bernardino County
Windy loam Amador County
Madera sandy loam Coachella Valley
Santa Lucia silt loam Santa Barbara County
Moclio silt loam Santa Barbara County
Linne clay . Santa Barbara County
0.1 10.8
1.4 19.5
1.9 3.3
8.7 6.0 6.7 5.6 7.9 7.5
94 51 60 34 18 37
1 40 28 42 58 25
*H. P. HEBMANSON and J. M. BIBLE, unpublished data {1967}
5 9 12 24 24 38
Saturation percentage
2L 54 27 94 45 48
STLCOPCB4016689
Fig. l--Microcoulometric GLC chromatograms. (A) represents 240 ng of Aroclor 1254. Chromatograms B through G each represents the injection of the extractives from 24 mg of soil fortified at ten ppm and stored for 4-12 months at 30'C and a soi] moisture of 40-4596 of saturation. (B) Windy loam after 12 months. Note essential superimposr^bility with A. (C) Linne clay after 12 months. Note relative decreased peak heights in earlier eluting peaks. (D) Mocho silt loam after 12 months. Overall recovery from this soil is about 20% less than from the others. Laveen loamy sand after (E) four, (F) eight, and (G) 12 months.
208
DSW 032728
STLCOPCB4016690
lH __ _
.
10
grams
ujjm isterc 2 . <C) mod > silt ; soil onrny
The groatort change in Avoclor 1.254 composition in the soil was with Laveon loamy sand (figs. 1E-1G) . After one yeax' the characteristic Aroclor 1254 elution pattern wan no longer readily recognizable. Although about 75VS of peak 7 remarried, almost all of the material comprising the first three peaks was gone, and losses; of material from the next three peaks were evident. The Laveen loamy sand samples were reanalyzed using an EC detector. Better resolution of the peaks was obtained as smaller quantities of material were injected, and the Stability of the EC detector greatly facilitated quantitation. The results are given in Table II.
TABLE II
Relative per cc-nt recovery of Aroclor 1254 from fortified Laveon loamy sand.
Mon t)i s Peak 1
Peak 2
Peak 3
Peak 5
Peak 7
0
100
100
100
100
100
2 48 75 81 102 96
4 17 48 60 89 89
8 0 20 32 04 86
12 0 13 23 66 02
BAGLEY et al. (1970), BIROS et al,. (1970) , and STALLING and 11UCKINS (1971) showed with glc- mass spectra data that Aroclor 1254 con sisted predominantly of tetra-, penta-, and hexacnlorobiphenyl isomers, and that the less chlorinated isomers eluted faster from the glc column. Thus, the material remaining in our Laveen loamy sand after one year consisted of mainly penta- and hexachlorobi.phenyl isomers.
DDT, added separately to the same soils in parallel for comparison purposes, gave the results shewn in Table III. Combined DDT and DDE remaining after one year accounted for over 70% of the DDT applied in each soil. With the possible exception of Aroclor 1254 added to loamy sand, the PCB and DDT-DDE residues were equally persistent in soil under our conditions.
NIMMO et n_l . (1971) demonstrated experimentally that Aroclor 1254 can enter the estuarine food chain from contaminated sediments either through organisms ingesting the sediment or absorbing the chemical
209
DSW 032729
TABLE III Recovery of residues from soils fortified at 10 ppm with DDT*.
Soil type
Soil moisture (% of saturation)
0 DDT
Santa Lucia silt loam
Windy loam
Madera sandv loam
Laveen loamy sand
Mocho silt loam
'
Linne clay
*ppm oven-dry weight
42 48 41 39 40 44
9.6 9.4 10.6 10.0 8.1 9.7
Months after fortification 1 4 8 8 12
DDT
DDT
PPM DDT
DDE
DDT
12 DDE
9.6 9.1 9.2
9.3 7.5 9.7
3.6 8.9 7.8 8.0 6.3 8.4
9.2 9.0 7.8 7.9 5.4 7.3
1.2 Trace 2.4 1.9
8.3 8.3 6.7 7.3 5.1 5.8
1.2 Trace
2.1
from the water; the ratio of individual Aroclor icomers maintained integrity in the sediment and tissues of test: animals. invecn loamy sand clearly showed the composition of the persisting PCD residues in this soil was altered with time. If volatilization is mainly responsible for the observed loss of lesser chlorinated PCBs, these materials may, through aerial fallout, become unidentified contaminants far from the source of origin. The composition of the residues remaining in the soil and those obtained elsewhere through aerial fallout will be quite different; biological magnification of these residues clearly will not yield a recognizable glc pattern. Thus, suspected environmental pesticide residues should be subjected to careful evaluation as they could be PCBs of altered composition.
Ackncwledgmcnt
This work was supported by Regional Research Project W-45.
BAGLEY. G. E., REICHEL, W. L. , and CROMARTIE, E., JAOAC 51, 251 (1970).
BIROS, F. J., WALKER, A. C., and KEDBERY, A.. Bull. Environ. Contam. Toxicol. !5, 317 (1970).
HERMANSON, H. P. and RIBLE, J. M., unpublished data (1967).
JENSEN, S., JOHNELS, A. G., OLSSON, M.. and OTTERLIND, G., Nature 22.4, 247 (1909).
KOLMAN, J. H., TEN NOEVER DE BRAUW, M. C., and DEVOS, R. H., Nature 221., 1126 (1969).
NIMMO, D. R., WILSON, P. D., BLACKMAN, R. R., and WILSON, A. J., JR., Nature 231, 50 (1971).
REYNOLDS, L. M., Bull. Environ. Contam. Toxicol. 4, 128 (1969).
RISEBROUGH, R. W., RIECI1E, P. , PEAKALL, D. B., HERMAN, S. G., and KIRVEN, M. N., Nature 220. 1098 (1968).
SAHA, J. G., BHAVARAJU, B., LEE. Y. W., and RANDELL, R. L., J. Agr. Food Chem. 17. 077 (1969).
STALLING, D. L. , and HUCKINS, J. N., JAOAC 54., 802 (1971).
211
DSW 032731
STLCOPCB4016693