A Physical Concept of Soil-Water Equilibria forNonionic Organic Compounds

Abstract. Soil-water equilibrium data suggest that the transfer of nonionic chem-icals from water to soil may be described in terms ofa hypothesis ofsolute partition-ing in the soil organic matter. This concept allows estimation of soil-water distribu-tion coefficients eitherfrom solvent-water partition coefficients or aqueous solubili-ties.

Earlier studies of the uptake by soil ofnonionic organic compounds from waterhave indicated that soil organic matter isthe principal adsorbent (1-3), and it hasbeen speculated that the high surfacearea of this organic matter is an impor-tant factor (4). We here present data forthe hypothesis that sorption by soil or-ganic matter is essentially a partitioningprocess. The supporting data consist of(i) our own determination of the iso-therms of seven organic compounds thatshow no indication of curvature even atconcentrations approaching saturationand (ii) extensive data (those of ours andothers) over seven orders of magnitudeof solubilities and with various soils, inwhich the partition coefficients are wellcorrelated by solubilities and the organiccontents of the soils.

Figure 1 shows the 20°C equilibriumisotherms of 1,2-dichloroethane, 1,2-di-chloropropane, 1,2-dibromoethane, 1,1,-2,2-tetrachloroethane, 1,1, 1-trichloro-ethane, tetrachloroethene, and 1,2-di-chlorobenzene from water, at pH 6.8,onto a Willamette silt loam (1.6 percentorganic matter, 26 percent clay, 3.3 per-cent sand, and 69 percent silt). Iso-therms plotted in micrograms of organicper gram of soil versus Ce [in parts permillion (ppm)] are essentially linear;moreover, the highest Ce is about 0.30 to0.50 of the aqueous solubility (S) for thefirst five compounds and 0.75 to 0.95 Sfor the last two compounds.

In addition to the fact that the iso-therms show no indication of curvatureeven at high relative concentrations, thesoil-water distribution coefficients (asdetermined from the slopes of the iso-therms) appear to be inversely propor-tional to the corresponding solubilities(Fig. 2). The isotherm data for poly-chlorinated biphenyl (PCB) isomersshow similar results (5). These findings,together with the lack of isotherm curva-ture, are consistent with the idea that theuptake of neutral organic chemicals bysoil is essentially a process of partition-ing (dissolution) rather than physical ad-sorption (6).

Since the distribution coefficient of anorganic chemical is known to dependlargely on the content of organic matter

SCIENCE, VOL. 206, 16 NOVEMBER 1979

(3), the above reasoning leads to the fur-ther postulate that the uptake of neutralorganics by soil is due mainly to parti-tioning in the soil organic matter as sug-gested by Swoboda and Thomas (7) forparathion in soil. Hartley (8) speculatedthat a solvent action of the oily constitu-ents of the organic matter might be re-sponsible for soil uptake. If we assume apartition concept, the reported correla-tion of soil organics-water distribution

-1200._

0

0

Etoo

c& 800

E

.0

_E

= 400

,

10

a.

6

5

4

co

3

2

-3

coefficients (G) with octanol-water parti-tion coefficients (K) for nonionic chem-icals (9) is presumably a result of similarequilibrium processes.The partitioning hypothesis, more-

over, is consistent with the observed en-thalpies. Adsorption of a trace com-ponent, as on activated carbon (10), re-quires a relatively high exothermic ARto balance the resultant decrease in en-tropy. By contrast, partitioning of a sol-ute may not be exothermic and wouldhave a nearly constant Al since

AH = AH -AH, (1)

where Allo and All, are the molar heatsof solution in the organic phase (in thiscase, soil organics) and in water, respec-tively. Low water solubilities of organiccompounds are usually associated with

2400400 800 1200 1600 2000

Equilibrium concentration, Ce (ppmn)

Fig. 1. Soil-water equilibrium isotherms at 20°C.

02,4,5,2'.4',5'-PCBmN DDT

-M 25'5'PCB

_ S (ppmn) G \ L ,4-C

B 3570 27C 3520 36D 3230 46 K, BHC IParathionF 1230(1) 75(1)G 148 180H 200 2101 24(16) 1160(7,12,14) H TetrachloroetheneJ 7.8(25)H17) 1730411) 6 1,2-Dichlo rob enzene 1,1 ,1TrichloroethaneK 5(18) 29001411)L 0.64(19) 8000(5) F 1,2-Brorno-3-chloroPropane D 1 1,2,2-Tetra-M 0.027(19) 47000(5) C 1,2-Dibromoethnen chloroethaneN 0.004(25'(20) 140000521) B 1,2-Dichloropropane

0.000?519) 220000(5) 1,2 hloroethaneI I I A,2Dclrthn-1 0 4 6

Log S ("mole/liter)

Fig. 2. Soil organics-water distribution coefficients (G) plotted as a function of the aqueoussolubilities (S) of selected nonionic compounds. Reference numbers are given in parenthesesafter the S and G values.

00368075179/1116-03100.50/0 Copyright 0 1979 AAAS I 831

on

Nov

embe

r 11

, 201

4w

ww

.sci

ence

mag

.org

Dow

nloa

ded

from

o

n N

ovem

ber

11, 2

014

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fr

om

large (positive) Alf. The correspondingAH0 terms should be much smaller, re-flecting the improved compatibility in anorganic phase. For compounds of lowwater soluoility, Al would be somewhatless negative (exothermic) than -Aff.This reaidily accounts for the Al valuesfor 8- and y-BHC (benzene hexachlor-ide) (11) and parathion (12) in soil-waterequilibria; Aft will be small for com-pounds with low AHR and may even bepositive if Af,, is negative. In our data,AHl for 1,2-dichlorobenzene is practical-ly zero as a result of the low AR,,(S = 133 ppm at 3.5°C; 148 ppm at 20C),and All for 1,1, 1-trichloroethane is posi-tive (> -H) because of the negativeAH, value for this compound (S = 1790ppm at 3.5C; 1360 ppm at 20°C). Theseresults are not compatible with an ad-sorption model.Consider now the relation between G

and S. Since S is a good estimator of theorganic-water partition coefficients forslightly soluble organic compounds (13),we may expect to find a correlation be-tween log G and log S, similar to that be-tween logG and log K. The G values cal-culated on the basis of the individual soilorganic matter contents for the sevencompounds from this work and othernonionic compounds from the literatureare shown in Fig. 2. The regression equa-tion is

may show substantial adsorptionthrough ion-exchange even though parti-tioning is not favored by their high watersolubilities. Surface binding (for ex-ample, hydrogen bonding) may poten-tially be a factor for highly polar com-pounds. The contribution of these effectsneeds to be considered separately.

It thus appears that the uptake of neu-tral chemicals by soils is consistent withthe hypothesis of solute partitioning tothe organic content of the soil. The re-sulting relationship between log G andlog S provides a means for estimating thesoil-water distribution.

CARY T. CHIOULouis J. PETERS

VIRGIL H. FREEDAgricultural Chemistry Department,Oregon State University,Corvallis 97331

References and Notes1. C. A. I. Goring, Annu. Rev. Phytopathol. 5, 285

(1967).2. J. W. Hamaker, in Environmental Dynamics of

Pesticides, R. Haque and V. H. Freed, Eds.(Plenum, New York, 1975), p. 115; G. W. Baileyand J. L. White, J. Agric. Food Chem. 12, 324(1964); M. Leistra, ibid. 18, 1124 (1970); J. W.Hamaker and J. M. Thompson, in OrganicChemicals in the Soil Environment, C. A. I.Goring and J. W. Hamaker, Eds. (Dekker, NewYork, 1972), p. 51.

3. S. M. Lambert, J. Agric. Food Chem. 16, 340(1968).

4. R. Haque, in Environmental Dynamics ofPesti-cides, R. Haque and V. H. Freed, Eds. (Plenum,New York, 1975), p. 97.

5. R. Haque and D. Schmedding, J. Environ. Sci.Health Bll (No. 2), 129 (1976).

6. M. Manes and L. J. E. Hofer, J. Phys. Chem.73, 584 (1969).

7. A. R. Swoboda and G. W. Thomas, J. Agric.Food Chem. 16, 923 (1968).

8. G. S. Hartley, in Herbicides and the Soil, E. K.Woodford and G. R. Sagar, Eds. (Blackwell,Oxford, 1960), p. 63.

9. G. G. Briggs, Proc. 7th Br. Insecticide Fungi-cide Con]. (1973), p. 83.

10. C. T. Chiou and M. Manes, J. Phys. Chem. 78,622 (1974).

11. A. C. Mills and J. W. Biggar, Soil Sci. Soc. Am.Proc. 33, 210 (1969).

12. B. Yaron and S. Saltzman, ibid. 36, 583 (1972).13. C. T. Chiou, V. H. Freed, D. W. Schmedding,

R. L. Kohnert, Environ. Sci. Technol. 11, 475(1977).

14. J. A. Leenheer and J. L. Ahlrichs, Soil Sci. Soc.Am. Proc. 35, 700 (1971).

15. S. Saltzman and B. Yaron, in Pesticide Chemis-try, A. S. Tahori, Ed. (London and Breach,New York, 1971), vol. 6, p. 87.

16. N. N. Mclnikov, Residue Rev. 1971, 36 (1971).17. L. Weil, G. Dure, K. E. Quentin,Z. WasserAb-

wasser Forsch. 7, 169 (1974).18. R. E. Slade, Chem. Ind. (N.Y.) 40, 314 (1945).19. R. Haque and D. Schmedding, Bull. Environ.

Contam. Toxicol. 14, 13 (1975).20. J. W. Biggar, G. R. Dutt, R. L. Riggs, ibid. 2, 90

(1967).21. Y. 0. Shin, J. J. Chodan, A. R. Wolcott, J.

Agric. Food Chem. 18, 1129 (1970).22. We thank Prof. M. Manes and Prof. I. J. Tinsley

for valuable discussion and J. L. Bronson fortechnical assistance. Work was supported byNSF/RANN grant AEN-76-17700 and NIH-ESgrants 00040 and 00210. Published with the ap-proval of the Oregon State Agricultural Experi-ment Station as Technical Paper 5062.

15 January 1979; revised 21 June 1979

log G = 4.040 0.038) -

0.557 (± 0.012) log S (2)

with r2 = 0.988 and n = 15 (where G isdimensionless and S is in micromoles perliter) and covers more than seven ordersof magnitude in S and four orders ofmagnitude in G.The close fit suggests that the makeup

of organic matter in soil is not critical indetermining log G values for neutralchemicals (14). Moreover, it impliesthat, although the uptake by other soilconstituents may involve other mecha-nisms, their contribution will be relative-ly small. For instance, the uptake byclays is considerably lower than that oforganic materials (11, 15).

Since a chemical's partition value de-pends on its relative solubilities in thetwo phases, the solution pH may be ex-pected to have a strong effect on the par-tition coefficients of organic acids andbases. At high pH, the dissociated anionof an organic acid should be poorly dis-tributed in soil organics because of itshigh water solubility and possibly the re-pulsion by the surface negative charge ofthe organic matter. At low pH, certaincationic species (for example, paraquat)

Widmanstaetten Patterns in Josephinite,a Metal-Bearing Terrestrial Rock

Abstract. Widmanstaetten patterns have been found in several specimens ofjo-sephinite, a complex, metamorphosed, metal-bearing rock from placers on serpen-tinized peridotite in southwest Oregon. The patterns, in interior less-altered regionsofthe specimens, are typical ofexsolution textures produced during slow cooling ofahomogeneous metal. The bulk composition of the metal phases indicates that thehomogeneous metal must have existed at temperatures above 5000C. JosephiniteWidmanstaetten patterns are the first known in terrestrial rock. We interpret them as

further evidence that josephinite was derivedfrom the mantle.

Josephinite, which is variable in com-position, contains alloys of Ni, Fe, andCo. Other phases include garnet, sul-fides, and arsenides; alteration phases in-clude magnetite and serpentine. Jose-phinite is found in placers of streamswhich originate on and traverse the Jose-phine Peridotite, which is the depletedmantle of an obducted ophiolite complex(1). The josephinite specimen describedin this report consists primarily of Fe,Ni, and Co. The specimen, 3 cm long and2 cm wide, has been sectioned and pol-ished for this study.X-ray diffractometer analyses of pol-

ished surfaces show that there are face-centered cubic (y) and body-centeredcubic (a) metal phases present. Observa-

tions of polished surfaces made with op-tical and scanning electron microscopesreveal that the a and y phases are inter-grown in a pattern similar to Widman-staetten patterns found in some meteor-ites. Figures 1 and 2 are scanning elec-tron microscope images of the Wid-manstaetten pattern from an unetchedsurface. The y lamellae are arrangedwith a symmetry similar to that foundin a (Ill) section of an octahedritemeteorite. This is a Widmanstaetten pat-tern in the metallurgical sense. How-ever, it differs from the Widmanstaettenpatterns found in meteorites in two sig-nificant ways. The lamellae in the jo-sephinite specimen have the y structure,whereas lamellae in meteorites consist of

SCIENCE, VOL. 206, 16 NOVEMBER 19790036-8075J7911 16-0832$00.5O0 Copyright 0 1979 AAAS832