Partitioning of polycyclic aromatic hydrocarbons (PAH) to water-soluble soil organic matter

European Journal of Soil Science, June 199.5, 46, 193-204

Partitioning of polycyclic aromatic hydrocarbons (PAH) to water-soluble soil organic matter

C.R. M A X I N & I . KOGEL-KNABNER* Department of Soil Science, University of Bayreuth, P.O. Box I 0 I 2 51, 95440 Bayreuth and *Soil Science Group, University of Bochum, NA 61134, 44780 Bochum, Germany

Summary

The mobility of polycyclic aromatic hydrocarbons (PAH) in soils can be influenced by the presence of dissolved organic matter. Partition coefficients of selected polycyclic aromatic hydrocarbons, ranging from 3-ring to 6-ring compounds, to water-soluble soil organic matter (WSSOM) were determined. Partition coefficients were determined for WSSOM obtained from two soils under agricultural use and forest and for commercially available humic acid (Aldrich), taking advantage of a reversed phase (C18) separation method. The WSSOM was characterised with regard to charge and hydrophilic/hydrophobic properties with a dissolved organic matter (DOM) fractionation method. No sorption to WSSOM was found for the tri- and tetracyclic PAH, whereas the penta- and hexacyclic PAH showed a significant binding to both types of WSSOM and to Aldrich humic acid. The affinity of penta- and hexacyclic PAH to WSSOM was considerably lower compared to the affinity to Aldrich humic acid. This is suggested to be due to the lower amount of hydrophobic fractions, c. 30%, in the natural WSSOM as compared to Aldrich humic acid. Effective partition coefficients (Koc,ff) for the sorption of PAH to bulk soil calculated from K D ~ ~ and DOM in the naturally occurring concentration range were only 60-70% of the Koc values in pure water. The impact of DOM on pollutant transport is further influenced by non-equilibrium behaviour of PAH in soils and by sorption of DOM to the solid-soil matrix. Several scenarios are described in which the effect of DOM on pollutant transport may become important.

Introduction sediments has been found by various authors (Gschwend &

Polycyclic aromatic hydrocarbons (PAH) are released into the environment mainly from anthropogenic sources, including fossil fuel combustion (Sims & Overcash, 1983) and forest and agricultural fires (Edwards, 1983; Freeman & Catell, 1990). They are distributed over long distances in the atmosphere (Windsor & Hites, 1979; Bjorseth & Olufsen, 1983; Masclet et al., 1988) and thus appear to be ubiquitous in the environment. Concentrations in soils vary between 50-500 Fg kg-' as background concentration in Europe and Northern America (Windsor & Hites, 1979; Hellmann, 1982; Edwards, 1983) to maximum values higher than 10000 pg kg-I in soils contaminated by vehicle (Blumer et al., 1977; Hellmann, 1982; Butler et al., 1984; Fleischmann & Wilke, 1991) or industrial emission (Davani et al., 1986; Tebaay et al., 1991). As microbial degradation is too slow to compensate the input rate to the soils, a significant accumulation in soils and

Hites, 1981; Jones et al., 1989). In recent years, growing concern has been attributed to the

effect of colloidal and dissolved organic matter on transport and distribution of PAH in the environment. It has been shown that organic matter dissolved or suspended in the aqueous phase may reduce bioavailability (McCarthy, 1983; Kukkonen et al., 1990) and enhance solubility (Carter & Suffet, 1982; Means & Wijayaratne, 1982; Chiou et al., 1986; Chiou et al., 1987) of hydrophobic organic chemicals. Furthermore, the ability of organic macromolecules to enhance the transport of hydrophobic compounds has been verified experimentally in soil columns (Caron & Suffet, 1985; Bengtsson et al., 1987; Enfield et al., 1989; Magee et al., 1991). The importance of colloidal transport in addition to water flow in macropores under field conditions has been stressed by Simmleit & Herrmann (1987).

Transport models include the convection/dispersion trans- Received 14 January 1994; revised 2 September 1994; accepted 4 November 1994 Correspondence: I. Kogel-Knabner.

Port equation and a retardation term describing the partitioning or sorption of the compound of interest to the solid phase (Rao et al., 1979; Valocchi, 1985; Harmon et al., 1989; Wagenet &

0 199.5 Blackwell Science Ltd. 193

194 C.R. Maxin & I . Kiigel-Knahner

Rao, 1990). This retardation factor is a function of the organic carbon (OC) content of the sorbent and the hydrophobicity of the compound (which is characterized by the partition coefficient KO,-). In the presence of dissolved organic matter (DOM) as mobile organic phase the apparent aqueous concentration may be enhanced. Therefore, the concentration of DOM (described by the dissolved organic carbon concentration DOC) and its affinity to the hydrophobic compounds (characterized by KDoc) must be included in the transport model.

Many studies have been performed in order to obtain partition coefficients of hydrophobic organic compounds such as PAH or chlorinated hydrocarbons, using dissolved organic macromolecules as sorbents. The methods used include solubility enhancement (Chiou et al., 1987), fluorescence quenching (Gauthier et a/., 1986; Backhus & Gschwend, 1990), headspace partitioning (Diachenko, 198 1; Jota & Hassett, 1991) and a reverse phase separation method (Landrum et al., 1984; Morehead et al., 1986). ,Most of the studies used commercial humic acid as model DOM, which cannot be considered to be representative of natural humic substances (Malcolm & MacCarthy, 1986). Humic and fulvic acids from soils and surface waters have also been investigated. However, soil humic acids are usually not mobile under field conditions. Water-extractable soil organic matter (WSSOM), the behaviour of which is important for chemical compound transport in soils, has been neglected so far. The objective of our study was to investigate the impact of mobile soilborne DOM on the mobility of several polycyclic aromatic compounds differing in hydrophobicity. Therefore, we used extracts of WSSOM as sorbent for the determination of partition coefficients. On the basis of these data, we aimed to calculate effective partition coefficients which describe the affinity of PAH to soils under the influence of DOM. These effective partition coefficients take into account the amounts of solid phase OC and liquid phase DOC that occur in soil under natural conditions.

Theoretical considerations

Hydrophobic substances are considered to be bound in soils mainly by partitioning to soil organic matter (Means et al., 1980; Chiou, 1989), clay minerals playing only a minor role in systems with low organic matter content (Karickhoff, 1984; Murphy et al., 1990). As a partitioning isotherm is linear until the saturation concentration is reached (Chiou, 1989), the soil concentration X s of a compound in equilibrium is related to the dissolved concentration Cw by the partition coefficient Koc:

Xs = KocCw. ( 1 )

Dissolved organic matter and organic colloids may form microscale hydrophobic environments in water, thus enhan- cing the apparent solubility (Chiou et al., 1986; 1987) and influencing the partition equilibrium (Curl & Keoleian, 1984; Gschwend & Wu, 1985; Voice & Weber, 1985). It can be

assumed that the interaction between the hydrophobic compound associated with the water soluble organic phase and the compound dissolved in water can be described by a partition process as well. The amount of substance associated with DOM (XDoc) is related to the freely dissolved concentration, Cw , by the partition coefficient KDoc:

XDOC = KDOC CW. (2)

The total amount of the substance of interest in the aqueous phase (Ck ) can be calculated as the sum of the substance partitioned to DOM and the freely dissolved substance using the experimentally-determined partition coefficient KD0c. To quantify the impact of DOM, the fraction partitioned to DOM (XDoc) has to be multiplied by the DOC concentration (D):

C & = Cw + CWKDOCD. (3)

The adsorbed concentration of equilibrium in a three-phase system can be described by an effective partition coefficient KOCeff, provided that the dissolved sorbent does not undergo adsorption onto soil (Enfield et al., 1989; Kan & Tomson, 1990; Kogel-Knabner et al., 1990):

(4)

Combining Equations (3) and (4) yields the relation between the three partition coefficients:

(5)

Good correlations between calculated and experimentally- derived effective partition coefficients of hexachlorobenzene have been reported by Enfield et al. (1989).

Materials and methods

Soil properties and extraction procedure for WSSOM

Two soils under forest (Allersdorf) and agricultural use (Neumarkt/Oberpfalz) were chosen for the extraction of WSSOM. Their properties are given in Table 1. In an attempt

Table 1. Properties of the soils used for extraction of water-soluble soil organic matter.

Site Neumarkt/Oberpfalz Allersdorf

Soil type" Aquic Haplumbrept Cumulic Hapludoll

Vegetation agricultural use forest Alnus glutinosa Fraxinus excelsior

Horizon A A Texture silty sand clayey loam Organic carbonig kg-lh 15 60 PH (HZO) 6.6 1.7 Water content/g kg-' 82 333

'According to Soil Survey Staff (1992). bDetermined in 1991 by dry combustion with a Carmhomat 8 ADG (Wosthoff, Bochum, Germany).

0 1995 Blackwell Science Ltd, European Journal of Soil Science, 46, 193-204

Partitioning polycyclic aromatic hydrocarbons to organic matter 195

DOC Electric conductivity Table 2. Properties of the investigated solutions. Extract

or solution Date /mg C dm-3 PH /mS m-'

(a) Neumarkt soil extracts Extracts used for sorption experiments N,,," 16 April 1991 27.2

21 May 1991 17.1 6 May 1991 11.2*0.2c

27 May 1991 17.5 3 June 1991 12.0k0.4

12June 1991 14.4 14 June 1991 17.2k0.2 24 June 1991 17.9k0.8 15 July 1991 14.9 & 0.1

7.2 6.5 7.2 7.3 7.4 7.3 n.d. 7.2 7.6

8.5 8.7 n.d.b n.d. n.d. 7.2 n.d. n.d. 8.1

Extract used for fractionation 25 August 1991 22.4 & 1 .5 7.5 n.d.

extraction ratio soil : water= 1 : 2. aN,ir : extraction of air-dried soil. bn.d. =not determined. - + = standard deviation.

(b) Allersdorf soil extracts Extracts used for sorption experiments

2 July 1991 27.8 f 2.7 8.1 9.8 23 July 1991 12.1 & 0.6 8.4 8.8

Extract used additionally for fractionation 14 August 1991 21.9 *2.3 7.9 13.1

extraction ratio soil : water = 1 : 10.

(c) Aldrich humic acid 21 May 1991 21.7k 1.3 7.5 2.2 23 May 1991 18.6k2.6 n.d. n.d. 4 June 1991 18.7 * 3.9 n.d. n.d.

24June 1991 20.4 * 0.1 n.d. n.d. 4 July 1991 18.6 + 0.6 n.d. n.d.

15 July 1991 20.0 * 0.2 n.d. n.d.

obtained by dilution of stock solution (1 : 100)

to investigate WSSOM solutions with different composition and properties, the two soils chosen differed with regard to vegetation cover, organic matter content, and texture. A key factor influencing the DOM composition and thus the affinity of DOM for hydrophobic organic chemicals is the pH value. Due to limitations of the method used for the separation of free and bound phases, only soils with neutral pH were chosen. It will become evident below, however, that differences in soil properties other than pH do not result in severe changes in the compositional pattern of WSSOM.

To obtain representative and reproducible solutions of WSSOM, after sampling the soil was homogenised by sieving and stored at -20°C. Aliquots were extracted before each sorption experiment, This procedure proved to be necessary, as air drying resulted in an increasing content of WSSOM (Kaupenjohann & Franke, 1991). Freeze drying of the extracts was not possible because redissolving the freeze-dried extracts yielded flocculated solutions. Presumably due to the high

calcium contents (80-90 mg Ca dm-3), the WSSOM precipitate was no longer soluble.

The frozen samples were allowed to thaw in a refrigerator for 12 h, sieved ( < 2 mm) and extracted for 2 h with deionized water using a soil : water ratio of 1 : 2 and 1 : 10 for the loamy sand (Neumarkt) and clayey loam (Allersdorf), respectively. The extracts were centrifuged for 10 min at 11 OOOg, filtered though a Whatman (Maidstone, England) GFP5.5 glass-fibre filter and a Sartorius (Gottingen, Germany) SM 11106 0.45-pm cellulose acetate filter.

Before performing the sorption experiments, the extracts were analysed for pH by a pH electrode (WTW GmbH, Weilheim i. OB, Germany), for DOC by a Dohrmann DC-90 Carbon Analyser (Santa Clara, CA) and for electric conductivity by a Hanna Instruments conductivity electrode (Laborcenter, Niirnberg, Germany). The properties of the extracts used in the sorption experiments are given in Table 2. Variations in DOC content in spite of the standardised


196 C.R. Maxin & I . Kogel-Knabner

extraction procedure may occur due to differences in room temperature during extraction.

To minimize microbial activity, sodium azide was added to a concentration of 0.5 g dmP3. If the extracts could not be used immediately, storage in a refrigerator without visible alteration of the solution was possible up to a maximum of 4 d. Aldrich humic acid was used as received. 500 mg was weighed into a I-dm3 volumetric flask and brought to volume. The solution was homogenised in an ultrasonic bath, filtered through a Sartorius (Gottingen, Germany) SM 11 106 0.45-pm cellulose acetate filter and stored in the refrigerator. The solutions used in the experiments were obtained by diluting this stock solution 1 : 10.

Characterization of WSSOM

The composition of the WSSOM extract was characterized by a fractionation method based on different affinities to exchange resins (Leenheer, 198 1) which has been modified to a smaller scale. A detailed description is given by Vance & David (1991) and Guggenberger & Zech (1993a). Briefly, the method consists of the following steps.

An acidified (pH 2) solution is successively passed through three adsorption columns at a flow rate of 1 cm3 min-'. Hydrophobic acids (HOA) and hydrophobic neutrals (HON) are retained on the first column, containing the XAD-8 hydrophobic resin. Hydrophobic bases (HOB) and hydrophilic bases (HIB) are retained on a second, cation-exchange column (AG-MP-50 resin). The hydrophilic acids (HIA) are retained on the third anion-exchange column (Duolite A-7 resin). The DOC content is determined for all the solutions collected as described below. the DOC content of the efflux from the third column is measured after the void volume has passed the system. This value represents the fraction of hydrophilic neutrals (HIN), which pass all three columns. After removing the third column (Duolite A-7), the efflux contains the fractions of HIN and HIA. The concentration of HIA is then calculated by difference. In the same way the concentration of HIB and HOB is calculated by difference after removing the second (AG-MP-50) column. For the differentiation between HOA and HON, the fraction of HOA is desorbed from the first column (XAD-8) by 0.1 N NaOH. The fraction of HON is calculated by difference between the total DOC content and the sum of fractions. Thus, the relative percentage of hydrophilic acids, bases and neutrals as well as the relative percentage of hydrophobic acids, bases and neutrals in the original soil extract can be calculated. Relative standard deviations for the percentages of fractions are generally < 10%.

Sorption experiments

Polycyclic aromatic compounds covering a range of hydrophobicity were obtained in solid or liquid form. Phenanthrene,

anthracene, fluoranthene, pyrene and benzo(e)pyrene (all obtained from Aldrich, Steinheim, Germany) were each weighed in volumetric flasks and dissolved in methanol. Benzo(k)fluoranthene (Alltech, Unterhaching, Germany) was dissolved in methanol/dichloromethane (90 : 10 v : v). Ben- zo(ghi)perylene was obtained in a stock solution in dichloromethane from Supelco, Bad Homburg, Germany. All substances except fluoranthene (98%) were at least 99% pure.

The stock solutions used for the sorption experiments were obtained after a dilution step. Small amounts (20 mm3-100 mm3) of the stock solutions were pipetted into volumetric flasks (50-250 cm3) already half-filled with the solution of interest (WSSOM or a 0.5 g dm-3 NaN3 solution in deionized water). The pipette was rinsed twice with methanol. The volumetric content of methanol did not exceed 1% in any experiment, and it was assumed that a low content of methanol does not interfere with the process of solubility enhancement (Edwards et al., 1991). The concentration of each compound in the flasks did not exceed 50% of the compound solubility in water. The flasks were brought to volume and shaken on a horizontal shaker in the dark for 24 h. For the kinetic studies (benzo(e)pyrene and benzo(k)fluoranthene), they were shaken for 15 min, 2 h, 6 h, 12 h, 24 h and 48 h. Sorption isotherms were obtained using at least five initial concentrations. Each experiment was performed in triplicate. Benzo(e)pyrene and benzo(k)fluoranthene were investigated in mixed solutions. An experiment with the compounds as single solutes was performed in order to compare the results.

Separation of PAH fractions by C,a-columns

On account of the physico-chemical properties of PAH (low aqueous solubility, low vapour pressure, strong fluorescence), three of the above mentioned experimental approaches, namely the solubility enhancement, fluorescence quenching, and reverse-phase separation techniques, may prove success- ful. Each of them has specific advantages and shortcomings.

The solubility enhancement approach (Chiou et al., 1987) requires the experiments to be carried out at the solubility limit, hence over saturation of the solution may lead to erroneous results. Thus the results may not be representative for environmental conditions. Furthermore, this technique has not been used so far for investigating the DOM-PAH interaction. Therefore, it would not have been possible to compare our results with previously reported data.

The main advantage of the fluorescence quenching technique (Gauthier et al., 1986) is that the DOM-PAH interaction is not disturbed by the experiment. However, there is some evidence that the assumptions on which the calculation is based may not always be valid (Puchalski et al., 1992). This technique cannot be used to study PAH with more than four condensated rings (Stein, 1993), because of the low water solubility of these compounds in combination with a low fluorescence activity.



Table 3. Hydrophobic and hydrophilic fractions (%) of the extracts before and after passing the C18-column.

Allersdorf Neumarkt

Fraction Extract Effluent Extract Effluent

Hydrophilic acids and neutrals 62.9 66.7 68.4 64.8 Hydrophilic and hydrophobic bases 2.2 0.3 0.0 1.7 Hydrophobic acids and neutrals 39.4 33.0 32.9 33.5

Therefore, the reverse-phase separation technique (Landrum et al., 1984; Morehead et al., 1986) was chosen to determine ‘freely dissolved’ and ‘bound’ phases of the compound. The solution containing PAH and DOC is passed through a SPE column (3 cm3) filled with reverse-phase material (c18). Whereas the phase of ‘freely dissolved’ PAH is retained on the solid phase, the association with DOC prevents the ‘bound’ phase from being sorbed to the C18-material. This fraction appears in the effluent of the C18-column. In contrast to Landrum et al. (1984), we did not use radiolabelled substances to compare solution activity before and after passing the Clg-column. We chose to elute the fraction sorbed to the Clg-material with an appropriate organic solvent. The comparison of Kooc values for Aldrich humic acid with those from Landrum et al. (1984) and Gauthier et al. (1986, 1987) shows that the results obtained in this study are in good agreement with those of other methods (see results).

The two fractions are operationally defined, as it is not clear whether the interaction between hydrophobic parts of WSSOM and the PAH really is a ‘binding’ or rather a partitioning process. Regard has to be given to several aspects when applying this method.

The flow rate has to be high enough not to disrupt the weak interactions between DOM and the hydrophobic compounds (Morehead et al., 1986), as one basic assumption is that the dissociation kinetics of adsorption of unassociated PAH into the C18 phase is much faster than the dissociation kinetics of PAH bound to dissolved organic matter. Morehead et al. (1986) found a small flow-rate dependence as a result of these competing rates. They also observed that at flow rates greater than 6 cm3 min-’ the errors are small relative to the range of KDoc. The dissolved organic substances must not adsorb to the c18 material during the passage. At pH <5, acidic compounds of the investigated DOM may be protonated and retained on the C18-column as well as the PAH.

Complete sorption of the hydrophobic compounds to the Cls-material in absence of DOM has to be assured. The recovery from the column should be quantitative. For this reason, borosilicate glass columns with PTFE-frits were preferred to one-way cartridges (made of polypropylene with frits of polyethylene).

Baker columns (3 cm3) were filled with 500 mg Cl8-solid phase (40 pm/60 A, J.T. Baker Inc., Gross-Gerau, Germany). The columns were conditioned with 10 cm3 deionized water. The aqueous solution containing PAH was passed though the column at a flow rate of 12 cm3 min-’ using a Baker SPE

system. 50-cm3 and 100-cm3 flasks were rinsed with deionized water, which was also applied to the column (30 cm3 and 60 cm3, respectively). The DOC in the effluent was >95% of the organic carbon that had entered the column. The composition of both solutions regarding hydrophobic and hydrophilic substances slightly changed (Table 3), but considering the accuracy of the fractionation method (10%) the changes can not be regarded as significant.

After the last rinse had passed the column, the application of vacuum was continued until the material was completely dry (> 10 min). Tri- and tetracyclic PAH was eluted by slowly passing 5 cm3 of methanol through the columns. These solutions were analyzed by RP-HPLC without preconcentra- tion.

Penta- and hexacyclic PAH were eluted with 5 cm3 of hexane. 1 cm3 of hexane was applied to the top of the column and allowed to soak in for 2 min. Reduced pressure was applied until the hexane layer reached the top of the frit. This procedure was repeated four times. Samples in hexane were concentrated under a gentle stream of nitrogen and taken up in methanol for RP-HPLC analysis. The acetonitrilelwater program (6 rnin isocratic 55% CH3CN, 6-18 rnin gradient to 80% CH3CN, 18-24 rnin gradient to 100% CH3CN, 24-32 rnin isocratic 100% CH3CN, 32-35 rnin gradient to 55% CH3CN and 10 min re-equilibration time) was run by a Gynkothek GmbH 480 High Precision Pump (Gynkothek, Germering, Germany) on a ET 150/8/4 Nucleosil 5 ClX-PAH Macherey & Nagel (Diiren, Germany) column. A variable wavelength fluorescence detector (82 LFP, JASCO Inter- national Co. Ltd., Tokyo, Japan) was programmed to detect the compounds at the following emission and extinction wavelengths: phenanthrene and anthracene at 260/380 nm, fluoranthene and pyrene at 270/420 nm, benzo(e)pyrene at 270/380 nm, benzo(k)fluoranthene at 260/420 nm (all gain 100/attenuation 16) and benzo(ghi)perylene at 360/420 nm (gain 1000/attenuation 16). In order to keep the retention time constant the column temperature was kept constant at 19°C.

The recoveries of the individual compounds from 0.5 g dm-3 sodium azide solution are given in Table 4. If WSSOM is present in the solution applied to the C18 column, the recoveries of PAH will be lower, because the ‘bound’ phase is not retained on the column. The recovered PAH obtained from HPLC determination of PAH in WSSOM solution is operationally defined as the ‘freely dissolved fraction’. The ‘bound fraction’, which passed the column, was calculated by



Table 4. Recovery of PAH from aqueous solution of sodium azide (0.5 g dm-3).

Compound n p Recovery/% SD"

Phenanthrene 3 3 99.0 5.8 Fluoranthene 2 312 94.2 10.4 Anthracene 4 3 97.5 5.1 Pyrene 5 3 95.4 6.3 Benzo(e)pyrene 10 3 89.5 10.4 Benzo(k)fluoranthene 10 3 89.8 9.5 Benzo(ghi)per y lene 5 3 83.1 4.7

n =number of experiments; p = 2: in duplicate; p = 3: in triplicate. "Standard deviation.

Table 5. Hydrophobic and hydrophilic fractions (96) of the investigated extracts.

Sample Concentration Site /mg C dm-3 HIN" HIA HIB+HOB HOA HON

Allersdorf 2 1.9 21 42 2 24 11 Neumarkt 22.4 12 57 0 33 0

~~~

"HIN, HIA, HIB = hydrophilic neutrals, acids, and bases; HON, HOA, HOB =hydrophobic neutrals, acids, and bases.

difference from the recovery in sodium azide solution. The amount of 'bound fraction' was normalised to the DOC content of the solution in order to obtain sorption isotherms, the slopes of which give the KDoc values. The error range was determined as the difference between slope and the minimum and maximum slope resulting from data standard deviation.

Results and discussion

Characterization of WSSOM

Both solutions of WSSOM showed a similar composition (Table 5). Compounds with acidic character, hydrophilic and hydrophobic acids, dominate in the solutions. This is in accordance to the results of Vance & David (1991) and Guggenberger & Zech (1993b), who found high organic-acid contents in soil solutions of different origin.

Kinetic studies

The process of interaction between PAH and WSSOM is too fast to be observed by the method used in this study. Figure 1 shows that the equilibrium is reached within less than 15 min, which is tlie time required for the extraction of the aqueous solution. The slight decrease in the observed truly dissolved aqueous concentration with longer equilibration times may be caused by artefacts such as sorption to glass walls or microbial degradation. In order to minimize these effects, an equilibration period of 24 h was chosen for the sorption experiments.

0.00 6 0 10 20 30 40

Timelh 3

Fig. 1. Partition kinetics of polycyclic aromatic hydrocarbons to soilborne DOM (Neumarkt) and Aldrich humic acid: (a) benzo(e)pyrene; (b)\ benzo(k)fluoranthene; 0, Aldrich, A, Neumarkt.

Determination of KDoc

Tri- and tetracyclic PAH (phenanthrene, fluoranthene, anthracene, pyrene) did not show any measurable interaction with the WSSOM solutions. The method is insensitive to bound PAH concentrations lower than 1 pg dmp3. Therefore, KDoc values below 3.5 approximately cannot be determined.

The values of log KDOC obtained for anthracene and pyrene versus Aldrich humic acid were 4.14 and 4.4, respectively. The latter results are in good agreement with values obtained by Landrum et al. (1984) (~thracene/4.0/CIs-method) and Gauthier et al. (1987) (pyrene/5.O/fluorescence quenching). Partition coefficients obtained by different methods should be compared with caution, as the fluorescence-quenching method often yields higher partition coefficients than the C I method (Gauthier et al., 1986). In preliminary experiments with extracts of air-dried soil, a small but not significant interaction with anthracene and pyrene was found ('bound' concentration < 1 pg dm-3). Partition coefficients were determined to be 3.5 and 3.4, respectively. As extracts from air-dried soil stored under room temperature in the laboratory for several weeks are supposed not to represent naturally occurring WSSOM (see above), they were not investigated further.


Partitioning polycyclic aromatic hydrocarbons to organic mutter 199

100-

80 -

60 -

40 -

20 -

T

- . 0.0 0:5 1 :o 1.5 2.0 2.5

T

0.00 0.05 0.10 0.15 0.20 0.25 0.30

0 0.00 0.01 0.02 0.03 0.04 0.05 0.06

Freely dissolved/pg d ~ n - ~

Fig. 2. Partition isotherms of polycyclic aromatic hydrocarbons to DOM from different sources: (a) benzo(e)pyrene; (b) benzo(k)fluoranthene; (c) benzo(ghi)perylene. 0 , Aldrich; A, Allers- dorf; A, Neumarkt. The arrow marks values in single solute solutions.

Sorption isotherms were determined for benzo(e)pyrene, benzo(k)fluoranthene and benzo(ghi)perylene. They are shown in Fig. 2. The KDoc values are listed in Table 6. Partition coefficients (log KDoc) for Aldrich humic acid were 5.18 (benzo(e)pyrene), 5.09 (benzo(k)fluoranthene) and 5.79 (benzo(ghi)perylene). These results are in good agreement with the findings of other authors. Values of log KDOC for pentacyclic PAH associated with Aldrich humic acid have been found to range from 5.3 (Morehead et al., 1986: CIS- method/benzo(a)pyrene) through 5.9 (Backhus & Gschwend, 1990: fluorescence quenchinglperylene) to 6.3 (McCarthy &

Table 6. Slope (KDoc), intercept (a) and regression coefficients (r2) of the sorption isotherms, XDOC = a + K ~ O C . C,v. The slope is also given in log form as KDOC value.

Error Solution U KDOC r2 IogKDOC range"

Benzo(e)pyrene

Neumarkt +2.504 30336 Aldrich -7.558 152040

Allersdorf -1.029 30020

B e n z ~ ~ k ~ ~ u ~ r a n t h e n e Aldrich +0.127 123840 Neumark t +0.720 51750 Allersdorf +0.405 43 180

Benzo(ghi)perylene

Neumarkt +0.617 85710 Aldrich -2.260 614900

Allersdorf -0.512 109570

0.937 0.981 0.977

0.953 0.896 0.997

0.876 0.979 0.912

5.18 4.48 4.48

5.09 4.71 4.63

5.79 4.93 5.04

0.22 0.15 0.14

0.19 0.11 0.13

0.11 0.09 0.09

"Deviation from log KDoc value under the assumption of minimum or maximum sorption, respectively, as characterized by mean value& standard deviation.

Jimenez, 1985: dialysis/benzo(a)pyrene). In general, partition coefficients for natural humic substances are slightly lower. Morehead et al. (1986) have determined log K D ~ C of benzo(a)pyrene with natural waters to vary between 4.2 and 5.0 (C18-method). Backhus & Gschwend (1990) found log KDOC values for perylene and ground water between 5.4 and 6.0 (fluorescence quenching). Obviously, WSSOM has a distinctly lower affinity for PAH as log KDoc values were 4.48 (benzo(e)pyrene), 4.63 -4.7 1 (benzo(k)fluoranthene) and 4.93 - 5.04 (benzo(ghi)perylene).

Comparisons between solutions of single solute versus multi solute

Comparison of the sorption data for benzo(e)pyrene and benzo(k)fluoranthene reveals no difference between single solute and multi-solute solutions (single solute values are marked by an arrow in Fig. 2a,b). The absence of competition in a multi-solute system indicates that sorption does not occur at distinct sorption sites. This is in accordance with the hypothesis that the mechanism for the interaction between WSSOM and the hydrophobic solute is a partition process.

Reproducibility of the DOM-extraction method

Extracts with very low DOM content compared with average values sometimes showed increased affinity to PAH and were not included in the regression (data not shown). Variations between different extractions of WSSOM may be caused by differences in microbial activity.



.................................................................................

2 200, I # #

9 150- C

'e : 100- s

9 X 5 0 -

B

S n

S

u

+A- TP I

0.5 1 .o 1.5 2.0 2.5

Freely dissolvedipg drns

Fig. 3. Partition isotherm of benzo(e)pyrene, normalized to the hydrophobic fractions of DOM, which are regarded as sorbing entity. 0, Aldrich A, Allersdorf; A, Neumarkt.

Comparison of natural DOM with Aldrich humic acid

Both solutions of WSSOM showed a similar partitioning ability for PAH. The affinity to Aldrich humic acid, however, proved to be about three times higher. Taking the composition of DOM into consideration, one finds that the natural DOM solutions contain 33% (Neumarkt) and 35% (Allersdorf) of HOA and HON. The composition of Aldrich humic acid has so far not been documented, but it can be assumed that it contains mainly hydrophobic organic acids. If we further assume that the hydrophobic moieties are the sorbing entity (Kukkonen et al., 1990), the bound fraction can be normalised to the concentration of HOA and HON. The results of this calculation regarding benzo(e)pyrene as example are shown in Fig. 3. The affinity of PAH to the hydrophobic fractions of soilborne DOC is comparable to the affinity to Aldrich humic acid.

Isotherm shape

In contrast to theoretical considerations (Chiou, 1989), the isotherm for the sorption between benzo(e)pyrene and benzo(k)fluoranthene to Aldrich humic acid is not linear when the solution concentration exceeds 10% of the solubility in water. Above this concentration range, artefacts caused by incompletely dissolved substances can occur in sorption experiments with hydrophobic organic compounds. The partition behaviour at higher concentration is, therefore, interpreted with caution.

The isotherm shape at higher concentrations can be attributed either to experimental artefacts (enhanced retention or extraction efficiency on C 1 8-columns) or to decreasing affinity to DOM at higher PAH concentrations. These two processes cannot be distinguished here. Recently, Schlautmann & Morgan (1993) found that the partitioning model may not be valid in all cases, as there is some evidence for the DOM conformation and polarity influencing the binding process.

The sorption isotherms for WSSOM, however, are linear up to 30% of the solubility in pure water. Thus the non-linearity of the isotherm seems to be related to the specific properties of Aldrich humic acid.

The observation that at a concentration of 30% of the water solubility the isotherm of benzo(k)fluoranthene versus Neumarkt WSSOM seems to become non-linear as well as neglected in calculating the effective isotherms. For the calculation of the slope only the linear part of the isotherm was used. Although the effective isotherm may as well be calculated on the basis of a non-linear partition process of PAH to DOM, for environmental implications the linear part of the isotherm is fully sufficient. PAH at higher aqueous concentration levels will most probably not occur in natural environments.

Implications for pollutant sorption in soils

Effective Koc values as indicators for the importance of DOM as carrier for pollutants

The experiments have shown a measurable interaction between WSSOM and penta- and hexacyclic PAH. The partition coefficients derived from these experiments can be used in combination with solid soil organic matter partition coefficients (Koc) to calculate effective partition coefficients according to Equation (5) . As a result, the partitioning behaviour at equilibrium of PAH in a three-phase system containing solid soil, water and DOM can be predicted. Figure 4 shows that the influence of DOM on the effective partition coefficient increases with the hydrophobicity of the compound. There is practically no influence of DOM even at high concentrations on the partitioning of anthracene (KDoc of extracts from air- dried soil). Concentrations of DOM in the naturally occurring range up to 100 mg DOC dmP3 may lower the amount of benzo(ghi)perylene bound to soil by about one order of magnitude.



Still assuming equilibrium, these results can be used to calculate effective KP values, i.e. partition coefficients not normalized to the organic-carbon content of the soil. Obviously, the calculated amount of PAH sorbed to a soil with an organic-carbon content of 60 g kg-' is six times higher than that of a soil with 10 g kg-' of organic carbon at the same solution concentration. At the same PAH concentration in soil, the presence of DOM at 50 mg C dm-3 may enhance the amount in the aqueous phase to twice the amount in pure water (Fig. 5) .

The possible effect of DOM in natural soils

The calculation of effective partition coefficients is performed under two assumptions which may not hold for natural soils. It becomes clear that no general statement can be made about the effect of DOM on the mobility of PAH in soils. Pathways of PAH and DOM input have to be considered as well as the soil physical and chemical properties. We will, therefore, discuss several scenarios in which DOM may influence pollutant mobility in surface soils.

First, it should be mentioned that equilibrium is not reached in natural soils. Whereas the partition between DOM and water occurs rapidly, the exchange process between the soil matrix and the aqueous phase is limited by diffusion (Karickhoff, 1980; Brusseau & Rao, 1989; Pignatello, 1989). As the mass- transfer process between solid and aqueous phase is slower than the convective flow of water through the pores, nonequilibrium conditions continue to exist in sorption and desorption. In aggregated soils, this phenomenon is enhanced by the percolation of water through macropores, so that only part of the bulk soil is in contact with water. This aspect is particularly important with regard to agricultural soils preferentially ameliorated with mineral fertilizer, which do not receive organic manure or sewage sludge. They receive atmospheric deposition mainly during the cold season of the year when vegetation cover is scarce. After ploughing in spring, the deposited PAH are distributed homogeneously in the plough layer. Corresponding to low contents of organic carbon in the bulk soil, the DOM concentration in the soil solution is rather low. Together with retarded desorption from the soil matrix this scenario leads to a negligible impact of DOM on PAH mobility.

There are situations, however, in which large amounts of DOM are released into the soil solution, for example after liming or an increase in temperature. This is of particular importance in forest ecosystems, which are effective accumu- lators for airborne pollutants (Matzner et al., 1981; Buckley, 1982), due to the surface properties of leaves and needles (Reischl et al., 1987; Safe et al., 1992) and leaf area indexes of between 3 and 12 (Whittaker & Likens, 1975). In the annual cycle, the temperature increase in spring causes increased concentrations of WSSOM in the soil solution because of litter biodegradation. As the accumulated micro-pollutants are

350 DOCIrng drn-3

300 - 250-

$ 'm 200- - Y m

150-

e 6 100-

O W 0.000

/ O

0.001 0.002 0.003 In aqueous phase/pg drn-3

Fig. 5. Calculated sorption of benzo(e)pyrene to soils of different content of organic carbon in and without presence of DOC at different concentrations. KOC values are obtained according to Karickhoff (1984) and Kooc values from Table 6. - - - 10, - 60 g Corg kg-'.

released at the same time, it seems likely that they associate with the nascent WSSOM, resulting in increased mobility in soil.

A similar situation arises if organic manure or sewage sludge, both of which contain large amounts of DOM and probably considerable amounts of PAH (Witte et al., 1988; Fricke & Vogtmann, 1989), are applied to agricultural soils. Repeated long-term organic matter amendment of an agricultural soil resulted in the DOM-concentration being enhanced to 73 mg C dmP3 in percolation and batch experiments (Kalbitz, 1993, personal communication). The input of PAH and DOM at the same time together with rate- limited sorption of PAH to soil may lead to an enhanced movement of PAH in the soil profile. Whether this actually takes place depends on infiltration properties (pore-size distribution and preferential-flow systems) of the site as well as on the presence of DOM-sorbing constituents which are discussed below.

The second assumption which is not generally fulfilled under natural conditions is that the carrier DOM does not undergo sorption onto the solid phase. In the soil, DOM is currently sorbed and desorbed. Especially after drying, desorption from the bulk soil matrix is enhanced and may lead to co-desorption of other hydrophobic compounds (Curl & Keoleian, 1984). In contrast, there are often inorganic sorbents, especially iron and aluminium oxides, in the subsoil which are able to sorb DOM efficiently from soil solution. In this process, the hydrophobic constituents of DOM which are assumed to be responsible for the major part of the interaction with hydrophobic organic chemicals, are bound preferentially (Jardine et al., 1990; Guggenberger & Zech, 1993~) . Jardine et a1 (1990) investigated transport of DOM in an isolated soil pedon. They have shown that a major part of the organic layer output (9 mg C dm-3) was sorbed in the B horizon, the percentage of hydrophobic compounds decreasing from 60% to



25%. The output of the mineral soil contained 1-2 mg C dmP3. Furthermore, it has been shown that the sulphate anion and negatively-charged dissolved organic carbon are competing sorbates for the sorption sites on iron oxides. As a consequence, enhanced input of sulphate (acid rain) is likely to favour the mobility of dissolved organic substances and thus of hydrophobic organic chemicals as well. Guggenberger & Zech (1993b) compared DOM flow in three soil ecosystems with different acid-rain inputs. In healthy or moderately damaged ecosystems, the mineral soil input of about 27 mg C dm-3 was effectively sorbed onto iron oxides so that the mineral soil output did not exceed 1-2 mg C dmP3. In a soil severely disturbed by input of airborne acids, the output of DOM reached a much higher concentration, 1 1 mg C dmP3, even though the net amount of sorbing iron oxides was highest at this site.

These results demonstrate that the affinity of WSSOM from natural soils for PAH is high enough to influence the sorption equilibrium. Several scenarios have been discussed in order to evaluate the impact of the solubility enhancement on PAH mobility in soils, which may be considerable under some conditions. The involvement of DOM in pollutant transport is strongly influence by specific soil chemical and physical conditions. The conclusions drawn from the laboratory batch sorption experiments will have to be verified in column experiments and field studies.

Acknowledgements

The work reported in this paper was financially supported by the ‘Deutsche Forschungsgemeinschaft’ (KO 1035/1-2). We thank G. Badewitz for technical assistance and B. Glaser for performing the DOC measurements. We further are indebted to H. Deschauer, G. Guggenberger and J. Tolls for valuable discussions and suggestions.

References Backhus, D.A. & Gschwend, P.M. 1990. Fluorescent polycyclic

aromatic hydrocarbons as probes for studying the impact of colloids on pollutant transport in groundwater. Environmental Science and Technology, 24, 1214-1223.

Bengtsson, G., Enfield, C.G. & Lindqvist, R. 1987. Macromolecules facilitate the transport of trace organics. The Science of the Total Environment, 67, 159-164.

Bjorseth, A. & Olufsen, B.S. 1983. Long-range transport of polycyclic aromatic hydrocarbons. In: Handbook of Polycyclic Aromatic Hydrocarbons (ed. A. Bjorseth), pp. 507-524. Marcel Dekker, New York.

Blumer, M., Blumer, W. & Reich, T. 1977. Polycyclic aromatic hydrocarbons in soils of a mountain valley: correlation with highway traffic and cancer incidence. Environmental Science and Technology, 11, 1082-1084.

Brusseau, M.L. & Rao, P.S.C. 1989. The influence of sorbate-organic matter interactions on sorption nonequilibrium. Chemosphere, 18, 169 1 - 1706.

Buckley, E.H. 1982. Accumulation of airborne polychlorinated biphenyls in foliage. Science, 216, 520-522.

Butler, J.D., Butterworth, V., Kellow, S.C. & Robinson, H.G. 1984. Some observations on the polycyclic aromatic hydrocarbon (PAH) content of surface soils in urban areas. The Science of the Total Environment, 33, 75-85.

Caron, G. & Suffet, I.H. 1985. Effect of DOC in the environmental distribution of nonpolar organic chemicals. Chemosphere, 14, 993- 1000.

Carter, C.W. & Suffet, I.H. 1982. Binding of DDT to dissolved humic materials. Environmental Science and Technology, 16, 735 -740.

Chiou, C.T. 1989. Theoretical considerations of the partition uptake of nonionic organic compounds by soil organic matter. In: Reactions and Movements of Organic Chemicals in Soils, Special Publication 22 (eds B.L. Sawhney & K. Brown), pp. 1-29. Soil Science Society of America, Madison, WI.

Chiou, C.T., Malcolm, R.L., Brinton, T.I. & Kile, D.E. 1986. Water solubility enhancement of some organic pollutants and pesticides by dissolved humic and fulvic acids. Environmental Science and Technology, 20, 502-508.

Chiou, C.T., Kile, D.E., Brinton, T.I., Malcolm, R.L. & Leenheer, J.A. 1987. A comparison of water solubility enhancements of organic solutes by aquatic humic material and commercial humic acids. Environmental Science and Technology, 21, 1231- 1234.

Curl, R.L. & Keoleian, G.A. 1984. Implicit-adsorbate model for apparent anomalies with organic adsorption on natural adsorbents. Environmental Science and Technology, 18, 9 16-922.

Davani, B., Lindley, K. & Eiceman, G.A. 1986. Polycyclic aromatic hydrocarbons in soil at groundwater level near an earthen pit for produced water in the Duncan oil field. International Journal of Environmental Analytical Chemistry, 25, 299-31 1.

Diachenko, G.W. 1981. Sorptive interactions of selected volatile halocarbons with humic acids from different environments. Ph.D. dissertation, University of Maryland, College Park, MD.

Edwards, D.A., Luthy, R.G. & Liu, Z. 1991. Solubilization of polycyclic aromatic hydrocarbons in micellar nonionic surfactant solutions. Environmental Science and Technology, 25, 127- 133.

Edwards, N.T. 1983. Polycyclic aromatic hydrocarbons (PAHs) in the terrestrial environment. Journal of Environmental Quality, 12,

Enfield, C.G., Bengtsson, G. & Lindqvist, R. 1989. Influence of macromolecules on chemical transport. Environmental Science and Technology, 23, 1278-1286.

Fleischmann, S . & Wilke, B.-M. 1991. PAKs in Strapenrandboden. Mitteilungen der Deutschen Bodenkundlichen Cesellschaf, 63, 99- 102.

Freeman, D.J. & Cattell, F.R. 1990. Woodburning as a source of atmospheric polycyclic aromatic hydrocarbons. Environmental Science and Technology, 24, 1581-1585.

Fricke, K. & Vogtmann, H. 1989. Organische Schadstoffe in Bioabfallkomposten. Mull und Abfall, 9/89, 472-48 1 .

Gauthier, T.D., Seitz, W.R. & Grant, C.L. 1987. Effects of structural and compositional variations of dissolved humic materials on pyrene KOC values. Environmental Science and Technology, 21, 243-248.

Gauthier, T.D., Shane, E.C., Guerin, W.F., Seitz, W.R. & Grant, C.L. 1986. Fluorescence quenching method for determining equilibrium constants for polycyclic aromatic hydrocarbon binding to dissolved humic materials. Environmental Science and Technology, 20,

427 -44 1.

1 162- 1 1 66.



Gschwend, P.M. & Hites, R.A. 1981. Fluxes of polycyclic aromatic hydrocarbons to marine and lacustrine sediments in the North- eastern United States. Geochimica et Cosmochimica Acta, 45,

Gschwend, P.M. & Wu, S.C. 198.5. On the constancy of sediment- water partition coefficients of hydrophobic organic pollutants. Environmental Science and Technology, 19, 90-96.

Guggenberger, G. & Zech, W. 1993a. Zur Dynamik geloster organischer Substanzen (DOM) in Fichtenokosystemen-Ergeb- nisse analytischer DOM-Fraktionierung. Zietschrift f i r Pflan- zenernahrung und Bodenkunde, 156, 341 -347.

Guggenberger, G. & Zech, W. 1993h. Dissolved organic carbon control in acid soils of the Fichtelgebirge (FRG) as revealed by distribution patterns and structural composition analyses. Geo- derma, 59, 109-129.

Harmon, T.C., Ball, W.P. & Roberts, P.V. 1989. Nonequilibrium transport of organic contaminants in groundwater. In: Reactions and Movements of Organic Chemicals in Soils, Special Publication 22 (eds B.L. Sawhney & K. Brown), pp. 405-437. Soil Science Society of America, Madison, WI.

Hellmann, H. 1982. Polyzyklische Aromatische Kohlenwasserstoffe in Acker-und Waldboden und ihr Beitrag zur Gew'asserbelastung. Deutsche Gewasserkundliche Mitteilungen, 26, 63 -69.

Jardine, P.M., Wilson, G.V., McCarthy, J.F., Luxmoore, R.J., Taylor, D.L. & Zelazny, L.W. 1990. Hydrogeochemical processes controlling the transport of dissolved organic carbon through a forested hillslope. Journal of Contaminant Hydrology, 6, 3- 19.

Jones, K.C., Stratford, J.A., Waterhouse, K.S., Furlong, E.T., Giger, W., Hites, R.A., Schaffner, C. & Johnston, A.E. 1989. Increases in the polynuclear aromatic hydrocarbon content of an agricultural soil over the last century. Environmental Science and Technology,

Jota, M.A.T. & Hassett, J.P. 1991. Effects of environmental variables on binding of an PCB congener by dissolved humic substances. Environmental Toxicology and Chemistry, 10, 483-491.

Kan, A.T. & Tomson, M.B. 1990. Ground water transport of hydrophobic organic compounds in the presence of dissolved organic matter. Environmental Toxicology and Chemistry, 9,

Karickhoff, S.W. 1980. Sorption kinetics of hydrophobic pollutants in natural sediments. In: Contaminants and Sediments, Vol. 2 (ed. E.A. Baker), pp. 193-205. Ann Arbor Science Publishers, Ann Arbor, MI.

Karickhoff, S.W. 1984. Organic pollutant sorption in aquatic systems. Journal qf Hydraulic Engineering, 110, 707-734.

Kaupenjohann, M. & Franke, G. 1991. EinfluP von Probenvorbe- handlung und Temperatur auf die DOC-Sorption von Boden. Mitteilungen der Deutschen Bodenkundlichen Grsellschafi, 66, 353-354.

Kogel-Knabner, I., Knabner, P. & Deschauer, H. 1990. Enhanced leaching of organic chemicals in soils due to binding to dissolved organic carbon. In: Contaminated Soil '90 (eds F. Arendt, M. Hinsenveld & W.J. van den Brink), pp. 323-329. Kluwer Academic Publishers, Dordrecht.

Kukkonen, J., McCarthy, J.F. & Oikari, A. 1990. Effects of XAD-8 fractions of dissolved organic carbon on the sorption and bioavailability of organic micropollutants. Archives of Environ- mental Contamination and Toxicology, 19, 551 -557.

2359-2367.

23, 95-101.

253 -263.

Landrum, P.F., Nihart, S.R., Eadie, B.J. & Gardner, W.S. 1984. Reverse-phase separation method for determining pollutant binding to Aldrich humic acid and DOC of natural waters. Environmental Science and Technology, 18, 187-192.

Leenheer, J.A. 1981. Comprehensive approach to preparative isolation and fractionation of dissolved organic carbon from natural waters and wastewaters. Environmental Science and Technology, 15,

Magee, B.R., Lion, L.W. & Lemley, A.T. 1991. Transport of dissolved organic macromolecules and their effect on the transport of phenanthrene in porous media. Environmental Science and Technology, 25, 323-33 1.

Malcolm, R.L. & MacCarthy, P. 1986. Limitations in the use of commercial humic acids in water and soil research. Environmental Science and Technology, 20, 904-91 1.

Masclet, P., Pisikopulos, P., Beyne, S . & Mouvier, G. 1988. Long range transport and gas/particle distribution of polycyclic aromatic hydrocarbons at a remote site in the Mediterranean Sea. Atmos- phere and Environment, 22, 639-650.

Matzner, E., Hubner, D. & Thomas, W. 1981. Content and storage of polycyclic aromatic hydrocarbons in two forested ecosystems in northern Germany. Zeitschrift fur Pflanzenernahrung und Boden- kunde, 144, 283-288.

McCarthy, J.F. 1983. The role of particulate organic matter in decreasing accumulation of polynuclear aromatic hydrocarbons by Daphnia magna. Archives of Environmental Corztamination and Toxicology, 12, 559-568.

McCarthy, J.F. & Jimenez, B.D. 1985. Interactions between polycyclic aromatic hydrocarbons and dissolved humic material: binding and dissociation. Environmental Science and Technology,

Means, J.C. & Wijayaratne, R. 1982. Role of natural colloids in the transport of hydrophobic pollutants. Science, 215, 968-970.

Means, J.C., Wood, S.G., Hassett, J.J. & Banwart, W.L. 1980. Sorption of PAH by sediments and soils. Environmental Science and Technology, 14, 1524-1528.

Morehead, N.R., Eadie, B.J., Lake, B., Landrum, P.F. & Bemer, D. 1986. The sorption of PAH onto dissolved organic matter in Lake Michigan waters. Chemosphere, 15, 403-412.

Murphy, E.M., Zachara, J.M. & Smith, S.C. 1990. Influence of mineral-bound humic substances on the sorption of hydrophobic organic compounds. Environmental Science and Technology, 24, 1507-1516.

Pignatello, J.J. 1989. Sorption dynamics of organic compounds in soils and sediments. In: Reactions and Movements of Organic Chemicals in Soils, Special Publication 22 (eds B.L. Sawhney & K. Brown), pp. 45-80. Soil Science Society of America, Madison, WI.

Puchalski, M.M., Morra, M.J. & von Wandruszka, R. 1992. Fluorescence quenching of synthetic organic compounds by humic materials. Environmental Science and Technology, 26, 1787- 1792.

Rao, P.S.C., Davidson, J.M., Jessup, R.E. & Selim, H.M. 1979. Evaluation of conceptual models for describing nonequilibrium adsorption/desorption of pesticides during steady-flow in soils. Soil Science Society of America Journal, 43, 22-28.

Reischl, A., Reissinger, M. & Hutzinger, 0. 1987. Occurrence and distribution of atmospheric micropollutants in conifer needles. Chemosphere, 16, 2647-2652.

578-587.

19, 1072-1067.



Safe, S., Brown, K.W., Donnelly, K.C., Anderson, C.S., Markiewicz, K.V., McLachlan, M.S., Reischl, A. & Hutzinger, 0. 1992. Polychlorinated dibenzodioxins and dibenzofurans associated with wood preserving sites: biomonitoring with pine needles. Environ- mental Science and Technology, 26, 394-346.

Schlautmann, M.A. & Morgan, J.J. 1993. Effects of aqueous chemistry on the binding of polycyclic aromatic hydrocarbons by dissolved humic materials. Environmental Science and Technology,

Simmleit, N. & Herrmann, R. 1987. The behaviour of hydrophobic micropollutants in different karst water systems. I. Transport of micropollutants and contaminant balances during the melting of snow. Water, Air and Soil Pollution, 34, 79-95.

Sims, R.C. & Overcash, M.R. 1983. Fate of PAH in soil-plant systems. Residue Reviews, 88, 2-68.

Soil Survey Staff 1992. Keys to Soil Taxonomy, 5th edn, SMSS Technical Monograph No. 19. Pocahontas Press, Blacksburg, VA.

Stein, C. 1993. Sorption von polyzyklischen aromatischen Kohlen- wasserstoffen an gelostes organisches Material saurer Waldboden. Diplomarbeit am Lehrstuhl fur Bodenkunde und Bodengeographie, Universitlt Bayreuth.

Tebaay, R.H., Welp, G. & Brummer, G.W. 1991. Gehalte an poly- cyclischen aromatischen Kohlenwasserstoffen in Boden unterschied- licher Belastung. Mitteilungen der Deutschen Bodenkundlichen Gesellschaft, 63, 157-160.

27, 961-969.

Valocchi, A.J. 1985. Validity of the local equilibrium assumption for modelling sorbing solute transport through homogenous soils. Water Resources Research, 21, 808-820.

Vance, G.F. & David, M.B. 1991. Forest soil response to acid and salt additions of sulphate: 111. Solubilization and composition of dissolved organic carbon. Soil Science, 51, 297-305.

Voice, T.C. & Weber, W.J. 1985. Sorbent concentration effects in liquid/solid partitioning. Environmental Science and Technology,

Wagenet, R.J. & Rao, P.S.C. 1990. Modeling pesticide fate in soils. In: Pesticides in the Soil Environment (ed. M.M. Cheng), pp. 351-399. Soil Science Society of America, Madison, WI.

Whittaker, H. & Likens, G.E. 1975. The biosphere and man. In: Primary Production of the Biosphere, Ecological Studies 14 (eds H. Leith & H. Whittaker), pp. 305-328. Springer Verlag, Berlin.

Windsor, J.G. & Hites, R.A. 1979. Polycyclic aromatic hydrocarbons in Gulf of Maine sediments and Nova Scotia soils. Geochimica et Cosmochimica Acta, 41, 22-33.

Witte, H., Langenohl, T., Augustin, S.T. & Offenbacher, G. 1988. Untersuchungen zum Eintrag von organischen Schadstoffen in Boden und Pflanze durch die landwirtschaftliche Klarschlamm- verwertung. A: Belastung des Klarschlamms mit organischen Schadstoffen. Korrespondenz Ahwasser, 5188, 440-448.

19,789-796.


Partitioning of polycyclic aromatic hydrocarbons (PAH) to water-soluble soil organic matter

Documents