Influence of Humic Substances on the Removal of Pentachlorophenol

Faculty of Agriculture and Life Science, Hirosaki University,. Hirosaki 036-8561, Japan. To investigate some basic aspects of soil remediation using b...
0 downloads 0 Views 128KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 1031-1036

Influence of Humic Substances on the Removal of Pentachlorophenol by a Biomimetic Catalytic System with a Water-Soluble Iron(III)-Porphyrin Complex MASAMI FUKUSHIMA,* AKIRA SAWADA, MIKIO KAWASAKI, HIROYASU ICHIKAWA, KENGO MORIMOTO, AND KENJI TATSUMI National Institute of Advanced Industrial Science and Technology (AIST), 16-1 Onogawa, Tsukuba-West, Tsukuba 305-8569, Japan

FIGURE 1. Peroxide shunt pathway for organic substrates in the Fe(III)-porphyrin/KHSO5 system: AH, substrate (e.g., PCP); Por, porphyrin ligand.

MASAKAZU AOYAMA Faculty of Agriculture and Life Science, Hirosaki University, Hirosaki 036-8561, Japan

To investigate some basic aspects of soil remediation using biomimetic catalysts, the effects of humic substances (HSs) on the removal of xenobiotics, such as pentachlorophenol (PCP), were investigated. The use of a biomimetic catalytic system using tetra(p-sulfophenyl)porphineiron(III) (Fe(III)-TPPS) and potassium monopersulfate (KHSO5) resulted in the disappearance of PCP, accompanied by dechlorination. In addition, this process was enhanced by the presence of several types of HSs. The degrees of enhancement (% δ(PCP)60) achieved by the presence of HSs from peat and compost soils were larger than those in the presence of other types of HSs (tropical peat, brown forest, and ando soils). In control experiments, no PCP disappearance and dechlorination were observed in the presence of only KHSO5, only Fe(III)-TPPS, or combinations of HSs and either KHSO5 or Fe(III)-TPPS. To better understand the role of added HS in enhancing or inhibiting PCP disappearance, correlations between the chemical parameters of the HSs and % δ(PCP)60 were investigated. The most effective HSs had lower carboxylic acid contents and lower degrees of unsaturation. The carboxylic acid content and degree of unsaturation increase with the extent of humification. Therefore, HSs of a lower degree of humification would be predicted to be more useful in enhancing the disappearance of PCP in an Fe(III)-TPPS/ KHSO5 system.

Introduction Soil contamination by xenobiotics represents a serious environmental problem. To overcome such problems, the mechanisms involved in the transformations of xenobiotics in soil environments need to be better understood. To mimic the transformations of xenobiotics via ligninases in soil environments, reactions using an iron(III)-porphyrin complex (Por-Fe(III)) as a catalyst have been examined with respect to the oxidation of chlorinated organic compounds, * Corresponding author telephone: +81-298-61-8328; fax: +81298-61-8326; e-mail: [email protected]. 10.1021/es020681t CCC: $25.00 Published on Web 02/04/2003

 2003 American Chemical Society

such as 2,4,6-trichlorophenol (2,4,6-TrCP) (1-4). The oxidation of organic substrates (AH) by Por-Fe(III) appears to proceed via the peroxide shunt, as illustrated in Figure 1 (5). In the catalytic cycle, single-oxygen donors, such as KHSO5, are required to produce the active oxidant (high-valent ironoxo species) for the organic substrate, which is commonly referred to as an oxoiron(IV)-porphyrin radical cation (Por•+-Fe(IV)dO). Consequently, the reduction of Por•+Fe(IV)dO by AH can yield Por-Fe(IV)dO and a radical species derived from the substrate (A•). The second electron transfer from the substrate to Por-Fe(IV)dO results in the return of Por-Fe(III) to the initial state. On the other hand, in soil environments, humic substances (HSs) play important roles in remedial processes of organic pollutants via a variety of interactions, including solubilization and electron transfer (6, 7). HSs are heterogeneous polyelectrolytes, which contain a variety of functional groups and structural moieties (8). Thus, the chemical characteristics of HSs vary with soil types. Although such differences can be related to the degree of humification (9), this is an obscure concept. However, the degrees of humification for HSs can be characterized on the basis of chemical parameters, such as atomic ratios (e.g., H/C and O/C), contents of acidic functional groups (e.g., carboxylic and phenolic hydroxyl groups), and ratios of absorptivity at 400 or 465 nm to 600 or 665 nm (e.g., log E400/E600 and log E465/ E665) (10-14). These parameters vary significantly for HSs found in different soil types. For example, ando soil with a higher degree of humification shows a lower H/C ratio, but peat soil with a lower degree of humification has a higher H/C (10). In radical-based oxidative reactions (e.g., enzymatic, Fenton, and MnO2 oxidations) such as a Por-Fe(III)/KHSO5 system, the removal of organic pollutants may be enhanced or inhibited in the presence of HSs (15-18). This enhancement can be attributed to the coupling of radical species from organic pollutants with HSs (15, 17). In contrast, inhibition can be attributed to HSs serving as a scavenger of active oxygen species, which can act as a strong oxidant to organic pollutants (18). Thus, HSs may also influence the removal of organic pollutants via biomimetic catalytic systems involving Por-Fe(III). Nevertheless, the influence of HS on the removal of organic pollutants in biomimetic catalytic systems has not been examined in any detail. In addition, the chemical characteristics of HSs, as these relate to the enhancement or inhibition of the removal of organic pollutants via biomimetic catalytic systems, have not been elucidated. VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1031

TABLE 1. Origin of Soils, Elemental Composition, and % δ(PCP)60 for Humic Substances elemental composition origin of soils

abbreviation

%C

%H

%N

%O

%S

% ash

% δ(PCP)60a

Shinshinotsu peat, HA Bibai peat 1, HA Bibai peat 2, HA Inogashira ando soil, FA Dando brown forest soil, FA Inogashira ando soil, HA Dando brown forest, HA Tohro ando soil, HA tropical peat (G. Obos), HA tropical peat (Bukit Batu), HA AIST ando soil, HA compost from rice straw, HA wheat straw-applied soil, HA commercial (Aldrich), HA

SHA BHA1 BHA2 IFA DFA IHA DHA THA GOHA BBHA AIHA CHA WHA AHA

50.02 57.24 50.61 45.54 46.43 53.16 51.03 49.25 58.00 56.43 39.40 53.10 45.65 51.45

5.26 5.70 5.37 4.44 4.39 4.14 5.23 4.60 5.53 5.28 3.01 5.15 4.91 4.77

2.77 2.44 2.78 2.10 1.08 3.89 4.22 1.06 1.11 1.07 1.87 3.68 3.69 0.91

36.89 31.30 37.28 46.00 47.81 38.25 38.23 38.87 34.64 34.26 30.30 36.69 36.04 37.48

0.75 0.58 0.63 0.40 0.29 0.56 0.56 2.60 0.23 0.28 0.45 0.80 0.68 3.30

4.31 2.74 3.33 1.52 ndb nd 0.73 3.62 0.49 2.68 24.97 0.58 9.03 2.09

9.6 ( 0.4 18 ( 0.6 10 ( 0.1 -15 ( 1.0 -21 ( 1.8 -7.9 ( 1.0 -0.1 ( 0.1 -8.2 ( 0.3 -3.8 ( 0.4 -5.4 ( 0.8 -16 ( 1.2 15 ( 0.4 2.6 ( 0.3 1.2 ( 0.1

a

The test for PCP disappearance was conducted to triplicate.

b

nd, not detected.

We first studied the influences of HSs on the removal of organic pollutants via a biomimetic catalytic system that combines a water-soluble Por-Fe(III) with an oxygen donor. Tetra(p-sulfophenyl)porphine-iron(III) (Fe(III)-TPPS) and KHSO5 were used as the Por-Fe(III) and the oxygen donor, respectively. Pentachlorophenol (PCP), which is listed as a priority pollutant by the U.S. EPA (19), has been used in wood preservation and as a herbicide, and this use has led to extensive soil contamination (20). Thus, PCP was used as a model organic pollutant in the present study. Second, to elucidate which factors in HSs play a role in enhancing or inhibiting PCP oxidation, correlations between the degree of PCP disappearance and the chemical characteristics of HSs were investigated.

Experimental Section Materials. Fe(III)-TPPS was prepared by the metalation of free TPPS (Dojindo Laboratory) with FeSO4 following the method of Fleischer et al. (21) and was purified by cationexchange chromatography using Dowex 50W X8-100 (50100 mesh). The elemental analyses are as follows: Calculated for Na3C44H25N4O12S4Fe‚2H2O; C, 48.45%; H, 2.68%; N, 5.14%; S, 11.76%; Fe, 5.12%. Found; C, 48.48%; H, 2.65%; N, 4.94%; S, 11.71%; Fe, 5.20%. An aqueous solution of Fe(III)-TPPS (acetate buffer at pH 4.5) had an absorption maximum at 395 nm, as previously reported in the literature (22). KHSO5 was purchased from Merck. PCP (99.0% purity) was purchased from Nacalai Tesque, and a stock solution (0.01 M) was prepared by dissolution in acetonitrile. Humic Substances. The HSs, which were derived from four types of soils (peat, tropical peat, brown forest, and ando soils), were extracted and purified according to the protocol of the International Humic Substances Society (23). The sources of these materials and their elemental analyses are summarized in Table 1. The humic (HA) and fulvic (FA) acids from Inogashira ando soil and Dando brown forest soil were purchased from the Japanese Humic Substances Society (24). The sodium salt of HA, a commercially available material, was purchased from Aldrich. This was dissolved in aqueous 0.1 M NaOH, and the solution was then treated with a mixture of HF and HCl. The resulting precipitate was transferred to a dialysis tube (molecular mass cutoff 500 Da). After dialysis, the slurry was freeze-dried, resulting in the formation of powders. Analysis of HSs. Elemental Analysis. For all HS samples, elemental compositions (C, H, N, S, and ash contents) were analyzed at the Center for Instrumental Analysis at Hokkaido University (Sapporo, Japan). The percentage of oxygen was calculated by subtracting the sum of the percentages of C, 1032

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003

H, N, S, and ash from 100%. Before the analyses, the powdered HS samples were dehydrated by vacuum-drying for 12 h and then stored in a desiccator over silica gel. Acidic Functional Groups. Total acidity and carboxylic acid content were determined by the Ba(OH)2 and Ca(CH3COO)2 methods, respectively (25). The contents of phenolic hydroxyl groups were calculated by subtracting the carboxylic acid content from the total acidity. Spectroscopic Parameters. The UV-vis absorption spectra of the HSs were determined using a Jasco V-550 type spectrophotometer. The absorptivity (E) of a 50 mg L-1 solution of HS in 0.02 M phosphate buffer at pH 8, was calculated as

E (L cm-1 (g of C)-1) )

absorbance (1) [HS (g L-1)] × % C/100

Gel Permeation Chromatography. The average molecular mass (M) was determined by gel permeation chromatography (GPC). Standard materials having the following molecular masses were used as standards: ovalbumin (44 000 Da), myoglobin (17 000 Da), pullulan P-10 (12 200 Da), pullulan P-5 (5800 Da), and vitamin B12 (1350 Da). A 20-µL aliquot of an aqueous solution of each HS (250 mg L-1, dissolved in 0.1 M phosphate buffer at pH 8) was injected into a Jasco PU-980 type HPLC system (Japan Spectroscopic Co., Ltd.). The mobile phase consisted of a mixture of 0.1 M phosphate buffer (pH 8.0) and 0.1 M NaCl, and the flow rate was set at 0.75 mL min-1. An Ultrahydrogel 250 column (7.6 mm i.d. × 300 mm, exclusion limit 80 000 Da, Waters) was used as the solid phase, and the column temperature was maintained at 25 °C. The void volume of the column was determined to be 5.37 ( 0.02 mL using a thyroglobulin standard (670 000 Da). An SPD-6 UV-vis detector (Shimadzu Co., Ltd.) was used for the detection of HSs and the standards, at a wavelength of 280 nm. A Shodex RI SE-61 refractive index (RI) detector (Showa Denko, Co., Ltd.) was used for detection of the pullulan standards. The delay time between UV-vis and RI detectors was within 0.01 min, as measured using vitamin B12. Test for PCP Disappearance. A 15-mL aliquot of 0.02 M NaH2PO4/Na2HPO4/citrate buffer at pH 6, which contained 50 mg L-1 HS, was placed in a 50-mL Erlenmeyer flask. A 75 µL aliquot of 0.01 M PCP in acetonitrile and a 375 µL aliquot of aqueous Fe(III)-TPPS (200 µM) were added to the buffer solution. Subsequently, 187.5 µL of aqueous 0.01 M KHSO5 was added, and the flask was then shaken in a T-22S-type thermostatic shaking water bath (Thomas Kagaku Co., Ltd.) at 25 ( 0.1 °C. After the reaction period, a 800 µL aliquot

of the test solution was mixed with 400 µL of 2-propanol. It was confirmed that 2-propanol prevented any further disappearance of PCP for periods of up to 10 h. Hence, 2-propanol was used to quench the reaction. To analyze the level of PCP in the test solution, a 20 µL aliquot of the mixture was injected into the HPLC system. The mobile phase consisted of a mixture of 0.08% aqueous H3PO4 and methanol (20/80 ) v/v). A 5C18-MS Cosmosil packed column (4.6 mm i.d. × 250 mm, Nacalai Tesque) was used as a solid phase, and the column temperature was maintained at 50 °C. PCP was measured by UV-vis absorption at a wavelength of 220 nm. The detection limit of PCP in the HPLC analysis was calculated to be 0.5 µM. The percentage of PCP disappearance (% PCP disappearance) can be calculated using the relationship shown below:

% PCP disappearance )

[PCP]0 - [PCP]t [PCP]0

× 100 (2)

where [PCP]0 and [PCP]t represent the PCP concentrations added initially and at an arbitrary reaction period, respectively. In addition, chloride ions (Cl-), released during the disappearance of PCP, were analyzed by means of ion chromatography (DX-500-type, Dionex) using an IonPac AS11 column. The detection limit for Cl- was calculated to be 7.7 µM. To identify PCP-derived byproducts, the reaction mixture at 60 min was extracted with n-hexane, and the extract was analyzed by GC/MS using a previously described method (17).

Results and Discussion Influences of Solution Conditions. Reaction conditions, such as pH, Fe(III)-TPPS, KHSO5, and HS concentrations, can influence the oxidation of PCP. We first investigated the influences of such the conditions on PCP disappearance and Cl- release. In our previous study, BHA1 was used for the oxidative degradation of PCP via photo-Fenton processes (17, 26). In these studies, PCP oxidation was enhanced by the presence of BHA1. This can be attributed to the covalent binding of reaction intermediates from PCP, such as chlorophenoxy radicals, to the phenolic moieties of BHA1. As shown in Figure 1, PCP disappearance in the Fe(III)-TPPS/ KHSO5 system is also due to radical-based oxidation. Thus, the addition of BHA1 would be expected to enhance the oxidation of PCP in the Fe(III)-TPPS/KHSO5 system. Therefore, in the following investigations of the influence of solution conditions, BHA1 was used as a reference HS. Kinetics of PCP Disappearance and Cl- Release. Figure 2a shows the effect of BHA1 concentration on the kinetics of PCP disappearance at pH 6.0. The rate of disappearance of PCP was rapid during the first 1 min and then gradually decreased. On the basis of our current knowledge of the oxidation of 2,4,6-TrCP via the Fe(III)-TPPS/KHSO5 system, the formation of 2,4-dichloroquinone via a chlorophenoxy radical constitutes the initial rapid step and occurs within 1 min (1, 3). Thus, the rapid part of the biphasic kinetic curve shown in Figure 2a can be attributed to the formation of quinone derivatives via the pentachlorophenoxy radical (PCP•). On the other hand, PCP can be oxidized to dimers, such as hydroxyl-nonachlorodiphenyl ethers (H-NCDEs) and octachlorodibenzo-p-dioxin (OCDD), via a reaction between PCP and PCP• (17). Thus, the slow part of the kinetic curve likely corresponds to the oxidation of PCP by organic radical species, such as PCP•. If PCP• were produced during the initial step of PCP oxidation, quinone derivatives, H-NCDEs, and OCDD should be present at detectable levels in the reaction mixture. We therefore attempted to identify such byproducts by GC/MS. o-Tetrachloroquinone (o-TeCQ), 2H-NCDE, 4HNCDE, and OCDD were detected in the reaction mixture.

FIGURE 2. Effect of BHA1 concentration on % PCP disappearance (a) and on the release of Cl- (b) in the Fe(III)-TPPS/KHSO5 system. [PCP]0, 50 µM; [Fe(III)-TPPS], 5 µM; [KHSO5], 125 µM; pH 6.0. These results are consistent with PCP• serving as the initial oxidation intermediate. In addition, as shown in Figure 2b, Cl- was simultaneously released at a rapid rate during the first 1 min of the reaction. This tendency in Figure 2b is consistent with the data shown in Figure 2a. These results indicate that the disappearance of PCP can be attributed to a catalytic oxidative dechlorination by the Fe(III)-TPPS/ KHSO5 system. The molar ratios of [Cl-] released to the amount of [PCP] disappeared were in the range of 1.31.6. Therefore, the formation of byproducts, in which 1-2 chlorine atoms have been released from PCP, would be expected. Therefore, the byproducts described above are reasonable. We also investigated the effect of added BHA1 on the percentage of PCP disappearance. In the absence of BHA1, 63% of the PCP disappeared at 60 min, increasing to 87% for [BHA1] ) 50 mg L-1 and to 95% for [BHA1] ) 100 mg L-1. The [Cl-] at 60 min also increased with BHA1 concentration. However, in the presence of only BHA1, no PCP disappearance or Cl- release was observed up to 60 min. Thus, the disappearance of PCP and the release of Cl- via the Fe(III)TPPS/KHSO5 system are enhanced by the addition of BHA1. Because a measurement of the extent of PCP disappearance within 1 min was not possible, the kinetic interpretation of the data shown in Figure 2a was difficult. Thus, the efficiency of PCP removal is expressed as the percentage of the complete PCP disappearance. To quantitatively evaluate the effects of HSs on PCP disappearance, the percentage of PCP disappearance at 60 min in the absence of HSs was subtracted from that in the presence of HSs. Therefore, we introduced the % δ(PCP)60 value, which indicates the degree of enVOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1033

FIGURE 3. Effect of pH on % PCP disappearance (a) and on the concentration of Cl- (b). [PCP]0, 50 µM; [Fe(III)-TPPS], 5 µM; [KHSO5], 125 µM; [BHA1], 50 mg L-1; reaction time, 60 min. hancement of PCP disappearance as the result of the addition of HS, as follows:

% δ(PCP)60 ) (% PCP disappearance in the presence of HS at 60 min) - (% PCP disappearance in the absence of HS at 60 min) (3) Influence of pH. The pH of soil has been reported to be in the range of 3-7 (27). Therefore, influence of pH on PCP disappearance and Cl- release was investigated in that pH range. Figure 3a,b shows the effects of pH on the percentage of PCP disappearance and on the release of Cl-. In the presence and absence of BHA1, the percentage of PCP disappearance and the [Cl-] increased with increasing pH. Because the pKa value of PCP is known to be 4.75, the majority of the PCP would be dissociated to PCP- at pH values above 5. Moreover, Fe(III)-TPPS also has an anionic charge over a wide range of pH values. It would, therefore, be expected that the reactivity of PCP would be reduced at a higher pH because of electrostatic repulsion between PCP- and Fe(III)-TPPS. However, the results shown in Figure 3 were at variance with this expectation. As described in the literature (1, 3, 5), Por•+-Fe(III) may initially attack to PCP via electrophilic addition. Therefore, the increase in PCP disappearance with pH can be attributed to PCP• rather than PCP serving as an electron donor to Por•+-Fe(III). An enhancement in PCP disappearance in the presence of BHA1 was observed over a relatively wide pH range. The degree of enhancement of PCP disappearance as the result of the addition of BHA1, % δ(PCP)60, increased up to pH 6 and then decreased at pH 7. Influence of Fe(III)-TPPS and KHSO5 Concentrations. Figure 4a shows the effect of Fe(III)-TPPS concentration on the percentage of PCP disappearance and on the % δ(PCP)60 at pH 6. Control data in the absence of Fe(III)-TPPS 1034

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003

FIGURE 4. Effect of Fe(III)-TPPS (a) and KHSO5 (b) concentrations on % PCP disappearance. [PCP]0, 50 µM; pH 6; [BHA1], 50 mg L-1; reaction time, 60 min. (a) [KHSO5], 125 µM; (b) [Fe(III)-TPPS] 5 µM. confirmed that no PCP disappearance and dechlorination occurred in the presence of KHSO5 only or in the presence of both KHSO5 and BHA1 (data points at [Fe(III)-TPPS] ) 0 are not shown in Figure 4a). As shown in Figure 4a, the percentage of PCP disappearance increases with increasing concentrations of Fe(III)-TPPS up to 10 µM in both the absence and the presence of BHA1. However, the percentage of PCP disappearance and % δ(PCP)60 significantly decreased at [Fe(III)-TPPS] ) 50 µM. Shukla et al. (3) reported that the rate of degradation of 2,4,6-TrCP increased linearly with increasing concentrations of Fe(III)-TPPS, in which the ratio of substrate to catalyst was in the range of 170-2500. In our studies, these ratios were in the range of 1-100, and the amounts of catalyst were relatively larger than those reported in previous studies (3). This indicates that PCP is more resistant than 2,4,6-TrCP to oxidation. Moreover, we observed a decrease in the absorbance of the Soret band for Fe(III)TPPS (395 nm) after the reaction. This indicates that PCP oxidation competes with the degradation of Fe(III)-TPPS in the present system. Therefore, the decrease in PCP disappearance at [Fe(III)-TPPS] ) 50 µM in Figure 4a can be attributed to the inhibition of PCP oxidation as the result of competition between PCP oxidation and Fe(III)-TPPS degradation. Figure 4b shows the effect of variations in KHSO5 concentration on the reaction. In the control experiment, in the absence of KHSO5, no PCP disappearance and dechlorination were observed in the presence of Fe(III)-TPPS only or in the presence of both BHA1 and Fe(III)-TPPS (data points at [KHSO5] ) 0 are not shown in Figure 4b). The percentage of PCP disappearance increased with KHSO5 concentration. In the case of variations in KHSO5 concentration, an enhancement in PCP disappearance by the presence of BHA1 was consistently observed. Therefore, the above

FIGURE 5. Correlation between % δ(PCP)60 and O/C ratio. findings lead to the conclusion that PCP disappearance is enhanced by the addition of BHA1 to the Fe(III)-TPPS/ KHSO5 system. Enhancement and Inhibition of PCP Disappearance by HSs. The enhancement effect of BHA1 on the disappearance of PCP may be related to interactions between some moieties in HSs and PCP or Fe(III)-TPPS. We had expected that the oxidative coupling between phenolic moieties in BHA1 and radical species from PCP might enhance the disappearance of PCP. Moreover, it has been reported that the addition of phenolic moieties in HS, such as protocatechuic acid, enhances the photo-Fenton and peroxidase oxidations of PCP (15, 26). Thus, we examined the addition of model phenolic acids (125 µM), such as catechol, p-hydroquinone, gallic, caffeic, ferulic, syringic, vanillic, and protocatechuic acids, to the Fe(III)-TPPS/KHSO5 system at pH 6. However, no significant enhancements in PCP disappearance were observed (% δ(PCP)60 ) -1.5 ( 2.7%). This suggests that, in the Fe(III)-TPPS/KHSO5/BHA1 system, the enhancement of PCP disappearance cannot be due to oxidative coupling between PCP-derived radicals and phenolic moieties in BHA1. Thus, the degrees of enhanced or inhibited PCP disappearance are controlled by the types of HSs present. We examined a variety of HSs extracted from different soil origins. The % δ(PCP)60 values for 14 types of HSs are summarized in Table 1. The negative and positive values for the % δ(PCP)60 represent the inhibition and enhancement of PCP disappearance, respectively. Inhibition was observed in the case of AIHA, DHA, DFA, IHA, IFA, GOHA, BBHA, and THA, the origins of which were ando soil, brown forest soil, and tropical peat. In contrast, enhancement was clearly observed for BHA1, BHA2, SHA, and CHA, the origins of which were peat and compost soils. The addition of AHA (commercial) and WHA (wheat straw-applied soil) led to a small enhancement in PCP disappearance. Correlation Studies. To elucidate the roles of HSs in enhancing or inhibiting PCP disappearance, we investigated correlations between % δ(PCP)60 and the chemical characteristics of each HS. The characteristics of HSs can be characterized by their degree of humification. For example, the H/C or O/C ratio decreases or increases with the extent of humification (10, 11). Thus, we initially investigated correlations between % δ(PCP)60 and the above atomic ratios. The correlation between H/C and % δ(PCP)60 indicates a tendency for % δ(PCP)60 to increase with increasing hydrogen content (Supporting Information, Figure SI-1). Such a tendency suggests that the degrees of humification and unsaturation are factors in the enhancement of PCP disappearance. The O/C ratio was negatively correlated to % δ(PCP)60 (Figure 5). In addition, the O/C ratio is known to serve as one of the indices for the content of oxygencontaining functional groups, such as carboxylic, carbonyl, alcoholic and phenolic hydroxyl, and methoxyl moieties (11).

FIGURE 6. Correlation between % δ(PCP)60 and carboxylic group content (COOH).

FIGURE 7. Correlation between % δ(PCP)60 and log E400/E600. Thus, a negative correlation between the % δ(PCP)60 and the O/C ratio suggests that %δ(PCP)60 increases with a decrease in the content of oxygen-containing functional groups. To explain the correlation with the O/C ratio shown in Figure 5 more precisely, relationships between % δ(PCP)60 and the content of acidic functional groups were investigated. Tsutsuki et al. (11) reported that, during humification, the content of carboxyl groups increased while that of phenolic hydroxyl groups decreased. The content of carboxylic groups was negatively correlated to % δ(PCP)60 (Figure 6), indicating that an enhanced PCP disappearance tends to increase with decreasing carboxylic group content. However, the content of phenolic hydroxyl groups did not correlate to % δ(PCP)60 (Supporting Information, Figure SI-2). Phenolic hydroxyl groups are operationally defined (25) and can include some moieties of aliphatic hydroxyl groups (e.g. carbohydratelike). This suggests that the relation to actual phenolic hydroxyl groups is obscured in the present experiment. Thus, the tendency of an increase of % δ(PCP)60 with decreasing O/C can be attributed to a decrease in the content of carboxylic groups. These results also support the finding that enhanced PCP disappearance is related to the degree of humification of HSs. To clarify the effect of the degree of humification on the enhancement of PCP disappearance, a possible correlation with a spectroscopic parameter, log E400/E600, was examined. In general, the spectroscopic parameters (log E400/E600 and log E465/E665) decrease with the extent of humification (12, 14). As shown in Figure 7, the log E400/E600 values were positively correlated to % δ(PCP)60. However, the data points corresponding to the FA samples (IFA and DFA) were eliminated from the correlation interpretation. The spectroscopic parameter (log E465/E665) increases with increasing average molecular mass of HA (13). In our data, the logarithm VOL. 37, NO. 5, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

1035

of the average molecular mass (log M) was positively correlated to log E400/E600, except for the FA samples (Supporting Information, Figure SI-3). This proves that log E400/E600 increases with increasing log M. In the same soil origin, the molecular weight of FA can generally be lower than that of HA, but the spectroscopic parameter (log E465/ E665) for FA is reported to be higher than that for HA (14). This indicates that the spectroscopic parameters for FA cannot be used as indices of the degree of humification. Therefore, data points corresponding to the FA samples were outliers in the correlation between log E400/E600 and % δ(PCP)60. However, the positive correlation between % δ(PCP)60 and log E400/E600 shows that the enhancement in PCP disappearance decreases with the extent of humification in a series of HA samples. The interpretations described above lead to the conclusion that the addition of HSs of a lower degree of humification and unsaturation can be effective in enhancing PCP disappearance in the Fe(III)-TPPS/KHSO5 system. This suggests that favorable HSs in enhancing PCP disappearance are less aromatic in nature. 13C and 1H NMR studies have confirmed that the logarithm of the absorptivity at 280 nm (log E280) is positively correlated to the content of aromatic moieties in HS (28, 29). This indicates that log E280 increases with increasing content of aromatic moieties in HS. Thus, a correlation between % δ(PCP)60 and log E280 was examined. As a result, log E280 was negatively correlated to % δ(PCP)60, except for IFA and DFA (r2 ) 0.58, Supporting Information, Figure SI-4). This indicates that. % δ(PCP)60 tends to increase with decreasing content of aromatic moieties. For the correlation between % δ(PCP)60 and log E280, the data points from FAs were also outliers. In our data set, log E280 was negatively correlated to H/C ratios, except for IFA and DFA (Supporting Information, Figure SI-5). This suggests that log E280 decreases with increasing aliphatic moieties in HSs. Because FAs are generally more aliphatic nature than HAs, the log E280 values of FAs are lower than those of HAs. As shown in Figure 6, the content of carboxylic groups in FAs were larger than those in HAs, and % δ(PCP)60 decreased with increasing contents of carboxylic groups. Thus, despite of lower log E280, data points for the FA samples were outliers in the correlation between log E280 and % δ(PCP)60. For enhancing PCP removal in the Fe(III)-TPPS/KHSO5 system, favorable types of HSs are from peat and compost soils, and such HSs contain lower amounts of aromatic moieties and carboxylic groups. In contrast, 8 of the HSs samples caused the inhibition of PCP disappearance. Such HSs were derived from tropical peat, brown forest, and ando soils, and their characteristics include higher degrees of unsaturation and humification. It is known that peat soil HAs, which are of a lower degree of humification, include large amounts of polysaccharide-like moieties (30, 31). Thus, such aliphatic portions of HSs may contribute to the enhancement of some interactions between PCP and Fe(III)-TPPS.

Acknowledgments This work was supported by Grants-in-Aid for Scientific Research in Japan Society for the Promotion of Science (14380283). We thank Dr. S. Pongpiajun (The University of Birmingham, U.K.) for providing HA samples from tropical peat (BBHA and GOHA).

Supporting Information Available Five figures showing different correlations. This material is available free of charge via the Internet at http://pubs.acs.org.

1036

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 5, 2003

Literature Cited (1) Labat, G.; Seris, J.-L.; Meunier, B. Angew. Chem., Int. Ed. Engl. 1990, 29, 1471-1473. (2) Hasan, S.; Cho, J.-G.; Sublette, P. D.; Maule, A. J. Biotechnol. 1992, 24, 195-201. (3) Shukla, R. S.; Robert, A.; Meunier, B. J. Mol. Catal. A 1996, 113, 45-49. (4) Chen, S.-T.; Stevens, D. K.; Kang, G. Water Res. 1999, 33, 36573665. (5) Sheldon, R. A. In Metalloporphyrins in Catalytic Oxidations; Dekker: New York, 1994; pp 12-13. (6) Perminova, I. V.; Grechishcheva, N. Y.; Kovalevski, D. V.; Kudryavtsev, A. V.; Petrosyan, V. S.; Motorin, D. N. Environ. Sci. Technol. 2001, 35, 3841-3848. (7) Curtis, G. P.; Reinhard, M. Environ. Sci. Technol. 1994, 28, 23932401. (8) Gaffney, J. S.; Marley, N. A.; Clark, S. B. In Humic and Fulvic Acids. Isolation, Structure, and Environmental Role; Gaffney, J. S., Marley, N. A., Clark, S. B., Eds.; ACS Symposium Series 651; American Chemical Society; Washington, DC, 1996; Chapter 1. (9) Tsutsuki, K.; Kondo, R. Soil Sci. Plant Nutr. 1995, 41, 515-527. (10) Kuwatsuka, S.; Tsutsuki, K.; Kumada, K. Soil Sci. Plant Nutr. 1978, 24, 337-347. (11) Tsutsuki, K.; Kuwatsuka, S. Soil Sci. Plant Nutr. 1978, 24, 547560. (12) Kumada, K. Soil Plant Food 1955, 1, 29-30. (13) Power, J. F.; Langford, C. H. Anal. Chem. 1988, 60, 842-846. (14) Chefetz, B.; Tarchitzky, J.; Benny, N.; Hatcher, P. G.; Bortiatynski, J.; Chen, Y. In Humic Substances. Structure, Properties and Uses; Davies, G., Ghabbour, E., Eds.; The Royal Society of Chemistry; Cambridge, 1998; pp 69-78. (15) Morimoto, K.; Tatsumi. K. Chemosphere 1997, 34, 1277-1283. (16) Paciolla, M. D.; Davies, G.; Jansen, S. A. Environ. Sci. Technol. 1999, 33, 1814-1818. (17) Fukushima, M.; Tatsumi, K. Environ. Sci. Technol. 2001, 35, 1771-1778. (18) Klausen, J.; Haderlein, S. B.; Schwarzenbach, R. P. Environ. Sci. Technol. 1997, 31, 2642-2649. (19) Seronero, L. C.; Laespada, M. E. F.; Pavo´n, J. L. P.; Cordero, B. M. J. Chromatogr. A 2000, 897, 171-176. (20) Laine, M. M.; Ahtiainen, J.; Wågman, N.: O ¨ berg, L. G.; Jørgensen, K. S. Environ. Sci. Technol. 1997, 31, 3244-3250. (21) Fleischer, E. B.; Palmer, J. M.; Srivastava, T. S. Chatterjee, A. J. Am. Chem. Soc. 1971, 93, 3162-3167. (22) Nomura, T.; Orita, M. Bunseki Kagaku 1979, 28, 377-379. (23) Swift, R. S. In Methods of Soil Analysis Part 3. Chemical Methods; Sparks, D. L., Page, A. L., Helmke, P. A., Loeppert, R. H, Soltanpour, P. N., Tabatabai, M. A., Johnson, C. T., Summer, M. E. Eds.; Soil Science Society of America: Madison, WI, 1996; pp 1018-1020. (24) Watanabe, A.; Itoh, K.; Kuwatsuka, S. Soil Sci. Plant Nutr. 1994, 40, 601-608. (25) Schnitzer, M.; Khan, S. U. In Humic Substances in the Environment; Dekker: New York, 1972; pp 37-43. (26) Fukusima, M.; Tatsumi, K.; Morimoto, K. Toxicol. Environ. Chem. 2001, 79, 9-21. (27) Thomas, G. W. In Methods of Soil Analysis Part 3. Chemical Methods; Sparks, D. L., Page, A. L., Helmke, P. A., Loeppert, R. H., Soltanpour, P. N., Tabatabai, M. A., Johnson, C. T., Summer, M. E. Eds.; Soil Science Society of America: Madison, WI, 1996; Chapter 16. (28) Chin, Y.-P.; Aiken, G.; O’Loughlin, E. Environ. Sci. Technol. 1994, 28, 1853-1858. (29) Tanaka, S.; Oba, K.; Fukushima. M.; Nakayasu, K.; Hasebe, K. Anal. Chim. Acta 1997, 337, 351-357. (30) Killopes, S. D.; Killopes, V. J. In An Introduction to Organic Geochemistry; Longman Scientific & Technical: London, 1993; pp 93-106. (31) Tsutsuki, K.; Kondo, R. Jpn. J. Soil Sci. Plant Nutr. 1997, 68, 45-51.

Received for review April 9, 2002. Revised manuscript received December 12, 2002. Accepted December 18, 2002. ES020681T