Suppressive effect of soil application of carbonaceous adsorbents on dieldrin uptake by cucumber fruits

Abstract

The use of aldrin and dieldrin as pesticides was prohibited in 1975 in Japan. However, some of the soils still remain contaminated with dieldrin, because aldrin is easily oxidized to dieldrin and dieldrin is extremely stable in soil. In recent years, dieldrin at concentrations exceeding the limit set by the Food Sanitation Law of Japan (dieldrin < 0.02 mg kg^–1 fresh weight) has been detected in cucumber fruits produced in some areas of Japan. We examined the effect of the soil application of selected adsorbents on reducing dieldrin concentrations in cucumber fruits in three steps of pot experiments. Among the three types of biochar made from wood chip, rice husk, and bamboo, wood chip charcoal was found to be the most effective (pot experiment 1). The effect of wood chip charcoal was enhanced by high-temperature burning and crushing (pot experiment 2). However, the effect of activated carbon was superior to that of optimized (high-temperature-treated and crushed) wood chip charcoal (pot experiment 3). Therefore, activated carbon was selected as the most effective adsorbent. The effect of activated carbon to reduce dieldrin concentrations in cucumber fruits was confirmed in a field experiment, and the effect continued to a certain extent for at least four years after the application. We calculated the cost of activated carbon necessary to maintain a sufficient suppressive effect in the field, and this cost would appear to be acceptable to cucumber farmers. Consequently, application of activated carbon to dieldrin-contaminated soils can be considered a promising practical technique for reducing dieldrin concentrations in cucumber fruits.

Keywords:

• dieldrin
• cucumber
• activated carbon
• adsorbent
• biochar

Introduction

Aldrin and dieldrin, categorized as a group of persistent organic pollutants (POPs) (United Nations Environment Program 2001), are cyclodiene-type insecticides used on crops. In Japan, aldrin and dieldrin were registered in 1954, and were used extensively on arable land for insect control. Although they were not manufactured in Japan, 3300 t of aldrin and 683 t of dieldrin were imported from 1958 to 1972 (Japan Plant Protection Association 1958–1972). Because of their extreme persistence in the environment (Nash and Woolson 1967) and the fact that their accumulation in crops poses a potential threat to human health (World Health Organization 1989), the Japanese Government banned the use of these chemicals for food crops in 1971, and their use as pesticides was prohibited in 1975. Aldrin is easily oxidized to dieldrin in soil (Barrie et al. 1992), and dieldrin is extremely stable in soil (Ritter et al. 1998; Meijer et al. 2001). As a consequence, soils continue to remain contaminated with dieldrin (Seike et al. 2007), even though aldrin and dieldrin have not been used for the past 40 years. In recent years, dieldrin at concentrations exceeding the limit set by the Food Sanitation Law of Japan (dieldrin < 0.02 mg kg⁻¹ fresh weight) has been detected in cucumber fruits produced in some areas of Japan (Kondo et al. 2003). It is known that cucurbits take up considerable amounts of dieldrin from the soil (Lichtenstein and Schulz 1965; Lichtenstein et al. 1965; Yamamoto et al. 1973; Suenaga 1973; Nakamura 1990). In our previous study, we found that only Cucurbitaceae plants took up appreciable amounts of dieldrin from the soil among 17 plant families (Otani et al. 2007). Therefore, changing from cucumber to non-cucurbits would appear to be an effective strategy for avoiding dieldrin pollution in food crops. However, cucumber is a profitable crop, and therefore farmers require an alternative means of reducing dieldrin uptake in cucumber plants.

It is known that organic pollutants are strongly adsorbed to carbonaceous sorbents such as black carbon and biochar (Bucheli and Gustafsson 2000; Cornelissen et al. 2006; Koelmans et al. 2006; Ghosh 2007; Yu et al. 2009). Further, some reports have suggested that activated carbons (ACs) are effective in suppressing dieldrin uptake by cucumber (Hashimoto 2007; Hilber et al. 2009). Cucumber farmers consider that AC is relatively expensive compared with those of biochar such as wood chip charcoal. Therefore, it is necessary to compare the suppressing effect of biochar with AC in reducing dieldrin concentrations in cucumber fruits. And it is also important to know how long the effect of the adsorbent continues after the application in a dieldrin contaminated field. However, there was little information about these problems.

Thus, in pot experiments, we investigated the capacity of carbonaceous adsorbents (biochars and AC) to sequester dieldrin, attempted to optimize the capacity of a biochar, and compared the suppressive effect of the optimized biochar with that of AC on dieldrin uptake in cucumber fruits. On the basis of the results of these experiments, we selected a practical adsorbent and then examined its effect in a field experiment for four years. And we also calculated the cost of the selected adsorbent necessary to maintain a sufficient suppressive effect in the field in order to consider its practicality.

Materials and Methods

Adsorbents

Four types of carbonaceous adsorbents (AC, wood chip charcoal, rice husk charcoal, and bamboo charcoal) were used in this study. Powdered AC, SS1®, was purchased from Ajinomoto Fine-Techno Co., Inc., Kanagawa, Japan. Wood chip charcoal burned at 450°C [(WC450), MC450®] and 1000°C [(WC1000), MC1000®] were purchased from Date Forestry Association, Fukushima, Japan. Crushed wood chip charcoal burned at high temperature (WC1000C) was prepared by grinding MC1000® in a mortar. Rice husk charcoal (RC) and bamboo charcoal (BC) produced by the Fukushima Agricultural Technology Center were also used. These adsorbents were passed through sieves of increasing mesh size (75, 106, 250, 425, 850, and 2000 µm) in order to determine the particle size distributions (JIS A1204). The pH (in H₂O) of these adsorbents was measured at a 1 : 10 (solid/water) ratio. The carbon and nitrogen contents were measured using the dry combustion method (Sumigraph NC Analyzer NC-220F; Sumika Chemical Analysis Service, Ltd, Osaka, Japan). The phosphorus contents were measured by wet digestion (nitric-perchloric acids) and vanadomolybdate methods. The potassium contents were measured by wet digestion (hydrofluoric-perchloric acids) and determined by an atomic absorption spectrophotometer (AA280FS; Varian Technologies Japan Ltd, Tokyo, Japan). The specific surface areas (SSAs) were determined by nitrogen adsorption at −196°C using a surface area analyzer (Autosorb-1; Quantachrome Instruments, FL, USA).

Glasshouse pot experiments

The contaminated soil, an Andosol (FAO/UNESCO), was collected from the plowed layer (0–15 cm depth) of a farmland in Japan in 2003. The field was used to cultivate white asparagus, and received regular application of aldrin for insect control from the late 1960s to the early 1970s. The collected soil was air-dried, fully mixed, and passed through a 2 mm sieve. The resultant preparation was used for the pot experiments. Dieldrin concentration in the soil was 0.11 mg kg⁻¹ dry weight (dw); however, no aldrin was detected (<0.001 mg kg⁻¹ dw).

A sample of soil (10.2 kg dw) was placed in a plastic pot (volume 15 L; diameter 25.2 cm; height 30 cm). And tested carbonaceous adsorbents were then applied to the soil. The treatments of adsorbents application will be described later in detail. These adsorbents were fully mixed with the soil. The soil in each pot was fertilized with a high analysis compound fertilizer (N–P₂O₅–K₂O: 15–15–15) at a rate of 66.7 g per pot. Grafted cucumbers, consisting of a Cucurbita maxima Duch. cv. Tokiwa PowerZ2 rootstock and a Cucumis sativus L. cv. Kinsei114 scion, were used as test plants. We used grafted cucumbers because in Japan cucumber is generally cultivated using the grafting technique with Cucurbita sp. as a rootstock. The tested plants of grafted cucumber were grown in the pots with or without the carbonaceous adsorbents application, and the cucumber fruits were harvested from the seventh to the ninth nodes on the main shoots. The detail about the cultivation way will be also described later. All the pot experiments were conducted in a glasshouse at Fukushima Agricultural Technology Center under natural sunlight conditions. All the treatments were replicated three times. Fruit samples were stored at −20°C until extraction for dieldrin analysis.

Pot experiment 1

This experiment was conducted from April 27 to June 25, 2005. During the experiment, the mean air temperature was 21°C. Three kinds of biochars as carbonaceous adsorbents were applied to the soil. The amounts of adsorbents applied to the soil were as follows: 0.22 kg dw of WC450, 0.22 kg dw of RC, and 0.23 kg dw of BC per pot. The pots of no adsorbents application were also prepared.

Seeds of the tested plants were germinated in uncontaminated culture soil. Nine days (Cucumis sativus) and eight days (Cucurbita maxima) after sowing, grafted plants were made from the Cucurbita maxima as a rootstock and Cucumis sativus as a scion, and grown in the uncontaminated culture soil. Twelve days after grafting, one grafted plant was transplanted into each pot. Twenty-eight to 38 days after transplanting, the cucumber fruits were harvested.

Pot experiment 2

This experiment was conducted from May 8 to July 6, 2006. During the experiment, the mean air temperature was 22°C. Three kinds of wood chip charcoal with different treatments were applied to the soil. The amounts of adsorbents applied to the soil were as follows: 0.11 kg dw of WC450, WC1000, and WC1000C per pot. The pots of no adsorbents application were also prepared.

Seeds of the tested plants were germinated in uncontaminated culture soil. Eight days (Cucumis sativus) and seven days (Cucurbita maxima) after sowing, grafted plants were made from the Cucurbita maxima as a rootstock and Cucumis sativus as a scion, and grown in the uncontaminated culture soil. Fifteen days after grafting, one grafted plant was transplanted into each pot. Twenty-eight to 37 days after transplanting, the cucumber fruits were harvested.

Pot experiment 3

This experiment was conducted at the same time as pot experiment 2. Different rate of AC and WC1000C were applied to the soil. The amounts of adsorbents applied to the soil were as follows: 6.25, 12.5, 25, and 50 g dw of AC and 54.5, 109, and 218 g dw of WC1000C per pot. The pots of no adsorbents application were also prepared.

Seeds of the tested plants were germinated in uncontaminated culture soil. Eight days (Cucumis sativus) and seven days (Cucurbita maxima) after sowing, grafted plants were made from the Cucurbita maxima as a rootstock and Cucumis sativus as a scion, and grown in the uncontaminated culture soil. Fifteen days after grafting, one grafted plant was transplanted into each pot. Twenty-eight to 37 days after transplanting, the cucumber fruits were harvested.

Field experiment

A field study was conducted in farmland in Japan. The soil used in the pot experiments was collected from this field. The soil from the surface to a depth of 100 cm (20 cm intervals) in this field is an Andosol with a pH 5.0–5.5, carbon content of 36 to 38 g kg⁻¹, and soil texture of clay loam (from 0 to 80 cm) and light clay (from 80 to 100 cm). Dieldrin residue in the soil was found to a depth of 80 cm, with concentrations ranging from 0.006 to 0.08 mg kg⁻¹ (Fig. 1). Activated carbon (AC), which was effective in suppressing dieldrin uptake by grafted cucumbers in the pot experiment, was applied at 0, 0.25, and 0.5  kg dry weigth m⁻² in this field in plots of 2.5 m × 5.0 m (12.5 m²) on May 31, 2006. We used just a single application of AC. The experiment of the grafted cucumber cultivation was conducted in 2006, 2007, and 2009. In 2008, the tested field was kept as fallow condition. The field was then fully plowed three times from the surface to a depth of 15 cm, and the soil in each plot was fertilized with a high analysis compound fertilizer (N–P₂O₅–K₂O: 15–15–15) at a rate of 200 g m⁻² in each year of cucumber cultivation.

Figure 1. The vertical distribution of dieldrin residue in soil on this farm. Error bars represent ± standard error (n = 3).

The grafted cucumbers consisted of a Cucurbita maxima Duch. cv. Hikari Power Gold as a rootstock and a Cucumis sativus L. cv. Frontier as a scion. Seedlings of the tested plants were made in a glasshouse at Fukushima Agricultural Technology Center under natural temperature and light conditions. On May 8, 2006, May 7, 2007, and May 3, 2009, seeds of Cucumis sativus were sown in uncontaminated culture soil. One day after Cucumis sativus sowing, seeds of Cucurbita maxima were also sown in the same manner. Eight days (Cucumis sativus) and seven days (Cucurbita maxima) after sowing, grafted plants were made from the Cucubita maxima as a rootstock and Cucumis sativus as a scion, and grown in the uncontaminated culture soil. The grafted seedlings were then transplanted in the field on May 31, 2006 and 2007, and June 7, 2009. The density was 15 plants per plot. The grafted cucumbers were grown under open culture conditions. The cucumber fruits were harvested from the seventh to the ninth nodes on the main shoots from July 6 to July 10, 2006, June 25 to July 3, 2007, and June 29 to July 7, 2009. All the treatments were replicated three times. Fruit samples were prepared for analysis as in the pot experiments.

Soil and plant analysis

Dieldrin contents

Soil samples were air-dried and sieved (<2 mm). A 10 g sample of dried soil mixed with 10 mL of water was shaken with 150 mL of acetone (Wako Pure Chemical Industries, Osaka, Japan) for 30 minutes at room temperature. The extract was filtered through a filter paper (No. 5B; Kiriyama Glass, Tokyo, Japan) and passed through a column filled with Celite 545® (Kanto Kagaku, Tokyo, Japan). Cucumber samples were chopped finely, and 20 g fresh weight (fw) was shaken with 150 mL of acetone for 30 minutes at room temperature. The extract was filtered through a filter paper. The soil and plant extracts were concentrated to 20 mL at 40°C using a rotary evaporator and subsequently applied to a column containing diatomaceous earth (Chem Elut®; Varian, Palo Alto, CA). Dieldrin was eluted with 120 mL of n-hexane (Wako Pure Chemical Industries, Osaka, Japan). The eluent solution was concentrated using a rotary evaporator and then applied to a column containing florisil (5 g) (SepPak FL®; Waters, Milford, MA). Dieldrin was eluted with 50 mL of 15% (v/v) diethylether (Wako Pure Chemical Industries)–n-hexane. The solution was concentrated to 2 mL under a gentle stream of nitrogen for measurement of dieldrin.

The amount of dieldrin was measured using a gas chromatography/µ electron capture detector (GC/µECD, HP6890; Agilent Technologies, Santa Clara, CA) equipped with a capillary column (ENV-8MS, 30 m × 0.25 mm i.d. × 0.25 µm film thickness; Kanto Kagaku, Tokyo, Japan). The GC/µECD measurement conditions are shown in Table 1. The limits of detection for dieldrin in soil and cucumber were 0.001 mg kg⁻¹ dw and 0.001 mg kg⁻¹ fw, respectively. The recovery of initially added dieldrin standard (0.5 µg per sample) was 87.9 ± 5.3% in soil, and 83.2 ± 3.2% in fruit, respectively. The reference material for dieldrin in soil (JSAC0441), obtained from the Japan Society for Analytical Chemistry, was analyzed for quality assurance in triplicate. The result was 0.076 ± 0.003 mg kg⁻¹ dw (coefficient of variance, 0.27%), which was within the range of the specified value (0.076 ± 0.014 mg kg⁻¹ dw).

Table 1. Measurement conditions for dieldrin by the gas chromatography/µ electron capture detector (GC/µECD)

GC condition
Column	ENV-8MS, 30 m × 0.25 mm i.d. × 0.25 µm film thickness
Over temperature	80°C–5°C min^−1 → 280°C (20 minute hold)
Inlet temperature	250°C
Carrier gas	Helium (2 mL min^−1, purity 99.999%)
µECD condition
Detector temperature	300°C
Make up gas	Nitrogen (30 mL min^−1, purity 99.999%)

Chemical properties of the soil

The chemical properties of the soil used for the pot experiments are shown in Table 2, following the methods described by the Editorial Boards of Methods for Soil Environment Analysis (1997). Soil pH (in H₂O) was measured at a 1 : 2.5 (solid/water) ratio. Total carbon and nitrogen contents were determined by the dry combustion method (Sumigraph NC Analyzer NC-220F; Sumika Chemical Analysis Service, Ltd., Osaka, Japan). Total phosphorus content was measured by wet digestion (nitric-perchloric acids) and vanadomolybdate methods. Total potassium content was determined using wet digestion (hydrofluoric-perchloric acids) and an atomic absorption spectrophotometer (AA280FS; Varian Technologies Japan Ltd, Tokyo, Japan). Ammonium nitrogen and nitrate nitrogen were extracted by 2 mol L⁻¹ potassium chloride and determined by an auto analyzer (Bran+Luebbe AutoAnalyzer 3; BLTEC, Osaka, Japan). Available phosphorus was determined using the Truog method. Exchangeable potassium, calcium, magnesium, and the cation exchange capacity (CEC) were determined using the semi-micro Schollenberger method and the atomic absorption spectrophotometer.

Table 2. Chemical properties of the soil used for the pot experiments

								Exchangeable cations
pH (H2O)	Total carbon (g kg^−1)	Total nitrogen (g kg^−1)	Total phosphorus (g kg^−1)	Total potassium (g kg^−1)	Ammonium nitrogen, NH4−N (mg kg^−1)	Nitrate nitrogen, NO3-N (mg kg^−1)	Truog phosphorus (mg kg^−1)	Potassium (cmol kg^−1)	Calcium (cmol kg^−1)	Magnesium (cmol kg^−1)	Cation exchange capacity (CEC)
5.2	37.7	3.3	2.8	10.4	17.8	50.4	39.7	1.7	8.1	3.0	22.4

Statistical analysis

Statistical analyses were performed using the Statview 5.0 J statistical software package for Windows (SAS Institute, Berkeley, CA). Analysis of variance (ANOVA) followed by Tukey's multiple comparison test was used to determine which samples differed, using a pairwise comparison matrix.

Results

Physicochemical properties of adsorbents

The physicochemical properties of the tested adsorbents are listed in Table 3. The pH (in H₂O) values of the biochar adsorbents (WC450, RC, and BC) were neutral to weakly alkaline. The total carbon contents of WC450, RC, and BC were 75%, 40%, and 72%, respectively. The total nitrogen, phosphorus, and potassium contents of them were about 2–5%, 0.3–3%, and 5–14%, respectively. The SSAs of WC450, RC, and BC were 2, 46, and 2 m² g⁻¹, respectively, all of which were below 100 m² g⁻¹. The particle sizes of WC450, RC, and BC were distributed mainly within the ranges 425–2000 µm, 850–2000 µm, and 100–425 µm, respectively. High-temperature heating of wood chip charcoal (WC1000) did not significantly alter the pH, carbon content or particle size when compared with WC450. However, the SSA increased to two-orders of magnitude higher (149 m² g⁻¹). Further, as a result of crushing, WC1000C had a finer particle size (<75 µm) and considerably larger SSA (215 m² g⁻¹) than WC1000. For AC, the pH (11.2) was alkaline, and the carbon content (66%) was not much different from that of the wood chip charcoal, RC, and BC. The nitrogen and potassium contents were also not much different from those of the biochars, but phosphorus content was higher. The SSA of AC was relatively large (780 m² g⁻¹), and the particle size was mostly <75 µm.

Table 3. Physicochemical properties of the adsorbents

						Particle size (µm)
Materials	pH(H2O)	Total carbon (g kg^−1)	Total nitrogen (g kg^−1)	Total phosphorus (g kg^−1)	Total potassium (g kg^−1)	75–106	106–250	250–425	425–850	850–2000	2000–
WC450	8.0	746	2.7	0.32	11.5	5.8%	12.5%	14.7%	25.7%	28.3%	10.2%
WC1000	8.0	818	2.4	0.15	4.7	0.6%	3.1%	9.3%	36.5%	46.2%	12.6%
WC1000C	7.9	833	2.5	0.15	4.7	5.5%	15.3%	18.6%	18.8%	3.4%	0.0%
RC	6.6	404	5.2	0.85	12.4	0.0%	0.0%	0.7%	0.7%	59.3%	28.9%
BC	7.2	719	2.7	1.39	13.9	5.4%	29.1%	39.2%	39.2%	0.0%	0.0%
AC	11.2	858	2.2	6.3	11.6	3.7%	1.6%	0.0%	0.0%	0.0%	0.0%

Suppression of dieldrin uptake in cucumber fruits by some biochars

First, we compared the suppression of dieldrin uptake by cucumber using biochars of three different materials, namely, WC450, RC, and BC (pot experiment 1) (Fig. 2). In this experiment, the differences of the growth and fruit bearing of the test plants were not observed among the treatments. The dieldrin concentration in the cucumber fruits grown in the control soil (no application of adsorbents) was 0.055 mg kg⁻¹ fw. With application of WC450 (0.22 kg dw per pot) to the soil, the concentration decreased to 0.005 mg kg⁻¹ fw. However, the dieldrin concentrations following the application of RC (0.22 kg dw per pot) and BC (0.22 kg dw per pot) were both 0.045 mg kg⁻¹ fw. Thus, WC450 application had the highest suppressive effect, and the applications of RC and BC had little or no effect.

Figure 2. Suppression of dieldrin uptake by cucumber using different charcoal. WC450 [wood chip charcoal, non-crush, put into pot, 0.22 kg dry weight (dw)], RC (rice husk charcoal, non-crush, put into pot, 0.22 kg dw), BC (bamboo charcoal, crush, put into pot, 0.23 kg dw). Error bars represents ± standard error (n = 3). Columns with the same letter are not significantly different at p < 0.05 according to an analysis of variance (ANOVA)-protected Tukey's multiple range test.

Suppression of dieldrin uptake in cucumber fruits by optimized wood chip charcoal in the pot experiment

We compared the suppression of dieldrin uptake by cucumber using wood chip charcoal subjected to different treatments, namely, burning at relatively low temperature (WC450), burning at high temperature (WC1000), and burning at high temperature followed by crushing (WC1000C) (pot experiment 2) (Fig. 3). In this experiment, the differences of the growth and fruit bearing of the test plants were not observed among the treatments. The dieldrin concentration in the cucumber fruits grown in the control soil was 0.055 mg kg⁻¹ fw. With the application of WC450 (0.11 kg dw per pot) to the soil, the concentration decreased to 0.040 mg kg⁻¹ fw. However, WC1000 decreased the concentration to 0.020 mg kg⁻¹ fw, and WC1000C decreased the concentration to 0.004 mg kg⁻¹ fw. We thus found that the high temperature burning and crushing treatment of wood chip charcoal could enhance the reduction of dieldrin concentrations in cucumber fruits.

Figure 3. Suppression of dieldrin uptake by cucumber using wood chip charcoal with different treatment [input into pot, 0.11 kg dry weight (dw)]. WC450 (burning temperature 450°C, non-crush), WC1000 (burning temperature 1000°C, non-crush), WC1000C (burning temperature 1000°C, crush treatment). Error bars represents ± standard error (n = 3). Columns with the same letter are not significantly different at p < 0.05 according to an analysis of variance (ANOVA)-protected Tukey's multiple range test.

Comparison of the effects of activated carbon and optimized wood chip charcoal in suppressing dieldrin uptake in cucumber fruits

We compared the suppression of dieldrin uptake in cucumbers between AC and the optimized (burning at high temperature and crushing) wood chip charcoal (WC1000C) using different application amounts (pot experiment 3) (Fig. 4). In this experiment, the differences of the growth and fruit bearing of the test plants were not observed among the treatments. The dieldrin concentration in the cucumber fruits grown in the control soil was 0.049 mg kg⁻¹ fw. When 6.25, 12.5, 25, and 50 g dw of AC was applied to pots, the dieldrin concentration in the cucumber fruits decreased to 0.020, 0.001, <0.001, and <0.001 mg kg⁻¹ fw, respectively. However, when 54.5, 109, and 218 g dw of WC1000C was applied to pots, dieldrin concentration in the cucumber fruits decreased to 0.019, 0.001, and <0.001 mg kg⁻¹ fw, respectively. We thus found that the suppressive effect of the optimized wood chip charcoal was considerably lower than that of AC.

Figure 4. Concentration of dieldrin in cucumber fruits: adsorbent was applied with AC (○),WC1000C (□). Error bars represent ± standard error (n = 3).

Suppression of dieldrin uptake in cucumber fruits by powdered activated carbon in the field experiment

We examined the effect of AC on the reduction of dieldrin concentration in cucumber fruits in a field experiment (Fig. 5). In this experiment, the differences of the growth and fruit bearing of the test plants were not observed among the treatments. The dieldrin concentration in the cucumber fruits grown in the control soil was 0.022 mg kg⁻¹ fw, a concentration that exceeded the residue limit (0.02 mg kg⁻¹). When 0.25 or 0.5 kg dw m⁻² of AC was applied to the field, the concentrations of dieldrin in cucumber fruits grown in both plots in 2006 were 0.001 mg kg⁻¹ fw. We examined whether the suppressive effect of AC was maintained in subsequent years. In 2007, one year after AC application, dieldrin concentration of the control cucumbers was 0.026 g kg⁻¹ fw, whereas the concentrations in cucumbers grown in the plots that received 0.25 and 0.5 kg dw m⁻² of AC were both 0.004 mg kg⁻¹ fw. In 2009, three years after AC application, the dieldrin concentration in control cucumbers was 0.020 mg kg⁻¹ fw, whereas the concentrations in cucumber grown in the plots that received 0.25 and 0.5 kg dw m⁻² of AC were 0.008 mg kg⁻¹ fw and 0.006 mg kg⁻¹ fw, respectively.

Figure 5. Suppression of dieldrin uptake by cucumber using powder activated carbon materials in the field experiment. Error bars represent ± standard error (n = 3). Columns with the same letter are not significantly different at p < 0.05 according to an analysis of variance (ANOVA)-protected Tukey's multiple range test.

Discussion

It has been reported that various adsorbent materials in soils can effectively suppress the uptake of organic pollutants, such as pesticides, by plants. With the application of charcoal to soils, the uptake of carbofuran and chlorpyrifos by spring onion (Allium cepa) was reduced (Yu et al. 2009). Similarly, compost reduced the uptake of aldrin and dieldrin by turnip (Brassica rapa L. var. rapifera) (Nakamura 1990). Inputs of amorphous carbons (Guo et al. 1991; Martinez-iñigo and Almendros 1992; Sluszny et al. 1999) and condensed carbons (Guo et al. 1991) have been demonstrated to increase the capacity of soils to absorb pesticides. Furthermore, AC has been found to reduce dieldrin uptake by cucumber (Hilber et al. 2009; Hashimoto 2007) and heptachlor epoxide uptake by pumpkin (Murano et al. 2009). However, AC is usually more expensive than biochars. Therefore, we examined whether an optimized biochar could effectively suppress dieldrin uptake by cucumber fruits.

Of the three types of biochar examined, wood chip charcoal (WC450) was the most effective in reducing dieldrin concentration in cucumber fruits (Fig. 2). It has been considered that the effect of organic compound adsorption by carbonaceous adsorbents is related to the SSA of the adsorbent materials (Chun et al. 2004; Yu et al. 2006). Although in the present study the SSA of RC was higher than that of wood chip charcoal and BC, the suppressive effect of this material on dieldrin uptake by cucumber was small. We speculate that the low carbon contents of the RC (Table 2) was caused by the presence of silicate salt, which is hydrophilic and has no capacity to adsorb dieldrin. With regard to particle size distribution, we observed no relationship between the particle fineness of biochars and the suppression of dieldrin uptake by cucumber.

The wood chip charcoal, which was the most effective of the three biochars, was burned at high temperature (1000°C) and crushed to enhance its suppressive ability. With high-temperature treatment (WC1000), the SSA of the wood chip charcoal increased 75 times (Table 2), and its application resulted in a 50% decrease in the dieldrin concentration in cucumber fruits compared with the normally heated WC450 (Fig. 3). Further, by crushing the high-temperature-treated wood chip charcoal (WC1000C), the SSA increased 1.4 times, and its application caused an 80% decrease in the dieldrin concentration of cucumber fruits compared with the non-crushed WC1000. Similarly, it has been reported that the capacity of wood char to adsorb chlorpyrios and carbofuran increased when this material was heated at high temperature (Yu et al. 2009). We found that the performance of biochar was optimized by heating at high temperature and crushing.

Next, we compared the suppressive effect of optimized wood chip charcoal (WC1000C) on the dieldrin concentration in cucumber fruits with that of AC. Both adsorbents decreased dieldrin concentration in the cucumber fruits with increasing amounts of adsorbents applied to the soil (Fig. 4). However, AC appeared to have a superior ability to suppress dieldrin uptake by cucumber. To reduce levels to below the threshold limit (dieldrin < 0.02 mg kg⁻¹ fw), 54.5 g per pot of WC1000C was required, whereas only 6.25 g per pot of AC was required. Because the SSA of WC1000C was only about 25% that of AC, it was expected that the amounts of WC1000C needed to suppress dieldrin uptake in cucumber would be four times greater than those of AC. However, considerably smaller amounts of AC showed the same level of suppression as WC1000C. The superior ability of AC was not explicable merely in terms of SSA; other factors such as pore size and surface functional groups may also influence dieldrin adsorption ability.

On the basis of the pot experiment results, AC was selected as the most effective adsorbent for suppressing dieldrin uptake by cucumber. We accordingly attempted to confirm the suppressive effect of AC in a dieldrin-contaminated field. In the pot experiment, AC was completely mixed with the soil. In contrast, in the field experiment, AC was mixed only to a depth of 15 cm. However, the dieldrin residues in this field not only accumulated in the plow layer (0–15 cm depth) in which the AC was mixed but extended to a depth of 80 cm (Fig. 1). Nevertheless, application of 0.25 or 0.5 kg dw m⁻² AC in the field could suppress dieldrin concentration in cucumber fruits in 2006 as well as in the pot experiment (Fig. 5). It is speculated that the root distribution of the cultivated cucumber plants was mainly in the plow layer, and therefore the application of AC was sufficient to suppress dieldrin uptake by cucumbers. In the following years, the suppressive effect appeared to decrease with each successive year. There are two possibilities that might explain this phenomenon. One is that the surface of AC becomes saturated with organic matter in the soil (humic acid, fulvic acid, etc.) and root exudates (organic acids, amino acids, sugars, etc.), which cause a decrease in dieldrin adsorption. The other is that AC applied to the soil was scattered by plowing and was accordingly lost from the test plots. Although the suppressive effect tended to decrease with time, it remained substantial—compared with the soil that received no application, the dieldrin concentration in cucumber fruits grown in soil receiving 0.25 kg dw m⁻² of AC was decreased by 85% in 2007 and by 60% in 2009; similarly, the dieldrin concentration in cucumbers grown in the soil receiving 0.5 kg dw m⁻² of AC was decreased by 85% in 2007 and by 70% in 2009. From a practical point of view, it is very important that the suppressive effect of AC continues to some extent for at least four years after application. Further, we could not observe the effect of AC application on the growth and fruit bearing of the test plants in each year, even in the plot receiving 0.5 kg dw m⁻² of AC. We already reported that AC adsorbents application to soil did not affect the pH and CEC of the soil and the growth of young pumpkin plants (Murano et al. 2009). Therefore, the amounts of AC applied to the soil in our experiment seem to have no or little effect on the nutrient availability and the plant growth.

We also considered the usefulness of adsorbents from the viewpoint of cost. The price per kg dw of wood chip charcoal (WC450) is ¥169, the price of optimized wood chip charcoal (not crushed) (WC1000) is higher at ¥261 per kg dw, and the price of AC is ¥500 per kg dw, the highest of all the tested adsorbents. However, the suppressive abilities of WC450, RC, and BC were not sufficient. Therefore, we compared the costs of AC and WC1000C necessary to obtain the same effect. The exact price of WC1000C is not available, since we prepared WC1000C ourselves by crushing WC1000. Therefore, the price of WC1000C we used is adopted from that of WC1000. To achieve a greater than 50% reduction in dieldrin concentration (0.02 mg kg⁻¹ fw) compared with non-application (0.055 mg kg⁻¹ fw), the cost of WC1000C required (54.5 g dw per pot) was ¥14.3, whereas that of AC (6.25 g dw per pot) was ¥3.1. Therefore, the cost of AC required to suppress dieldrin contamination in cucumber fruits is considerably lower than that of WC1000C. Further, we calculated the cost of the AC application necessary to obtain a sufficient suppressive effect in the field. On the basis of our field experiment results, it appears that 0.25 kg dw of AC is sufficient to suppress dieldrin concentration in cucumber fruits. Accordingly, the cost of AC required amounts is ¥125,000 per 10 a, which seems expensive for an agricultural material. However, since the suppressive effect continues for at least four years, the cost per annum would be approximately ¥30,000 per 10 a. This is a cost that appears to be acceptable to cucumber farmers.

In conclusion, application of AC to contaminated soil showed an immediate and continued reduction of dieldrin concentration in cucumber fruits, and the cost per year was not that high. Therefore, this seems to be a promising practical technique for cucumber cultivation in dieldrin-contaminated fields. However, since we only examined a single soil type (Andosol) in this study, further studies using other types of soil are needed to confirm the effect of AC in reducing dieldrin concentrations in cucumber fruit. Furthermore, it is also necessary to examine the amount of AC required to obtain a sufficient suppressive effect, and the durability of the effect in different soil types.

Acknowledgments

This work was supported in part by a Grant-in-Aid for Research Projects for Ensuring Food Safety from Farm to Table (PO-2222) from the Ministry of Agriculture, Forestry and Fisheries of Japan.