Effects of liming on Cr(VI) reduction and Cr phytotoxicity in Cr(VI)-contaminated soils

Abstract

Liming is the most common approach for the amelioration of soil acidity in agriculture, and is widely used to reduce the mobility and bioavailability of heavy metals in soil. The purpose of this study was to investigate the hexavalent chromium [Cr(VI)] reduction and chromium (Cr) phytotoxicity in Cr(VI)-contaminated soils at different pH levels as a result of liming. Calcium carbonate (CaCO₃) was added to two acid agricultural soils of Taiwan (Neipu and Pinchen) at their natural pH of about 4, to adjust soil pH to approximately 5 and 6, respectively. The soils were then spiked with six levels of Cr(VI) (0, 150, 300, 500, 1000 and 1500 mg kg⁻¹). X-ray absorption near edge structure spectroscopy (XANES) of Cr was used to determine the extent of the reduction of Cr(VI) at different pH levels. At the same time, extractions of Cr by Dowex M4195 and Chelex 100 resins were carried out to determine the availability of Cr(VI) and trivalent chromium [Cr(III)] in the Cr(VI)-spiked soils, respectively. Also, a pot experiment with wheat (Triticum vulgare) seedlings was carried out to test the phytotoxicity of the Cr(VI)-spiked soils. The results showed that for Cr(VI)-contaminated soils which contain a high amount of organic matter, such as the Neipu soil, the effect of liming on Cr(VI) reduction and Cr phytotoxicity is insignificant. However, for Cr(VI)-contaminated soils which have a low amount of organic matter, such as the Pinchen soils, liming could decrease the extent of Cr(VI) reduction and increase the availability of Cr(VI), thereby enhancing the phytotoxicity of Cr.

Keywords:

• Cr(VI) reduction
• Cr phytotoxicity
• resin-extractable Cr(VI)
• resin-extractable Cr(III)
• X-ray absorption near edge structure spectroscopy (XANES).

Introduction

Soil chromium (Cr) contamination as a result of the discharge of industrial wastes containing hexavalent chromium [Cr(VI)] is increasingly becoming a cause of concern. Cr(VI) is highly soluble, mobile, and bioavailable compared to the barely soluble trivalent Cr species, Cr(III). Because Cr(III) is considered to be more stable than Cr(VI) and of lower phytotoxicity in soil, Cr(VI) toxicity can be moderated through the reduction of Cr(VI) to Cr(III) (Deiana et al. 2007). In soil, the mobile forms of Cr(VI) ( and ) can be reduced by different organic reducers (Kotaś and Stasicka 2000; Bolan et al. 2003b). Organic compounds are expected to be the primary reductants for Cr(VI) species in most surface soils (Bartlett and James 1988). The reaction to reduce Cr(VI), called dechromification, was reported to be catalyzed by Fe(II) and S⁻ (Bartlett and James 1988). The catalytic role of Mn(II) in the reduction of Cr(VI) by organic acids was also highlighted (Li et al. 2007; Tian et al. 2010). In the root-soil system, Cr(VI) can be reduced to Cr(III) by cytoplasmatic reductants on root surfaces (Becquer et al. 2003) and by several components of root exudates in the rhizosphere (James and Bartlett 1984).

Chromium(VI) reduction by organic material is most rapid under acidic conditions (Zhilin et al. 2004; Deiana et al. 2007). The most favorable solution pH for Cr(VI) reduction by organic materials ranges from 1 to 4 (Wittbrodt and Palmer 1996). Zhilin et al. (2004) showed that Cr(VI) was reduced by peat and coal humic substances at pH 5.4, but not at pH 9.2. Deiana et al. (2007) emphasized that at pH > 4.2 the reduction of Cr(VI) by caffeic acid occurred slowly. Qafoku et al. (2010) also found that sediments rich in Fe(II)-bearing minerals exposed to acidic waste liquids would result in releasing ferrous ions and ultimately reducing Cr(VI). In addition, because Cr(VI) has lower redox potential values at higher pH, the higher pH is detrimental to the reduction of Cr(VI) (Park et al. 2004; Su and Ludwig 2005).

The infertility of acid soils is a major limitation to crop production in highly weathered soils. Application of lime materials is the most common approach used for the amelioration of soil acidity (Orndorff et al. 2008). Liming of contaminated soils is also a remediation treatment for reducing the mobilization and bioavailability of heavy metals (Bolan et al. 2003a; Han et al. 2007; Klitzke and Lang 2009; Hong et al. 2010). However, increase of pH due to lime application is adverse not only to reduction of Cr(VI) but also to adsorption of Cr(VI). Liming to raise soil pH results in more negative charges on soil surfaces, and thus the desorption of Cr(VI) by soil solids will be enhanced. Therefore, lime application might raise the availability of Cr(VI) in soils and increase the phytotoxicity of Cr in Cr(VI)-contaminated soils by hindering Cr(VI) reduction and enhancing Cr(VI) desorption. On the other hand, liming might reduce the availability of Cr(III) that is formed from Cr(VI) reduction in soils and thus decrease the contribution of Cr(III) to the injury of plants in acid soils (Chen et al. 2008). Therefore, the effect of liming on Cr phytotoxicity in acidic Cr(VI)-contaminated soils is worthy of investigation.

A distinct spectral feature, the pre-edge peak, occurs in the X-ray absorption near edge structure (XANES) spectra for Cr(VI), which is used to quantify the proportion of total Cr as Cr(VI) in soil (Shaffer et al. 2001; Qafoku et al. 2009). The pre-edge peak at 5992 eV in the Cr K-edge XANES spectra is a unique characteristic of Cr(VI) but not of Cr(III) (Bajt et al. 1993; Szulczewski et al. 1997). As reported in many previous studies (Wei et al. 2002; Bang and Hesterberg 2004; Lee et al. 2006), XANES spectroscopy has been successfully applied to identify the oxidation state of Cr. Brown and Sturchio (2002) suggested that XANES spectroscopy provides a nondestructive measurement of the oxidation states of chromium on the soil surface and is a useful tool to estimate the degree of Cr(VI) reduction. Thus, XANES spectroscopy has been used to examine the extent of Cr(VI) reduction after amendment by organic materials by determining the amounts of Cr(VI) remaining in soils (Chiu et al. 2009; Chen et al. 2010). In our previous study, the exchanger resins Dowex M4195 and Chelex 100 were used to measure the availability of Cr(VI) and Cr(III) in soils respectively and then to assess their contributions to the phytotoxicity of soil Cr on wheat seedlings (Chen et al. 2010).

Extending from our previous work of assessing the phytotoxicity of Cr in Cr(VI)-spiked soils (Chen et al. 2010), the objective of this study was to investigate the effects of increase of pH due to liming on Cr(VI) reduction and Cr phytotoxicity in Cr(VI)-contaminated soils. The XANES method was used to examine the extent of the reduction of Cr(VI) in soils at different pHs modified with lime amendment. In addition, both Dowex M4195 and Chelex 100 resin extractions were used to simultaneously determine the availability of Cr(VI) and Cr(III) in Cr(VI)-spiked soils. A pot experiment with wheat seedlings was also carried out to assess the Cr phytotoxicity of Cr(VI)-spiked soils as influenced by liming.

Materials and Methods

Soil properties, CaCO₃ addition, and preparation of Cr(VI)-spiked soils

Two major agricultural acid soils, those of Pinchen and Neipu in Taiwan, were used in this study. Soil samples were air-dried, ground, and sieved (<2 mm), and then stored at room temperature for further use. Soil pH was measured at a 1:1 ratio of soil to water (McLean 1982). Organic matter was determined by the Walkley-Black method (Nelson and Sommers 1982). Particle size analysis was carried out using a hydrometer (Gee and Bauder 1986). Extractable iron (Fe), manganese (Mn) and aluminum (Al) were determined by dithionite-citrate-bicarbonate (DCB) (Mehra and Jackson 1960). The selected soil properties are summarized in Table 1. The pH values of the Neipu and Pinchen soils are 4.1 and 4.3, which may be out of range for growing general crops (Von Uexküll and Mutert 1995), and thus the application of liming materials should be considered in order to correct the soil acidity and thus enhance plant growth (Mokolobate and Haynes 2002).

Table 1. Selected properties of the studied soils

Soil	Taxonomy	pH (1:1)	Particle size fraction (g kg^−1)				g kg^−1
			Sand	Silt	Clay	Texture	OM†	FeDCB‡	MnDCB
Neipu	Typic Dystrudept	4.1	337	395	267	Loam	93.7	13.20	0.05
Pinchen	Tropeptic Halorthox	4.3	168	422	410	Silty clay	27.3	27.73	0.03
†Organic matter
‡Dithionite-citrate-bicarbonate extraction
Fe, iron; Mn, manganese.

In this study, we would like to highlight the effects of increase of pH due to liming on Cr(VI) reduction. In order to control soil samples at specific pHs, a treatment sequence of calcium carbonate (CaCO₃) addition prior to Cr(VI) spiking was carried out for the soil sample preparation. Two rates of CaCO₃ were applied to two tested soils to adjust pH from the natural level of 4 to approximately 5 and 6, respectively. The rates for Neipu soils were 3.0 and 6.0 g CaCO₃kg⁻¹. The rates for Pinchen soil were 1.0 and 3.0 g CaCO₃kg⁻¹. The two CaCO₃ amended and single non-amended soil samples were mixed thoroughly with distilled water to reach their respective water holding capacity. The samples were then incubated at room temperature for 1 month. After liming, the pH values of the CaCO₃ non-amended and amended soil samples were measured as 4.3, 5.4, and 6.4 for Neipu soil, and 4.5, 5.0, and 6.2 for Pinchen soil, respectively.

Chromium(VI)-spiked soils were prepared by adding potassium dichromate (K₂Cr₂O₇) into each soil sample to reach six levels of 0 (as control), 150, 300, 500, 1000, and 1500 mg Cr(VI) kg⁻¹, respectively, after CaCO₃ amendment. Distilled water was added to the Cr(VI)-spiked soils to reach their water holding capacity and mixed thoroughly (i.e., wetted). They were then air-dried at room temperature. The air-dried Cr(VI)-spiked soils were then passed through a 10 mesh sieve to ensure that the Cr(VI) added in each soil sample was mixed homogeneously. The wetting and drying procedures were repeated three times within a period of 3 weeks to mimic field conditions. After three wetting-drying cycles, the Cr(VI)-spiked soil samples were ground and sieved through a 10 mesh sieve. The pH values of Cr(VI)-spiked soil samples were measured to assess the change of pH resulting from Cr(VI) addition, and the degree of Cr(VI) reduction in soils was determined by XANES analysis.

XANES for soil Cr(VI) analysis

The Cr(VI)-spiked soils were re-ground to pass through a 230 mesh sieve and then stored in sealed glass bottles for 1 week prior to the XANES analysis. The Cr standards were prepared with mixtures of chromium oxide Cr₂O₃ (98% pure, Aldrich Chemicals) and K₂Cr₂O₇ (98% pure, Merck Chemicals) powders and boron nitride (99% pure, Aldrich Chemicals). All of the standards contained 5% total Cr by weight but with various Cr(VI) percentages (10, 20, 30, 40, 50, 70, 85, and 100%) (Lee et al. 2006). Chromium standards and soil samples were then mounted on sample holders and sealed with transparent tape for XANES analysis. The chromium K-edge XANES analysis was implemented with the Wiggler beam line BL17C1 at the National Synchrotron Radiation Research Center in Taiwan. The electron storage ring was operated with an energy of 1.5 GeV and a current of 120–200 mA. XANES spectra were drawn from bulk soil samples to determine the Cr valence states in soils. Data were collected in transmission mode for Cr standards, and in fluorescence mode for soil samples. And the normalized XANES spectra were plotted using the software program Origin Pro 7.0 (Cheng et al. 2009).

Extractions of Cr(VI) by Dowex M4195 resin and Cr(III) by Chelex 100 resin

DOWEX M4195 in a copper (Cu)-saturated form has a high affinity toward anion ligands such as chromate, oxalate, arsenate, and phosphate (Zhao et al. 1998). The DOWEX M4195 resin was proposed for assessing the phytotoxicity of soil Cr(VI) in our previous study (Lee et al. 2006). The ion exchange resin, Dowex M4195 (particle size >425 µm), was washed with 1 M hydrochloric acid (HCl) and immersed in deionized water for 2 d. Dowex M4195 resin was then saturated with 500 mg L⁻¹ CuCl₂ to convert it into the Cu-saturated form. The Cu-saturated resin was then retained in 80 mesh polypropylene (PP) bags (Lee et al. 2006). The Chelex 100 resin (with a particle size ranging from 297 to 840 µm) was also washed with 1 M HCl and immersed in deionized water for 2 d. Chelex 100 resin was saturated using 2 M calcium chloride (CaCl₂) to convert it into the Ca-saturated form. The calcium (Ca)-saturated Chelex 100 resin was retained in 60 mesh PP bags (Chen et al. 2008).

Ten gram Cr(VI)-spiked soil samples were placed in 300-mL flasks, and then 100 mL distilled water was added. Then 1 g of Chelex 100 and 2 g Dowex M4195, which were enclosed in PP bags, were each added into separate flasks. The flasks were then shaken at 25°Cfor 24 h. Then, 2 M sulfuric acid (H₂SO₄) and 10% sodium chloride (NaCl) were used respectively to desorb Cr(VI) from Dowex M4195 and Cr(III) from Chelex 100. Chromium concentrations in the desorbed solutions were determined using ICP-AES (Inductively Coupled Plasma Atomic Emission Spectroscopy, Perkin Elmer, Optima 2000DV). The Cr desorbed from Dowex M4195 was regarded as the extractable Cr(VI) and the Cr desorbed from Chelex 100 was regarded as the extractable Cr(III).

Pot experiment

A modified Neubauer test of the growth of the wheat seedlings (Triticum vulgare, variety Taichuang select No. 34) (Yu et al. 2004) was used to investigate the Cr phytotoxicity of Cr(VI)-spiked soils at all pH tested levels. A series of plastic bottles (10 cm diameter ×7.5 cm high) were filled with 90 g of Cr(VI)-spiked soil and mixed with 50 g of acid-washed quartz sand. Three mL of nutrient solution containing 2.4% (NH₄)₂SO₄, 0.9 % KNO₃, 3.3% MgSO₄, and 1.2 % KH₂PO₄ were added to the mixture in each bottle. Distilled water was then added to bring each soil to field capacity. The bottles were kept at room temperature (25°C) for 2 d to allow the solution and the solids mixture to reach equilibrium. Then, 100 g of acid-washed quartz sand was spread on the solid surface. One hundred seeds were placed in each pot. Another 50 g of acid-washed quartz sand was added to cover the seeds, and distilled water was added as required. There were three pots as replicates for each Cr(VI)-spiked soil sample. The pot experiment was carried out in the Phytotron of the National Taiwan University. The environmental conditions for plant growth were provided in detail in our previous study (Chen et al. 2010). After 25 d of plant growth, the average plant height of the wheat seedlings in each pot was measured. The seedlings were harvested, washed, and oven-dried at 70°C for 2 d. Afterwards the dry weights of the root and shoot of the plants were measured separately for each pot.

In order to assess the effect of Cr spiking on soil extractable Cr and plant growth responses given specific pH changed with liming, a completely random design (CRD) was carried out in this study. Because pH change was numerical instead of categorical after liming, it would not be suitable for assessing the effect of interaction of Cr spiking and pH change. Thus, one-way ANOVA was used for the CRD experiments and the Duncan multiple range test was performed to infer the significance of effects of Cr spiking on soil extractable Cr and plant growth responses at different pH separately.

Results and Discussion

Effects of soil pH on Cr(VI) reduction and proton consumption with Cr(VI) reduction

The effects of soil pH on the reduction of Cr(VI) to Cr(III) were investigated using the XANES spectra of Cr(VI)-spiked soils (Fig. 1). It was found that there was no detectable peak of Cr(VI) in the XANES spectra for Neipu soils at pH 6.4, 5.4, and 4.3 after Cr(VI) spiking (Fig. 1a; only the spectra of 1500 mg Cr(VI) kg⁻¹-spiked soil are shown). This showed that the Cr(VI) added in Neipu soil could be reduced almost completely to Cr(III) at all the three tested pH values. However, there were distinct Cr(VI) peaks in the XANES spectra of Pinchen soils after Cr(VI) spiking with 1500 mg Cr(VI) kg⁻¹ at pH 6.2, 5.0, and 4.5 respectively (Fig. 1b), indicating that the added Cr(VI) in Pinchen soil was partially reduced. In addition, the Cr(VI) peak area was in the order of pH 6.2 > 5.0 > 4.5, suggesting the amounts of Cr(VI) left in the Pinchen soils increased with the increase in pH value. This suggests that the increase in soil pH as a result of CaCO₃ addition will increasingly inhibit Cr(VI) reduction in the Pinchen soil. These results are similar to those found in many other studies (Zhilin et al. 2004; Deiana et al. 2007), indicating that reductions of Cr(VI) by organic materials can be accelerated at low pH. As the pH of Pinchen soils increased to 5.0 and 6.2 from 4.5 by the adding of CaCO₃, this probably took them out of the most favorable pH range for Cr(VI) reduction (Wittbrodt and Palmer 1996; Schlautman and Han 2001). However, for the Neipu soils, there was no dependence of pH on the extent of Cr(VI) reduction. Under these circumstances, the effect of pH on the extent of Cr(VI) reduction is insignificant. This might be due to the fact that there was a much higher percentage of organic matter in Neipu soil than in Pinchen soil (Table 1). The Neipu soil thus could supply relatively larger amounts of organic matter as the reductant for Cr(VI) reduction, compared to the Pinchen soil. Since the Cr(VI)-spiked soils were not prepared in anaerobic conditions, it is unlikely that Fe(II) or sulfide can exist and act as reductants for added Cr(VI).

Fig. 1. The XANES spectra of (a) Neipu and (b) Pinchen soils spiked with 1500 mg hexavalent chromium [Cr(VI)] kg⁻¹ at different pH levels.

The consumption of protons during Cr(VI) reduction results in an increased pH (Park et al. 2004; Liu et al. 2008). Hence, the change of soil pH is another index for determining the extent of Cr(VI) reduction after Cr(VI) spiking. Fig. 2 shows the changes in pH for Neipu and Pinchen soils after Cr(VI) spiking at different initial pH values and after three wetting and drying cycles. The pH of Neipu soil was increased with the levels of Cr(VI) spiking. Compared with the control, the pH of 1500 mg Cr(VI) kg⁻¹ spiked Neipu soil was increased more than 1.5 units. The pH increase after Cr(VI) spiking could be attributed to the reduction of Cr(VI) and corresponds to the results (Fig. 1a) showing that that the Cr(VI) added in Neipu soil was reduced almost completely to Cr(III) at all three tested pH values. Therefore, as amounts of spiked Cr(VI) were increased, the extents of Cr (VI) reduction and soil pH were increased in Neipu soil. On the other hand, in Pinchen soil, there was only an imperceptible increase in soil pH values with increased levels of Cr(VI) spiking. The smaller extent of pH increase after Cr(VI) spiking in Pinchen soil compared to that in Neipu soil results from the fact that Cr(VI) added in the Pinchen soil was only partially reduced, as supported by the XANES spectra shown in Fig. 1b. Therefore, the significant difference of the change in pH between the two soils after Cr(VI) spiking could be attributed to the different extent of Cr(VI) reduction. However, the extent of pH change after Cr(VI) spiking might also be influenced by pH buffer capacity of soils. Neipu soils have higher pH buffer capacity than Pinchen soils as indicated by the fact that larger amounts of CaCO₃ had to be applied to Neipu soil than Pinchen soil to reach similar pH. Hence, it should be noted that the extent of pH change after Cr(VI) spiking in Neipu soils due to Cr(VI) reduction was buffered because Neipu soils had higher pH buffer capacity. The above results support the finding that large amounts of organic matter in the Neipu soil act as electron donors to reduce the spiked-Cr(VI) while simultaneously consuming protons, resulting in an increase in soil pH. On the other hand, there was not enough organic matter content in the Pinchen soil to promote Cr(VI) reduction and thus the soil pH did not increase with the increase in the level of Cr(VI) spiking.

Fig. 2. Changes of pH for the Neipu and Pinchen soils after spiking with various amounts of hexavalent chromium [Cr(VI)] and after three wetting-drying cycles at different initial soil pH levels.

Soil resin-extractable Cr(VI) and Cr(III) at different pH

The extracted amounts of Cr by Dowex M4195 resin and Chelex 100 resin from the Neipu and Pinchen soils represented the available Cr(VI) and Cr(III) respectively (Chen et al. 2010) at the different pH levels as shown in Table 2. In the Neipu soil, there was no detectable available Cr(VI) at any of the three pH levels, and the available Cr(III) was up to about 8.6 mg kg⁻¹ at pH 4.3 when Cr(VI) spiked at 1500 mg kg⁻¹. The results are consistent with the XANES spectra (Fig. 1a) where most of the added Cr(VI) was reduced to Cr(III). Moreover, Stewart et al. (2003) pointed out that Cr(III) sorption would be enhanced by higher soil pH and carbonates. At a pH higher than 5, Cr(III) is relatively insoluble (Anderson 1997), and at pH 6.0, Cr(III) is almost completely precipitated (Fendorf 1995; James et al. 1997). For Neipu soil the increase of pH due to liming might promote Cr(III) fixation through adsorption and precipitation and thus there was a small amount of extractable Cr(III) by Chelex 100.

Table 2. Chromium (Cr) extractions by Chelex 100 and Dowex M4195 resins from the Cr(VI)-spiked soils at different soil pH levels

Soil	Cr(VI) added (mg kg^−1)	Resin extraction of Cr (mg kg^−1)
		pH 4.3		pH 5.4		pH 6.4
Neipu		Chelex 100	Dowex M4195	Chelex 100	Dowex M4195	Chelex 100	Dowex M4195
	0 (control)	ND†	ND	ND	ND	ND	ND
	150	l.l^c§ ± 0.1‡	ND	1.0^c ± 0.2	ND	1.3^b ± 0.3	ND
	300	2.8^c ± 0.4	ND	2.0^c ± 0.4	ND	1.7^b ± 0.2	ND
	500	5.1^b ± 0.9	ND	4.0^b ± 0.4	ND	2.5^b ± 0.7	ND
	1000	8.4^a ± 1.1	ND	8.2^a ± 1.2	ND	6.0^a ± 0.3	ND
	1500	8.6^a± 2.3	ND	8.3^a± 1.7	ND	6.1^a± 1.8	ND
Pinchen		pH 4.5		pH 5.0		pH 6.2
	150	0.7^e ± 0.0	6.5^d ± 0.1	0.3^d ± 0.0	6.2^d ± 0.1	ND	6.1^c± 0.1
	300	1.3^d ± 0.1	6.1^d ± 0.1	1.5^d ± 0.0	7.7^d ± 0.1	0.2^d ± 0.0	6.2^c ± 0.1
	500	3.7^c ± 0.1	9.2^c ± 0.6	4.1^c ± 0.0	17.8^c ± 0.7	1.5^c ± 0.1	13.1^c ± 0.1
	1000	11.6^b ± 0.4	40.7^b ± 0.0	12.7^b ± 2.3	62.0^b ± 1.7	9.2^b ± 0.0	65.2^b ± 0.0
	1500	20.9^a± 0.1	96.3^a± 1.8	22.0^a± 0.5	130.5^a± 3.0	24.0^a± 1.1	223.8^a± 7.9
†Lower than the detection limit (0.05 mg kg^−1)
‡Data shown are means ± SD (n = 3)
§Values followed by the same letter, within each column for the same soil, are not significantly different at the 0.05 level according to the Duncan multiple range test

Nevertheless, compared to the Neipu soil, there were higher amounts of extractable Cr(VI) by Dowex M4195 and extractable Cr(III) by Chelex 100 in the Pinchen soil, especially under the Cr(VI) spikes of 1000 and 1500 mg kg⁻¹ (Table 2). For the 1500 mg Cr(VI) kg⁻¹ spiked Pinchen soil at different pHs, the amounts of extractable Cr(VI) by Dowex M4195 were increased by raising the pH with added CaCO₃ (Table 2). The above results can be explained by the fact that the Cr(VI) added in the Pinchen soil was not completely reduced to Cr(III) and thus the increase in pH adversely affected the reduction of Cr(VI), as demonstrated in the XANES spectra shown in Fig. 1b. The increase in amounts of extractable Cr(VI) caused by raising the pH might be partly due to dissociation of proton ions from weakly acidic functional groups of organic matter and some clay minerals, which in turn increases the net negative charge on the soil's surface, thereby reducing the adsorption of Cr(VI) (Bolan and Thiagarajan 2001), if liming increases soil pH higher than the zero points of charge (ZPC) of the soils. Nevertheless, the pH values did not significantly affect the amounts of soil extractable Cr(III) by Chelex 100 in the Pinchen soil. This might be due to the fact that the Cr(III) produced from Cr(VI) reduction was adsorbed or precipitated within the range of tested pHs, and thus the amounts of resin-extractable Cr(III) were not affected by the soil pH.

Phytotoxicity of Cr in the Cr(VI)-spiked soils affected by pH

Three growth indices, the dry weight of the root and the shoot as well as the plant height, for wheat seedlings grown in the Cr(VI)-spiked Neipu soils at different pH values are shown in Table 3. These three growth indices of the wheat seedlings showed similar responses to the amounts of Cr(VI) spiking and level of soil pH. The dry weight and the plant height increased with the increase in Cr(VI), but no growth retardation indicating phytotoxicity of Cr was found in any of the treatments. The results were consistent with the very low amounts of soil extractable Cr(III) and Cr(VI) by Chelex 100 and Dowex M4195 resins respectively (Table 2). This was probably due to the fact that most of the Cr(VI) spiking in the Neipu soil was reduced to Cr(III) (Fig. 1b) thereby raising the soil pH as a result of the proton consumption during the Cr(VI) reduction (Fig. 2), which in turn promotes Cr(III) fixation and facilitates seedling growth. The increase in seedling growth could also be attributed to increased levels of potassium addition resulting from potassium dichromate spiking.

Table 3. Dry weights of shoot and root and plant height of wheat seedlings grown in the hexavalent chromium-[Cr(VI)]-spiked Neipu soils at different soil pH levels

pH	Cr(VI) added (mg kg^−1)	Dry weight (% of control)		Plant height (% of control)
		Shoot	Root
4.3	0 (control)	100.0^c‡ ± 0.0†	100.0^d± 0.0	100.0^c± 0.0
	150	106.5^c± 3.6	141.2^ab± 10.1	106.9^c± 2.2
	300	105.8^c± 1.4	110.2^cd±  3.9	121.4^b± 7.6
	500	121.9^b± 2.9	125.1^bc± 9.7	123.3^b± 7.9
	1000	130.7^b± 9.8	123.7^c± 13.5	135.2^a± 6.1
	1500	148.4^a± 9.8	152.2^a± 11.5	143.4^a± 8.2
5.4	0 (control)	100.0^b± 0.0	100.0^c± 0.0	100.0^c± 0.0
	150	110.4^b± 2.8	121.9^b± 0.8	101.1^c± 2.3
	300	113.3^b± 6.1	114.4^bc± 11.4	109.2^b± 5.9
	500	104.1^b± 17.0	97.6^c± 8.8	110.1^ab± 3.5
	1000	111.2^b± 4.3	106.9^bc± 0.5	115.7^ab± 2.8
	1500	136.8^a± 7.0	158.6^a± 16.7	117.9^a± 5.2
6.4	0 (control)	100.0^bc± 0.0	100.0^b± 0.0	100.0^b± 0.0
	150	96.5^bc± 16.7	73.9^c± 19.7	108.0^ab± 5.7
	300	105.6^bc± 5.2	90.1^bc± 7.7	115.7^a± 12.4
	500	92.7^c± 4.6	76.0^c± 3.7	110.9^ab± 3.3
	1000	108.4^b± 3.3	102.5^b± 4.9	117.3^a± 2.4
	1500	125.2^a± 5.5	127.6^a± 18.7	114.4^a± 1.1
†Data shown are means ± SD (n = 3)
‡Values followed by the same letter, within each column at the same pH, are not significantly different at the 0.05 level according to the Duncan multiple range test

As shown in Table 4, the dry weights of the root and the shoot as well as the plant height of the wheat seedlings grown in the Cr(VI)-spiked Pinchen soils were significantly reduced when the soil was spiked with over 500 mg Cr(VI) kg⁻¹soil, and the extent of the growth reduction was larger at a higher pH. This might be due to the fact that the amount of soil extractable Cr(VI) increased as the pH increased, as shown in Table 2, thereby inhibiting the growth of the wheat seedlings. Since the amount of soil extractable Cr(III) by Chelex 100 was not affected by the soil pH (Table 2), the effect of liming on the phytotoxicity of Cr on wheat seedlings grown in the Cr(VI)-spiked Pinchen soil was mainly a result of the larger amounts of extractable Cr(VI) present at the higher pH.

Table 4. Dry weights of shoot and root and plant height of wheat seedlings grown in the hexavalent chromium-[Cr(VI)]-spiked Pinchen soils at different soil pH levels

pH	Cr(VI) added (mg kg^−1)	Dry weight (% of control)		Plant height (% of control)
		Shoot	Root
4.5	0 (control)	100.0^a* ± 0.0*	100.0^b± 0.0	100.0^a± 0.0
	150	105.7^a± 3.8	138.0^a± 15.1	97.0^a± 1.9
	300	97.9^a± 8.8	145.9^a± 16.0	94.7^a± 7.2
	500	65.0^b± 3.6	93.7^b± 24.6	63.5^b± 7.9
	1000	36.7^c± 5.8	45.3^c± 17.7	36.7^c± 5.6
	1500	24.8^d± 9.9	44.7^c± 18.4	20.9^d± 3.7
5.0	0 (control)	100.0^b± 0.0	100.0^b± 0.0	100.0^a± 0.0
	150	108.6^a± 3.0	139.1^a± 12.1	102.2^a± 1.7
	300	73.4^c± 4.0	54.6^c± 5.3	77.5^b± 3.3
	500	51.7^d± 4.8	28.4^d± 6.2	54.9^c± 5.4
	1000	28.7^e± 3.3	14.6^e± 5.1	25.8^d± 1.9
	1500	23.8^e± 3.7	23.9^de± 3.4	22.3^d± 2.5
6.2	0 (control)	100.0^ab± 0.0	100.0^a± 0.0	100.0^ab± 0.0
	150	111.1^a± 3.2	105.2^a± 4.9	97.0^b± 1.5
	300	95.9^b± 12.5	106.3^a± 7.5	106.3^a± 8.9
	500	54.6^c± 0.2	37.7^b± 4.7	54.3^c± 1.9
	1000	26.5^d± 8.9	16.4^c± 5.4	24.3^d± 3.5
	1500	15.7^d± 4.0	13.6^c± 2.1	17.4^d± 1.3
†Data shown are means ± SD (n = 3)
‡Values followed by the same letter, within each column at the same pH, are not significantly different at the 0.05 level according to the Duncan multiple range test

Conclusion

This study showed that the Cr(VI) spiking of the Neipu soil, which contains a high level of organic matter, was mostly reduced to Cr(III) at the three tested pH levels. Thus, there was no growth retardation in the wheat seedlings grown in the Cr(VI)-spiked Neipu soil. On the contrary, most of the Cr(VI) spiked Pinchen soil, which contains a low level of organic matter, was not reduced to Cr(III), and the extent of the Cr(VI) reduction decreased with the increase in pH, as demonstrated by the XANES spectra. Moreover, the amount of soil extractable Cr(VI) by Dowex M4195 resin increased with the increase in pH as a result of liming. In addition, the extent of the growth retardation in the wheat seedlings grown in the Cr(VI)-spiked Pinchen soil increased with the increase in pH level. These results indicate that for Cr(VI)-contaminated soils which contain a high amount of organic matter, such as the soil in Neipu, the effect of liming on Cr(VI) reduction and Cr phytotoxicity is insignificant. However, for Cr(VI)-contaminated soils which contain only a low amount of organic matter, such as the soil in Pinchen, liming can decrease the extent of Cr(VI) reduction and increase the availability of Cr(VI), thereby enhancing the phytotoxicity of Cr in the soil.

Acknowledgments

This research was sponsored by the National Science Council, Taiwan, Republic of China, under grant No. NSC 95-2313-B-002-076-MY3 and NSC 97-2313-B-451-009-MY3.