Evaluation of the potential redistribution of chromium fractionation in contaminated soil by citric acid/sodium citrate washing

1 Introduction

Chromium (Cr) is widely used in various important industrial applications, such as leather tanning, dyeing and chromium plating (Reddy et al., 1997; Rai et al., 2005), and as a result has been released into the agricultural environment. In soil, chromium can occur as Cr(III) and/or Cr(VI), mainly depending on pH and redox conditions. The two forms behave quite differently, with Cr(III) being much less soluble and therefore less mobile than Cr(VI) (Fibbi et al., 2012). The hexavalent form of Cr usually exists as soluble anionic species over a wide pH range, such as chromate ( ${\text{CrO}}_4^{2-}$ ), hydrochromate ( ${\text{HCrO}}_4^{-}$ ), and dichromate ( ${\text{Cr}}_2{\text{O}}_7^{2-}$ ), and possesses significantly higher levels of toxicity than other valence states (Sharma and Forester, 1995; Bartlett, 1991). Compared to other heavy metals, the removal of chromium is more complex due to changes of valence states and fractionations. Generally, owing to the high toxicity of Cr(VI), in soil, particular attention has been paid to changes in the soluble and exchangeable Cr fraction due to its bioavailability (Bhattacharyya et al., 2005; Barrera-Diaz et al., 2012).

Washing has been considered as a successful method for the removal of heavy metals from contaminated soil, a process that usually employs different chelating agents (Schramel et al., 2000; Reddy and Chinthamreddy, 2000). Among the agents, citric acid (CA) has been a major focus for soil remediation research because of its ability and efficiency in mobilizing metal cations, coupled with only a minor impact on the physical and chemical properties of the soil (Romkens et al., 2002). Compared with the effect of inorganic acid on slags compound waste (Moutsatsou et al., 2006.), the effect of organic acid on polluted soil appears to be better. The results of Wen et al. (2009) indicated that CA can be rapidly degraded, with 20% degradation occurring between 1 and 4 d, depending on the level of soil contamination, and with 70% degradation occurring within 20 d. In contrast, the authors found ethylenediaminetetraacetic acid (EDTA) to be more persistent in the soils; only 14% of the EDTA was degraded after 20 d. Several studies have been published in which CA is reported as advantageous in the use of chelate-assisted phytoextraction, because it is biodegradable and rapidly degrades to carbon dioxide and water (Evangelou et al., 2007; Yan et al., 1996; Huang et al., 1998b). Jean et al. (2007) carried out batch experiments to investigate the mobilization of Cr in soil, and indicated that CA was the most effective for Cr mobilization. In addition, Peters (1999) also showed that CA was effective, while other chelating agents like gluconate, oxalate, and ammonium acetate were ineffective at removing heavy metals from Aberdeen Proving Ground soils.

The most serious issue associated with the use of CA as a washing agent is that the process may change the distribution of metals' chemical fractionation. This can lead to the release of a soluble and exchangeable fraction, which poses a great environmental risk because of its known characteristic of bioavailability (Kotaś and Stasicka, 2000). In general, sequential extraction schemes have been used to determine the fractionation of Cr in soils. The BCR sequential extraction procedure, which has been developed under the auspices of the European Community Bureau of Reference (Rauret et al., 1999; Davidson et al., 1998), involves the separation of elements into four main fractions: “acid extractable”, “reducible”, “oxidizable” and “residual”. In fact, the approach has been certified as a useful tool for monitoring the relative changes in Cr element partitioning (Reddy et al., 2001).

A study of the mobilization of Cr from a contaminated soil was conducted by Jean et al. (2007) to examine the effect of CA on the distribution of metals, and concluded that only slight modification could be found. However, the authors also reported that Cr was extracted from the fractions R3 and R4 and was redistributed in both fractions R1 and R2. This was attributed to the Cr bound to iron-oxides possibly not being extracted effectively from the R2 fraction, and thus was extracted with the last fraction (R4), but also fraction R3. This part of Cr was therefore redistributed in fractions R1 and R2 during the mobilization with chelants. However, no mechanistic details were uncovered to help understand the observed effects. In addition, few studies have assessed the effects of CA washing on the fractionation of Cr. In this context, our study used a four-step washing experiment to investigate the effects of CA on changes in Cr chemical fractionation in actual contaminated soil, as well as to evaluate the potential risk of acid extractable Cr.

2 Materials and methods

2.2

2.2 Experimental procedure

Citric acid/sodium citrate (CA/SC) washing of Cr in the soil was carried out in batch experiments. In order to investigate the redistribution of Cr fractionation, a four-step experiment was designed (Fig. 1). A 100.00 g homogeneous soil sample was used in the experimental procedure.

Close

Figure 1

Schematic representation of the washing and extraction of fractions one-by-one. Black panes indicate the extracted fractions.

Export to PPT

Step 1:

First, a 10.00 g soil sample was used to analyze the content of four fractions (R1, R2, R3 and R4). Then, another 10.00 g soil sample was washed using CA/SC solution. The main reason for this was to reduce test error during the preliminary experiment. The residual 80.00 g sample was prepared for Step 2.

The washing tests were conducted in 100 mL polyethylene tubes. The tubes containing the 10.00 g sample and a measured volume of 0.2 mol L⁻¹ CA/SC (20 mL) were agitated using a Vortex shaker (QILINBEIER, QL-861, Jiangsu). The tubes were then placed in a shaking incubator (ZDP-250, Shanghai) at a speed of 150 rpm at room temperature (25 ± 2 °C). Each tube was shaken first for 5 h, then 10 h, then 15 and 20 h. The suspensions were centrifuged at 4500 rpm for 20 min and the supernatants were then filtered through a 0.45 μm membrane for Cr analysis. The residual solid was extracted and the fractionations of Cr were analyzed. The four fractions were named as R1_s1, R2_s1, R3_s1 and R4_s1.

Step 2:

Following Step 1, R1 was extracted from the residual 80.00 g soil sample. Then, 10.00 g of the treated soil sample was used to analyze the content of the four fractions (R1, R2, R3 and R4). Another 10.00 g soil sample was washed using CA/SC solution, and the residual 60.00 g was prepared for Step 3. The process of CA/SC washing in Step 2 was the same as in Step 1. Then, the residual solid after washing was analyzed and the fractions were named as R1_s2, R2_s2, R3_s2 and R4_s2.

Step 3:

Following Step 2, R2 was extracted from the residual 60.00 g soil sample. A 10.00 g soil sample from this 60.00 g was then used to analyze the content of the four fractions (R1, R2, R3 and R4). Another 10.00 g soil sample was then washed using CA/SC solution. The residual 40.00 g was prepared for Step 4. The process of CA/SC washing in Step 3 was the same as in Step 1. Then, the residual solid after washing was analyzed and the fractions were named as R1_s3, R2_s3, R3_s3 and R4_s3.

Step 4:

Following Step 3, R3 was extracted from the residual 40.00 g soil sample. A 10.00 g soil sample from this 40.00 g was then used to analyze the content of the four fractions (R1, R2, R3 and R4). And another 10.00 g soil sample was used to be washed by CA/SC solution. The process of CA/SC washing in Step 4 was the same as in Step 1. Then, the residual solid after washing was analyzed and the fractions were named as R1_s4, R2_s4, R3_s4 and R4_s4.

3 Results and discussion

For tests conducted during Step 1, the total amounts of Cr recovered during the BCR procedure for the initial soil were 98.43, 221.00, 153.20, and 231.19 mg kg⁻¹, respectively. The distribution of Cr chemical fractionation after CA/SC (0.2 mol/L) washing is also shown in Fig. 2. As indicated, the whole washing procedure was considered as two phases by taking 5 h as boundaries, and it is clear that there was a dramatic difference between them. The removal efficiency of R2, R3 and R4 was extremely quick during the beginning of the washing process, but after 5 h washing, they showed slight changes. However, it is important to note that, for the washed soil, the percentage of R1 increased from 12.83% to 28.54%. After 10 h, the ratio of R1 (R1_s1) with Cr(total) was the biggest percentage at 30.88%. Moreover, it was 2.41 times that for the controlled sample. In addition, the contents of R2_s1, R3_s1 and R4_s1 were lower than those before washing. Therefore, the effect of washing for R2 and R3 was to achieve satisfactory results in a short period of time. At initial concentrations of 221.00, 153.20 mg kg⁻¹, the removal ratios of R2 and R3 were 53.62% and 33.62% respectively by washing for 5 h, and 65.43% and 58.77% respectively by washing for 20 h. For R4, the removal ratio was 53.57% by washing for 20 h.

Close

Figure 2

Variation of the four fractions of the BCR sequential extraction procedure for Cr in the contaminated soils by CA/SC washing (R1, Acid soluble; R2, Reducible; R3, Oxidizable; R4, Residual).

Export to PPT

During Step 2, although R1 had been removed by acetic acid, the amounts of Cr recovered in R1, R2, R3 and R4 were 16.07, 257.68, 312.05, and 125.36 mg kg⁻¹, respectively. The levels of R2 and R3 were 36.25% and 43.88% of the total Cr. The effect of CA/SC washing on the distribution of Cr fractionation is illustrated in Fig. 3. After 5 h of washing, the content of R1_s2 was found to increase distinctly (176.79 mg kg⁻¹). This value was close to 11 times that of the amount of R1 in the control sample in Step 2. Moreover, a small range of variance occurred in R4, whereas a large range of variance occurred in R2 and R3. Specifically, they were decreased from 257.68 to 79.02 (R2_s2) and 312.05 to 174.11 mg kg⁻¹ (R3_s2), respectively. It is thus reasonable to assume that much acid soluble Cr in R2 and R3 had been released and was able to be extracted by the first BCR procedure. After washing for 10 h, the concentrations of R1_s2, R2_s2 and R3_s2 decreased slightly compared to those after 5 h of washing.

Close

Figure 3

Variation of the four fractions of the BCR sequential extraction procedure for Cr in the contaminated soils by CA/SC washing after Step 2.

Export to PPT

Fig. 4 shows the distribution of Cr fractionation after Step 3. The amount of R2 extracted was 98.63 mg kg⁻¹. Owing to the experimental design, theoretically there was not any R1 or R2. This was perhaps as a result of the effect of acetic acid, which was used to determine the content of R1. In response to the release, after reacting for 16 h, acetic acid led to the redistribution of Cr fractionations. After CA/SC washing for 5 h, the contents of R2 and R3 were decreased from 98.63 and 272.24 (R2_s3) to 16.24 and 80.00 mg kg⁻¹ (R3_s3), respectively. However, R1 increased from 2.39 to 124.18 mg kg⁻¹ (R1_s3). Again, the content changes of R2 and R3 were similar in the Step 1 and Step 2.

Close

Figure 4

Variation of the four fractions of the BCR sequential extraction procedure for Cr in the contaminated soils by CA/SC washing after Step 3.

Export to PPT

During Step 4, theoretically only R4 remained in the soil. Fig. 5 shows the distribution of Cr fractionation, indicating the contents of R1, R2, R3 and R4 were 0, 11.56, 19.97 and 68.47 mg kg⁻¹, respectively. After washing for 5 and 10 h, the contents of R2 and R3 both decreased. R2_s4 decreased from 11.56 to 6.18 (5 h) and 6.57 mg kg⁻¹ (10 h). R3_s4 decreased from 19.97 to 9.71 (5 h) and 11.14 mg kg⁻¹ (10 h). The content of R1_s4 increased obviously, from 0 to 14.37 (5 h) and 11.92 mg kg⁻¹ (10 h). The content of R4_s4 decreased slightly after 5 h of washing.

Close

Figure 5

Variation of the four fractions of the BCR sequential extraction procedure for Cr in the contaminated soils by CA/SC washing after Step 4.

Export to PPT

The proportions of Cr chemical fractionations after washing in Step 2, Step 3 and Step 4 are shown in Fig. 6. It can be seen that the ratio of R4 remained almost constant by washing for either 5 or 10 h for the three steps. The proportions of R2 and R3 decreased significantly. After 5 h of washing, the proportion of R2 decreased from 36.23% to 14.46%, 19.27% to 5.03% and 11.56% to 6.18%. Furthermore, the proportion of R3 decreased from 43.88% to 31.86%, 53.20% to 24.76% and 19.97% to 9.76%. The proportion of R1 increased significantly after washing. Comparing the control sample and the sample washed for 5 h, the proportion of R1 increased from 2.26% to 32.35%, 0.47% to 38.43%, and 0 to 14.37%. In addition, there is a little change in the proportion of variation of Cr fractions from 5 to 10 h washing.

Close

Figure 6

Ratio of the four fractions of the BCR sequential extraction procedure for Cr in the contaminated soils by CA/SC washing after Steps 2, 3 and 4.

Export to PPT

Previous studies have shown that the efficiency of Cr fractionation by CA/SC washing is better than other forms of washing, such as DTPA (Diethylene triamine pentacetate acid) and EDTA (Pichtel and Pichtel, 1997; Sun et al., 2001; Hong et al., 2002; Mühlbachová, 2011). Here, during the washing process, the ratio of R1 with total Cr increased before 5 h and decreased after 5 h. This means that there was a redistribution of Cr fractionation, i.e. R1 increased from 98.43 to 141.20 mg kg⁻¹ and the amount recovered in R2 decreased from 221.00 to 76.40 mg kg⁻¹. A similar result was reported in the earlier work by our group (Liang et al., 2011) as well as by other researchers. Jean et al. (2007), for example, reported that the content of R1 was increased during the remediation of Cr-Ni-contaminated soil by CA. However, the ratio of R1 with total Cr in their samples was too low (0.069%) to verify the hypothesis. The authors mentioned that, during the sequential extractions, the Cr bound to the iron-oxides might not have been extracted effectively from the R2 fraction, and thus it could have been extracted with the last fraction (R4), but also in fraction R3. This part of Cr might therefore have been redistributed in the R1 and R2 fractions during the mobilization with chelants. In fact, other heavy metals have also been shown to decrease exchangeable fractions, Lei et al. (2008) found that after EDTA extraction, the concentrations of exchangeable Pb, Cd, Cu, and Zn increased. The concentrations of carbonate, iron and manganese oxides, organic matter, and the residue of heavy metals decreased. In addition, Manouchehri et al. (2006) reported that the stoichiometric ratio of reagent/total metal needs to be investigated with respect to all the extractable cations present in the soil. Therefore, the quantification analysis of Cr fractionation seems to be a better method for complex soil system.

In this study, the result of the BCR procedure could directly reflect the redistribution of Cr chemical fractionation. We observed that R4, R3 and R2 were decreased by washing. However, contrary to the literature (Wuana et al., 2010), CA removed most of the metals hitherto associated with the exchangeable and reducible fractions, such as Ni, Zn and Cd, extracted Cr in R4 using CA washing needs to be investigated in the soil. In this study, because there is a d-value between measured R4 and real R4, Step 4 should be considered as the key step for calculating the redistribution rate. According to Fig. 5, the content of R4 was almost constant. We can hypothesize that 86.00 mg kg⁻¹ (R4_s3) was the real value of R4 in the contaminated soil. $$\mathrm{\Delta }\text{R}4=\text{R}4(\text{in original soil})-\text{R}4(\text{R}{4}_{\text{s}3})$$ Whereas, we have known that residual fraction was difficult to be washed. Therefore, ΔR4 can be considered as the undetected content of R2 and R3. For Step3 and Step 2, the removal contents of total Cr were 188.6 and 164.7 mg kg⁻¹, respectively. However, when comparing the calculated actual decreased contents of R2 and R3, it was found that R1 increased 121.8 and 150.7 mg kg⁻¹. Moreover, after washing in Step 1, if the value of R1 decreased by the washing process, then R2 and R3 had been decreased and transformed into R1. Therefore, Step 4, Step 3 and Step 2, all proved true in converting R2 and R3. The slow leaching of Cr therefore corresponds to dissolution of the mineral matrix and most likely to oxide dissolution.

A previous report (Jean et al., 2008) indicated that the effectiveness of the chelants in solubilizing Cr is not directly related to its complexation constants. It can be explained by: (i) its ability to solubilize the mineral matrix containing the metals. According to Di Palma (2009), the mechanism of metal extraction involves a two-step dissolution-chelation process where, after metal salts dissolute due to the strong acidity of the leaching solution, chelation occurs. When the soil was acidified by citric acid, the dominant species of Cr(III) for pH < 4.5 are Cr³⁺(aq) and Cr(OH)²⁺ (Andjelkovic et al., 2012); and (ii) the competition of CA with Cr(VI) on the surface sites. As the fact that the oxidation and reduction of soluble chromium added to soils depend on the soil structure (Kozuh et al., 2000), the desorption of Cr(VI) also depends on the soil structure. However, as a complex process, CA/SC washing brought on the redistribution of Cr in the four fractions of the BCR procedure. Moreover, the fact that increase of R1 content may be attributed to the properties of minerals. In the initial stages of washing, water soluble and acid-leachable concentrations of Cr were removed. Subsequently, Cr adsorbed on minerals was released gradually by dissolution with washing agents. This process enhanced the ratio of R1 in total Cr.

4 Conclusions

CA/SC washing was successful for the removal of soil Cr, especially oxidizable and residual fractions of soil Cr, but the soluble and exchangeable Cr fraction increased. After washing for 5 h, the content of total Cr remained almost constant. Results of the four steps indicated that the agent can extract the R2 and R1 contained in R4 and R3.

Changes in Cr fractionation in contaminated soil by CA/SC washing mean that the process not only improves soil Cr removal, but also releases the soluble and exchangeable fraction of Cr in contaminated soil simultaneously. The main reason may be that the agent dissolved the minerals of the soil, and the washing procedure may affect the inaccessible Cr, which was bound-up in a soil crystal lattice.

This is an advisable choice for the remediation of Cr-polluted soil in the future, especially for decreasing the secondary pollution of the soluble and exchangeable fraction, which was contained in R4 and R3.