Introduction

The contamination of soils and aquatic environments by Cr has attracted considerable attention over the last years due to high levels of Cr waste released from a large variety of industrial processes. Cr(III) and Cr(VI) are two thermodynamically stable oxidation states of chromium present in natural soils, and their geochemical and biologic behaviors are drastically different (Kotaś and Stasicka 2000). Cr(III) is known to be relatively non-toxic and weakly mobile in soils, while Cr(VI) is highly toxic and commonly very mobile. Accordingly, reduction of Cr(VI) to Cr(III), and subsequent precipitation of Cr(III) is the most common method of minimizing Cr(VI) mobility.

Cr(VI) could be reduced by many reductants present in soils, such as organic C-based materials, Fe(II) and sulfide (Fendorf 1995), which could be potentially used in remediation of Cr(VI) contamination. The utilization of organic materials is more environmentally beneficial because they would not exhibit toxicity themselves and introduce additional environmental problems. However, reduction of Cr(VI) by organic compounds was very slow (Wittbrodt and Palmer 1996). The presence of clay minerals and hydrous oxides, such as TiO2, γ-Al2O3, α-FeOOH, has been confirmed to be capable of catalyzing Cr(VI) reduction by organic compounds(Deng and Stone 1996a, 1996b; Deng et al. 2003; Tzou et al. 2003). These studies have also indicated that surface-catalyzed Cr(VI) reduction might occur in soils, while the mechanisms have not been carefully examined till date.

Oxalic and tartaric acids are organic acids existing ubiquitously in soil solutions (Jones 1998), which could affect geochemical behavior of Cr(VI) in soils through competitive adsorption for reactive surface sites or redox reactions involving soils. Therefore, the purposes of this research were to examine the influence of organic acids on Cr(VI) adsorption by a Fe-rich soil as a function of pH and to assess whether Cr(VI) reduction by organic acids could be catalyzed by the soil surfaces. The red soil was selected because it is widely spread in south China and it has high contents of Fe oxides. Results obtained from the whole soil would be more helpful when developing remedial techniques for chromium site remediation.

Materials and methods

The Fe-rich soil sample was collected from the 100- to 120-cm layer of red soil in Gaoan County (115°0′–115°34′E, 28°02′–28°38′N), Jiangxi Province, air-dried, and ground to pass 100-mesh sieve prior to being used in this study. The depth was pre-selected because the content of soil organic matters (SOM) was very low in this layer so that the interaction between Cr(VI) and soil organic matters could not be considered. Soil pH measurement was made in H2O at 1:2.5 (w/v) soil solution ratio using a pH glass electrode. Organic carbon was determined according to the Walkley and Black wet oxidation procedure (Nelson and Sommers 1996). Organic matter concentration of the soil was taken to be organic carbon concentration multiplied by 1.724. The cation exchange capacity (CEC) was determined by 1.0 M NH4OAC buffered at pH 7 (Chapman, 1965). Dithionate-citrate-bicarbonate extractable Fe was measured according to the method of Mehra and Jackson (1960). Amorphous or poorly Fe oxides were determined by extracting 1 g of soil in 30 mL of acidified ammonium oxalate buffer [0.175M (NH4)2C2O4 + 0.1M H2C2O4, pH 3.0] for 2 h (Loepper and Inskeep 1996).To analyze the total dissolved iron in extracts, hydroxylamine hydrochloride was first added to reduce Fe3+ to Fe2+ in samples, and then concentration of Fe2+ was measured using 1,10-phenanthroline method. The selected properties of the soil are presented in Table 1. All solutions and suspensions were prepared using deionized water. All the glassware were soaked in 5% HNO3 overnight and then thoroughly rinsed with deionized water prior to use. Analytical reagent grade chemicals were used for preparation of all solutions throughout the study. Stock solution containing 0.01 M Cr(VI) was prepared from K2CrO4, and organic acids stock solutions were prepared daily.

Table 1 Selected properties of the soil used in this study
Full size table

Batch experiments involved adding 1.000 g of the soil into a 50-mL plastic tube, followed by sequential addition of an organic acid and Cr(VI) to a total volume of 30 mL with 0.01 M NaNO3 as background electrolyte. Initial suspensions contained 0.1 mM Cr(VI) with the organic acid concentration of 0, 0.1 or 0.2 mM, oxalic acid and 0 or 0.1 mM tartaric acid. Solution pH ranging from 3.00 to 6.00 was obtained by adding 0.1 or 0.01 M HNO3 or NaOH. Each tube was then sealed and shaken at room temperature (25–30°C) for 24 h. After pH of the suspension was measured, the solution was separated from the solid phase by centrifugation at 4,000 rpm for 10 min. The concentration of Cr in solutions was measured with flame atomic absorption spectrophotometer for total Cr and the colorimetric analysis for Cr(VI) using 1,5-diphenylcarbazide indicator (James and Barlett 1983). Cr(III) was calculated as the difference between total Cr and Cr(VI). Adsorbed Cr(VI) was calculated from the difference between the Cr(VI) initially added to the system and that remaining in the solution after equilibration. All the experiments were triplicated, and the mean values were presented.

Results and discussion

Interaction between Cr(VI) and the soil in the presence of oxalic acid

As shown in Fig. 1, in the absence of oxalic acid, the decrease in adsorption of Cr(VI) by the soil with the increase in pH was observed. This is because the amphoteric Fe-rich soil surfaces typically would bear more negative charges with the increase in pH, then adsorption of oxyanions like CrO 2−4 onto the soil would be inhibited by electrostatic repulsion. Additionally, direct reduction of Cr(VI) by the soil components could be precluded because the soil used in the study was highly weathered. So the observed decrease in [Cr(VI)]aq could only be attributed to its adsorption onto soil surfaces in the absence of organic acids, which was in agreement with previous study (Zhou et al. 2001).

Fig. 1
figure 1

Interaction between Cr(VI) and the soil in the presence of oxalic acid

Full size image

As shown in Fig. 1, when coexisting with oxalic acid, Cr(VI) adsorption by the soils was decreased with more Cr(VI) existing in solutions, which was more pronounced as the initial concentration of this competitive anion was increased from 0.1 to 0.2 mM. Previous studies have also shown that the adsorption of chromate onto goethite was diminished in the presence of oxalate (Mesuers and Fish 1992). Oxalic acid may exert influence on Cr(VI) adsorption through competition for binding sites with Cr(VI) species or through the alteration of surface properties of the soil. Studies have confirmed that adsorption of oxalic acid onto Fe oxide surfaces through formation of inner-sphere complexes (Persson and Axe 2005), so oxalic acid might also associate with surface functional groups on surfaces of the soil minerals such as Fe oxides, forming strong inner-sphere surface complexes. However, adsorption of Cr(VI) by clay minerals was often regarded through electrostatic attraction with out-sphere complex formation (Zachara et al. 1988; Fritzen et al. 2006). Additionally, the presence of oxalate would result in more negative surface charge of the Fe-rich soils and a decrease in ξ potential (Xu et al. 2003, 2004), leading to the increase in electrostatic repulsion between Cr(VI) and the negatively charged surface sites. Therefore, the two mechanisms described above jointly resulted in the inhibition of Cr(VI) adsorption by the soil in the presence of oxalic acid.

Interaction between Cr(VI) and the soil in the presence of tartaric acid

In addition to two carboxylic acid groups, tartaric acid also possesses two α-hydroxyl function groups, thereby leading to reactions occurring in the soil–Cr(VI)–tartaric acid system considerably different from that in the soil–Cr(VI)–oxalic acid system. As shown in Fig. 2, when tartaric acid was introduced to the soil–Cr(VI) system, Cr(VI) concentration in equilibrium solutions after 24 h was far less than that of tartaric acid-free system across the pH range investigated, which was more remarkable especially at low pH. For example, Cr(VI) existing in equilibrium solution of the ternary soil–Cr(VI)–tartaric acid system was 10.5 μM, with 53.2 μM in equilibrium solution of the soil–Cr(VI) system at the most acidic pH examined in this study. This indicated that soil surfaces catalyzed Cr(VI) reduction by tartaric acid occurred since there was no evidence for enhancement of Cr(VI) adsorption by the soil in the presence of tartaric acid. Because homogeneous reduction of Cr(VI) by tartaric acid in solutions across the pH wide in this experiment proceeds very slowly and could be ignored, the observed acceleration of reduction reaction was owing to catalytic effects from the soil surfaces. Zhou et al. (2001) have also suggested that the soil colloids could act as a catalysts to speed up the reduction of Cr(VI) by p-methoxypheonl. The corresponding concentration of Cr(III) in equilibrium solutions was presented in Fig. 3.

Fig. 2
figure 2

Interaction between Cr(VI) and the soil in the presence of tartaric acid

Full size image
Fig. 3
figure 3

Concentration of Cr(III) in equilibrium solutions of Cr(VI)–soil–tartaric acid system

Full size image

Several mechanisms have been advanced for the catalytic effect of metal oxides and clay minerals (Stone and Torrents 1995; Deng and Stone 1996a, 1996b; Deng et al. 2003; Smolen et al. 2003). These included: (1) increasing the reactant reactivity through adsorption onto the surfaces; (2) facilitating both reactants encounter and subsequent electron transfer through metal coordination; (3) redox cycle of active metals, like Fe(III)–Fe(II), acting as an electron shuttle between reductant and oxidant. Fe-oxides and clay minerals are included in the soil studied, and it might also be able to catalyze Cr(VI) reduction by tartaric acid through similar mechanisms mentioned above, which would be demonstrated as the following.

At present, it is still not possible to directly investigate the nature of adsorbed species pertinent to surface catalysis reactions (Stone and Torrents 1995), hence phosphate was chosen as a probe to test whether Cr(VI) reduction would be inhibited when the surface sites available for reactants adsorption were decreased by the competitive adsorption between reactants and phosphate. The results in turn could provide some indirect evidences for the mechanism that reactants adsorption was always a necessary step for the soil surface catalyzed Cr(VI) reduction to take place. Phosphate was selected because it has high affinity for soil surfaces and was commonly found in soil solutions. As shown in Fig. 2, although Cr(VI) concentration in equilibrium solutions of soil–Cr(VI)–tartaric acid system with addition of phosphate was still less than that of soil–Cr(VI) system at pH < 4.84, it was clearly higher than that of soil–Cr(VI)–tartaric acid system across the pH range examined. Studies have shown that phosphate suppressed Cr(VI) adsorption (Aide and Cummings 1997). So the observed inhibition of soil surfaces catalytic effect was obviously caused by the decrease in adsorption of Cr(VI) and tartaric acid. At pH > 4.84, both Cr(VI) and tartaric acid adsorption would be further suppressed because the shortage of available adsorption sites for either Cr(VI) or tartaric acid in the presence of phosphate might be intensified by the surfaces shifting to more negative. Therefore, Cr(VI) concentration in equilibrium solutions of soil–Cr(VI)-tartaric acid system with addition of phosphate was even higher than that of soil–Cr(VI) system at pH > 4.84 (Fig. 2), which obviously indicated that Cr(VI) reduction reactions might not occur virtually. Similar results were also obtained that the rate of TiO2-catalyzed reduction of Cr(VI) by mandelic acid was decreased in the presence of phosphate, which was believed to be caused by the decrease in surface sites available for reactants adsorption (Deng and Stone 1996a).

As shown in Fig. 4, a number of ferrous ions were released to solutions after the reaction between 1 g soil and 0.1 mM tartaric acid across the pH wide from 3.0 to 4.3 for 24 h. Total ferrous ions concentration in solutions ranged from 21.6 to 5.4 μM, and concentration of Fe(II) could be up to 13.8 μM. Thus, Cr(VI) could be directly reduced by Fe(II) existing in the soil solutions. According to Stumm and Sulzberger (1992), Fe(III) could facilitate electron transfer between organic compounds and Cr(VI) through the formation of strong complexes with organic ligands and the existence of the Fe(III)–Fe(II) redox couple, acting as electron shuttles. Because high content of Fe oxides were included in the soil, the tartaric acid could promote release of Fe(III) to solution as well as the redox cycle of Fe(III)–Fe(II) through the reductive dissolution of Fe oxides by tartaric acid especially at low pH, which might be regarded as another mechanism for the soil surface catalytic effect. Previous studies have also demonstrated that reductive transformation of 4-cyanonitrobezene would be promoted through the reductive dissolution of goethite by ascorbic acid (Smolen et al. 2003).

Fig. 4
figure 4

Release of ferrous ions from tartaric acid promoted dissolution of the soil (Suspensions containing 1 g soil and 0.1 mM tartaric acid with total volume of 30 mL and reaction time was 24 h)

Full size image

To further examine the role of Fe(III)–Fe(II) redox cycle in soil surface catalytic effect, a separate experiment was performed according to the similar procedures as those in Fig. 1, in which silica, instead of the soil, was used as adsorbent, with the suspensions containing silica + 0.1 mM Cr(VI), silica + 0.1 mM Cr(VI) + 0.1 mM tartaric acid, or silica + 0.1 mM Cr(VI) + 0.1 mM tartaric acid + 0.05 mM Fe(III). As shown in Fig. 5, the percentage of Cr(VI) adsorbed by silica in the absence of tartaric acid was below 10%, and Cr(VI) adsorption and reduction in ternary silica + Cr(VI) + tartaric acid systems was less than 20%, which indicated that the catalytic effect of silica surfaces was less than that observed with soil. While after addition of 0.05 mM Fe(III) to the ternary silica–Cr(VI)–tartaric acid systems, [Cr(VI)]aq was largely decreased, even below analytical detection limit of the method at low pH (3.35). So it was evident that the reduction of Cr(VI) by tartaric acid was very efficient in the presence of Fe(III), which in turn proved that Fe(III)–Fe(II) redox cycle could facilitate Cr(VI) reduction by tartaric acid in soils.

Fig. 5
figure 5

Interaction between Cr(VI) and silica in the presence of tartaric acid and Fe(III)

Full size image

Conclusions

Results from this study show that addition of oxalic acid would inhibit adsorption of Cr(VI) by the Fe-rich soil, thereby enhancing the mobility of Cr(VI) in the soil and ecological risk posed by the contaminant. Cr(VI) reduction by tartaric acid could be effectively catalyzed by the soil surfaces across the pH range (2.5–5.5) examined, which indicates that such biological reducant commonly present in soils could be used in remediation of Cr(VI)-contaminated soil.