Hexavalent chromium Cr(VI) is of particular environmental concern due to its toxicity, mobility, and challenging removal from industrial wastewater. It is a strong oxidizing agent that is carcinogenic and mutagenic and diffuses quickly through soil and aquatic environments. Moreover, it does not form insoluble compounds in aqueous solutions; therefore, separation by precipitation is not feasible. While Cr(VI) oxyanions are very mobile and toxic in the environment, trivalent Cr(III) cations are the opposite and, like many metal cations, Cr(III) forms insoluble precipitates. Thus, reducing Cr(VI)–Cr(III) simplifies its removal from effluent and also reduces its toxicity and mobility. Permeable reactive barriers (PRBs) with zero-valent iron (ZVI) have been used to remediate contaminated groundwater with metals, but using ZVI in remediation of contaminated groundwater or wastewater is limited due to its lack of stability, easy aggregation, and difficulty in separation of iron from the treated solution. Thus, the technology used in the present study is developed to address these problems by placing a layer of bentonite after the PRB layer to remove iron from the treated water. The removal rates of Cr(VI) under different values of pH were investigated, and the results indicated the highest adsorption capacity at low pH.

INTRODUCTION

The increasing contamination of urban and industrial wastewaters by toxic metal ions is a worrying environmental problem. Chromium, which is one of the most toxic and important heavy metals, is commonly found in wastewater from industrial activities mainly through tanning and electroplating industries (Gupta & Babu 2009).

Chromium generally exists in water with two stable oxidation states: hexavalent Cr(VI) and trivalent Cr(III) (Weng et al. 2007), whereas Cr(VI) is reported to have toxic effects on humans and it is considered to be genotoxic and carcinogenic in nature (Cheuhan & Sankararamakrishnan 2011). The permissible limit of Cr(VI) in industrial effluents is set at 0.5 mg/L by the Ministry of Environment and Forests, Government of India (Panda et al. 2011).

The most probable Cr(VI) species in aqueous solution are , , and , the relative distribution of which depends on the solution pH, on the Cr(VI) concentration and on the redox potential (Barrera-Díaz et al. 2003). However, none of these Cr(VI) species form insoluble precipitates, making separation through a direct precipitation method not feasible (Violeta et al. 2010), but Cr(III) is less toxic and can be readily precipitated out of the solution in the form of (Singh et al. 2011). Thus, to form a chromium solid phase, which can be easily separated from the aqueous media, it is necessary to change the oxidation state. Passive, in situ reactive barriers have proved to be viable, cost-effective systems for remediation of Cr-contaminated groundwater at some sites.

Permeable reactive barriers (PRBs) are installed in the flow-path of groundwater, most typically as vertical treatment walls (Blowes et al. 2000). The most common PRBs are those containing granular zero-valent iron (ZVI) and have typically been used to treat contaminated groundwater (Jiang et al. 2013). Redox-active solids used in PRBs, such as ZVI, promote rapid removal of redox-sensitive contaminants, such as Cr, by various mechanisms including adsorption and reductive precipitation (Shi et al. 2011; Jiang et al. 2014).

Using ZVI in remediation of contaminated groundwater or wastewater is limited due to its lack of stability, large amounts of iron, and residual sludge generated and difficulty in separating it from the treated solution as well as the associated cost (Fua et al. 2014). To address these issues, PRB(ZVI) has been supported by the layer of bentonite after the PRB layer to remove iron and residual sludge and also metal cations Cr(III) from treated water.

Bench top laboratory studies with micro ZVI have been carried out to evaluate treatment effectiveness and reaction processes.

MATERIALS AND METHODS

Materials and characterizations

Distilled water was used in all preparations. Potassium dichromate and deionized water were used to prepare synthetic chromium, and the solution pH was adjusted to the desired value by adding or NaOH.

Uncontaminated silica sand with non-uniform size (mean diameter d50 = 0.65 ± 0.2 mm) was used as porous medium: the bulk density of sand (rb = 1.703 g/cm3), particle density (rs = 2.62 g/cm3), average porosity of sand (n = 0.35), and average saturated hydraulic conductivity (K = 0.52 cm/s). In the control runs with no addition of ZVI, no removal of Cr(VI) was found over the time period of typical experiments.

The sample of bentonite used in this work was sieved to an average particle diameter of 0.42 mm, washed several times with deionized water, and dried at 110 °C for 24 h. This bentonite was mainly composed of montmorillonite. The chemical composition of the bentonite is SiO2 65–69.5%, Al2O3 11.88%, Fe2O3 1.73%, MgO 2–4%, CaO 1.5–4%, Na2O 2–4%, and TiO2 0.1%.

Methods

Preparation of the solution of chromium

A stock solution of 1.0 g/L was prepared by dissolving 2.8298 g of potassium dichromate in 1.0 L of double distilled water, which was further diluted for the preparation of test solutions. Several solutions with different initial concentrations of potassium dichromate were prepared. The required pH was adjusted by drop addition of 0.4 N , depending on the acidity of the sample.

Preparation of the uncontaminated sand

Sand was soaked in water for 24 h to dissolve the lumps and then washed on the No. 200 sieve until the wash water coming through the sieve is clear and then dried at 105 °C. The sand used in this study is poorly graded sand.

Preparation of the reactive media

Micro ZVI may not be suitable for use in PRB because the small particle size may result in low permeability. To eliminate this shortcoming, several methods might be used to enhance the permeability of reactive micro scale materials.

The reactive barrier was composed of sand and micro ZVI that were mixed with each other and exposed to nitrogen during the mixing to minimize oxidation of Fe(0) because it is a redox-sensitive element. Then, it was put directly in the layer of PRB. The properties of micro ZVI are shown in Table 1. The average diameter of used sand was between 0.15 and 0.6 mm.

Table 1

The main characteristics of micro zero-valent iron

 Specification 
Assay (cerimetric) ≥99.0% 
Substances insoluble in hydrochloric acid ≤0.1% 
Substance soluble in water ≤0.1% 
Chloride (Cl) ≤0.002% 
Heavy metals (as Pb) ≤0.02% 
As (arsenic) ≤0.0005% 
Residual solvents (Ph. Eur./USP/ICH) Excluded by production process 
Particle size (<10 μm) ≥65% 
Particle size (>45 μm) ≤35% 
Average specific surface area 14.5 m2/g 
 Specification 
Assay (cerimetric) ≥99.0% 
Substances insoluble in hydrochloric acid ≤0.1% 
Substance soluble in water ≤0.1% 
Chloride (Cl) ≤0.002% 
Heavy metals (as Pb) ≤0.02% 
As (arsenic) ≤0.0005% 
Residual solvents (Ph. Eur./USP/ICH) Excluded by production process 
Particle size (<10 μm) ≥65% 
Particle size (>45 μm) ≤35% 
Average specific surface area 14.5 m2/g 

The thickness of the reactive barrier containing sand and micro ZVI was 1.5 cm and the micro ZVI-to-sand ratio was 1:5. Also, to enhance the permeability of bentonite layer, sand with average diameter of 0.6–1.18 mm was mixed with bentonite.

Apparatus and instrumentation

The pH values of the solutions were measured by Hanna pH meter using a combined glass electrode; the pH meter was standardized using buffer solutions of pH values 4, 7, and 10.

The metal ions Cr(VI) were determined by UV-vis spectrophotometer (HACH DR 4000).

The concentration of total Cr in solution was determined using a flame atomic absorbance spectrometer.

Analytical procedure

The total chromium concentration in solution samples was analysed by atomic absorption (AA) spectroscopy method. The analysis of samples was carried out using Smith–Heiftje AA spectrophotometer. The light source used for the analysis was a hollow cathode lamp with a current of 6.0 mA. The wavelength required is 357.9 nm with a band pass of 0.5 nm. The total chromium measurement was done following the Environmental Protection Agency (EPA) protocols under method #7000 (US-EPA 1986). The analysis of Cr(VI) was carried out using spectrophotometer (HACH DR 4000; program: number 1560) according to the 1.5-diphenylcarbohydrazide method using a single dry powder formulation called chroma ver 3 chromium reagent for Cr(VI). With the same method the total Fe was determined by the FerrorVer method using the spectrophotometer (HACH DR 4000; program: number 2165).

Samples taken for total chromium were analysed with a method having a method detection limit (MDL) of 1.0 μg/L, and MDL for Cr(VI) is 0.006 mg/L and for total Fe is 0, and the maximum concentration of Cr(VI) and total Fe that can be measured and reported with 99% confidence is 0.7 and 3 mg/L, respectively. Subtracting the Cr(VI) values from the total chromium gave the Cr(III) concentration for each sample.

EXPERIMENTAL SETUP

The continuous adsorption studies reduction of Cr(VI) by micro ZVI were conducted in a bench-scaled laboratory setup as shown in Figure 1, schematically. Uncontaminated sand with non-uniform size (mean diameter d50 = 0.65 ± 0.2 mm) was used as a porous medium. The length of the porous medium after the PRB position was 3.5 cm and after bentonite layer was 18 cm. There are several empty compartments separated by movable and perforated walls, one of them before the position of PRB and the other after it for monitoring pH, concentration of Cr(VI) in these reservoirs during the experimental periods; there is also a compartment for bentonite layer.

Figure 1

Schematic laboratory setup: (1) water tank, (2) plastic mesh, and (3) perforated walls.

Figure 1

Schematic laboratory setup: (1) water tank, (2) plastic mesh, and (3) perforated walls.

The laboratory model was made of Plexi-glass; a main reservoir was provided to supply influent water with a certain Cr(VI) concentration into porous medium. A reservoir up-stream and one down-stream of porous medium were considered to create a steady-state condition of flow through sand. A certain pore velocity through sand was made with regulation of water level in these reservoirs.

EXPERIMENTAL PROCEDURE

The simulated polluted groundwater with a Cr(VI) (initial concentration 25 mg/L) was prepared from a stock solution using deionized water and supplied to the entrance compartment. Then it was collected at a compartment after both PRB layer and bentonite layer with appropriate interval where it is analysed for Cr(VI) concentration, total Cr, total Fe and pH solution. The temperature during the operation was 25 ± 5 °C.

Initial pH of Cr(VI) solutions was adjusted to the desired value in the range from 3 to 6, using either or NaOH.

RESULTS AND DISCUSSION

Micro ZVI, as a Cr(VI) reducing agent, is commonly used for the reduction of Cr(VI) to the trivalent state, which usually takes place under acidic conditions and subsequently precipitates as Cr(III) hydroxide. Since Cr(III) hydroxide is precipitated, the resulting effluent will contain little or no residual chromium.

Experiments show that better reduction rates are achieved at low pH values (Welch et al. 2005; Chen et al. 2007); this is possibly due to charge distribution and spatial configuration changes as hydro complexes; this agrees with recent reports in which the reduction of Cr(VI)–Cr(III) seems to be slow at pH > 3.5, so values of pH less than 3 are needed to accelerate the reduction reaction in aqueous solution (Selvarani & Prema 2012).

Current methods of treating Cr(VI) are by chemical reduction to Cr(III) under different pH values. The reaction mechanisms of Cr(VI) with micro ZVI are believed to involve instantaneous adsorption of Cr(VI) on the surface of micro ZVI where electron transfer takes place and Cr(VI) is reduced to Cr(III) with the oxidation of ZVI to Fe3+, followed by the subsequent precipitation of mixed Cr and Fe hydroxides (Equations (4) and (5)).

The corrosion process induced by water or dissolved oxygen in the influent groundwater reacting with the Fe0 material leads to the formation of Fe2+, hydrogen gas, and (Equations (1)–(3)). The release of causes an increasing pH within the reactive media and pH value greater than 6 is observed after PRB, which is illustrated in Figure 2; in these conditions iron salts are not soluble and precipitate as ferric oxyhydroxides (ferrihydrite), amakinite (Fe(OH)2) or magnetite (Fe3O4).

Figure 2

Variation of the pH values along the reservoir after PRB (ZVI) layer over time in every test.

Figure 2

Variation of the pH values along the reservoir after PRB (ZVI) layer over time in every test.

No more dissolved Cr(III) was detected during the reaction, which indicated that most of the Cr(III) was coprecipitated with Fe(III). This observation is consistent with a former observation (Suponik & Blanco 2014). 
formula
1
 
formula
2
 
formula
3
 
formula
4
 
formula
5
 
formula
6

The reducing agents are usually ZVI; however, these reduction methods have their respective shortcomings; when ZVI is used as the reducing agent it is found that ferric hydroxide is produced as a solid waste, which requires subsequent disposal. Also, large amounts of iron cations are generated after PRB(ZVI) but only small amounts of Cr3+ cations because it precipitates as .

To solve these problems and to optimize the use of ZVI, produced materials should be removed from treated water, so we attempted it by placing a layer of bentonite after PRB(ZVI) layer, which carries a strong negative charge bonding the positive charge in many toxins.

The bentonite layer contributes to remove produced materials from treated water, also removing a little residual Cr(VI).

It is seen from Figure 3 that the appearance of red pigmented sediments in the solution in compartments after PRB means that it contains iron-rich precipitates. The change in colour from this compartment to the other compartment after a bentonite layer indicates reduction in iron concentration in the aqueous solution.

Figure 3

Laboratory setup.

Figure 3

Laboratory setup.

Figures 4(a), 5(a), and 6(a) illustrate the adsorption process of Cr(VI) in PRB(ZVI) which shows the highest adsorption capacity at lower pH. By increasing the pH, the adsorption of Cr(VI) decreases, due to the competition between the anions and (Ponder et al. 2000).

Figure 4

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 3).

Figure 4

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 3).

In this work, from Figures 456, it can be seen that higher amounts of iron are produced in a water with a lower initial pH value than in a water which was characterized by higher pH value. Thus, it can be concluded that the oxidation of ZVI proceeds faster in low pH. This occurs mostly due to the processes described by reactions (1) and (2). As a result of reaction (3) and others, there is a high probability that Fe(II) may be oxidized to Fe(III), which is then (under suitable conditions) precipitated in various forms. Also, it was noted from these figures that the concentrations of total Cr, Cr(III), and total Fe decreased after the bentonite layer, which means most of the chromium and iron species were removed by bentonite.

Figure 5

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 4.5).

Figure 5

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 4.5).

Figure 6

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 6).

Figure 6

Concentrations of (a) Cr(VI), (b) total Cr, (c) Cr(III), and (d) total Fe, after PRB and after bentonite layer (initial pH = 6).

Also, it can be noted from the experiments that the flow of the solution decreases over time, as illustrated in Figure 7. This implies that the velocity of solution in this setup decreases over time. The velocity can be represented according to Darcy's law as follows: 
formula
7
where k is the hydraulic conductivity and dh/dl is the hydraulic gradient.
Figure 7

The variation of flow of chromium solution through the setup over time.

Figure 7

The variation of flow of chromium solution through the setup over time.

Figure 1 shows that Δh = 0.6 cm = constant and also the length of setup = constant, so that mean dh/dl (hydraulic gradient) = constant. Also, as mentioned on page 7, it can be noted from the experiments that the flow of the solution decreases over time, as illustrated in Figure 7. This implies that the velocity of solution in this setup decreases over time. So it is concluded from Darcy's law V = k dh/dl that k decreases over time. k is the hydraulic conductivity for all materials in the setup, but the hydraulic conductivity of sand does not decrease because its permeability is high and the size of pores in sand is large, so that the mean hydraulic conductivity of bentonite in bentonite layer and reactive media in PRB(ZVI) decreases over time due to the production of and precipitated in these layers. Decreasing of hydraulic conductivity of bentonite and PRB(ZVI) layers leads to decrease in velocity of the solution in these barriers. This consequently leads to increment in the contact time. Increasing the contact time through the PRB(ZVI) should enhance the efficiency of reducing Cr(VI). But the reverse is the case because of the changes in the Fe0 surface over time, leading to a decrease in electron transfer and Fe0 reactivity. Likewise, contact time in the bentonite layer increases, but the effect of this increase on removal of hydroxides and cations is negligible because the particles of clay were covered by them. From Figure 7 it was noted that by increasing pH, the flow of solution in the setup decreased, which was due to the formation of more hydroxide species in higher pH levels.

CONCLUSION

The removal mechanism of contaminants Cr(VI) by micro ZVI concerns the directional transfer of electrons from ZVI to the contaminants, which transforms the latter into non-toxic or less toxic species. The chromate is reduced to the non-toxic/insoluble chromic ion (Cr3+) which presumably forms an insoluble phase consisting of Cr hydroxide. It was found that Cr(VI) removal efficiency increased significantly with decreasing pH, mainly because in acid conditions the accelerated corrosion of Fe0 enhanced the reaction rate. Most of the ferric and chromium hydroxides is produced as a solid waste, and also iron cations and Cr(III) are generated after PRB(ZVI). These problems were solved by placing a layer of bentonite after PRB(ZVI) and the quality of treated water was improved. Laboratory tests show that the Cr(VI) removal by PRB(ZVI) is improved while pH of contaminated water decreases, but at low pH values a large amount of iron and residual sludge is generated which may be removed using bentonite layer. Thus, it is possible to adjust pH of contaminated groundwater to low values before reaching PRB on site by adding and then placing bentonite layer after PRB.

REFERENCES

REFERENCES
Barrera-Díaz
C.
Palomar-Pardavé
M.
Romero-Romo
M.
Martínez
S.
2003
Chemical and electrochemical considerations on the removal process of hexavalent chromium from aqueous media
.
J. Appl. Electrochem.
33
,
61
71
.
Blowes
D. W.
Ptacek
C. J.
Benner
S. G.
McRae
C.
Bennett
T. A.
Puls
R. W.
2000
Treatment of inorganic contaminants using permeable reactive barriers
.
J. Contam. Hydrol.
45
,
123
137
.
Jiang
W.
Pelaez
M.
Dionysiou
D. D.
Entezari
M. H.
Tsoutsou
D.
O'Shea
K.
2013
Chromium(VI) removal by maghemite nanoparticles
.
Chem. Eng. J.
222
(
15
),
527
533
.
Jiang
W.
Cai
Q.
Xu
W.
Yang
M.
Cai
Y.
Dionysiou
D. D.
O'Shea
E. K.
2014
Cr(VI) adsorption and reduction by humic acid coated on magnetite
.
Environ. Sci. Technol.
48
(
14
),
8078
8085
.
Ponder
S. M.
Darab
J. G.
Mallouk
T. E.
2000
Remediation of Cr(VI) and Pb(II) aqueous solutions using supported, nanoscale zero-valent iron
.
Environ. Sci. Technol.
34
(
12
),
2564
2569
.
Selvarani
M.
Prema
P.
2012
Removal of toxic metal hexavalent chromium [Cr(VI)] from aqueous solution using starch – stabilized nanoscale zero valent iron as adsorbent: equilibrium and kinetics
.
Int. J. Environ. Sci.
2
(
4
),
1962
1975
.
Suponik
T.
Blanco
M.
2014
Removal of heavy metals from groundwater affected by acid mine drainage
.
Physicochem. Probl. Miner. Process.
50
(
1
),
359
372
.
US-EPA
1986
Method # 7000, SW-846
,
3rd edn
.
US Environmental Protection Agency
,
Washington, USA
.
Violeta
L. L.
Carlos
B. D.
Bryan
B.
Patricia
B. H.
Fernando
U. N.
Víctor
S. M.
2010
Cr(VI) reduction in wastewater using a bimetallic galvanic reactor
.
J. Hazard. Mater.
176
,
418
425
.
Welch
C. M.
Nekrassova
O.
Compton
R. G.
2005
Reduction of hexavalent chromium at solid electrodes in acidic media: reaction mechanism and analytical applications
.
Talanta
65
,
74
80
.