an electrokinetic/fe0 permeable reactive barrier system for the treatment of nitrate-contaminated...

7
An electrokinetic/Fe 0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils Tasuma Suzuki*, Yukinori Oyama, Mai Moribe, Masakazu Niinae Department of Environmental Science and Engineering, Graduate School of Science and Engineering, Yamaguchi University, 2-16-1 Tokiwadai, Ube, Yamaguchi 755-8611, Japan article info Article history: Received 12 September 2011 Received in revised form 12 November 2011 Accepted 17 November 2011 Available online 27 November 2011 Keywords: Nitrate Electrokinetic remediation Permeable reactive barrier Zero-valent iron Iron oxides abstract Effective nitrate removal by Fe 0 permeable reactive barriers (Fe 0 PRB) has been recognized as a challenging task because the iron corrosion product foamed on Fe 0 hinders effective electron transfer from Fe 0 to surface-bound nitrate. The objectives of this study were (i) to demonstrate the effectiveness of an electrokinetic/Fe 0 PRB system for remediating nitrate- contaminated low permeability soils using a bench-scale system and (ii) to deepen the understanding of the behavior and fate of nitrate in the system. Bench-scale laboratory experiments were designed to investigate the influence of the Fe 0 content in the permeable reactive barrier, the pH in the anode well, and the applied voltage on remediation effi- ciency. The experimental results showed that the major reaction product of nitrate reduction by Fe 0 was ammonium and that nitrate reduction efficiency was significantly influenced by the variables investigated in this study. Nitrate reduction efficiency was enhanced by either increasing the Fe 0 content in the Fe 0 reactive barrier or decreasing the initial anode pH. However, nitrate reduction efficiency was reduced by increasing the applied voltage from 10 V to 40 V due to the insufficient reaction time during nitrate migration through the Fe 0 PRB. For all experimental conditions, nearly all nitrate nitrogen was recovered in either anode or cathode wells as nitrate or ammonium within 100 h, demonstrating the effectiveness of the system for remediating nitrate-contaminated subsurface soils. ª 2011 Elsevier Ltd. All rights reserved. 1. Introduction Nitrate contamination of drinking water sources is a world- wide environmental issue. The conventional technologies used in drinking water purification plants for nitrate removal include reverse osmosis and ion exchange, which are some- times combined with biological denitrification for the subse- quent treatment of concentrated nitrate. Although these technologies are capable of removing nitrate effectively, in situ approaches are highly desirable because in situ remediation has the advantage of removing, degrading, or immobilizing contaminants before contaminants extensively migrate in the environment. Permeable reactive barriers using zero-valent iron (Fe 0 PRBs) as the reactive media represent a cost-effective in situ technology for remediation of contaminated subsurface regions (Henderson and Demond, 2007; Cundy et al., 2008; Thiruvenkatachari et al., 2008). In fact, monitoring the long- term performance of field-scale or pilot-scale Fe 0 PRB systems has demonstrated that Fe 0 PRB systems effectively remediate groundwater contaminated by various contami- nants including organic compounds (McMahon et al., 1999; * Corresponding author. Tel.: þ81 836 85 9690; fax: þ81 836 85 9601. E-mail address: [email protected] (T. Suzuki). Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate/watres water research 46 (2012) 772 e778 0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2011.11.048

Upload: tasuma-suzuki

Post on 30-Oct-2016

217 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

ww.sciencedirect.com

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8

Available online at w

journal homepage: www.elsevier .com/locate /watres

An electrokinetic/Fe0 permeable reactive barrier system forthe treatment of nitrate-contaminated subsurface soils

Tasuma Suzuki*, Yukinori Oyama, Mai Moribe, Masakazu Niinae

Department of Environmental Science and Engineering, Graduate School of Science and Engineering, Yamaguchi University,

2-16-1 Tokiwadai, Ube, Yamaguchi 755-8611, Japan

a r t i c l e i n f o

Article history:

Received 12 September 2011

Received in revised form

12 November 2011

Accepted 17 November 2011

Available online 27 November 2011

Keywords:

Nitrate

Electrokinetic remediation

Permeable reactive barrier

Zero-valent iron

Iron oxides

* Corresponding author. Tel.: þ81 836 85 969E-mail address: [email protected]

0043-1354/$ e see front matter ª 2011 Elsevdoi:10.1016/j.watres.2011.11.048

a b s t r a c t

Effective nitrate removal by Fe0 permeable reactive barriers (Fe0 PRB) has been recognized

as a challenging task because the iron corrosion product foamed on Fe0 hinders effective

electron transfer from Fe0 to surface-bound nitrate. The objectives of this study were (i) to

demonstrate the effectiveness of an electrokinetic/Fe0 PRB system for remediating nitrate-

contaminated low permeability soils using a bench-scale system and (ii) to deepen the

understanding of the behavior and fate of nitrate in the system. Bench-scale laboratory

experiments were designed to investigate the influence of the Fe0 content in the permeable

reactive barrier, the pH in the anode well, and the applied voltage on remediation effi-

ciency. The experimental results showed that the major reaction product of nitrate

reduction by Fe0 was ammonium and that nitrate reduction efficiency was significantly

influenced by the variables investigated in this study. Nitrate reduction efficiency was

enhanced by either increasing the Fe0 content in the Fe0 reactive barrier or decreasing the

initial anode pH. However, nitrate reduction efficiency was reduced by increasing the

applied voltage from 10 V to 40 V due to the insufficient reaction time during nitrate

migration through the Fe0 PRB. For all experimental conditions, nearly all nitrate nitrogen

was recovered in either anode or cathode wells as nitrate or ammonium within 100 h,

demonstrating the effectiveness of the system for remediating nitrate-contaminated

subsurface soils.

ª 2011 Elsevier Ltd. All rights reserved.

1. Introduction contaminants before contaminants extensivelymigrate in the

Nitrate contamination of drinking water sources is a world-

wide environmental issue. The conventional technologies

used in drinking water purification plants for nitrate removal

include reverse osmosis and ion exchange, which are some-

times combined with biological denitrification for the subse-

quent treatment of concentrated nitrate. Although these

technologies are capable of removing nitrate effectively, in situ

approaches are highly desirable because in situ remediation

has the advantage of removing, degrading, or immobilizing

0; fax: þ81 836 85 9601..jp (T. Suzuki).ier Ltd. All rights reserve

environment.

Permeable reactive barriers using zero-valent iron (Fe0

PRBs) as the reactive media represent a cost-effective in situ

technology for remediation of contaminated subsurface

regions (Henderson and Demond, 2007; Cundy et al., 2008;

Thiruvenkatachari et al., 2008). In fact, monitoring the long-

term performance of field-scale or pilot-scale Fe0 PRB

systems has demonstrated that Fe0 PRB systems effectively

remediate groundwater contaminated by various contami-

nants including organic compounds (McMahon et al., 1999;

d.

Page 2: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8 773

Vogan et al., 1999; Guerin et al., 2002; Wilkin et al., 2003) and

heavy metals (Puls et al., 1999a, 1999b; Wilkin et al., 2003;

Wilkin et al., 2005; Ludwig et al., 2009). However, effective

long-term nitrate removal by traditional Fe0 PRB systems has

been recognized as a challenging task (Thiruvenkatachari

et al., 2008) because, as noted by several researchers based

on the results obtained from a series of batch experiments

performed at near neutral initial pH (Cheng et al., 1997; Huang

et al., 1998, 2003; Choe et al., 1999; Huang and Zhang, 2002;

Huang and Zhang, 2005; Huang and Zhang, 2006; Suzuki

et al., unpublished results), the iron corrosion products

foamed on Fe0 hinders effective electron transfer from Fe0 to

surface-bound nitrate (i.e., slows down the reaction).

In this study, we developed an Fe0 PRB system that was

coupled with an electrokinetic process (EK/Fe0 PRB system). In

the EK/Fe0 PRB system, the EK process was expected to play

three important roles that may help overcome the shortcom-

ings of traditional Fe0 PRB systems. First, one of the main

strengths of EK processing is its ability to induce themovement

of dissolved compounds through low permeability soils by

electroosmosis and ion migration (Acar and Alshawabkeh,

1993; Acar et al., 1995; Kim et al., 2002; Reddy and

Chinthamreddy, 2003; Reddy and Saichek, 2004; Saichek and

Reddy, 2005; Niinae et al., 2008; Alshawabkeh, 2009; Reddy

et al., 2009; Reddy, 2010). Therefore, the EK/Fe0 PRB system

could be a novel remediation system that is capable of

producing more uniform nitrate transport in heterogeneous

subsurface soils compared to traditional Fe0 PRB systems.

Second, in the EK/Fe0 PRB system, acid generated at the anode

byelectrochemical reactions couldprevent the formationof the

passive layer of iron corrosion products on Fe0 particles.

Consequently, more effective long-term nitrate removal could

be achieved by the EK/Fe0 PRB system. Finally, the EK/Fe0 PRB

system is capable not only of converting toxic nitrate into less

toxic substances but also of recovering the dissolved reaction

products.

The specific objectives of this study were (i) to demonstrate

the effectiveness of the EK/Fe0 PRB system for the treatment of

nitrate-contaminated low permeability soils using a bench-

scale system and (ii) to deepen the understanding of the

behavior and fate of nitrate in the EK/Fe0 PRB system by

ascertaining the effects of the Fe0 content of the PRB, the pH of

the anode well, and the applied voltage. Insights obtained

from this study provide fundamental information that is

useful for optimizing remediation conditions.

2. Materials and methods

2.1. Materials and chemicals

All solutions were prepared with distilled deionized (DDI)

water with a resistivity greater than 18 MU cm�1 (WA200,

Yamato Scientific Co. Ltd., Japan). All chemical reagents

(guaranteed reagent grade) were purchased either from Naca-

lai Tesque, Inc. or Wako Pure Chemical Industries, Ltd. and

used without further purification. The zero-valent ion (Fe0)

purchased from Nacalai Tesque, Inc. was in the size range of

20e60mesh, and themeasuredBET surfaceareawas 1.1m2g�1

(Jemini 2375, Micromeritics Instrument Corp., GA). Kaolinite

from the Dixie Rubber Pit was purchased from Nichika Inc.

Particle size distribution analysis using a laser diffraction

particle size analyzer (SALD-2000J, Shimadzu Corp., Japan)

showed that the kaolinite contained approximately 63% clay-

size particles (<5 mm), and the remaining material was silt.

Additionally, preliminary experiments showed that the water

content of the saturated kaolinite was 50% by weight.

2.2. EK/Fe0 PRB experimental setup and procedures

A schematic of the EK/Fe0 PRB system is shown in Fig. 1. A PVC

column measuring 2 cm in diameter and 20 cm in length was

loaded with non-contaminated saturated kaolinite (Soil A),

artificially nitrate-contaminated saturated kaolinite (Soil B),

and non-contaminated Fe0 PRB (Soil C). Then, both sides of the

PVC column were sealed with a 0.2 mm-mixed cellulose ester

typemembrane filter to prevent outflowof soil particles. Soil A

was prepared by mixing an equal weight of kaolinite and DDI

water using a shaker (SA300, Yamato Scientific Co. Ltd.,

Japan). In a similar way, Soil B was prepared by mixing an

equal weight of kaolinite and an aqueous NaNO3 solution. Soil

C was prepared bymixing kaolinite, Fe0, and excess DDI water

and then discarding the supernatant after centrifuging the

mixture at 1220 g for 20 min using a benchtop centrifuge (CT4,

Hitachi Koki Co. Ltd., Japan).

Once loaded with Soils AeC, the PVC column was con-

nected to anode and cathode wells in which graphite elec-

trodes were placed. The graphite electrodes were then

connected to a DC power supply (GP060-10R, Takasago Ltd.,

Japan) providing constant voltage across the column. Anode

and cathode wells were filled with a 0.05 mol/L NaCl solution.

To avoid the water flow driven by the difference in static fluid

pressure, thewater levels of the anode and cathodewellswere

maintained constant by overflowing the solutions to reser-

voirs. The pH in the cathodewell was held constant during the

experiments at 6.2 using an automated titrator (AUT-701,

DKK-TOA Corp., Japan). The pH in the anode well was

adjusted before the experiments to either 2.5 or 4.5 using HCl,

and there was no pH control during the experiments.

The current was monitored using a direct current (DC)

current sensor (CMD-4-DC01-SC, U.R.D. Co. Ltd., Japan) and

recorded using a data logger (XL111-M, Yokogawa Electric

Corp., Japan).Water sampleswere collected from reservoirs at

appropriate time intervals and analyzed for NO�3 , NO�

2 , and

NHþ4 using an HPLC system with conductivity detection

(Prominence series, Shimadzu Corp., Japan). After energizing

for 100 h, the power of the DC power supply was turned off,

and the final pH in the Fe0 PRB was measured using a spear

type pH electrode (GST-5724C, DKK-TOA Corp., Japan). The

experimental conditions for each experiment are summarized

in Table 1.

3. Results and discussion

3.1. Electrochemical redox reactions at graphiteelectrodes

Electrolysis under DC current creates a reducing environment

at the cathode and an oxidizing environment at the anode.

Page 3: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

Fig. 1 e Schematic diagram of the EK/Fe0 PRB system.

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8774

Therefore, before NO�3 contaminated soil was treated using

the EK/Fe0 PRB system, blank experiments were performed to

investigate the possible generation of nitrogen species at

graphite electrodes (Tests 2e3 in Table 1). The Fe0 content in

the PRB and initial anode pH were 5 wt% and 2.5, respectively.

The results of these trials are shown in Fig. 2. It is clear from

Fig. 2 that the generation of NO�3 and NHþ

4 was negligible at

10 V. However, when the applied voltage was increased to

40 V, the amount of NHþ4 generated at the cathode increased

Table 1 e Experimental conditions and results for tests 1e8.a

Test Initial conditionsb

NO�3 in Soil B

(mgN/kg wet soil)Fe0 in PRB

(wt%)Applied

voltage (V)Init

anode

1 150 0 10 2.5

2 0 5 10 2.5

3 0 5 40 2.5

4 150 80 10 2.5

5 150 50 10 2.5

6 150 5 10 2.5

7 150 5 10 4.5

8 150 5 40 4.5

a For all tests, both the anode and cathode wells were filled with 0.05

energization time was 100 h.

b The initial pH value of Soil Bwas ca.5.5. The initial pH value of Fe0 PRBsw

dissolution.

c The pH in the Fe0 PRB varied depending on the location, generally the l

values shown here were obtained from the soils located at approximatel

d Energy consumption for remediation was calculated by summing the p

1.5 min. For the experiments performed at 10 V (Tests 4e7), the remediatio

from contaminated Soil B, was determined to be 100 h according to the

anode well in 20 h (Fig. 5). Therefore, the remediation time was assumed t

with time during the 100 h-experiment. The generation of

NHþ4 at the cathode was expressed by Eqn. (1) representing the

reaction between dissolved nitrogen gas and hydrogen gas

generated at the cathode via water electrolysis.

3H2ðgÞ þN2ðgÞ þ 2Hþ/2NHþ4 (1)

Consequently, although it is apparent that the NHþ4 gener-

ation rate at the cathode varies depending on electrolyte

Anode pHafter 100 h

PRB pHafter 100 hc

Energy consumptionfor remediationd (Wh)

ialpH

2.2 2.4 e

2.4 6.1 e

1.9 4.8 e

2.4 6.7 8.9 (in 100 h)

2.4 5.9 6.9 (in 100 h)

2.3 5.2 8.0 (in 100 h)

3.5 7.0 7.3 (in 100 h)

2.2 5.0 26.4 (in 40 h)

mol/L NaCl solution. The pH in the cathode well was 6.2, and the

as higher than that for Soil B (6.0e7.5) due to acid consumption by Fe0

owest for the anode side and the highest for the cathode side. The pH

y 1 cm from the anode.

roduct of the applied voltage and the current with the time interval of

n time, which was defined as the time required to recover all NO�3 eN

results shown in Figs. 3e5. For Test 8, NO�3 finished migrating to the

o be 40 h considering the time for NHþ4 migration to the cathode well.

Page 4: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

0

2

4

6

8

10

0 20 40 60 80 100

NH4+ (10 V, cathode)

NO3- (10 V, anode)

NH4

+ (40 V, cathode)

NO3

- (40 V, anode)

HN

4+O

N ro 3-

)Ng

m( sllew ni ssa

m

Time (hours)

Fig. 2 e Mass of NOL3 and NHD

4 formed through

electrochemical redox reactions at electrodes. The Fe0

content in the PRB and initial anode pHwere 5 wt% and 2.5,

respectively. See Test 2 (for 10 V) and Test 3 (for 40 V) in

Table 1 for more detailed experimental conditions.

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8 775

conditions in wells, it was concluded that the generation of

NO�3 and NHþ

4 during the 100 h-experiment was negligible at

the applied voltage of 10 V, while significant amounts of NH4þ

were generated in the cathode at 40 V. Accordingly, when

experiments were performed at 40 V (Test 8 in Table 1),

0

20

40

60

80

100

0 20 40 60 80 100

Cathode NO 3 -

Cathode NH 4 +

Anode NO 3 -

AnodeNH 4

+

Total N

) %

( y r e v o c e r n e g o r t i

N

Time (h)

ba

0

20

40

60

80

100

0 20 4

)%( yrevocer negorti

N

c

Fig. 3 e Influence of the Fe0 content in the PRB on the NO3- redu

and initial pH in the anode well were 10 V and 2.5, respectively. (

content in the PRB [ 50 wt% (Test 5), (c) Fe0 content in the PRB

nitrogen recovery (NR) from the NO�3 contaminated Soil B,

which was defined by Eqn. (2), was not calculated for NHþ4 in

the cathode well.

NR ¼ Nitrogen species recovered in wells ðmgNÞTotal NO�

3 in the contaminated soil ðmgNÞ (2)

3.2. Influence of the Fe0 content in the PRB

To investigate the influence of the Fe0 content in the PRB on

NO�3 reduction efficiency and nitrogen recovery (NR), three

sets of experiments, each of which had a different Fe0 content,

were performed (Tests 4e6 in Table 1). The applied voltage

and initial pH in the anode well were 10 V and 2.5, respec-

tively. As shown in Fig. 3, when the Fe0 content was higher

than 50 wt%, nearly 100% of NO�3 eN was recovered as NHþ

4 in

the cathode well over the course of 100 h. However, when the

Fe0 content was decreased to 5 wt%, because of an insufficient

amount of Fe0, only 60% of NO�3eN was recovered as NHþ

4 in

the cathode well in 100 h. The remaining NO�3eN penetrated

through the Fe0 PRB and was recovered in the anode well.

Therefore, it was demonstrated that the Fe0 content in the PRB

was an important factor in controlling the performance of the

EK/Fe0 PRB system.

These results showed that the major reaction product of

NO�3 reduction by the Fe0 PRB was NHþ

4 , which is consistent

with the previous studies that investigated the reduction of

NO�3 by Fe0 using batch systems (Cheng et al., 1997; Huang

et al., 1998, 2003; Choe et al., 1999; Huang and Zhang, 2002,

0

20

40

60

80

100

0 20 40 60 80 100

) %

( y r e v o c e r n e g o r t i

N

Time (h)

0 60 80 100

Time (h)

ction efficiency and nitrogen recovery. The applied voltage

a) Fe0 content in the PRB[ 80 wt% (Test 4 in Table 1), (b) Fe0

[ 5 wt% (Test 6).

Page 5: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

0

20

40

60

80

100

0 20 40 60 80 100

Cathode NO3

-

Anode NO3

-

Anode NH4

+

Nitr

ogen

rec

over

y (%

)

Time (h)

Fig. 5 e Nitrogen recovery in wells as a function of

energization time (Test 8 in Table 1). Experimental

conditions: Fe0 content in the PRB [ 5 wt%, applied

voltage [ 40 V, initial anode pH [ 4.5. The data for NHD4 in

the cathode well is not shown because a significant

amount of NHD4 was formed through electrochemical N2

reduction (Fig. 2). Correspondingly, the data for total

nitrogen is also not shown.

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8776

2005, 2006; Huang and Zhang, 2004; Hwang et al., 2011; Suzuki

et al., unpublished results). These results also clarified the

behavior and fate of NO�3 in the EK/Fe0 PRB system. Under the

electric potential gradient, the negatively charged NO�3 ion

migrated to the anode, reaching the Fe0 PRB and being reduced

to NHþ4 by reacting with Fe0 in the PRB. The positively charged

NHþ4 ion then migrated to the cathode. However, if the NO�

3

reduction rate in the Fe0 PRB was not fast enough, a portion of

NO�3 penetrated the Fe0 PRB without being reduced into NHþ

4

and was recovered in the anode well.

3.3. Influence of the initial pH in the anode well

To investigate the influence of the pH in the anode well, the

initial anode pHwas increased from 2.5 to 4.5 with holding the

Fe0 content in the PRB and the applied voltage the same as in

Test 6 (Tests 7 in Table 1). The obtained results are shown in

Fig. 4. The comparison between Figs. 3(c) and 4 showed that

the percentage of NO�3 penetrating through the Fe0 PRB

increased from 30 % to 60 % accompanying the increase in the

initial anode pH from 2.5 to 4.5. This result indicated that the

reaction rate between NO�3 and Fe0 in the PRB was enhanced

by decreasing pH, which is also consistent with the results of

previous studies in which batch experiments were performed

to investigate the effects of aqueous pH on reaction kinetics

(Cheng et al., 1997; Huang et al., 1998; Choe et al., 1999; Huang

and Zhang, 2004; Suzuki et al., unpublished results).

3.4. Influence of the applied voltage

Another set of experiments (Test 8 in Table 1) was performed

to investigate the influence of the applied voltage. In this

experiment, the applied voltage was increased from 10 V to

40 V, while the Fe0 content in the PRB and the initial anode pH

0

20

40

60

80

100

0 20 40 60 80 100

Cathode NO3

-

Cathode NH4

+

Anode NO3

-

AnodeNH4

+

Total N

Nitr

ogen

rec

over

y (%

)

Time (h)

Fig. 4 e Nitrogen recovery in wells as a function of

energization time (Test 7 in Table 1). Experimental

conditions: Fe0 content in the PRB [ 5 wt%, applied

voltage [ 10 V, initial anode pH [ 4.5.

were held the same as in Fig. 4 (5 wt% and pH ¼ 4.5). The ob-

tained results are shown in Fig. 5. A comparison between Figs.

4 and 5 showed that the proportion of NO�3 penetrating

through the Fe0 PRB increased from 60 % to 90 % accompa-

nying the increase in the applied voltage from 10 V to 40 V.

It is important to discuss the reason behind the deterio-

ration in NO�3 reduction efficiency caused by the increase in

the applied voltage because increasing the applied voltage

concurrently influences several factors controlling the NO�3

reduction efficiency. Increasing the applied voltage enhances

the electrolysis of water at the anode, which results in greater

acidification in the Fe0 PRB. In fact, the pH in the Fe0 PRB after

100 h of energization for the 40 V experiment (Test 8) was 5.0,

which was lower than the corresponding value (7.0) obtained

from the 10 V experiment (Test 7). Because the NO�3 reduction

rate by Fe0 was accelerated by decreasing pH (Cheng et al.,

1997; Huang et al., 1998; Choe et al., 1999; Huang and Zhang,

2004; Suzuki et al., unpublished results), the NO�3 reduction

efficiency was expected to be improved by increasing the

applied voltage from 10 V to 40 V. However, increasing the

applied voltage also accelerated the NO�3 migration rate

toward the anode well. As illustrated in Fig. 3(c) and Fig. 4, the

NO�3eN mass recovered in the anode reached a plateau in

roughly 60 h, showing that 60 h was required for NO�3 mole-

cules located at the far left column to migrate all the way to

the anodewell at the applied voltage of 10 V. In contrast, at the

applied voltage of 40 V, 20 h was sufficient to reach a plateau

as shown in Fig. 5. This discrepancy clearly demonstrated that

NO�3 molecules migrated faster in the Fe0 PRB at higher

applied voltages and indicated that adequate reaction time

between NO�3 and Fe0 might not be obtained at 40 V.

In summary, increasing the applied voltage could lead to

improvements in the NO�3 reduction efficiency because the Fe0

Page 6: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8 777

PRB is more effectively acidified, but could alternatively lead

to the deterioration of the NO�3 reduction efficiency because

the reaction time between NO�3 and Fe0 is reduced. Under the

experimental conditions investigated, the NO�3 reduction

efficiency was decreased by increasing the applied voltage

from 10 V to 40 V, indicating that the reduced contact time had

a larger negative impact compared to the positive impact

resulting from stronger soil acidification.

4. Discussion

The pH in the Fe0 PRB after 100 h of energization was in the

range of 5.0e7.0 (Tests 4e8 in Table 1), which was high

compared to the pH in the anodewell (pH 2.2e3.5). However, it

is important to note that the pH discrepancy did not mean

that the acid generated at the anode was not effectively used

to acidify Fe0 PRBs. In fact, when Fe0 was absent in the PRB

(Test 1), the pH in the PRB successfully dropped to 2.4, which

was about the same as the pH in the anode well (pH 2.2). This

result demonstrated that an acid front actually reached the

Fe0 PRB in 100 h. Thus, the observed pH discrepancy between

Fe0 PRBs and anode well in Tests 4e8 was a result of acid

consumption by dissolving Fe0 particles (Fe0 þ 2Hþ / Fe2þ þH2[). Because of the reaction, Fe0 particles in the PRB retained

ametallic glaze after 100 h of energization, and a black passive

coating that was visually confirmed in our study (Suzuki et al.,

unpublished results) was not observed. These experimental

results demonstrated that the EK/Fe0 PRB system has the

capability to effectively utilize the acid generated at the anode

to regenerate the Fe0 surface, which allows the system to treat

nitrate-contaminated subsurface water flows for a protracted

period of time. This is clearly a technical advantage of the EK/

Fe0 PRB system over traditional Fe0 PRB systems.

This study also demonstrated that the Fe0 content in the

PRB, the anode pH, and the applied voltage were all important

factors in controlling the NO�3 reduction efficiency. However,

more research is required to fully understand the complex

relationship between these factors. Furthermore, in addition

to the NO�3 reduction efficiency and nitrogen recovery (both of

which were evaluated in this study), there are several other

aspects that need to be considered in determining the reme-

diation conditions. First, this study confirmed that NH4þ was

generated at 40 V via N2 reduction (Eqn. (1) and Fig. 2).

Although NHþ4 is less toxic than NO�

3 , and for this reason, the

current study attempted to convert NO�3 into NHþ

4 , it is still

desirable to operate the EK/Fe0 PRB system at the lower

applied voltage to avoid or at least to minimize the generation

of NHþ4 . Second, ferrous ion (Fe2þ) that is released as a result of

Fe0 dissolution by acid gives the water a red rust coloring (i.e.,

red water) when exposed to oxidants (e.g., oxygen). In fact,

reddish-brown precipitation was visually observed in the

cathode well when the experiment was performed at 40 V

(Tests 3 and 8 in Table 1). Therefore, the initial anode pH and

the applied voltage need to be carefully determined to avoid

excessive acidification and subsequent Fe0 dissolution.

Finally, as shown in Table 1, the total energy consumption

strongly depended on the applied voltage. Under the experi-

mental conditions investigated in this study, the total energy

consumption was reduced by choosing a lower voltage,

although applying a higher voltage clearly shortened the

remediation time. Therefore, proper choice of the remediation

time and the applied voltage is essential for the EK/Fe0 PRB

system because these variables directly affect the total energy

consumption and the remediation cost.

Acknowledgement

The authors gratefully acknowledge the partial financial

support for this study by the Japan Society for the Promotion

of Science (JSPS), Grant-in-Aid for Scientific Research (B),

No.22360380, 2010.

r e f e r e n c e s

Acar, Y.B., Alshawabkeh, A.N., 1993. Principles of electrokineticremediation. Environmental Science and Technology 27,2638e2647.

Acar, Y.B., Gale, R.J., Alshawabkeh, A.N., Marks, R.E., Puppala, S.,Bricka, M., Parker, R., 1995. Electrokinetic remediation: basicsand technology status. Journal of Hazardous Materials 40,117e137.

Alshawabkeh, A.N., 2009. Electrokinetic soil remediation:challenges and opportunities. Separation Science andTechnology 44, 2171e2187.

Cheng, I.F., Muftikian, R., Fernando, Q., Korte, N., 1997. Reductionof nitrate to ammonia by zero-valent iron. Chemosphere 35,2689e2695.

Choe, S., Chang, Y.Y., Hwang, K.Y., Hwang, K.J., 1999. Kinetics ofreductive denitrication by nanoscale zero-valent iron.Chemosphere 41, 1307e1311.

Cundy, A.B., Hopkinson, L., Whitby, R.L., 2008. Use of iron-basedtechnologies in contaminated land and groundwaterremediation: a review. Science of the Total Environment 400,42e51.

Guerin, F.T., Horner, S., McGovern, T., Davey, B., 2002. Anapplication of permeable reactive barrier technology topetroleum hydrocarbon contaminated groundwater. WaterResearch 36, 15e24.

Hwang, Y.H., Kim, D.G., Shin, H.S., 2011. Mechanism study ofnitrate reduction by nano zero valent iron. Journal ofHazardous Materials 185, 1513e1521.

Henderson, A.D., Demond, A.H., 2007. Long-term performance ofzero-valent iron permeable reactive barriers: a critical review.Environmental Engineering Science 24, 401e423.

Huang, C.P., Wang, H.W., Chiu, P.C., 1998. Nitrate reduction bymetallic iron. Water Research 32, 2357e2364.

Huang, Y.H., Zhang, T.C., 2002. Kinetics of nitrate reduction byiron at near neutral pH. Journal of Environmental Engineering128, 604e611.

Huang, Y.H., Zhang, T.C., 2004. Effect of low pH on nitratereduction by iron powder. Water Research 38, 2631e2642.

Huang, Y.H., Zhang, T.C., 2005. Effects of dissolved oxygen onformation of corrosion products and concomitant oxygen andnitrate reduction in zero-valent iron systems with or withoutaqueous Fe2þ. Water Research 35, 1751e1760.

Huang, Y.H., Zhang, T.C., 2006. Effects of surface-bound Fe2þ onnitrate reduction and transformation of iron oxide(s) in zero-valent iron systems at near-neutral pH. Journal ofEnvironmental Engineering 132, 527e536.

Huang, Y.H., Zhang, T.C., Shea, P.J., Comfort, S.D., 2003. Effects ofoxide coating and selected cations on nitrate reduction by ironmetal. Journal of Environmental Quality 32, 1306e1315.

Page 7: An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

wat e r r e s e a r c h 4 6 ( 2 0 1 2 ) 7 7 2e7 7 8778

Kim, S.-S., Ham, S.-J., Cho, Y.-S., 2002. Electrokinetic remediationstrategy considering ground strata: a review. GeoscienceJournal 6, 57e75.

Ludwig, R.D., Smyth, D.J.A., Blowes, D.W., Spink, L.E.,Wilkin, R.T., Jewett, D.J., Weisener, C.J., 2009. Treatment ofarsenic, heavy metals, and acidity using a mixed ZVIcompost PRB. Environmental Science and Technology 43,1970e1976.

McMahon, B.P., Dennehy, F.K., Sandstrom, W.M., 1999. Hydraulicand geochemical performance of a permeable reactive barriercontaining zero-valent iron, Denver Federal Center. GroundWater 37, 396e404.

Niinae, M., Nishigaki, K., Aoki, K., 2008. Removal of lead fromcontaminated soils with chelating agents. MaterialsTransactions 49, 2377e2382.

Puls, W.R., Blowes, W.D., Gillham, W.R., 1999a. Long-termperformance monitoring for a permeable reactive barrier atthe U.S. coast guard support center, Elizabeth City, NorthCarolina. Journal of Hazardous Materials 68, 109e124.

Puls, W.R., Paul, J.C., Powell, M.R., 1999b. The application of in situpermeable reactive (zero-valent iron) barrier technology forthe remediation of chromate-contaminated groundwater:a field test. Applied Geochemistry 14, 989e1000.

Reddy, K.R., 2010. Technical challenges to in-situ remediation ofpolluted sites. Geotechnical and Geological EngineeringJournal 28, 211e221.

Reddy, K.R., Chinthamreddy, S., 2003. Sequentially enhancedelectrokinetic remediation of heavy metals in low bufferingclayey soils. Journal of Geotechnical & GeoenvironmentalEngineering 129, 263e277.

Reddy, K.R., Maturi, K., Cameselle, C., 2009. Sequentialelectrokinetic remediation of mixed contaminants in lowpermeability soils. Journal of Environmental Engineering 135,989e998.

Reddy, K.R., Saichek, R.E., 2004. Enhanced electrokinetic removalof phenanthrene from clay soil by periodic electric potentialapplication. Journal of Environmental Science and Health, PartA 39, 1189e1212.

Saichek, R.E., Reddy, K.R., 2005. Electrokinetically enhancedremediation of hydrophobic organic compounds in soils:a review. Critical Reviews in Environmental Science&Technology 35, 115e192.

Suzuki, T., Moribe, M., Oyama, Y., Niinae, M. Nitrate reduction byzero-valent iron: Equilibrium and kinetics studies. ChemicalEngineering Journal, unpublished results.

Thiruvenkatachari, R., Vigneswaren, S., Naidu, R., 2008.Permeable reactive barrier for groundwater remediation.Journal of Industrial and Engineering Chemistry 14, 145e156.

Vogan, L.J., Focht, M.R., Clark, K.D., Graham, L.S., 1999.Performance evaluation of a permeable reactive barrier forremediation of dissolved chlorinated solvents in groundwater.Journal of Hazardous Materials 68, 97e108.

Wilkin, R.T., Puls, R.W., Sewell, G.W., 2003. Long-termperformance of permeable reactive barriers using zero-valentiron: geochemical and microbiological effects. Ground Water41, 493e503.

Wilkin, T.R., Su, C., Ford, G.R., Paul, J.C., 2005. Chromium-removalprocesses during groundwater remediation by a zero valentiron permeable reactive barrier. Environmental Science andTechnology 39, 4599e4605.