research article geochemical modeling of trivalent...

Research ArticleGeochemical Modeling of Trivalent ChromiumMigration in Saline-Sodic Soil during Lasagna ProcessImpact on Soil Physicochemical Properties

Salihu Lukman1 Alaadin Bukhari2 Muhammad H Al-Malack2

Nuhu D Mursquoazu3 and Mohammed H Essa2

1 Department of Civil Engineering ACHB King Fahd University of Petroleum and Minerals Hafar Al-Batin 31991 Saudi Arabia2Department of Environmental Engineering University of Dammam Dammam Saudi Arabia3 Environmental Engineering Department University of Dammam Dammam 31451 Saudi Arabia

Correspondence should be addressed to Salihu Lukman salihulukmanyahoocom

Received 16 March 2014 Accepted 27 May 2014 Published 24 July 2014

Academic Editor Claudio Cameselle

Copyright copy 2014 Salihu Lukman et alThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Trivalent Cr is one of the heavymetals that are difficult to be removed from soil using electrokinetic study because of its geochemicalproperties High buffering capacity soil is expected to reduce the mobility of the trivalent Cr and subsequently reduce the remedialefficiency thereby complicating the remediation process In this study geochemical modeling and migration of trivalent Cr insaline-sodic soil (high buffering capacity and alkaline) during integrated electrokinetics-adsorption remediation called the Lasagnaprocess were investigated The remedial efficiency of trivalent Cr in addition to the impacts of the Lasagna process on thephysicochemical properties of the soil was studied Box-Behnken designwas used to study the interaction effects of voltage gradientinitial contaminant concentration and polarity reversal rate on the soil pH electroosmotic volume soil electrical conductivitycurrent and remedial efficiency of trivalent Cr in saline-sodic soil that was artificially spiked with Cr Cu Cd Pb Hg phenol andkerosene Overall desirability of 0715 was attained at the following optimal conditions voltage gradient 036Vcm polarity reversalrate 1763 hr soil pH 100 Under these conditions the expected trivalent Cr remedial efficiency is 6475

1 Introduction

In early 1992 a discussion took place between the thenMonsanto Chief Executive Officer (CEO) and Administra-tor of the United States Environmental Protection Agency(USEPA) which ultimately led to the invention of the Lasagnaprocess [1] In the late 1993 Brodsky and Ho of Monsantofiled the first Lasagna US patent followed by a secondone all published in 1995 [2 3] In the Lasagna processcontaminated soil is remediated by creating at least oneliquid permeable zone within a contaminated soil regionand turning it into treatment zone Appropriate materials(sorbents catalytic agents microbes oxidants and buffers)are then introduced into the treatment zone An electrodeis placed at the first end of the contaminated soil regionand another of opposite charge is placed at the opposite end

of the contaminated soil region A direct electric currentis then transmitted through the contaminated soil regionbetween the two electrodes This causes movement of waterand dissolved organic and inorganic materials in subsurfacesoils from one electrode (anode) to the other (cathode) underelectroosmosis as a result of currentmovement from anode tocathode In 1802 electroosmosis was first observed detailedstudy of the mechanism was done by Reuss [4] in his classicexperiment reported by Abramson [5] In 1909 Freundlichand Neumann [6] provided the general name ldquoelectrokineticphenomenardquo to refer to the electrically driven mass flowof dissolved contaminants and pore fluid transport in soilsinduced by an applied DC voltage It is made up of transportof pore fluid via electroosmosis (EO) and transport of ionsor charged species via electromigration [7] The direction

Hindawi Publishing Corporatione Scientific World JournalVolume 2014 Article ID 272794 20 pageshttpdxdoiorg1011552014272794

2 The Scientific World Journal

12

1

08

06

04

02

0

minus02

minus04

minus06

0 2 4 6 8 10 12 14

Eh (V

)

Cr3+

CrOH2+ Cr(OH)2+

CrO42minus

Cr(OH)30

Cr(O

H) 4

minus

HCrO4minus

PH2 = 1bar

Po2 = 1bar

pH

Figure 1 Redox potential (Eh)-pH diagram for CrndashOndashH system[12]

and quantity of contaminantmovement are influenced by thecontaminant concentration solubility speciation degree ofhydrophobicity soil type and structure and the mobility ofcontaminant ions as well as the interfacial chemistry and theconductivity of the soil pore fluid [8]The remedial efficiencygenerally depends on the nature of the contaminants andsoil properties such as pH permeability adsorption capac-ity buffering capacity and geochemical processes (such asacidbase reactions and migration dissolutionprecipitationredox reactions complexation and speciation) [7 9] Saline-sodic soils possess electrical conductivity above 4 dSmsoil paste pH greater than 82 and exchangeable sodiumpercentage greater than 15 [10 11]

First application of electrokinetics took place in India inthe 1930s It was used to remove excess salts from alkali soilsin order to restore it to arable condition [13] Following itsinvention in 1993 extensive studies started in 1994 in bench-scale [14] then scaled up in a pilot-scale under laboratoryconditionsThefirst field test called ldquoPhase I Small Field Testrdquowas conducted in 1995 at the Paducah gaseous diffusion plant(PGDP) site whose soil was contaminatedwith trichloroethy-lene (TCE) Full-scale remediation using the Lasagna tech-nology was undertaken at two other contaminated sites inthe United States [15] The specific details of all the Lasagnaprocess implementations are presented in Tables 1 and 2 Itis noteworthy that only two of the studies have consideredsimultaneous removal of contaminant mixture All othershave dealt with only a single organic compound or heavymetal It has already been observed that contaminated soilsdo not contain single contaminants only but usually severalpollutants appear in the soil in mixed components [16ndash19]

The geochemical properties of the most stable forms ofCr that is trivalent and hexavalent Cr under electrokineticremediation have been extensively studied in different types

of soils (kaolin glacial till etc) by Reddy and his coworkersand other investigators [28ndash36] The trivalent Cr thoughconsidered relatively nontoxic compared to the hexavalentCr exists in the subsurface environments as cation Cr3+ andin the following hydroxocomplex forms Cr(OH)

4

minus CrOH2+and Cr(OH)

3

0 Cr3+ and CrOH2+ ions are mostly prevalentat soil pH values less than 6 while Cr(OH)

4

minus and Cr(OH)3

0

ions prevail when pH is greater than 118 The redox statealso affects the Cr state with reduced state favoring thepresence of the trivalent Cr while the oxidized state favorsthe existence of the hexavalent Cr Most of the trivalent Crspecies are less mobile because of their low solubility overwide pH range (lt12) and may be readily adsorbed by thenegatively charged clay surfaces There exists redox state inthe subsurface environment because of the generation ofoxygen and hydrogen gases at the electrodes in addition tothe possible presence of iron (reducing agent) manganese(oxidizing agent) or microorganisms The redox potential(Eh) and soil pH determine the possible oxidation of Crfrom the trivalent to hexavalent form as shown in Figure 1Chinthamreddy and Reddy [28] have found no significantoxidation of trivalent Cr in high buffering capacity soil suchas glacial till

Empirical modeling using response surface methodology(RSM) offers great and numerous advantages which includelarge amount of information from a small number of exper-iments evaluation of simultaneous interaction effects of theindependent parameters on the responses and simultaneousoptimization of multiple factors and responses for obtainingoptimal conditions [37 38]The key success of RSM is uncov-ering interactions of factors which cannot be achieved usingthe traditional one-factor-at-a-time (OFAT) optimizationapproach [39] Fundamental understanding of the physicsand chemistry which governs the process is essential in deter-mining the influential factors to be investigated and theirlevels or ranges are necessary for successful implementationof RSM for any process modeling and optimization Basicallythere exist four different experimental designs for RSMimplementation 3-level factorial design (3FD) Box-Behnkendesign (BBD) central composite design (CCD) andDoehlertdesign (DD) Bezerra et al [38] have reviewed each of thesedesign methods The Box-Behnken design is obtained bycombining two-level factorial designs with incomplete blockdesigns followed by adding a specified number of replicatedcenter points BBD is preferred when investigating three (3)factors using RSM because it will give enough informationfor analyzing factor-response interactions from the leastexperimental runs when compared to 3FD and CCD

Some microbially-driven biotransformation processesmay affect the soil physicochemical properties after elec-trokinetic remediation because of the passage of electriccurrent and development of pH gradients [40] These leadto original soil mineral degradation and alteration viabiotransformation While biotransformation deals with thebioweathering and alteration or degradation of clay mineralsbiomineralization refers to the formation of amorphousand crystalline materials from aqueous ions by biologi-cally mediated processes In addition to current and pH

The Scientific World Journal 3

Table1Ap

plications

ofLasagn

aprocessatbench-scales

from

inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Cell

dimensio

ns(le

ngthtimes

widthtimesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)current

(mA)

Power

consum

ption

kWhrm

3

Runtim

edays

Electro

osmotic

cond

uctiv

ity

cm2Vminus1sminus1

(times10minus5)

Treatm

ent

zone

spacing

cm

Reference

AClowast+sand

bacteria+AC

+sawdu

stp-nitro

phenol

Kaolinite

10cm

ID

216cm

long

Yes

continuo

us90ndash9

91ndash73(con

stant)

1020

25

6[14

20]

AC(Bam

boo

charcoal)

Cd

Sand

yloam

24cmtimes10cm

times8c

mYes

continuo

us796

17ndash27

mdash12

mdash10

[21]

AC(Bam

boo

charcoal)

Cd

Kaolin

24cmtimes10cm

times8c

mNo

continuo

us93

13ndash23

mdash8

mdash10

[22]

AC(Bam

boo

charcoal)

24-

dichloroph

enol

andCd

Sand

yloam

24cmtimes10cm

times10cm

Yes

continuo

us

7597(C

d)

5492(24-

dichloroph

enol)

1Variable

1219

1ndash12848

105

mdash16

[23]

GAClowastlowast

CrC

dCu

Pb

HgZn

ph

enol

kerosene

Salin

e-sodic

clay

24cmtimes10cm

times12cm

No

continuo

us

759(C

r)344

(Cd)41(Cu

)558(Pb)9249

(Hg)268(Zn)

100(pheno

l)498(kerosene)

06ndash

1880

1777ndash4

273

21425

6[24]

lowastAc

tivated

carbon

lowastlowastGranu

lara

ctivated

carbon


Table2Ap

plications

ofLasagn

aprocessatpilot-andfield-scalesfrom

inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Sitedimensio

ns(le

ngthtimeswidth

timesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)Cu

rrent

(A)

Power

consum

ption

kwhm

3

Runtim

emon

th

Electro

osmotic

cond

uctiv

ity

cm2Vminus1sminus1(times10minus5)

Treatm

ent

zone

spacingcm

Reference

AC+sand

p-nitro

phenol

Kaolin

Kaolinite

clayloam

122mtimes061m

times061m

Yes

continuo

us98

1(constant)962

(based

oncurrentd

ensity)

513

056ndash17

3556

[20]

GAC

1TC

E2Clay

loam

46mtimes3m

times46m

Yes

continuo

us99

035ndash0

4540

(con

stant)

mdash4

1260

[25]

Iron

filings

+kaolin

TCE

Clay

loam

64mtimes92

mtimes137m

Yes3-week

95ndash9

9023ndash0

31(con

stant)

110ndash200

mdash12

1260

amp150

[26]

Iron

filings

+kaolin

TCE

Clay

loam

274mtimes22

mtimes135m

Nopu

lsemod

e99

015ndash0

26

(con

stant)

500ndash

700

mdash24

mdash150

[1527]

Iron

filings

+kaolin

TCE

Clay

loam

33mtimes24

mtimes75

mmdash

60(afte

r1year)

016

(con

stant)

250ndash

400

mdash24

mdash150

[15]

1Granu

lara

ctivated

carbon

2Trichloroethylene


Gas ventGas vent

ValveValve

Filterpaper

Filterpaper

Cathodiccompartment

Cathodeelectrode

Anodecompartment

Anodeelectrode

Granularactivatedcarbon(GAC)

Contaminatedsoil

B

C

F G

D



Power supply multimeter

Figure 2 Coupled electrokinetics-adsorption experimental setup

14

12

10

8

6

4

2

0R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

pH

Figure 3 Weekly pH variation

Table 3 Codification and ranges of factors

Variable Designation Units Coded variable levelsminus1 0 +1

Polarity reversal 119860 hr 0 24 48Voltage gradient 119861 Vcm 02 06 1Concentration 119862 mgkg 20 60 100

gradients heavy metals also cause to affect the followingbiological assays soil microbial biomass carbon enzymeactivity basal soil respiration and earthworm assays andseed assays [41ndash44] Given the aforementioned intricacies

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

250

200

150

100

50

0

Elec

tric

al co

nduc

tivity

(dS

m)

Figure 4 Weekly soil electrical conductivity variation

of the geochemical behavior and migration of trivalent Crin soil during electrokinetic remediation this study wasaimed at investigating trivalent Cr migration and remedialefficiency in high buffering capacity and alkaline soil duringelectrokinetic study in addition to the impacts of the soilremediation on the physicochemical properties of the soil Acarefully designed experiment using BBD was used to studythe interaction effects of voltage gradient initial contaminantconcentration and polarity reversal rate on the trivalent Crremedial efficiency in saline-sodic soil that was artificiallyspiked with Cr Cu Cd Pb Hg phenol and kerosene usingRSMmodeling and optimization tools


Table 4 Design of experimental runs using the Box-Behnken design

Run order Polarity reversal 119860 (hr) Voltage gradient 119861 (Vcm) Concentration 119862 (mgkg) Remedial efficiency 1 0 06 20 0002 48 06 20 0003 24 1 20 0004 24 1 100 0005 24 06 60 79976 0 1 60 72737 24 02 20 0008 0 02 60 36939 48 1 60 656610 48 06 100 00011 0 06 100 255012 24 02 100 00013 48 02 60 3488

Table 5 Comparing electrical current with voltage gradient and soilpH for all tests

Run Current A Voltage gradient Vcm pHR6 302 1 129R9 265 1 126R3 225 1 126R4 204 1 127R2 132 06 109R1 117 06 102R10 112 06 119R5 103 06 112R11 061 06 120R8 021 02 98R12 015 02 83R7 014 02 81R13 013 02 104

Table 6 A sample mass balance analysis of trivalent Cr for Runs 811 and 13

Runs Run 8 Run 11 Run 13Initial concentration mgkg 3720 7795 3720Residual concentration mgkg 2346 5808 2423

1st GAC chamber FInitial concentration mgkg 690 690 690Residual concentration mgkg 2115 000 1970

2nd GAC chamber GInitial concentration mgkg 690 690 690Residual concentration mgkg 1448 000 2530Mass balance 12175 7451 14899

2 Materials and Methods

21 Characterization Natural saline-sodic clay obtainedfrom Al-Hassa Oasis Saudi Arabia was used in this studyThe soil has the following characteristics pH (83) mois-ture content (391) soil organic matter (259) electrical

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13Experimental runs

3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)

Figure 5 Cumulative electroosmotic volume for each test

conductivity (1524 dSm) specific surface area (907m2g)pore volume (0014 cm3g) pore size (6255 A)mdashmineralogyfrom X-ray diffraction (XRD) quartz (SiO

2) (874) calcite

(CaCO3) (52) and dolomite (CaMg(CO

3)2) (74) X-ray

fluorescence spectroscopy (XRF) revealed that the soil con-sists of the following elements Ca (3764) Si (3473) Fe(1041) Al (76) K (342) Mg (248) Pd (285) andTi (086)These properties were determined usingmethodsof the American Society of Testing and Materials (ASTM)standards and were reported elsewhere [45] The granularactivated carbon (GAC) used in the present study whosesurface area is 952m2g was produced locally from datepalm pits using phosphoric acid impregnation method Itscharacterization and properties have been reported elsewhere[46 47]

22 Adsorption Testing Single and competitive adsorption offive heavy metals (Cr Cd Cu Zn and Pb) were performedto determine the selectivity sequence and to understand theadsorption behavior of these metals under different pH con-ditions This is particularly important to this study because


Table 7 Comparing trivalent Cr remedial efficiency with factors and some responses

Runs Remedial Current Residual Electroosmotic volume Polarity reversal Voltage gradient Initial Crefficiency A pH mL rate hr Vcm concentration mgkg

R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100

Table 8 Experimental validation of trivalent Cr remedial efficiency and soil pH using voltage gradient = 1Vcm average concentration =4415mgkg and polarity reversal rate = 0 hr

Response Experimental result Model prediction Prediction error 90 CIlowast 90 CI 90 PIlowastlowast 90 PIlow high low high

Cr remedial efficiency 7588 5111 3264 3117 7595 1836 10000Residual soil pH 123 126 235 117 135 108 140lowastConfidence intervallowastlowastPrediction interval

Table 9 Optimal factor levels required to maximize remedialefficiency of trivalent Cr

Item ValuePolarity reversal hours 1763Voltage gradient Vcm 036Concentration mgkg 6000Expected remedial efficiency of trivalent Cr 6475Expected residual soil pH 1000Desirability 0715

Table 10Values of soil surface area and pore volume and size beforeand after treatment

Description BETlowast surface area Pore volume Pore sizem2g cm3g A

Before 907 0014 6255After 1121 0045 16324lowastBET Brunauer-Emmett-Teller

Table 11 Soil mineralogical transformations before and after treat-ment

Phase name Before After Quartz SiO2 874 553Calcite CaCO3 52 447Dolomite CaMg(CO3)2 74 mdash

Table 12 Values of constituent soil elements before and aftertreatment

Element Before After Ca 3764 4206Si 3473 2342Fe 1041 1506Al 76 955K 342 461Mg 248 249Pd 285 146Ti 086 135

soil mineralogy affects heavy metal adsorption behavior andselectivity sequence under different pH conditions Lukmanet al [45] reported the detailed procedures carried out for thecompetitive adsorption testing

23 Coupled Electrokinetics-Adsorption Study Fifteen (15)bench-scale experiments each having a 21-day run timewere designed and performed to investigate the migrationand distribution of trivalent Cr in a contaminant mixtureusing the coupled electrokinetics-adsorption technique andto understand the operating variablesrsquo effects on saline-sodicsoil


minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation

Deviation from reference point (coded units)

Soil

pH3

rd w

eek

(ove

rall)

B voltage gradient Vcm = 060

Actual factors

C concentration = 6000

A polarity reversal hours = 2400

(a)

1291

814

02

06

10

20

60

100

C concentration (mgkg)B voltage gradient (Vcm)

89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor

X1 = C concentration

X2 = B voltage gradient (Vcm)


(b)

Actual factorsA polarity reversal hours = 2400



minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)



X1 = C concentrationX2 = B voltage gradient (Vcm)

(d)

Figure 6 Perturbation plots showing the relative significance of factors on soil pH (a) and electrical conductivity (c) (left) 3D responsesurface and contour plots showing how the influential factors affect soil pH (b) and electrical conductivity (d) (right)

24 Reactor Design and Experimental Procedures The Plexi-glas reactor total volume was about 2268 cm3 made of sevenchambers The overall reactor dimensions are 24 cm (long)times 10 cm (width) times 12 cm (depth) Approximately 1 kg of localKSA soil was artificially spiked with kerosene heavy metals(Cu Cr Cd Pb Zn and Hg) and phenol at predeterminedconcentrationsThoroughmixingwas done usingmechanicalmixer (GilsonCompany Inc) so as to achieve a homogeneousdistribution of the contaminants around the soil matrix Themixed spiked soil was placed in a fume-hood for dryingover a period of time necessary to evaporate the solvents(hexane and distilled water) Distilled water was added toadjust the final moisture content of the soil to about 33ndash70 The initial conditions of the soil pH moisture contentorganic matter and electrical conductivity were measured aswell as the actual initial concentrations of the contaminantsThen the uniformlymixed contaminated soil was placed intothe cell layer by layer Each layer was compactedwith stainless

steel spatula so that the amount of void space wasminimizedThe reactor used for the experiments consists of the cell twographite electrodes serving as anode and cathode DC powersupply (LG GP-505) processing fluid reservoirs heavy dutyrecirculation pump (BVP Instratec) portable data logger(TDS-303 Tokyo Sokki Kenkyujo Co Ltd) for real-timemonitoring of temperature electric current and voltageacross the system (Figure 2) The two electrode compart-ments with 240mL working volume placed at each end ofthe cell were isolated from the soil zone by a porous Perspexplate and filter paper The conditioning of the electrolytewas controlled using anolyte (2N NaOH) and catholyte (1NHNO

3)The pH of the processing fluids was monitored every

8 hr for the 21-day duration of each test Based on the pHand volume the processing fluids remaining in the electrodechambers complete replacement or refill were carried outaccordingly Two planar-shaped electrodes 10 cm times 10 cm times05 cm were used to generate a uniform electric field Within


14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D

Spatial pH distribution in the reactor

Figure 7 pHprofilewith twoGAC treatment zones for investigatingbipolar effects (R11)

the described cell two treatment zones that cut across the cellvertically bracketing the spiked soil compartment were filledwith the GAC The data monitoring system was recordingelectric current variation applied voltage and temperature ofthe soil compartments online following a 30min preset timestep and automatically stores them for subsequent retrievalusing floppy disc which can be read using personal computerfor easy data and energy consumption analysis The powersupply provides a constant DC electric voltage for the elec-trokinetic tests Every week fractions of the soil specimenswere taken at the center of each chamber to determine theresidual concentrations of the contaminants soil pH watercontent organic matter and electrical conductivity Uponthe completion of each test the electrode assemblies weredisconnected and the soil specimen was extruded from thecell sectioned into parts weighed and preserved in glass vialsfor organic extraction heavymetal digestion and subsequentanalyses using the analytical procedures outlined below

25 Analytical Procedures for ContaminantExtraction and Analysis

Heavy Metals Extraction of heavy metals from soil sampleswas performed according to guidelines spelt out in EPAmethod 3050B for acid digestion of soils sediments andsludges [48] and analyzed using flame atomic absorptionspectrometry (AAnalyst 700 Perkin Elmer) All soil sam-ples were extracted in duplicate EPA method 7000B [49]was employed for heavy metal analysis using flame atomicabsorption spectrometry except for mercury which wasanalyzed using mercury analyzer (Solid Mercury AnalyzerSMS 100 Perkin Elmer) according to EPA method 7473

[50] Visual MINTEQ 30 [51] was employed to model themetal ion speciation using its dissolved concentration pHtemperature and ionic strength

Kerosene and Phenol A mixture of methylene chloride andhexane (1 1) (vv) was used as the extraction solvent Soilsamples were extracted using pressurized fluid extractionaccording to EPAmethod 3545 procedures [52] using acceler-ated solvent extractor (ASE 200 Dionex) Volume of extractgenerated was then injected into the GC-MS (Clarus 580Perkin Elmer) equipped with autosampler for analysis TPHquantification was done by using the total chromatographicarea counts using retention time range for the elution ofhydrocarbon within the kerosene range C

8ndashC16 Guidelines

spelt out in EPA mMethod 8270D [53] for the quantificationof semivolatile organics by GC-MS were adhered to

251 Data Reliability QC Protocols Accuracy and PrecisionTo evaluate reliability of the analytical procedures duplicatesamples were analyzed for each sample Quality control (QC)protocols spelt out in EPA method 7000B [49] were usedThese include the use of initial calibration blank (ICB)initial calibration verification (ICV) continuous calibrationverification (CCV) and continuous calibration blank (CCB)The accuracy of the spiked soil samples was evaluated usingpercent recovery set at about plusmn30 of the spiked value [49]Repeatability of the experimental results was assessed byensuring that the precision obtained using the relative percentdifference (RPD) was not above 30

26 Testing Program and Mathematical Model DevelopmentBox-Behnken design (BBD) was chosen for the experimentaldesign because of its advantages over central compositedesign (CCD) and 3-level factorial design when dealingwith only three factors In BBD the experimental pointsare hyperspherically arranged equidistant from the centralpoint [38] Response surface methodology (RSM) was usedin modeling optimization and interpretation of the resultswith the help of Design-Expert version 8 (Stat-Ease Inc)platform [39 54] The investigated variables (called factorsin RSM) are the polarity reversal rate voltage gradient andinitial contaminant concentration designated as A B andC respectively These variables were selected based on theirknown influence on contaminant remedial efficiency andwere coded and varied according to Table 3 Based on thefactor levels and the chosen number of central points (3)a total of fifteen (15) experiments were randomly designedusing the BBD (Table 4) and subsequently conducted Onlyone central point is shown in Table 4

This experimental design was preliminarily evaluatedusing variance inflation factor (VIF) to check for orthogo-nality (independence of factors) and leverage which quanti-tatively measures the potential of a design point to have sig-nificant influence on model fit [54] These were determinedusing (1) and (2) respectively VIF value of 1 indicates that thefactor is orthogonal to all other factors in the design In a casewhereby factors are highly correlated then R2 value becomesa poor indicator of modelrsquos predictive ability and it becomes


minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102

A polarity reversal hours (hr)B voltage gradient (Vcm)

X1 = A polarity reversal (hours)X2 = B voltage gradient (Vcm)


(b)

Figure 8 (a) Perturbation plot showing the relative significance of factors on electroosmotic volume (b) 3D response surface and contourplots showing the influence of voltage gradient on cumulative electroosmotic volume

10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)

Figure 9 Comparing variations of electric current with soil temperature (a) current (b) temperature

more and more difficult to unravel how each of the investi-gated factors affect the response Experimental points havinghigh leverage values close to 1 should influence the model fitby carrying any error (experimental or measurement) intothe model as such they should be conducted more carefully[39]

VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)

where 1198772119894is the coefficient of determination 119901 is the number

of model terms and 119899 is the number of experimentsFollowing design evaluation the responses were fitted to

a quadratic model which was fine-tuned by removing anyinsignificant term This will maximize R2 and minimize lack

of fit The general quadratic equation for fitting models inRSM is

119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)

where 119910 is the response or dependent variable 119896 is the num-ber of factors 120573o 120573i 120573ii and 120573ij are the coefficients to be fittedusing regression for constant term linear quadratic andinteraction parameters respectively and 119909 is the variables

The developed models were evaluated using the richdiagnostic tools provided in Design-Expert which includenormal plot of residuals (to test the assumption of normal-ity of residuals) predicted versus actual plot (to test theassumption of constant variance) Box-Cox plot (to checkthe need for data transformation) and externally studentizedresiduals (to check the presence of any outlier in the data)


minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor


X1 = C concentrationX2 = B voltage gradient Vcm


(b)

Figure 10 (a) Perturbation plot showing the relative significance of factors on average electric current (b) 3D response surface and contourplots showing the influence of voltage gradient on average electric current

Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0

Soil and GAC chambers

1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C

Figure 11 Trivalent Cr distribution andmigration from the contam-inated chamber to the GAC chambers after 13 tests

The effects of factors were compared at a particular pointin the design space using the perturbation plot Responsesurface and contour plots were then generated

27 Use of Desirability Function in Numerical Optimiza-tion Desirability function being one of the mathematical

0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)

Figure 12 Speciation diagram for trivalent Cr species at differentweekly pH values

methods for computation of critical values (maximum orminimum) andmeasuring overall success of optimizingmul-tiple responses using geometric mean was employed for theoptimization of trivalent Cr remedial efficiency A search fora combination of factor levels which simultaneously satisfiesthe goals imposed on factors and responses is first carried outfollowed by combining these goals into an overall desirabilityfunction that ranges from 0 (outside of the optimizationlimits) to 1 (at the goal) Combining all responses into overalldesirability eliminates favoring one response over another



1st week2nd week3rd week

100

80

60

40

20

0

Rem

oval

()

Figure 13 Weekly percentage removal of trivalent Cr for 13 tests

The aim is not to clinch a desirability value of 1 but to find agood set of conditions that will meet all the set goals for eachfactor and response [55] This is achieved using numericaloptimization algorithms [52]

3 Results and Discussion

Discussion of the monitored results obtained after perform-ing thirteen (13) tests with 3 centre points will be focusedon geochemical processes affecting sorptiondesorptionand migrationremoval mechanisms such as the develop-ment of acidbase fronts migration and reactions dissolu-tionprecipitation oxidationreduction reactions complexa-tion and metallic ion speciation In addition presentation ofthe developed mathematical models and discussion on howthe factors affect the respective responses will follow

31 Single and Competitive Adsorption of Heavy Metals onClay Lukman et al [24 45] have discussed the physicochem-ical characteristics of the saline-sodic soil Additional discus-sion will be provided in the subsequent sections Lukman etal [45] have found out that the adsorptive capacities of Cuand Zn ions are higher in the multicomponent adsorptionscenario than in the single component scenario The adsorp-tion selectivity sequences obtained using the coefficient ofdistribution for the single and multicomponent scenariosare Cr gt Pb gt Cu gt Cd gt Zn and Cr gt Cu gt Pb gt Cdgt Zn respectively [45] Yong et al [40] have identified thegeneral factors that influence selectivity sequence to be ionicsize or activity first hydrolysis constant soil type and pHof the system From the multicomponent desorption studytrivalent Cr ions were tightly held by the soil surface thushaving the least percentage desorption followed by Cd andCu ions Reddy and his coworkers [28 30 32] have reportedthat trivalent Cr ions adsorb highly to soil solids and formcationic species that are insoluble over a wide range of pH

This is in line with the present findings by Lukman et al [45]which revealed high selectivity for the trivalent Cr duringmulticomponent adsorption and desorption tests

32 Soil pH Distribution Electrical Conductivity BipolarEffects Electroosmotic Flow and Current

Soil pH andElectrical ConductivityThe soil pH (83) indicatesthat it contains appreciable soluble salts capable of undergo-ing alkaline hydrolysis such as sodium carbonate [11] Thehydrolysis of calcite and dolomite may be limited by theirlow solubility thus producing a pH of about 8ndash82 in soilsIn addition Na+ ions do not strongly compete with H+ ionsfor exchange sites as do Ca2+ ions that are strongly and moretightly held on the soil surface The inability of the displacedNa+ ions to inactivate OHminus ions results in increased soil pHwhich is usually greater than 82 Moreover for a soil whosepH is greater than 82 its exchangeable sodium percentagehas to be greater than 15 [11] Presence of calcite and dolomitecoupled with alkaline hydrolysis of sodium carbonate giveshigh electrical conductivity to the soil (1524 dSm)

The saline-sodic nature of the soil necessitates the use ofprocessing fluids (2N NaOH and 1N HNO

3) to continuously

neutralize the rapidly generated H+ and OHminus ions at theanode and cathode respectivelyThese fluids were monitoredevery 8 hours and replaced as they degraded HNO

3and

NaOH are strong acid and base respectively and dissociatecompletely according to the following reactions

HNO3(l) 997888rarr H+ (aq) +NO

3

minus(aq) (4)

NaOH (aq) 997888rarr Na+ (aq) +OHminus (aq) (5)

Because of the electrochemical decomposition of waterOHminus and H+ ions are produced at the cathode and anoderespectively as shown in (6) and (7)

4H2O (l) + 4eminus 997888rarr H

2(g) + 4OHminus (aq) (6)

2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)

The electrochemically generated H+ andOHminus ions due towater electrolysis at the anode and cathode respectively areneutralized to form water molecules (8) because of the OHminusand H+ ions produced from the dissociation of the catholyteand anolyte respectively as shown in (5) and (4)

H+ (aq) +OHminus (aq) 997888rarr H2O (l) (8)

The oxygen and hydrogen gases generated may be ventedout while some amount may go into the soil and alterthe redox chemistry [56] Na+ and NO

3

minus ions migrate intothe soil to the opposite electrodes thereby increasing theelectrical conductivity as the treatment process progressesA sustained and variable electroosmotic flow was observeddue to the migration of the Na+ ions which could enhancethe migration of the double layer complexes toward thecathode while nitrate ions could be involved in complexformation with the cations [28] This electroosmotic flowwill lead to decreasing volume of the anolyte and increasing


minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation

Deviation from reference point (coded units)Actual factors

Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor

B voltage gradient (Vcm)C concentration (mgkg)

X1 = B voltage gradient (Vcm)X2 = C concentration


(b)

Figure 14 (a) Perturbation plot showing the relative significance of factors on trivalent Cr remedial efficiency (b) 3D response surface andcontour plots showing the influence of initial contaminant concentration on trivalent Cr remedial efficiency

10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000

A polarity reversal (hours)B voltage Gradient (Vcm)

C concentrationSqrt (Zn 3rd week (overall))

Pb 3rd week (Overall)Sqrt (Cu 3rd week (overall))Sqrt (Cd 3rd week (overall))Sqrt (Cr 3rd week (overall))Ln (Hg 3rd week (overall))Phenol 3rd week (overall)

TPH 3rd week (overall)Ln (energy comsumption)

Sqrt (average current)

Sqrt (electroosmotic volume)Sqrt (flushed pore volume)

Sqrt (electroosmotic conductivity)Sqrt (anolyte refill)

Sqrt (catholyte refill)Anolyte replacement

Catholyte replacementElectrical conductivity 3rd week

Soil pH 3rd week (overall)Combined

Figure 15 Combined and individual response desirability values for all responses and factors

100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

X1 = B voltage gradient (Vcm)X2 = C concentrationActual factorA polarity reversal hours = 1765

B voltage gradient (Vcm) C concentration

Figure 16 3D surface plot of the overall desirability variationrelative to influential factors

volume of the catholyte over time Hence refilling the anolyteis necessary if it has not degraded completely In additionsince the processing fluids are finite in volume and theelectrochemical decomposition of water at the electrodesis continuous for the test duration then a time will bereached when all the ions in the processing fluids havebeen exhausted Consequently rise and fall in catholytepH and anolyte pH respectively are expected before thecomplete replacement of the processing fluids Now OHminusions generated at the cathode according to (6) migrate intosoil toward the anode In this migration process soil pHrises (Figure 3) andmetal hydroxides are formedwhich couldprecipitate and reduce the electrical conductivity (Figure 4)and increase current consumption near the cathode [41]At the same time soluble hydroxocomplexes are formedwith the cations due to complexing property of the hydroxylions [24 57] On the other hand hydrogen ions generatedat the anode (7) migrate toward the cathode This process


may lead to soil protonation or desorption of indigenousand spiked heavy metals and hence increase the electricalconductivity (Figure 4) [32] Given the presence of calcite anddolomite in the soil minerals the developing acid front maybe buffered by the carbonate mineral thereby hindering anyfall in the soil pH (Figure 3) From the forgoing discussionit is clear that there will be an overall increase in the soilpH and electrical conductivity (Figure 4) as the integratedelectrokinetics-adsorption remediation progresses Resultsobtained for electrokinetic remediation of high bufferingcapacity glacial till by Reddy and hiscoworkers [30ndash32 58]have corroborated these findings The transient nature of theacidbase front migration and reactions may be responsiblefor the lower final values of some pH and EC than thepreceding 1st or 2nd week values In addition the elec-troosmotic flow (Figure 5) will undoubtedly vary spatiallyand temporally as it also depends on the soil zeta potentialprocessing fluids pH pore fluid viscosity and permittivity[7 59ndash61]

It was observed from Figure 3 that the average initial soilpH after spiking is within the range 77ndash8 lower than theoriginal soil pH (83) while the final pH ranges from 8 to 129The lower initial pHwas due to the acidity of the contaminantsolutions while the higher final pH values resulted from thehigh buffering capacity of the soil which neutralized the gen-erated acidic front from (7) but allowed the migration of thebasic front generated from (6) In addition the hydroxyl ionsgenerated from the dissociation of NaOH (5) and reductionof water at the cathode (6) aid in neutralizing the generatedacidic front Consequently all weekly pH values are higherthan the initially spiked soil pH for all the tests Additionallylow pH rise (8ndash104) was observed for all the tests conductedusing 02 Vcm (R7-8 R12-13) whereas highest pH (126ndash129)was recorded for all tests conducted using 1 Vcm (R3-4R6 and R9) consistently High voltage gradient leads to thepassage of high amount of current which increases the rateof the electrochemical decomposition of the electrolyte andenhances subsequentmigration of the basic front into the soilThis basic front migration is responsible for raising the soilpH This observed effect of the voltage gradient on the soilpH has been successfully modeled mathematically and thecoded linear model equation at 5 significant level (005 119875value) is presented in (9) while the graphical presentation ofthe significant influential factors together with 3D responsesurface and contour plots is given in Figures 6(a) and 6(b)

Soil pH = 1107 + 0097 lowast 119860 + 177 lowast 119861 + 039 lowast 119862 (9)

where A is the polarity reversal hr B is the voltage gradientVcm and C is the concentration mgkg

Anderson and Whitcomb [39] have reported that R2is biased hence a more accurate less biased and bettergoodness-of-fit statistic called adjusted R2 was computed forevaluating the model accuracy The modelrsquos R2 and adjustedR2 (unbiased estimate of the coefficient of determination)are 07725 and 07105 respectively High values of R2 areessential for modeling the experimental design space whilein identification of significant factors R2 value does notmatter and for significant factors will remain significant [39]

It is very clear that model equation perturbation and 3Dresponse surface plots have shown the significant influenceof voltage gradient on the soil pH over the other factors(polarity reversal rate and initial contaminant concentration)The relative contribution or effect of any given model termis directly proportional to its coefficient Perturbation plot(Figure 6(a)) revealed a sequence of relative influence ofthe operating parameters on the target response as followsvoltage gradient gt concentration gt polarity reversal

Bipolar EffectsThe two treatment zones F andGcontain 100granular activated carbon which may be used as electrodematerial due to its electrical conducting properties [14]The sides of the GAC chambers facing anode and cathodeelectrodes tend to behave as bipolar electrodes by acting ascathode and anode while the inner sides behave as anodeand cathode respectively These bipolar electrodes would beexpected to generateH+ andOHminus ions depending onwhetherthe side is acting as anode or cathode [20] and may beexpected to alter the pH distribution in the soil profileThesebipolar effects were investigated at the end of R11 and the pHprofile is presented in Figure 7The pHprofile shows the vari-ation of pH within the unspiked chambers B and D spikedchamber C and GAC chambers F and G The pH rangesfrom 119 (near the anode) to 126 (near the cathode) whichsuggest that bipolar effects did not manifest due to the pres-ence of carbonate minerals that impact high acid bufferingcapacity

Sparks [62] posited that electrical conductivity (EC) is thebest index for the assessment of soil salinity As important asthis parameter is most works on electrokinetic remediationfailed to at least report the soil electrical conductivity letalone monitor its variation over the treatment durationElectrical conductivity greatly influences electrokinetic reme-diation because it determines the amount of current flowingthrough the soil The usual voltage gradient of 1 Vcm forbench-scale studies [63] when applied to saline-sodic soilswould lead to high electric current flow Lukman et al [24]have reported that this would lead to excessive soil heatingreduction in the soil moisture content high energy andprocess fluid consumption high electroosmotic flow rate(Figure 5) and in some cases higher percentage removal ofcontaminants EC is simultaneously influenced by many soilproperties viz water content soluble salts grain size humustemperature texture and cation exchange capacity (CEC)[64] The 1st week of EC data shows that tests conductedusing 1 Vcm (R3 9 6) possess the highest EC values withR1 (06Vcm) coming second highest No discernible trendwas visible in the case of initial contaminant concentrationdespite its influence on the EC as depicted in Figure 6(c)Similar trend was observed for the 3rd week where R9and 6 have the highest EC values (Figure 4) A generalincrease of EC with time and voltage gradient (Figures 6(c)and 6(d)) was observed (except for R11) The reason forthis observation has been elaborately discussed above Thesevariations and impacts of the influential investigated factorshave been modeled and presented in the 3D response surfaceplot in Figure 6(d) Perturbation plot (Figure 6(c)) revealed


a sequence of relative influence of the operating parameterson the soil electrical conductivity as follows concentration gtvoltage gradient gt polarity reversal

Electroosmotic Flow The cumulative electroosmotic vol-ume for all the tests presented in Figure 5 shows that R2(20mgkg) R5 (60mgkg) and R4 (100mgkg) have thehighest values Other parameters that may influence elec-troosmotic flow are clay zeta potential voltage gradient andtime-dependent fluid properties such as dielectric constantand viscosity [65] Equation (10) shows the electroosmoticvelocity as derived according to Helmholtz-Smoluchowski(H-S) theory

V119890=120576119904120577

120578119864 = 119896

119890119864 (10)

where v119890is the electroosmotic velocity 120576

119904is the pore fluid

permittivity 120578 is the pore fluid viscosity 120577 is the soil zetapotential 119896

119890is the coefficient of electroosmotic conductivity

and E is the voltage gradientThese parameters make the measured electroosmotic

volume for all the tests to vary temporally The reduction ofthe thickness of the diffuse double layer resulting from higherionic concentration with subsequent higher ionic strengthcauses reduction in the electroosmotic flow [66] hencehigher concentrations usually yield lower electroosmoticvolume (Figure 5) Reddy et al [66] have observed similartrend The electroosmotic volume usually decreases withtime because of the increase in electrical conductivity withtime (Figure 4) that leads to higher ionic strength as the treat-ment proceedsMoreover voltage gradient has been observedto be most influential to the electroosmotic flow (Figure 8)The least electroosmotic volumes recorded belong to thelowest voltage gradient used (02 Vcm) that is in the caseof R7 R12 R8 and R13 This is because high voltage gradientcauses the passage of high electric current which leads tohigh electromigration with subsequent substantial transferof momentum to the surrounding pore-fluid molecules [66]The soil zeta potential defined as the electrical potentialexisting at the junction between the fixed and mobile partsof the electrical double is influenced by the type andconcentration of dissolved ions in the pore fluid in addition tothe pore fluid chemistry Clay soils being negatively chargedusually possess negative zeta potential At low pH belowthe point of zero charge (PZC) zeta potential may becomepositive because of excessive protonation and increase inionic strength resulting from increased dissolution of metalions in the pore fluid and their subsequent adsorption ontothe soil particles and compression of the electrical doublelayer [67] Reversal of the zeta potential charge could reversethe direction of the electroosmotic velocity as shown in(10) At high pH values such as those encountered in thisstudy deprotonation andmetal hydroxide precipitation couldmaintain a negative zeta potential hence electroosmoticflow will remain unidirectional as observed in all the testsElectroosmotic flow has not been influenced by hydraulicgradient in this study as it occurs even under negativehydraulic Equation (11) presents the model equation (1198772 =

0946 and adjusted 1198772 = 09057) relating the electroosmoticvolume to the factors Voltage gradient appears to be themost influential followed by polarity reversal rate and initialcontaminant concentration (Figure 8(a)) At high voltagegradient (1 Vcm) the decrease in the electroosmotic volume(Figure 8(b))may be attributed to the development of bubbleswithin the electrode chambers due to temperature risewhich then seeps into the soil to reduce the soil saturationwith subsequent reduction in the electroosmotic volume [24]

Sqrt (Electroosmotic volumemL)

= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862

+ 526 lowast 119861 lowast 119862 minus 534 lowast 1198602

minus 2095 lowast 1198612

(11)

Current and Temperature Table 5 presents the average elec-tric current recorded for each test during the 3-week testduration in descending order of magnitude to show howit is influenced by the applied voltage gradient and howit correspondingly affects the soil pH Clearly the higherthe voltage gradient the more amount of current is passedthrough soil which results in rapid generation ofH+ andOHminusions and subsequent rise in soil pH (Table 5) The current isusually low at the beginning of the tests (Figure 9(a)) risesgradually as the tests continue and then declines sometimesto a stable value while in some instances keeps on fluctuat-ing according to the time-dependent geochemical processestaking place such as ionic dissolution and precipitation anddegradation of the processing fluids Study conducted byMaturi and Reddy [68] corroborated the fluctuating currenttrend Upon application of the driving force the voltagegradient the processing fluids and pore fluid migrate whilethe dissolved ions electromigrate to opposite poles Theseprocesses lead to increase in the ionic strength of the porefluid thereby increasing the current flow to amaximumvalueThe observed decline of the current to a stable value may beattributed to the electromigration of cations and anions to therespective electrode with subsequent possible precipitationof the cations due to increase in the soil pH as the testprogresses [66 69] Temporal geochemical processes suchas mineral and chemical dissolution and neutralization reac-tions taking place in the electrode chambers also contributeto the variation of the electric current A maximum value of513 A was recorded for R6 whose average current was 302AThis current is considered extremely high considering thefact that it is about two orders of magnitude greater thanthe recorded current values for other bench-scale studiesthat employed the Lasagna process (lt30mA) in other soilapart from saline-sodic soil as shown in Table 1 Otherstudies using electrokinetic remediation only using voltagegradient of 1 Vcm or higher have reported higher valuesbut usually less than 300mA [58 66 70] Using low voltagegradient of 02 Vcmhas only resulted in reducing the currentto about 130ndash210mA (Table 5) This unique and importantobservation may be explained by the high salinity and


sodicity of the investigated soil which provides large amountof dissolved salts and minerals (carbonates) in the pore fluidfor sustained high electrical conduction High current flowthrough the soil will significantly affect the soil temperatureelectroosmotic flow rate electrode material and processingfluids degradation soil pH geochemical processes remedialefficiency and energy consumption In a related study byLukman et al [24] they also recorded similar high current(28 A) To emphasize on the effect of the electric currenton the soil temperature current and temperature readingsrecorded using a time step of 30min is presented in Figure 9for R11 (voltage gradient = 06Vcm)This test has 061 A and2845∘C as the average current and temperature respectivelyThe maximum values were 091 A and 346∘C respectivelywhich were recorded under room temperature of 24∘C Itis clear from Figure 9 that low current leads to low soiltemperature and vice-versa In a preliminary study conductedby Lukman et al [24] using 1 Vcm 3634∘C and 47∘C werethe average and maximum soil temperatures indicating thatthe soil becomes very hot when using 1 Vcm While soilheating may be advantageous in increasing the volatility oforganics solubility of minerals (carbonates) and reductionin pore fluid viscosity which will increase electroosmoticflow it may also be undesirable since it will reduce thesoil moisture content due to pore fluid evaporation withsubsequent reduction in current and electroosmotic flowIn addition it will increase soil electrical conductivity andenergy expenditure [20] Previous studies have not reportedsignificant rise of soil temperature during bench-scale tests[20] A linear model was obtained (12) which relates thefactors to the average electric current whose respective R2and adjusted R2 are 09556 and 09435 The perturbationand response surface plots (Figure 10) also revealed thesignificant influence of the applied voltage gradient overinitial contaminant concentration and polarity reversal rate

Sqrt (Average current)

= 1 + 0020 lowast 119860 + 059 lowast 119861 minus 0059 lowast 119862(12)

33 Trivalent Chromium Migration Model Validationand Optimization Figure 11 presented the distribution andmigration of trivalent Cr from the contaminated chamber Cto the GAC chambers F and G for all the thirteen (13) testsThis migration becomes more pronounced for tests R5 R6and R9 In the case of R6 (no polarity reversal) significanttrivalent Cr migration took place from the contaminatedchamber C to the GAC chamber F near the anode Thisobservation may be attributed to the formation of highamount of negatively chargedmetal hydroxocomplexes at pH129 which are then attracted to the anode via electromigra-tion but become adsorbed onto theGAC in chamber F duringthe transport process VisualMINTEQ 30 [51] was employedto model the trivalent Cr ion speciation for R5 from theweekly monitoring data using the dissolved concentrationpH temperature and ionic strength The speciation diagrampresented in Figure 12 reveals the increasing dominance ofthe negatively charged complexCr(OH)

4

minus and the decreasingconcentration of aqueous Cr(OH)

3at pH 112 This explains

the greater movement of the trivalent Cr species toward theanode in R6 at pH 129 Pourbaix [71] and Chinthamreddyand Reddy [29] have already asserted that Cr(OH)

4

minus ionswill become the dominant species at pH values greater than118 thus trivalent Cr solubility increases However undernormal soil pH trivalent Cr has limited solubility and highlyadsorbs to soil [29 32] In a related study by Reddy andChinthamreddy [30] which involved an alkaline and highacid buffering soil called glacial till they did not observesignificant trivalent Cr migration and no removal Althoughthe soil redox statemay be dynamic because of the generationof oxygen and hydrogen gases at the electrodes in additionto the possible presence of iron (reducing agent) manganese(oxidizing agent) or microorganisms that can oxidize thetrivalent Cr to the hexavalent form oxidation of trivalent Crdoes not take place appreciably in high buffering capacitysoil such as saline-sodic soil [28] For this reason hexavalentCr was not studied Migration of the trivalent Cr fromthe contaminated chamber to the GAC chambers indicatedremarkable remedial efficiency for some of the tests (R5 R6and R9) while others indicated low or no removal at all (R1ndashR4 R7 R10 and R12) There is zero remedial efficiency whenthere was accumulation of the contaminant at the samplinglocation thereby having the residual concentration (C

119900) to be

greater than the initial (C) in which case C119900C gt 1 Hence

Figure 11 utilized C119900C to indicate the migration of trivalent

Cr whenC119900C lt 1 or its accumulation at any given location or

chamber when C119900C gt 1

Mass balance analyses of Cr were performed for Runs 811 and 13 From Table 6 the mass balance for Runs 8 11 and13 is 12175 7451 and 14899 respectivelyThese values wereobtained using the ratio between the residual Cr in the con-taminated chamber (C) plus any increase in Cr concentrationin theGAC chamber and the initial Cr concentration Amongother reasons for the discrepancies in mass balance that issometimes encountered during electrokinetic remediation asput forward by previous investigators [30 66 72] includeadsorption onto the electrode and geotextilematerials (whichhouses the GAC in the two chambers) and non-uniformdistribution of contaminants within the small soil sample(about 2 g) taken for acid digestion and analysis Takingdifferent samples spatially from the contaminated chamberwill help improve the mass balance

The tests were sorted in decreasing order of remedialefficiency (Table 7) to reveal some salient points that willhelp in providing adequate connection between factors andresponses Highest remedial efficiencies (7997ndash3488) wererecorded for tests involving 60mgkg initial trivalent Crconcentration whereas no removal was recorded for alltests involving 20mgkg Only one test involving 100mgkgrecorded some remedial efficiency (Table 7) Low remedialefficiency at 20mgkg may be attributed to the availability ofadsorption sites for trivalent Cr ions coupled with the highselectivity for Cr for this particular soil type [45] at the givenconcentration At higher concentrations (100mgkg) and pHtrivalent Cr may precipitate as Cr(OH)

3 thus rendering it

immobile [66] Evenwith low electric current electroosmoticflow and voltage gradient (02 Vcm) 3488 and 3693of the trivalent Cr was removed from the contaminated


chamber in tests R13 and R8 respectively Polarity reversalrate did not show any discernible pattern Hence there isneed for simultaneous optimization of these three factors foroptimal removal of the trivalent Cr It is important to notethat high voltage gradient (1 Vcm) or passage of high electriccurrent does not necessarily translate into high remedialefficiency but will definitely increase the energy expenditureAt high voltage gradient current is high leading to highelectroosmotic flow toward cathode This opposite flow mayinterfere with the electromigration of the anionic trivalent Crspecies that are migrating toward the anode thus reducingthe overall remedial efficiency Electromigration constitutethe major transport mechanism for charged species whoserate is 10ndash300 times higher than the advective electroosmotictransport [73] At low voltage gradient (02 Vcm) extremelylow electroosmotic flow takes place and sustained electromi-gration prevails The weekly percentage removal of trivalentCr from the contaminated chamber is presented in Figure 13The dynamic and temporal changes in the geochemical pro-cesses controlling the contaminant removal are attributableto the observed trends in the weekly percentage removal

Equation (13) relates the investigated factors to the reme-dial efficiency with 09335 and 08966 as the R2 and adjustedR2 values respectively

Sqrt (Cr remedial efficiency)

= 878 minus 071 lowast 119860 + 058 lowast 119861

+ 063 lowast 119862 minus 150 lowast 1198612minus 739 lowast 119862

2

(13)

Perturbation plot (Figure 14(a)) also supports the observedinfluence of the initial Cr concentration on the remedialefficiency followed by voltage gradient then polarity reversalrate The investigated factor levels can be used to determinethe optimal conditions required to achieve maximum reme-dial efficiency as depicted in the 3D response surface plot(Figure 14(b))

Model Validation To validate the practical applicability of thedeveloped models affecting the remedial efficiency (13) andsoil pH (9) additional experimental test was run at voltagegradient of 1 Vcm initial contaminant concentration of4415mgkg and without polarity reversal (Table 8) Resultsof the model validation showed that the experimental resultslie within 90 confidence interval (CI) and prediction inter-val (PI) with associated prediction error of 235 and 3264for soil pH and remedial efficiency respectively Since thevalidation results fall within the prediction interval then theoutcome of the confirmation test was a success [39] Hencethe models can provide good approximations necessary tomove in the proper direction

Optimization of Trivalent Chromium Removal Numericaloptimization was employed to find the optimal factor levelsthat will specifically target maximum remedial efficiencyof trivalent Cr while optimizing all the other contaminantremedial efficiencies and responses (Figure 15) An overalldesirability value of 0715 was obtained and its variationbased on the influential factors (initial concentration and

voltage gradient) is depicted in Figure 16 Optimal conditionsrequired to achieve effective trivalent Cr removal at 60mgkgare presented in Table 9 Overall desirability of 0715 wasattained at the following optimal conditions voltage gradient= 036Vcm polarity reversal rate = 1763 hr soil pH = 100Under these conditions the expected trivalent Cr remedialefficiency is 6475

34 Impacts of the Integrated Electrokinetic Remediationon Soil Physicochemical Properties Preceding sections haveelaborately discussed and modeled the impacts of the pro-posed remediation technique on the soil pH and electricalconductivity Additionally the passage of electric current andsoil pH gradients will result in the following physicochemicalinteractions (1) possible dissolution of the clay mineralsbeyond a pH range of 7ndash9 (2) dissolution of available soil saltssuch as carbonates (3) production of cementitious productsresulting from the precipitation of metal ions at pH valuescorresponding to their hydroxide solubility values and (4)soil structural changes which affect its engineering character-istics [41ndash44] Surface area pore volume and size (Table 10)mineralogical compositions (Table 11) and elemental con-stituents (Table 12) were analyzed before and after the test forR5 At the end of the test (pH = 112) the soil specific surfacearea has increased (907 to 1121m2g) with correspondingincrease in the pore volume and size These results haveconfirmed that some dissolution of the soil minerals hastaken place during the electrokinetic remediation process dueto variations in the pore fluid chemistry Soil pores are due tothe presence of interlayer spaces that becomes prominent in2 1 clay mineral types such as montmorillonite and smectite[40 62 74] Table 11 presents the mineral transformationwhere dolomite completely disappeared calcite and quartzwere altered and degraded respectively after the test Theconstituent soil elements were not spared as the amountof each one either increased or decreased after the test asshown in Table 12 These observations may be explainedby microbially-driven biotransformation processes involvingdissolution and precipitation which take place under bothaerobic anaerobic conditions This leads to mineral disso-lution and formation of new minerals from aqueous ions(biomineralization) as noticed in Table 11 [40] Yong et al[40] have asserted that the scientific basis for biomineraliza-tion is still not well understood

4 Conclusions

The study reported herein investigated the migration oftrivalentCr ions fromamultiple contaminated natural saline-sodic soil The soil salinity and sodicity which providedlarge amount of dissolved salts and minerals (carbonates)in the pore fluid for sustained high electrical conductionwere responsible for the extremely high electric currentflow This led to excessive soil heating high energy andprocess fluid consumption high electroosmotic volume andin some cases higher percentage removal of trivalent CrSignificant migration of Cr from the contaminated chamberto the granular activated carbon chamber was recorded


which led to highest remedial efficiencies (7997ndash3488) fortests involving 60mgkg initial trivalent Cr concentrationwhereas no removal was recorded for all tests involving20mgkg Even under low electric current electroosmoticflow and voltage gradient (02 Vcm) up to 3693 of thetrivalent Cr was removed from the contaminated chamberIt has been shown that high voltage gradient (1 Vcm) orpassage of high electric current does not necessarily translateinto high remedial efficiency Bipolar effects did not manifestdue to the presence of carbonate minerals that impact highacid buffering capacity For test without polarity reversaltrivalent Cr moved toward the anode due to the formationof high amount of anionic Cr(OH)

4

minus hydroxocomplex athigh pH which was further attracted to the anode via elec-tromigration Nonadsorption of this ion onto the negativelycharged clay soil due to the possession of similar chargeincreased its availability and mobility Speciation modelingusing Visual MINTEQ 30 reveals the increasing dominanceof the anionic Cr(OH)

4

minus and the decreasing concentrationof aqueous Cr(OH)

3at pH 112 Effects of voltage gradient

initial contaminant concentration and polarity reversal rateon the effective removal of Cr ions were experimentallystudied using the Box-Behnken Design of experiment andmathematically modeled and numerically optimized usingresponse surface methodology Results of the model vali-dation showed that the experimental results lie within 90confidence interval and prediction interval with associatedprediction error of 235 and 3264 for soil pH and trivalentCr remedial efficiency respectively Overall desirability of0715was attained at the following optimal conditions voltagegradient = 036Vcm polarity reversal rate = 1763 hr andsoil pH = 100 Under these conditions the expected trivalentCr remedial efficiency is 6475 Passage of electric currentand variations in the pore fluid chemistry led to soil mineraldissolution and alteration via biotransformation

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

The authors would like to acknowledge the support pro-vided by King Abdul-Aziz City for Science and Technology(KACST) through the Science amp Technology Unit at KingFahd University of Petroleum amp Minerals (KFUPM) forfunding this work through Project no 11-Env1669-04 as partof the National Science Technology and Innovation Plan

References

[1] S V Ho B M Hughes P H Brodsky J S Merz and L P EgleyldquoAdvancing the use of an innovative cleanup technology casestudy of Lasagnardquo Remediation Journal vol 9 pp 103ndash116 1999

[2] PH Brodsky and SVHo ldquoIn situ remediation of contaminatedsoilsrdquo US Patent 5398756 1995

[3] S V Ho and P H Brodsky ldquoIn-situ remediation of contami-nated heterogeneous soilsrdquo US Patent 5476992 1995

[4] F F Reuss ldquoSur un nouvel effet de lrsquoelectricite galvaniquerdquoMemoires de la Societe Imperiale des Naturalistes deMoscou vol2 pp 327ndash337 1809

[5] H A Abramson Electrokinetic Phenomena and Their Applica-tion to Biology andMedicine ACSMonograph Series ChemicalCatalog New York NY USA 1934

[6] I Ravina and D Zaslavsky ldquoNon-linear electrokineticphenomenamdashI review of literaturerdquo Soil Science vol 106 pp60ndash66 1968

[7] A T Yeung ldquoGeochemical processes affecting electrochemicalremediationrdquo in Electrochemical Remediation Technologies forPolluted Soils Sediments and Groundwater pp 65ndash94 JohnWiley amp Sons 2009

[8] J Virkutyte M Sillanpaa and P Latostenmaa ldquoElectrokineticsoil remediationmdashcritical overviewrdquo Science of the Total Envi-ronment vol 289 no 1ndash3 pp 97ndash121 2002

[9] S-S Kim S-J Han and Y-S Cho ldquoElectrokinetic remediationstrategy considering ground strate a reviewrdquo Geosciences Jour-nal vol 6 pp 57ndash75 2002

[10] V P Evangelou Environmental Soil and Water ChemistryPrinciples and Applications Wiley-Interscience New York NYUSA 1998

[11] I P Abrol J S P Yadav and F I Massoud ldquoSalt-affected soilsand their management food and agriculture Organization ofthe United Nationsrdquo FAO Soils Bulletin vol 39 1988

[12] C D Palmer and P R Wittbrodt ldquoProcesses affecting theremediation of chromium-contaminated sitesrdquo EnvironmentalHealth Perspectives vol 92 pp 25ndash40 1991

[13] A N Puri and B Anand ldquoReclamation of alkali soils byelectrodialysisrdquo Soil Science vol 42 no 1 pp 23ndash27 1936

[14] S V Ho P W Sheridan C J Athmer et al ldquoIntegratedin situ soil remediation technology the Lasagna processrdquoEnvironmental Science and Technology vol 29 no 10 pp 2528ndash2534 1995

[15] C J Athmer and S V Ho ldquoField studies organic-contaminatedsoil remediation with lasagna technologyrdquo in ElectrochemicalRemediation Technologies for Polluted Soils Sediments andGroundwater K R Reddy and C Cameselle Eds pp 625ndash646John Wiley amp Sons 2009

[16] K Maturi and K R Reddy ldquoSimultaneous removal of organiccompounds and heavy metals from soils by electrokineticremediation with a modified cyclodextrinrdquo Chemosphere vol63 no 6 pp 1022ndash1031 2006

[17] K R Reddy P R Ala S Sharma and S N Kumar ldquoEnhancedelectrokinetic remediation of contaminated manufactured gasplant soilrdquo Engineering Geology vol 85 no 1-2 pp 132ndash1462006

[18] T Li S Yuan JWan et al ldquoPilot-scale electrokineticmovementof HCB and Zn in real contaminated sediments enhanced withhydroxypropyl-120573-cyclodextrinrdquoChemosphere vol 76 no 9 pp1226ndash1232 2009

[19] M Elektorowicz ldquoElectrokinetic remediation of mixed metalsand organic contaminantsrdquo in Electrochemical RemediationTechnologies for Polluted Soils Sediments and Groundwater pp315ndash331 John Wiley amp Sons 2009

[20] S V Ho C J Athmer P W Sheridan and A P Shapiro ldquoScale-up aspects of the Lasagna process for in situ soil decontamina-tionrdquo Journal of HazardousMaterials vol 55 no 1ndash3 pp 39ndash601997

[21] J W Ma H Wang and Q Luo ldquoMovement-adsorption andits mechanism of Cd in soil under combining effects of


electrokinetics and a new type of bamboo charcoalrdquo ChineseJournal of Environmental Science vol 28 no 8 pp 1829ndash18342007 (Chinese)

[22] JWMaHWang and R R Li ldquoRemoval of cadmium in kaolinby electrokinetics-bamboo charcoal adsorptionrdquo Environmen-tal Chemistry vol 26 pp 634ndash637 2007 (Chinese)

[23] J WMa F YWang Z H Huang and HWang ldquoSimultaneousremoval of 24-dichlorophenol and Cd from soils by electroki-netic remediation combined with activated bamboo charcoalrdquoJournal of Hazardous Materials vol 176 no 1ndash3 pp 715ndash7202010

[24] S LukmanMH Essa N DMursquoazu and A Bukhari ldquoCoupledelectrokinetics-adsorption technique for simultaneous removalof heavy metals and organics from saline-sodic soilrdquo TheScientific World Journal vol 2013 Article ID 346910 9 pages2013

[25] S V Ho C Athmer P W Sheridan et al ldquoThe lasagnatechnology for in situ soil remediation 1 Small field testrdquoEnvironmental Science and Technology vol 33 no 7 pp 1086ndash1091 1999

[26] S V Ho C Athmer P W Sheridan et al ldquoThe lasagnatechnology for in situ soil remediation 2 Large field testrdquoEnvironmental Science amp Technology vol 33 no 7 pp 1092ndash1099 1999

[27] B D Swift and J J Tarantino ldquoApplication of the Lasagna soilremediation technology at the DOE paducah gaseous diffusionplantrdquo in Proceedings of the Waste Management Symposium(WM rsquo03) Tucson Ariz USA February 2003

[28] S Chinthamreddy and K R Reddy ldquoOxidation and mobilityof trivalent chromium in manganese-enriched clays duringelectrokinetic remediationrdquo Soil and Sediment Contaminationvol 8 no 2 pp 197ndash216 1999

[29] S Chinthamreddy and K R Reddy ldquoGeochemistry ofchromium during electrokinetic remediationrdquo in Proceedingsof the 4th International Symposium on EnvironmentalGeotechnology and Global Sustainable Development BostonMass USA 1998

[30] K K Reddy and S Chinthamreddy ldquoEffects of initial form ofchromium on electrokinetic remediation in claysrdquo Advances inEnvironmental Research vol 7 no 2 pp 353ndash365 2003

[31] K R Reddy and U S Parupudi ldquoRemoval of chromium nickeland cadmium from clays by in-situ electrokinetic remediationrdquoSoil and Sediment Contamination vol 6 no 4 pp 391ndash407 1997

[32] K R Reddy U S Parupudi S N Devulapalli and C Y XuldquoEffects of soil composition on the removal of chromium byelectrokineticsrdquo Journal of HazardousMaterials vol 55 no 1ndash3pp 135ndash158 1997

[33] K R Reddy S Chinthamreddy R E Saichek and T J CutrightldquoNutrient amendment for the bioremediation of a chromium-contaminated soil by electrokineticsrdquo Energy Sources vol 25no 9 pp 931ndash943 2003

[34] L Hopkinson A Cundy D Faulkner A Hansen andR Pollock ldquoElectrokinetic stabilization of chromium (VI)-contaminated soils electrochemical remediation technologiesfor polluted soilsrdquo in Sediments and Groundwater pp 179ndash1932009

[35] D B Gent R M Bricka A N Alshawabkeh S L Larson GFabian and S Granade ldquoBench- and field-scale evaluation ofchromium and cadmium extraction by electrokineticsrdquo Journalof Hazardous Materials vol 110 no 1ndash3 pp 53ndash62 2004

[36] P R Buchireddy R M Bricka and D B Gent ldquoElectrokineticremediation of wood preservative contaminated soil containing

copper chromium and arsenicrdquo Journal of Hazardous Materi-als vol 162 no 1 pp 490ndash497 2009

[37] R H Myers D C Montgomery and C M Anderson-CookResponse Surface Methodology Process and Product Optimiza-tion Using Designed Experiments John Wiley amp Sons 2009

[38] M A Bezerra R E Santelli E P Oliveira L S Villar and L AEscaleira ldquoResponse surface methodology (RSM) as a tool foroptimization in analytical chemistryrdquo Talanta vol 76 no 5 pp965ndash977 2008

[39] M J Anderson and P JWhitcombRSMSimplified OpitimizingProcesses Using Response Surface Methods for Design of Experi-ments Productivity Press 2005

[40] R N Yong M Nakano and R Pusch Environmental SoilProperties and Behaviour CRCPress NewYork NY USA 2012

[41] J Hamed Y B Acar and R J Gale ldquoPb(II) removal fromkaolinite by electrokineticsrdquo Journal of geotechnical engineeringvol 117 no 2 pp 241ndash271 1991

[42] M Pazos A Plaza M Martın and M C Lobo ldquoThe impactof electrokinetic treatment on a loamy-sand soil propertiesrdquoChemical Engineering Journal vol 183 pp 231ndash237 2012

[43] S H Kim H Y Han Y J Lee CW Kim and JW Yang ldquoEffectof electrokinetic remediation on indigenous microbial activityand community within diesel contaminated soilrdquo Science of theTotal Environment vol 408 no 16 pp 3162ndash3168 2010

[44] Q-Y Wang D-M Zhou L Cang and T-R Sun ldquoApplicationof bioassays to evaluate a copper contaminated soil before andafter a pilot-scale electrokinetic remediationrdquo EnvironmentalPollution vol 157 no 2 pp 410ndash416 2009

[45] S Lukman M H Essa N D Mursquoazu A Bukhari and CBasheer ldquoAdsorption and desorption of heavy metals ontonatural clay material influence of initial pHrdquo Journal of Envi-ronmental Science and Technology vol 6 no 1 pp 1ndash15 2013

[46] M H Essa and M A Al-Zahrani ldquoDate pits as potential rawmaterials for the production of activated carbons in Saudi Ara-biardquo International Journal of Applied Environmental Sciencesvol 4 no 1 pp 47ndash58 2009

[47] M H Essa M A Al-Zahrani and T N Suresh ldquoOptimizationof activated carbon production from date pitsrdquo InternationalJournal of Environmental Engineering vol 5 pp 325ndash338 2013

[48] EPAMethod 3050BmdashAcid Digestion of Sediments Sludges andSoils United States Environmental Protection Agency 1996

[49] USEPAMethod 7000BmdashFlame Atomic Absorption Spectropho-tometry United States Environmental Protection Agency 2007

[50] USEPA Method 7473mdashMercury in Solids and Solutions byThermal Decomposition Amalgamation and Atomic AbsorptionSpectrophotometry United States Environmental ProtectionAgency 2007

[51] J P Gustafsso ldquoVisual MINTEQ ver 30rdquo 2010 httpwww2lwrkthseEnglishOurSoftwarevminteqindexhtm

[52] USEPA Method 3545 Pressurized Fluid Extraction (PFE)United States Environmental Protection Agency 1996

[53] USEPA Method 8270DmdashSemivolatile Organic CompoundsBy Gas ChromatographyMass Spectrometry (GCMS) UnitedStates Environmental Protection Agency 2007

[54] Start-EaseDesign-Expert Version 8 Software forWindow Start-Ease Minneapolis Minn USA 2011

[55] D Derringer and R Suich ldquoSimultaneous optimization ofseveral response variablesrdquo Journal of Quality Technology vol12 pp 214ndash219 1980


[56] A Z Al-Hamdan andK R Reddy ldquoTransient behavior of heavymetals in soils during electrokinetic remediationrdquo Chemo-sphere vol 71 no 5 pp 860ndash871 2008

[57] A T Yeung and Y Gu ldquoA review on techniques to enhanceelectrochemical remediation of contaminated soilsrdquo Journal ofHazardous Materials vol 195 pp 11ndash29 2011

[58] K R Reddy and S Chinthamreddy ldquoEnhanced electrokineticremediation of heavy metals in glacial till soils using differentelectrolyte solutionsrdquo Journal of Environmental Engineering vol130 no 4 pp 442ndash455 2004

[59] A Alok R P Tiwari and R P Singh ldquoEffect of pH of anolytein electrokinetic remediation of cadmium contaminated soilrdquoInternational Journal of Engineering Research amp Technology vol1 pp 1ndash11 2012

[60] L M Vane and GM Zang ldquoEffect of aqueous phase propertieson clay particle zeta potential and electro-osmotic permeabilityimplications for electro-kinetic soil remediation processesrdquoJournal of Hazardous Materials vol 55 no 1ndash3 pp 1ndash22 1997

[61] K Beddiar T Fen-ChongADupas Y Berthaud andPDanglaldquoRole of pH in electro-osmosis experimental study on NaCl-water saturated kaoliniterdquo Transport in Porous Media vol 61no 1 pp 93ndash107 2005

[62] D L Sparks Environmental Soil Chemistry Academic PressNew York NY USA 2nd edition 2003

[63] A T Yeung ldquoMilestone developments myths and future direc-tions of electrokinetic remediationrdquo Separation and PurificationTechnology vol 79 no 2 pp 124ndash132 2011

[64] F E Asimakopoulou I F Gonos and I A StathopulosldquoMethodologies for determination of soil ionization gradientrdquoJournal of Electrostatics vol 70 no 5 pp 457ndash461 2012

[65] S Pamukcu ldquoElectrochemical transport and transformationsrdquoin Electrochemical Remediation Technologies for Polluted SoilsSediments and Groundwater pp 29ndash65 John Wiley amp SonsNew York NY USA 2009

[66] K R Reddy R E Saichek K Maturi and P Ala ldquoEffects ofsoil moisture and heavy metal concentrations on electrokineticremediationrdquo Indian Geotechnical Journal vol 32 pp 258ndash2882002

[67] R J Hunter and M James ldquoCharge reversal of kaolinite byhydrolyzable metal ions an electroacoustic studyrdquo Clays amp ClayMinerals vol 40 no 6 pp 644ndash649 1992

[68] K Maturi and K R Reddy ldquoCosolvent-enhanced desorptionand transport of heavy metals and organic contaminants insoils during electrokinetic remediationrdquo Water Air and SoilPollution vol 189 no 1ndash4 pp 199ndash211 2008

[69] JMDzenitis ldquoSteady state and limiting current in electroreme-diation of soilrdquo Journal of the Electrochemical Society vol 144no 4 pp 1317ndash1322 1997

[70] K R Reddy andM R Karri ldquoEffect of voltage gradient on inte-grated electrochemical remediation of contaminant mixturesrdquoLand Contamination and Reclamation vol 14 no 3 pp 685ndash698 2006

[71] M Pourbaix Atlas of Electrochemical Equilibria in AqueousSolutions 1974

[72] K R Reddy S Danda and R E Saichek ldquoComplicatingfactors of using ethylenediamine tetraacetic acid to enhanceelectrokinetic remediation of multiple heavy metals in clayeysoilsrdquo Journal of Environmental Engineering vol 130 no 11 pp1357ndash1366 2004

[73] Y B Acar and A N Alshawabkeh ldquoPrinciples of electrokineticremediationrdquo Environmental Science amp Technology vol 27 no13 pp 2638ndash2647 1993

[74] J K Mitchell Fundamentals of Soil Behavior John Wiley ampSons 1993

Submit your manuscripts athttpwwwhindawicom

Forestry ResearchInternational Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Environmental and Public Health

Journal of



EcosystemsJournal of


MeteorologyAdvances in

EcologyInternational Journal of


Marine BiologyJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Advances in


Environmental Chemistry

Atmospheric SciencesInternational Journal of



Waste ManagementJournal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics


Geological ResearchJournal of

EarthquakesJournal of


BiodiversityInternational Journal of


ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

OceanographyInternational Journal of


The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


ClimatologyJournal of


12

1

08

06

04

02

0

minus02

minus04

minus06

0 2 4 6 8 10 12 14

Eh (V

)

Cr3+

CrOH2+ Cr(OH)2+

CrO42minus

Cr(OH)30

Cr(O

H) 4

minus

HCrO4minus

PH2 = 1bar

Po2 = 1bar

pH

Figure 1 Redox potential (Eh)-pH diagram for CrndashOndashH system[12]

and quantity of contaminantmovement are influenced by thecontaminant concentration solubility speciation degree ofhydrophobicity soil type and structure and the mobility ofcontaminant ions as well as the interfacial chemistry and theconductivity of the soil pore fluid [8]The remedial efficiencygenerally depends on the nature of the contaminants andsoil properties such as pH permeability adsorption capac-ity buffering capacity and geochemical processes (such asacidbase reactions and migration dissolutionprecipitationredox reactions complexation and speciation) [7 9] Saline-sodic soils possess electrical conductivity above 4 dSmsoil paste pH greater than 82 and exchangeable sodiumpercentage greater than 15 [10 11]

First application of electrokinetics took place in India inthe 1930s It was used to remove excess salts from alkali soilsin order to restore it to arable condition [13] Following itsinvention in 1993 extensive studies started in 1994 in bench-scale [14] then scaled up in a pilot-scale under laboratoryconditionsThefirst field test called ldquoPhase I Small Field Testrdquowas conducted in 1995 at the Paducah gaseous diffusion plant(PGDP) site whose soil was contaminatedwith trichloroethy-lene (TCE) Full-scale remediation using the Lasagna tech-nology was undertaken at two other contaminated sites inthe United States [15] The specific details of all the Lasagnaprocess implementations are presented in Tables 1 and 2 Itis noteworthy that only two of the studies have consideredsimultaneous removal of contaminant mixture All othershave dealt with only a single organic compound or heavymetal It has already been observed that contaminated soilsdo not contain single contaminants only but usually severalpollutants appear in the soil in mixed components [16ndash19]

The geochemical properties of the most stable forms ofCr that is trivalent and hexavalent Cr under electrokineticremediation have been extensively studied in different types

of soils (kaolin glacial till etc) by Reddy and his coworkersand other investigators [28ndash36] The trivalent Cr thoughconsidered relatively nontoxic compared to the hexavalentCr exists in the subsurface environments as cation Cr3+ andin the following hydroxocomplex forms Cr(OH)

4

minus CrOH2+and Cr(OH)

3

0 Cr3+ and CrOH2+ ions are mostly prevalentat soil pH values less than 6 while Cr(OH)

4

minus and Cr(OH)3

0

ions prevail when pH is greater than 118 The redox statealso affects the Cr state with reduced state favoring thepresence of the trivalent Cr while the oxidized state favorsthe existence of the hexavalent Cr Most of the trivalent Crspecies are less mobile because of their low solubility overwide pH range (lt12) and may be readily adsorbed by thenegatively charged clay surfaces There exists redox state inthe subsurface environment because of the generation ofoxygen and hydrogen gases at the electrodes in addition tothe possible presence of iron (reducing agent) manganese(oxidizing agent) or microorganisms The redox potential(Eh) and soil pH determine the possible oxidation of Crfrom the trivalent to hexavalent form as shown in Figure 1Chinthamreddy and Reddy [28] have found no significantoxidation of trivalent Cr in high buffering capacity soil suchas glacial till

Empirical modeling using response surface methodology(RSM) offers great and numerous advantages which includelarge amount of information from a small number of exper-iments evaluation of simultaneous interaction effects of theindependent parameters on the responses and simultaneousoptimization of multiple factors and responses for obtainingoptimal conditions [37 38]The key success of RSM is uncov-ering interactions of factors which cannot be achieved usingthe traditional one-factor-at-a-time (OFAT) optimizationapproach [39] Fundamental understanding of the physicsand chemistry which governs the process is essential in deter-mining the influential factors to be investigated and theirlevels or ranges are necessary for successful implementationof RSM for any process modeling and optimization Basicallythere exist four different experimental designs for RSMimplementation 3-level factorial design (3FD) Box-Behnkendesign (BBD) central composite design (CCD) andDoehlertdesign (DD) Bezerra et al [38] have reviewed each of thesedesign methods The Box-Behnken design is obtained bycombining two-level factorial designs with incomplete blockdesigns followed by adding a specified number of replicatedcenter points BBD is preferred when investigating three (3)factors using RSM because it will give enough informationfor analyzing factor-response interactions from the leastexperimental runs when compared to 3FD and CCD

Some microbially-driven biotransformation processesmay affect the soil physicochemical properties after elec-trokinetic remediation because of the passage of electriccurrent and development of pH gradients [40] These leadto original soil mineral degradation and alteration viabiotransformation While biotransformation deals with thebioweathering and alteration or degradation of clay mineralsbiomineralization refers to the formation of amorphousand crystalline materials from aqueous ions by biologi-cally mediated processes In addition to current and pH


Table1Ap

plications

ofLasagn


from

inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Cell

dimensio

ns(le

ngthtimes

widthtimesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)current

(mA)

Power

consum

ption

kWhrm

3

Runtim

edays

Electro

osmotic

cond

uctiv

ity

cm2Vminus1sminus1

(times10minus5)

Treatm

ent

zone

spacing

cm

Reference

AClowast+sand

bacteria+AC

+sawdu

stp-nitro

phenol

Kaolinite

10cm

ID

216cm

long

Yes

continuo

us90ndash9

91ndash73(con

stant)

1020

25

6[14

20]

AC(Bam

boo

charcoal)

Cd

Sand

yloam

24cmtimes10cm

times8c

mYes

continuo

us796

17ndash27

mdash12

mdash10

[21]

AC(Bam

boo

charcoal)

Cd

Kaolin

24cmtimes10cm

times8c

mNo

continuo

us93

13ndash23

mdash8

mdash10

[22]

AC(Bam

boo

charcoal)

24-

dichloroph

enol

andCd

Sand

yloam

24cmtimes10cm

times10cm

Yes

continuo

us

7597(C

d)

5492(24-

dichloroph

enol)

1Variable

1219

1ndash12848

105

mdash16

[23]

GAClowastlowast

CrC

dCu

Pb

HgZn

ph

enol

kerosene

Salin

e-sodic

clay

24cmtimes10cm

times12cm

No

continuo

us

759(C

r)344

(Cd)41(Cu

)558(Pb)9249

(Hg)268(Zn)

100(pheno

l)498(kerosene)

06ndash

1880

1777ndash4

273

21425

6[24]

lowastAc

tivated

carbon

lowastlowastGranu

lara

ctivated

carbon


Table2Ap

plications

ofLasagn


inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Sitedimensio

ns(le

ngthtimeswidth

timesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)Cu

rrent

(A)

Power

consum

ption

kwhm

3

Runtim

emon

th

Electro

osmotic

cond

uctiv

ity


Treatm

ent

zone

spacingcm

Reference

AC+sand

p-nitro

phenol

Kaolin

Kaolinite

clayloam

122mtimes061m

times061m

Yes

continuo

us98

1(constant)962

(based

oncurrentd

ensity)

513

056ndash17

3556

[20]

GAC

1TC

E2Clay

loam

46mtimes3m

times46m

Yes

continuo

us99

035ndash0

4540

(con

stant)

mdash4

1260

[25]

Iron

filings

+kaolin

TCE

Clay

loam

64mtimes92

mtimes137m

Yes3-week

95ndash9

9023ndash0

31(con

stant)

110ndash200

mdash12

1260

amp150

[26]

Iron

filings

+kaolin

TCE

Clay

loam

274mtimes22

mtimes135m

Nopu

lsemod

e99

015ndash0

26

(con

stant)

500ndash

700

mdash24

mdash150

[1527]

Iron

filings

+kaolin

TCE

Clay

loam

33mtimes24

mtimes75

mmdash

60(afte

r1year)

016

(con

stant)

250ndash

400

mdash24

mdash150

[15]

1Granu

lara

ctivated

carbon

2Trichloroethylene


Gas ventGas vent

ValveValve

Filterpaper

Filterpaper

Cathodiccompartment

Cathodeelectrode

Anodecompartment

Anodeelectrode


Contaminatedsoil

B

C

F G

D





14

12

10

8

6

4

2

0R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

pH






R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

250

200

150

100

50

0

Elec

tric

al co

nduc

tivity

(dS

m)















3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)



2) (874) calcite


3)2) (74) X-ray






R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















Table1Ap

plications

ofLasagn


from

inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Cell

dimensio

ns(le

ngthtimes

widthtimesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)current

(mA)

Power

consum

ption

kWhrm

3

Runtim

edays

Electro

osmotic

cond

uctiv

ity

cm2Vminus1sminus1

(times10minus5)

Treatm

ent

zone

spacing

cm

Reference

AClowast+sand

bacteria+AC

+sawdu

stp-nitro

phenol

Kaolinite

10cm

ID

216cm

long

Yes

continuo

us90ndash9

91ndash73(con

stant)

1020

25

6[14

20]

AC(Bam

boo

charcoal)

Cd

Sand

yloam

24cmtimes10cm

times8c

mYes

continuo

us796

17ndash27

mdash12

mdash10

[21]

AC(Bam

boo

charcoal)

Cd

Kaolin

24cmtimes10cm

times8c

mNo

continuo

us93

13ndash23

mdash8

mdash10

[22]

AC(Bam

boo

charcoal)

24-

dichloroph

enol

andCd

Sand

yloam

24cmtimes10cm

times10cm

Yes

continuo

us

7597(C

d)

5492(24-

dichloroph

enol)

1Variable

1219

1ndash12848

105

mdash16

[23]

GAClowastlowast

CrC

dCu

Pb

HgZn

ph

enol

kerosene

Salin

e-sodic

clay

24cmtimes10cm

times12cm

No

continuo

us

759(C

r)344

(Cd)41(Cu

)558(Pb)9249

(Hg)268(Zn)

100(pheno

l)498(kerosene)

06ndash

1880

1777ndash4

273

21425

6[24]

lowastAc

tivated

carbon

lowastlowastGranu

lara

ctivated

carbon


Table2Ap

plications

ofLasagn


inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Sitedimensio

ns(le

ngthtimeswidth

timesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)Cu

rrent

(A)

Power

consum

ption

kwhm

3

Runtim

emon

th

Electro

osmotic

cond

uctiv

ity


Treatm

ent

zone

spacingcm

Reference

AC+sand

p-nitro

phenol

Kaolin

Kaolinite

clayloam

122mtimes061m

times061m

Yes

continuo

us98

1(constant)962

(based

oncurrentd

ensity)

513

056ndash17

3556

[20]

GAC

1TC

E2Clay

loam

46mtimes3m

times46m

Yes

continuo

us99

035ndash0

4540

(con

stant)

mdash4

1260

[25]

Iron

filings

+kaolin

TCE

Clay

loam

64mtimes92

mtimes137m

Yes3-week

95ndash9

9023ndash0

31(con

stant)

110ndash200

mdash12

1260

amp150

[26]

Iron

filings

+kaolin

TCE

Clay

loam

274mtimes22

mtimes135m

Nopu

lsemod

e99

015ndash0

26

(con

stant)

500ndash

700

mdash24

mdash150

[1527]

Iron

filings

+kaolin

TCE

Clay

loam

33mtimes24

mtimes75

mmdash

60(afte

r1year)

016

(con

stant)

250ndash

400

mdash24

mdash150

[15]

1Granu

lara

ctivated

carbon

2Trichloroethylene


Gas ventGas vent

ValveValve

Filterpaper

Filterpaper

Cathodiccompartment

Cathodeelectrode

Anodecompartment

Anodeelectrode


Contaminatedsoil

B

C

F G

D





14

12

10

8

6

4

2

0R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

pH






R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

250

200

150

100

50

0

Elec

tric

al co

nduc

tivity

(dS

m)















3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)



2) (874) calcite


3)2) (74) X-ray






R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















Table2Ap

plications

ofLasagn


inceptionto

date

Treatm

ent

zone

material

Con

taminant

Soiltype

Sitedimensio

ns(le

ngthtimeswidth

timesdepth)

Polarity

reversal

downtim

e

Removal

efficiency

Voltage

gradient

(Vcm)Cu

rrent

(A)

Power

consum

ption

kwhm

3

Runtim

emon

th

Electro

osmotic

cond

uctiv

ity


Treatm

ent

zone

spacingcm

Reference

AC+sand

p-nitro

phenol

Kaolin

Kaolinite

clayloam

122mtimes061m

times061m

Yes

continuo

us98

1(constant)962

(based

oncurrentd

ensity)

513

056ndash17

3556

[20]

GAC

1TC

E2Clay

loam

46mtimes3m

times46m

Yes

continuo

us99

035ndash0

4540

(con

stant)

mdash4

1260

[25]

Iron

filings

+kaolin

TCE

Clay

loam

64mtimes92

mtimes137m

Yes3-week

95ndash9

9023ndash0

31(con

stant)

110ndash200

mdash12

1260

amp150

[26]

Iron

filings

+kaolin

TCE

Clay

loam

274mtimes22

mtimes135m

Nopu

lsemod

e99

015ndash0

26

(con

stant)

500ndash

700

mdash24

mdash150

[1527]

Iron

filings

+kaolin

TCE

Clay

loam

33mtimes24

mtimes75

mmdash

60(afte

r1year)

016

(con

stant)

250ndash

400

mdash24

mdash150

[15]

1Granu

lara

ctivated

carbon

2Trichloroethylene


Gas ventGas vent

ValveValve

Filterpaper

Filterpaper

Cathodiccompartment

Cathodeelectrode

Anodecompartment

Anodeelectrode


Contaminatedsoil

B

C

F G

D





14

12

10

8

6

4

2

0R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

pH






R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

250

200

150

100

50

0

Elec

tric

al co

nduc

tivity

(dS

m)















3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)



2) (874) calcite


3)2) (74) X-ray






R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















Gas ventGas vent

ValveValve

Filterpaper

Filterpaper

Cathodiccompartment

Cathodeelectrode

Anodecompartment

Anodeelectrode


Contaminatedsoil

B

C

F G

D





14

12

10

8

6

4

2

0R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

pH






R1 R2 R3 R4 R5 R6 R7 R8 R9 R10R11R12R13

Initial1st week

2nd week3rd week

Experimental runs

250

200

150

100

50

0

Elec

tric

al co

nduc

tivity

(dS

m)















3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)



2) (874) calcite


3)2) (74) X-ray






R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics


























3000

2500

2000

1500

1000

500

0

Cum

ulat

ive e

lect

roos

mot

ic v

olum

e (m

L)



2) (874) calcite


3)2) (74) X-ray






R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics

















R5 7997 103 112 234450 24 06 60R6 7273 302 129 120150 0 1 60R9 6566 265 126 139950 48 1 60R8 3693 021 98 8100 0 02 60R13 3488 013 104 6300 48 02 60R11 2550 061 120 138784 0 06 100R1 000 117 102 172800 0 06 20R2 000 132 109 254250 48 06 20R3 000 225 126 81450 24 1 20R4 000 204 127 227250 24 1 100R7 000 014 81 39600 24 02 20R10 000 112 119 223650 48 06 100R12 000 015 83 32400 24 02 100
















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















minus1000 minus0500 0000 0500 1000

8

9

10

11

12

13

A A

B

B

C

C

Perturbation


Soil

pH3

rd w

eek

(ove

rall)


Actual factors



(a)

1291

814

02

06

10

20

60

100


89

10

11

1213

14

Soil

pH3

rd w

eek

(ove

rall)

Actual factor




(b)




minus1000 minus0500 0000 0500 1000

0

50

100

150

200

250

A

A

B

B

C

C

Perturbation


Elec

tric

al co

nduc

tivity

3rd

wee

k (d

Sm

)

(c)

2219

2564

Actual factor

02

06

10

20

60

100

0

50

100

150

200

250El

ectr

ical

cond

uctiv

ity

3rd

wee

k (d

Sm

)




(d)







14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















14

12

10

8

6

4

2

0

pH

1st

unsp

iked

cham

ber

B

1st

GAC

cham

ber

F

Spik

ed ch

ambe

r C

2nd

GAC

cham

ber

G

2nd

uns

pike

d ch

ambe

r D














minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















minus1000 minus0500 0000 0500 1000

A

A

B

B

CC

Perturbation


Elec

troos

mot

ic v

olum

e (m

L)

Actual factors

30

25

20

15

10

0

times102

5




(a)

02

06

10

0

24

48

257895

63

Actual factor

0

5

1015

20

25

30

Elec

troos

mot

icvo

lum

e (m

L)

times102




(b)


10

09

08

07

06

05

04

030 200 400 600 800 1000

Time step

Elec

tric

curr

ent (

A)

(a)

36

34

32

30

28

26

24

22

20

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Time step

(b)



VIF = 1(1 minus 1198772

119894)

(1)

Leverage =119901

119899 (2)





119910 = 120573119900+

119896

sum119894=1

120573119894119909119894+

119896

sum119894=1

1205731198941198941199092

119894+

119896

sum1le119894le119895

120573119894119895119909119894119909119895+ 120576 (3)




minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















minus1000 minus0500 0000 0500 1000

0

1

2

3

4

AA

B

B

CC

Perturbation


Aver

age c

urre

nt (A

)




(a)

02

06

10

20

60

100

0

05

1

15

2

Sqrt

(ave

rage

curr

ent) 173781

0360555

Actual factor




(b)


Run 1

Run 2

Run 3

Run 4

Run 9

Run 10

40

30

20

10

0


1st

unsp

iked

cham

bers

B

1st

GAC

cham

bers

F

Spik

edch

ambe

rs C

2nd

GAC

cham

ber

G

2nd

uns

pike

dch

ambe

r D

Run 8

Run 5

Run 6

Run 7

Run 11

Run 12

Run 13

C0C




0

minus5

minus10

minus15

minus20

minus25

minus30pH

=1004

-2

nd w

eek

pH=106

-1

st w

eek

pH=1122

-3

rd w

eek

pH=801

-in

itial

Cr(OH)2+1

Cr(OH)3(aq)Cr(OH)4

minus

Cr+3

Cr2(OH)2+4

Cr3(OH)4+5

CrOH+2

Log

(C)






100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics

















100

80

60

40

20

0

Rem

oval

()










3) to continuously


3and



3

minus(aq) (4)





2H2O (l) 997888rarr O

2(g) + 4H+ (aq) + 4eminus (7)




3



minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics















minus1000 minus0500 0000 0500 1000

0

20

40

60

80

100

A

A

B

B

CC

Perturbation


Cr3

rd w

eek

(ove

rall)

( re

mov

al)




(a)

937886

0

20

60

100

02

06

10

minus202468

1012

Sqrt

(Cr3

rd w

eek

(ove

rall)

)

Actual factor




(b)


10805346

09999910814957

05818620861739

0821322086525

09207430820122

07743170681556

0802811058509

0525934061289

03967030853617

0954150541784

05009640617458

0714843

Desirability

0000 0250 0500 0750 1000












100

000

20

60

10002

06

10

000

027

053

080

Des

irabi

lity

















V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics


























V119890=120576119904120577

120578119864 = 119896

119890119864 (10)









= 49 + 257 lowast 119860

+ 1168 lowast 119861 + 122 lowast 119862



(11)







4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics



















4




4


119900) to be







3 thus rendering it








2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics




















2

(13)






4 Conclusions




4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics
















4


4






Acknowledgments


References



















































































Journal of












Volume 2014

Advances in









Geophysics











































































Journal of












Volume 2014

Advances in









Geophysics














research article geochemical modeling of trivalent...

Documents