11

Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

  • Upload
    others

  • View
    13

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

Scientia Iranica F (2015) 22(3), 1346{1356

Sharif University of TechnologyScientia Iranica

Transactions F: Nanotechnologywww.scientiairanica.com

Experimental investigation of nano zero-valent ironmobility in porous media

M.R. Fadaei Tehrania;�, A. Shamsaia and M. Vossughib

a. Department of Civil Engineering, Sharif University of Technology, Tehran, P.O. Box 11365-11155, Iran.b. Institute of Biotechnology and Environment (IBE), Department of Chemical Engineering, Sharif University of Technology,

Tehran, Iran.

Received 5 January 2015; received in revised form 13 April 2015; accepted 16 June 2015

KEYWORDSRemediation;Mobility;Nano zero-valent iron;nZVI;Groundwater.

Abstract. In this study, the colloidal stability and mobility of Fe/Ni nano particles,concurrently synthesized and stabilized in the presence of starch (S-nZVI/Ni), wereinvestigated. In particular, the in uence of pore velocity (ranging from 7 to 85 m/d)and injected particle concentrations (0.3 and 3 g/l) was evaluated in a one-dimensionalcolumn. Experimental results exposed the �ne mobility of the S-nZVI/Ni particles inporous materials. According to the breakthrough curves and mass recovery, the S-nZVItravel distance was limited to the range of 0.2 to 0.4 m for low pore velocities (5 to 7 m/d),and in the order of 10 m at higher velocities (> 50 m/d). Moreover, increasing porevelocity enhanced the mobility of S-nZVI. Results also proposed that the mobility of S-nZVI suspension in sand media should be lower than in glass beads media. The cloggingphenomenon of the column and the pore pressure variations during the injection periodwere strongly a�ected by media type and injected particle concentration. Clogging, dueto the deposition of particles, was observed, in particular, for 3 g/l nZVI suspension, lowvelocities and sand media. Finally, the results indicated that starch stabilized iron nanoparticles have the potential to become an e�ective reactive material for in-situ groundwaterremediation.© 2015 Sharif University of Technology. All rights reserved.

1. Introduction

Contamination of groundwater with pollutants is athreat to the environment and human health, so, itis essential to remediate polluted aquifers. The reme-diation of groundwater resources has been a subjectof interest for several decades. However, conventionalremediation and treatment technologies, such as pumpand treat, have shown limited e�ectiveness in reducingcontamination, and are expensive [1]. Recently, in-

*. Corresponding author. Tel.: +98 21 66164201;Fax: +98 21 66036005E-mail addresses: [email protected] (M.R. FadaeiTehrani); [email protected] (A. Shamsai);[email protected] (M. Vossughi).

situ techniques for remediation, such as PermeableReactive Barriers (PRBs), have become promisingalternatives to ex-situ methods, owing to their loweroperating costs [2]. Nano zero-valent iron particles,nZVI, could be used as reducing agents in PRBs forremoving a wide range of pollutants. They promiseto be signi�cantly more e�ective than granular iron;the reaction rates are 20 to 30 times faster, and thesorption capacity is much higher compared with thegranular form [3].

The greater total surface area of nZVI particles,along with higher reactive site density, that are presenton the surface of the particles, and/or the morereactive nature of the sites, are considered causativefactors in the higher reactivity of nZVI [4]. Numerousstudies have utilized iron nano particles for remedia-

Page 2: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356 1347

tion of various contaminants, like chlorinated organiccompounds and metal ions [5], nitrate [6,7], carbontetrachloride, benzoquinone [8], metalloids, such asarsenic [9,10], and organic compounds [11]. However,there are some challenges in exploiting nZVI-basedremediation techniques, which have made it di�cultto engineer its application for optimal performanceor to assess its probable risk to human or ecologicalhealth.

Due to their small size, nano particles could beinjected into an aquifer as colloidal suspension, andare expected to penetrate easily in porous materials inorder to reach the plume of contaminants. Further-more, they should have adequate mobility to transportaway from the injection point. However, it has beenindicated that bare nZVI is hardly mobile and stablein laboratory and �eld-scale studies [12-14]. Crucialpoints are stability against aggregation, mobility insubsurface environments and longevity under subsur-face conditions [15]. Typically, nZVI mobility fromthe injection point must be in the order of 10 m ormore for �eld applications [16]. Modi�cation of surfaceproperties has been proposed as the main approach toenhance the colloidal stability of nZVI aqueous disper-sion, and, consequently, improved mobility in porousmedia. Natural and engineered polymers and anionicsurfactants, as well as other organic coatings, havebeen identi�ed as promising materials for improvingthe colloidal stability of nZVI [17].

So far, several researchers have synthesized var-ious stable nZVI suspensions, and characterized themobility properties of the particles in porous media [17-19]. Phenrat et al. prepared stable nZVI slurryusing 5 to 40 nm sized Fe�/Fe-oxide nanoparticles,along with several di�erent types of polyelectrolyte.According to their results, the order of e�ectivenessfor the prevention of rapid aggregation and dispersionstabilization was PSS70K (83%) > PAP10K (82%) >PAP2.5K (72%) > CMC700K (52%) [20].

In this study, the mobility of bimetallic Fe/Ninanoparticles in a one-dimensional column was ex-perimented. First, a stable suspension of nZVI wassynthesized and stabilized by starch on-site. Then,several series of column experiments were planned toevaluate the impact of pore water velocity and slurryconcentration on nZVI mobility. Two types of porousmaterial were used:

(a) Glass beads, as a typical transparent inertmedium;

(b) Packed sand, as a typical aquifer material.

Finally, the experimental data were modeled usinga classic one-dimensional Convection-Dispersion owEquation (CDE) in combination with the Colloid Fil-tration Theory (CFT).

2. Materials and methods

2.1. MaterialsChemicals used in this study, including FeCl3.6H2O,NaBH4 and HCl, were purchased from Merck (Ger-many). A Hach DR 5000 spectrophotometer wasutilized to measure total and ferrous iron concentra-tions, according to Hach methods of 8008 and 8146.The iron nanoparticles were prepared just before theexperiments in order to lessen their surface oxidation.nZVI particles were synthesized by gradually adding0.15 M NaBH4 solution, at a rate of 1 to 2 mL/min,to 0.1 M FeCl3.6H2O aqueous solution containing 10%starch. This was performed at ambient temperatureand the resulted suspension was vigorously stirred at400 rpm, according to the methodology of He andZhao [21]. During this reaction, sodium borohydridereduced ferric ions into black particles, according tothe following reaction [22]:

4Fe3+(aq) + 3BH4 + 9H2O! 4Feo

(s) # +3H2BO3

+ 12H+(aq) + 6H2(g) " : (1)

Experiments were performed in saturated porous mediausing a plexi-glass column with a length of 0.6 m andan inner diameter of 0.05 m (Figure 1). The columnwas packed wet and was fed a background electrolytesolution with a Heidolph peristaltic pump (multi-channel ow rates from 0.5 to 320 mL/min) for atleast 10 Pore Volumes (PV) for removing backgroundturbidity and providing a uniform collector surfacecharge.

Two types of porous media were used in columnexperiments, including glass beads and sand whoseproperties are listed in Table 1. Glass bead grains wereof a spherical shape and particle size ranging between0:4 � 10�3 and 0:6 � 10�3 m, and sand grains wereof a globular shape and particle size ranging between0:4�10�3 and 0:6�10�3 m. The source reservoir of S-nZVI was sonicated with ultrasonic equipment (Mediasafe, UK, LTD) to minimize both aggregation andsedimentation during slurry injection. A permeableglass di�user screen was used at the column inlet toprovide a uniform distribution in the cross section. Thepressure drop along the column during the experimentwas continuously measured at the inlet, middle, andoutlet of the column.

Prior to each test, the glass beads were soakedin a hydrogen peroxide solution (5%) for 10 hr toremove impurities, washed with de-ionized water and,�nally, baked at 105�C for 24 hr [23]. Also, the sandwas immersed overnight in concentrated hydrochloricacid, in order to remove the background iron oxides,rinsed with de-ionized water, and dried in an oven ata temperature of 105�C. In all experiments, the pH ofthe input water was adjusted to 7� 0:2.

Page 3: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

1348 M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356

Figure 1. Picture and schematic of transparent column model.

Table 1. Properties of porous media used for columnexperiment.

Symbol Porous media parameters Value

L Column length 60 cm

D Column inner diameter 5 cm

Collector diameter

d50Glass beads 0.50 mm

Sand 0.5 mm

Collector porosity

nGlass beads 0.35

Sand 0.31

Collector pore volume

PVGlass beads 20 mL

Sand 20 mL

Collector density

�bGlass beads 2.5 g/cm3

Sand 2.5 g/cm3

Collector dry bulk density

�bGlass beads 1.7 g/cm3

Sand 1.7 g/cm3

T Temperature 20� 2

2.2. MethodsColloid transportation in column experiments wasmodeled using a Classic Filtration Theory (CFT) [24]:

@C@t

= D@2C@x2 � v @C@x �KattC; (2)

where C is the nZVI concentration in the aqueous phase(M/L3), D is the hydrodynamic dispersion coe�cient(L2/T), v is the pore velocity (L/T), t is time (T),and Katt is the rate at which nZVI attaches or is

deposited on available collector sites (1/T), which couldbe described as [17]:

Katt =3(1� �w)

2d50��ovP ; (3)

where �w is the volumetric water content, d50 is themedian collector grain size (L), � is sticking or removale�ciency, de�ned as the ratio of particles that attachto the collector to those that strike the collector, and�o is the theoretical collision e�ciency, de�ned as theratio of particles striking the collector to those thatapproach the collector, which can be described by [25]:

�o = �D + �I + �G; (4)

where �D, �I and �G are di�usion, interception, andgravity terms, correspondingly.

In this study, the governing equations were solvedusing an analytical solution to Eq. (2), which could bedescribed by [26]:

C(x; t) =m0

4�ntpDx

exp�� ((x� xo)� vt)2

4Dxt� �t

�;

(5)

where m0 is the injected contaminant mass per verticalunit (M/L), t is the time since the start of injection(T), n is the porosity, v is the seepage velocity, and �is the decay coe�cient (1/T).

When the partitioning of the contaminant is de-scribed with a linear isotherm, retardation of the frontrelative to the ground water ow could be expressedusing a retardation factor, R:

R =vvR

= 1 +�bnkatt; (6)

where v is the velocity of the groundwater, vR is thevelocity of the C=Co = 0:5 point on the concentration

Page 4: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356 1349

pro�le of the retarded constituent, and �b is the bulkmass density of the porous medium. The determinationof attachment coe�cient, katt, is essential to theevaluation of compound mobility in water.

When the coordinate system is aligned with themean velocity vector, the longitudinal (DL) hydrody-namic dispersion coe�cient is related to the pore uidvelocity through the following equation:

DL = �LjV j+D0m; (7)

where D0m is the e�ective molecular di�usion coef-�cient of the solute in the porous medium, jV j isthe magnitude of the velocity vector, and �L is thelongitudinal component of the dispersivity tensor. �Lis considered the characteristic property of a region of aporous medium, and typically, is conveniently treatedas independent of the pore uid velocity and P�ecletnumber [27].

To evaluate whether prepared S-nZVI suspensioncould be used for in-situ injection and delivery, itstransport behavior was studied by conducting columnexperiments. The e�ects of some main parameters,including pore velocity and S-nZVI slurry concentra-tion, were investigated. The pore velocities appliedin the majority of column studies were much higherthan those typically found in �eld scale approaches [19].Results of laboratory tests at high water velocities mayoverestimate nZVI transmission in lower velocity, �eldscale applications, due to reduced nZVI attachment toporous media at higher velocities.

In this study, 25 loading conditions (trial) wereapplied, details of which are listed in Table 2. Priorto the main experiments by S-nZVI suspension, tocompare the transport behavior of an ideal solutewith iron nanosuspension through the used porousmedia, a series of tracer experiments using Rhodamine(C28H31N2O3Cl) were planned. Rhodamine is non-degrading, non-sobbing, rather inert, and non-reactive.In each trial, the stabilized nZVI slurry of one PVwas pumped into the column. In addition, columne�uents were sampled at certain time intervals andanalyzed for concentrations of total and ferrous iron.All experiments were conducted in duplicate, and theresults were averaged.

3. Results and discussion

3.1. Characteristics of S-nZVITo characterize the stabilized iron nano particles, X-Ray powder Di�raction (XRD), Scanning Electron Mi-croscopy (SEM), and Dynamic Light Scattering (DLS)were recorded, the results of which are shown in Fig-ure 2. The starched nZVI particles displayed much lessagglomeration than those prepared without a stabilizer,so that, while S-nZVI remained suspended in water forseveral hours, non-starched particles agglomerated andprecipitated within minutes. XRD results, obtainedby a D8 Advanced Bruker di�ractometer, indicatedthe presence of Feo (peaked at 2� = 42, 67, 82), andFe2O3 (2� = 35, 53). SEM analyses by a S4160 FE-SEM device denoted that the S-nZVI were presentas discrete particles as opposed to dendritic ocs fornon-starched nZVI. DLS was performed using a S-redBadge model ZEN1600. The mean particle size wasestimated to be 78 nm with a standard deviation of14 nm, which translated to a surface area of at least15 m2/g. This means that the synthesized particleswere appropriate, in terms of size, for remediationpurposes [12,13].

3.2. Column experiments3.2.1. Tracer transmissionThe breakthrough curves (BTC) of the tracer thoroughglass beads and packed sand for the �ve velocities areshown as colored points in Figure 3. By assuming thatthe attachment coe�cient, katt, is equal to zero and theretardation factor, R, is equal to 1, other parameterswere calculated using inverse �tting of the experimentaldata on the analytical solution o�ered in Eq. (5). Thetwo �tted parameters were dispersion coe�cient andpore velocity, and results of calibration are given inFigure 3(a) to (e). By assuming that longitudinaldispersivity coe�cient, �L, was independent of veloc-ity, it was concluded that �L was averagely equal to2:6� 10�3 m for glass beads media.

Further analysis of the breakthrough curvesshowed that, in the case of glass beads media, BTCshad a symmetrical shape and no tailing, thus, indicat-ing that the physical non-equilibrium processes, suchas rate-limited mass transfer into regions of immobile

Table 2. Details of column experiments.

Trial no Porous material Injected suspension Pore velocity m/d

1 to 5Transparent

media�

Tracer 9 21 27 44 74

6 to 10 0.3 g/L S-nZVI 9 21 27 44 74

11 to 15 3 g/L S-nZVI 9 21 27 44 74

16 to 20 Packedsand

Tracer 7 19 32 55 85

21 to 25 3 g/L S-nZVI 7 19 32 55 85

* Glass-beads porous media is called here transparent media.

Page 5: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

1350 M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356

Figure 2. Characteristic of synthesized S-nZVI: (a) XRD diagram; (b) SEM image; (c) DLS results; and (d) stabilitystatus.

water and preferential ow paths, were not presentduring transport through porous media.

In the case of sand media, BTCs had a rather non-symmetrical shape. Thus, R 6= 1 and �tted parameterswere katt, D, and v, and the results of calibrationare given in Figure 3(f) to (j). Similarly, dispersivitycoe�cients, �L, for packed sand media, are obtainedas 2:2� 10�3 m.

3.2.2. S-nZVI transmissionIt is known that the e�ective velocity of a particle in aporous medium is rather more than that observed forwater and solutes. However, under some conditions,the di�erence could be ignored and the e�ective veloc-ity of a particle could be approximated with the valuesobtained from tracer tests [18]. For the particle size and ow rates used in this work, the average pore diameterwas 5�10�4 m and the ratio of colloid to water ow ratewas less than 0.5%. These values were small enoughto support the use of parameters obtained from tracertests, such as dispersivity coe�cients, to assess S-nZVImobility in porous media. The results of S-nZVI/Niinjection in column tests could be found in the followingparts. But, as a general rule, experimental and

model curves were in full compliance, both for inverse-simulated tests and for the others. The calculated R-squared values, R2, had a range of 0.7 to 1, and theP -values had a signi�cance level of 75% under most cir-cumstances, except for a few cases of BTCs, when nZVIconcentration was equal to 3 g/l and pore velocity lessthan 15 m/s. A slightly more pronounced discrepancywas noticed for the higher nanoparticle concentrations,but, overall, the quantitative and qualitative agreementwas satisfactory.

a. Glass beads mediaThe observed BTCs of nanoparticles in a transparentcolumn, as a function of time, in di�erent experimentalconditions of pore water velocities (v = 9, 21, 27,44 and 74 m/d) and injected concentrations (Co =0:3 and 3 g/l), are presented in Figure 4 by coloredpoints. Parameters were calculated by using the inverse�tting of experimental data on the analytical solution.As indicated in Figure 4, both pore velocity andslurry concentration had a signi�cant impact on S-nZVI transmission in porous media. Using 0.3 g/lnZVI concentration, the retardation factor at 9 m/dpore velocity was about 1.3, but was nearly 1.1 when

Page 6: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356 1351

Figure 3. Tracer breakthrough cures in transparent column experiments.

Page 7: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

1352 M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356

Figure 4. S-nZVI breakthrough cures in transparent column experiments.

Page 8: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356 1353

Figure 5. S-nZVI breakthrough cures in sand column experiments.

the pore water velocity was over 70 m/d. In the samecases, using 3 g/l nZVI slurry, the retardation factor,at 9 m/d pore velocity, was about 1.6, but was nearly1.2 when the pore water velocity was over 70 m/d. Aretardation factor of 1.6 implies that the nZVI plumemoves 1.6 times slower than the pore water velocity.These phenomena could be attributed to an increase indrug forces and shear stress resulting from the increasein pore water velocities, where decline clogging and thedetachment phenomena occurred. Further analysis ofthe breakthrough curves showed that release during ushing increased, mainly, by increasing the injectedconcentration, Co, while ow rate had a minor impact.For low concentration, Co = 0:3 g/l, the decline ofthe BTCs at the beginning of ushing was abrupt,regardless of the ow rate. For high concentration,Co = 3 g/l, a slightly retarded decline of breakthroughconcentration was observed, suggesting that a fractionof the deposited particles were released. It seemsthat this behavior is associated with the value ofCo, rather than to the pore water velocity, as the ow rate a�ected the position of the BTC peak, butdid not have a relevant in uence on its height andshape.

b. Packed sand mediaUsing colored circles, Figure 5 illustrates the obtainedbreakthrough curves of S-nZVI in a sand column asa function of time. Experimental conditions wereadjusted as pore water velocities of 7, 19, 32, 55and 85 m/d, injected concentration of Co = 3 g/l,and assuming �L = 2:2 � 10�3 m. As shown,the breakthrough curves were unsymmetrical, whichimplied that attachment and detachment phenomenaoccurred. Using inverse �tting of the experimentaldata on the analytical solution, the retardation factorsfor the above mentioned pore velocities were indicatedas 3.15, 2.81, 2.15, 1.74 and 1.51, correspondingly.As expected, S-nZVI/Ni retardation decreased whenpore velocity increased. This is consistent with theliterature for colloid transport in general, and for iron-based nano-particles in particular [28]. A comparisonbetween Figures 4 and 5 suggests that lower S-nZVIconcentration in the e�uent compared to the feed couldbe exclusively attributed to the retention of nZVI in thesand medium.

In the present study, colloid �ltration and columnexperiment theory was used to estimate the e�ectiveand maximum travel distance of nZVI for application

Page 9: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

1354 M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356

in the �eld remediation projects. Therefore, it is nec-essary that ZVI slurry and porous media used duringcolumn tests have similar properties with the soil inthe �eld. Several researchers have previously describedthe mobility of iron nano-particles through sand withthe help of the colloid �ltration theory, as a processcontrolled by a �rst-order kinetics mechanism [18]. Anadditional term (�krC) was introduced in the CFTequation to account for the removal of nZVI, as shownin Eq. (5). The �rst-order rate kinetic constant, kr,could be calculated from the experimental column data,according to following equation [18]:

Kr = �(v=L) ln(Ce=Co); (8)

where kr is a �rst-order kinetics rate constant fornZVI retention, L is the length of the column, Cois the initial concentration of nZVI, and Ce is thesteady state concentration of nZVI in the e�uent.Eq. (8) could be rearranged, as shown in Eq. (9), tocalculate the e�ective travel distance, Re�, needed fornZVI concentration to be reduced to 10% of the initialconcentration:

Re� = �(v=Kr) ln(Ce�ective=Co): (9)

Re� was calculated for several values of pore velocity,ranging between 1 and 80 m/d and initial nZVI con-centration; the results appear in Figure 6. Regardingthe impact of pore water velocity on S-nZVI transport,Figure 6 shows that increasing the pore water velocitydecreases the tailing phenomenon and enhances mobil-ity. This is due to the fact that increasing pore watervelocity will increase drag momentum and enhance thedetachment of nZVIs [19]. It is noted that in low porevelocities, typical of groundwater ow velocities, thenZVI travel distance was limited to the range of 0.1 to0.5 m. At higher velocities, which could be the caseduring the injection of nZVI suspension in the �eldfor remediation purposes, travel distance may reach arange of 1 to 10 m. For instance, at pore velocities of25 and 50 m/d, the calculated travel distances reach1.5 and 5 m, respectively.

Figure 6. E�ective traveling distance of S-nZVI incolumn experiments.

He et al. [21] studied the penetrability of CMCstabilized nZVI in sand columns, using a rather dilutedsuspension. From their results, it could be calculatedthat, at a Darcy velocity of 1 m/d, nanoparticles wouldtravel a maximum distance of almost 1.1 m, which isapproximately 3 times longer compared to the traveldistance of S-nZVI/Ni produced in the current study,at the same Darcy velocity. However, mobility resultsare not directly comparable, because of the di�erencein modeling conditions and nZVI slurry concentrations.

4. Conclusions

In this study, starched nZVI, with an average particlesize of 74 nm, was synthesized and applied in columnexperiments. The in uence of pore velocity, porousmedia, grain size and type, and injection suspensionconcentration on S-nZVI transfer was investigated.The experimental approach was designed to be rep-resentative of conditions expected at �eld scale re-mediation projects. As expected, the calculationsshowed that at low pore water velocities, typical ofnatural groundwater ow rates (< 10 m/d), S-nZVIremained within a short distance from their initialposition (< 0:5 m). In addition, at higher velocities,which may prevail during the in-situ injection process,the nanoparticles were able to reach a longer distance;e.g. for a pore velocity of 50 m/d, the travel distancewas in the order of 5 m. It was also determined thatthe type of porous material and nZVI concentrationwere the main parameters in controlling the transportproperties of S-nZVI. Furthermore, one may concludethat bench scale test conditions should replicate, asclosely as possible, intended �eld scale conditions, withrespect to pore water velocity, grain type and size, S-nZVI slurry concentration, and injection duration.

List of abbreviations

nZVI nano Zero Valent Iron particlesS-nZVI Stabilized nano Zero Valent Iron

particlesPRB Permeable Reactive BarrierXRD X-Ray powder Di�ractionSEM Scanning Electron MicroscopyDLS Dynamic Light ScatteringBTC Breakthrough Curve

Acknowledgments

This work was supported by the Institute of Biotech-nology and the Environment (IBE) at Sharif Universityof Technology, Tehran, Iran.

Page 10: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356 1355

References

1. Savage N., D.M.S. \Nanomaterials and water puri�-cation: opportunities and challenges", J. NanoparticleRes., 7(4-5), pp. 331-342 (2005).

2. Li, L., Benson, C.H. and Lawson, E.M. \Model-ing porosity reductions caused by mineral fouling incontinuous-wall permeable reactive barriers", Journalof Contaminant Hydrology, 83, pp. 89-121 (2006).

3. Zhang, X., Lin, S., Lu, X.-Q. and Chen, Z.-l. \Re-moval of Pb(II) from water using synthesized kaolinsupported nanoscale zero-valent iron", Chemical Engi-neering Journal, 163, pp. 243-248 (2010).

4. Cundy, A.B., Hopkinson, L. and Whitby, R.L.D. \Useof iron-based technologies in contaminated land andgroundwater remediation: A review", Science of TheTotal Environment, 400, pp. 42-51 (2008).

5. Rangsivek, R. and Jekel, M.R. \Removal of dissolvedmetals by zero-valent iron (ZVI): Kinetics, equilibria,processes and implications for stormwater runo� treat-ment", Water Research, 39, pp. 4153-4163 (2005).

6. Fadaei, M.R., Vossoughi, M. and Shamsai, A. \In-situnitrate remediation using nano iron/nickel particles",Environment Protection Engineering, 40, pp. 75-86(2014).

7. Yang, G.C.C. and Lee, H.-L. \Chemical reduction ofnitrate by nanosized iron: kinetics and pathways",Water Research, 39, pp. 884-894 (2005).

8. Zhang, X., Deng, B., Guo, J., Wang, Y. and Lan, Y.\Ligand-assisted degradation of carbon tetrachlorideby microscale zero-valent iron", Journal of Environ-mental Management, 92, pp. 1328-1333 (2011).

9. Fadaei, M.R., Shamsai, A. and Vossughi, M. \In-situPb2+ remediation using nano iron particles", Journalof Environmental Health Science & Engineering, 13(1)(2015). DOI 10.1186/s40201-015-0157-3

10. Kim, K.-R., Lee, B.-T. and Kim, K.-W. \Arsenicstabilization in mine tailings using nano-sized mag-netite and zero-valent iron with the enhancement ofmobility by surface coating", Journal of GeochemicalExploration, 113, pp. 124-129 (2012).

11. Fagerlund, F., Illangasekare, T.H., Phenrat, T., Kim,H.J. and Lowry, G.V. \PCE dissolution and simul-taneous dechlorination by nanoscale zero-valent ironparticles in a DNAPL source zone", Journal of Con-taminant Hydrology, 131, pp. 9-28 (2012).

12. Noubactep, C. \A critical review on the mechanism ofcontaminant removal in Feo-H2O systems", Environ.Technol., 29, pp. 909-920 (2008).

13. O'Carroll, D., Sleep, B., Krol, M., Boparai, H. andKocur, C. \Nanoscale zero-valent iron and bimetallicparticles for contaminated site remediation", Adv.Water Resour., 51, pp. 104-122 (2013).

14. Zhang, W.X. \Nanoscale iron particles for environmen-tal remediation: an overview", J. Nanoparticle Res.,5(3-4), pp. 323-332 (2003).

15. Wang, D.J. \Transport behavior of humic acid-modi�ed nanohydroxyapatite in saturated packed col-umn: e�ects of Cu, ionic strength, and ionic compo-sition", J. Colloid Interface Sci., 360(2), pp. 398-407(2011).

16. Saleh, N., Kim, H.J., Phenrat, T., Krzystof, M.,Tilton, R. and Lowry, G. \Ionic strength and com-position a�ect the mobility of surface-modi�ed feo",Nanoparticles in Water-Saturated Sand Columns. En-viron. Sci. Technol., 42, pp. 3349-3355 (2008).

17. Kocur, C.M., O'Carroll, D.M. and Sleep, B.E. \Impactof nZVI stability on mobility in porous media", Journalof Contaminant Hydrology, 145, pp. 17-25 (2013).

18. Hosseini, S.M. and Tosco, T. \Transport and retentionof high concentrated nano-Fe/Cu particles throughhighly ow-rated packed sand column", Water Re-search, 47, pp. 326-338 (2013).

19. Fagerlund, H.J., Illangasekare, F., Tilton, T. andLowry, G.V. \Particle size distribution, concentration,and magnetic attraction a�ect transport of polymer-modi�ed Feo nanoparticles in sand columns", Envi-ronmental Science & Technology, 43, pp. 5079-5085(2009).

20. Phenrat, T., Saleh, N., Sirk, K., Kim, H.J., Tilton,R.D. and Lowry, G.V. \Stabilization of aqueousnanoscale zero-valent iron dispersions by anionic poly-electrolytes: adsorbed anionic polyelectrolyte layerproperties and their e�ect on aggregation and sedimen-tation", J. Nanopart Res., 10, pp. 795-814 (2008).

21. He, F. and Zhao, D. \Preparation and characterizationof a new class of starch-stabilized bimetallic nanopar-ticles for degradation of chlorinated hydrocarbons inwater", Environ. Sci. Technol., 39, pp. 3314-3320(2005).

22. Lee, C., Jee, Y.K., Won, I.L., Nelson, K.L., Yoon,J. and Sedlak, D.L. \Bactericidal e�ect of zero-valentiron nanoparticles on Escherichia coli", EnvironmentalScience and Technology, 42, pp. 4927-4933 (2008).

23. Kanel, S.R., Neppolian, B., Choi, H. and Yang, J.W.\Heterogeneous catalytic oxidation of phenanthrene byhydrogen peroxide in soil slurry: Kinetics, mechanism,and implication", Soil and Sediment Contamination,12, pp. 101-117 (2003).

24. Yao, K.M., Habibian, M.T. and O'Melia., C.R. \Waterand waste-water �ltration: Concepts and applica-tions", Environ. Sci. Technol., 5, pp. 1105-1112 (1971).

25. Tufenkji, N. and Elimelech, M. \Correlation equationfor predicting single-collector e�ciency in physiochem-ical �ltration in saturated porous media", Environ.Sci. Technol., 38, pp. 529-536 (2004).

26. Bedient, P.B., Rifai, H. S. and Newell, C.J., Ground-water Contaminant, Prentice Hall PTR, pp. 151-189(1999).

27. Gaganis, P., Kjeldsen, P. and Burganos, V.N. \Model-ing natural attenuation of multicomponent fuel mix-tures in the vadose zone: Use of �eld data andevaluation of biodegradation e�ects", Vadose Zone J.,3(4), pp. 1262-1275 (2004).

Page 11: Experimental investigation of nano zero-valent iron ...scientiairanica.sharif.edu/article_3726_87eab97bac3b12e5fcc8f1b1ff431b91.pdf · Experimental investigation of nano zero-valent

1356 M.R. Fadaei Tehrani et al./Scientia Iranica, Transactions F: Nanotechnology 22 (2015) 1346{1356

28. Tiziana, T., Julian, B., Rainer, M. and Rajan-drea, S. \Transport of ferrihydrite nanoparticles insaturated porous media, role of ionic strength and ow rate", Environmental Science & Technology,http://dx.doi.org/10.1021/es202643c (2012).

Biographies

Mohamad Reza Fadaei Tehrani received his BSdegree in Civil Engineering and MS degree in HydraulicStructures Engineering, in 2003 and 2005, respectively,from Sharif University of Technology, Tehran, Iran,where he is currently a PhD degree candidate in theDepartment of Civil Engineering. His research interestsinclude water and wastewater treatment, groundwaterremediation, image processing and nano-particles tech-nology.

Abolfazl Shamsai received his PhD degree from theDepartment of Civil Engineering at the University ofCalifornia, Davis, USA, in 1981, and then completedpost-doctoral studies at the University of California,Berkeley, USA, from 1988 to 1990. Dr. Shamsai is cur-

rently Professor in the Civil Engineering Department atSharif University of Technology (SUT), Tehran, Iran,where he has taught multiple courses in the �elds ofwater and dam engineering, and been advisor to manyMS and PhD students. In 1995, Dr. Shamsai waspresented with the national exemplary professor awardby SUT. Since 1971, he has 415 publications, including17 books, 160 conference papers, and 128 journalpapers, published in Iranian and foreign journals.

Manouchehr Vossoughi received a BS degree inChemical Engineering from Sharif University of Tech-nology (SUT), Tehran, Iran, in 1974, and MS andPhD degrees in Biotechnology from the Departmentof Chemical Engineering at the University of Toulouse,France, in 1979 and 1983. Dr. Vossoughi is currentlyfaculty member and Professor in Bio- and Nano-Technology in the Department of Chemical Engineeringand Petroleum at Sharif University of Technology,Tehran, Iran. His research interests include microbialenhanced oil recovery, new surfactants in petroleumengineering and application of nano technology in oilproduction enhancement.