Investigations on the capacity and mechanism of iron uptake by nano zero-valent iron particles
JISMY ANTONY1, V MEERA1,* and VINOD P RAPHAEL2
1Department of Civil Engineering, Government Engineering College, APJ Abdul Kalam Technological University, Thiruvananthapuram 695016, India
2Department of Chemistry, Government Engineering College, APJ Abdul Kalam Technological University, Thiruvananthapuram 695016, India
*Author for correspondence (vmeera@gectcr.ac.in) MS received 24 May 2020; accepted 21 July 2020
Abstract. Nano zero-valent iron (nZVI) particles have been identified as one of the potential candidate for the removal of various metal ions. This study addresses the effectiveness of nZVI on iron removal from an aqueous solution containing Fe2?/Fe3?in the ratio of 2.5:0.5, which is not studied so far. Liquid-phase reduction technique was utilized to synthesize the nZVI. The characterization of nZVI and its interaction with iron in aqueous solution was found using transmission electron microscopy (TEM), scanning electron microscopy–energy dispersive X-ray (SEM–EDX), X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), ultraviolet–visible spectroscopy (UV–Vis spectroscopy) and Fourier transform infrared spectroscopy (FTIR). It was found that the nZVI particles were largely spherical with size varying from 80–99 nm. Abundant oxygen functional groups detected in the FTIR spectrum serve as the accessible sites for iron adsorption. SEM and XRD studies provided a clear image of core-shell structure of nZVI. The study also investigated the role of pH, iron concentration, nZVI dosage and contact time on the uptake behaviour of contaminant iron. A removal efficiency of 63% was achieved in 3 h, with optimum nZVI loading of 5 g l-1, from a solution having initial iron concentration of 0.5 mg l-1at natural solution pH 6.85. This efficiency was increased to 70% at optimum pH 10. The batch study revealed that adsorption kinetics followed pseudo-second-order kinetics with R2 value of 0.99 and data followed Dubinin–Radushkevich isotherm. Outcomes of the study recommend that nZVI could turn into a promising adsorbent for iron removal from aqueous solutions.
Keywords. Nano zero-valent iron; adsorption; iron removal; XPS; liquid phase reduction; D–R isotherm.
1. Introduction
Water bodies get contaminated by iron either through dumping of household and industrial waste or through geogenic sources. The wellsprings of iron in surface water are mainly the contamination from steel and iron industries, metal erosion and mining [1]. The presence of 3 to 15 mg l-1 iron in surface and ground water has been reported in many parts of India [2]. High iron content weakens the organo-leptix characteristics of water thereby accelerates the growth of various microorganisms which are tolerant to chlorine [3]. The persistent utilization of water with high iron concentration may cause different health issues, for example, impairment of hematopoiesis, hemochromatosis, eye disorders and heart diseases [4].
Apart from these health issues, the high concentration of iron causes streaks and stains on plumbing fixtures and clothing.
The precipitation of iron may likewise cause clogging of pipeline and softeners and in turn create unwanted circum- stances in the water distribution systems [5].
The conventional methods such as ion-exchange, electrocoagulation, oxidation–precipitation–filtration pro- cess, separation through filter media and adsorption using calcium carbonate-based minerals show above 90% iron removal efficiency [1]. Even though ion-exchange is widely used technique for iron removal, clogs are formed due to oxidation and in turn make the process labour intensive [6].
Regular replacement of electrodes and high current densi- ties make the electrocoagulation process less effective for iron remediation [7]. The oxidant used in oxidation–filtra- tion–precipitation process is difficult to transport and store and it will corrode the components of the treatment system upon use [8]. Back washing and low temperature conditions are necessary in filter media separation and the process is ineffective for removing inorganic iron [9]. The presence of co-cations, natural organic matter and lower pH conditions reduce the efficiency of adsorption of iron using calcium carbonate-based minerals. These drawbacks of conventional methods led to some of the recent innovative techniques, which include use of various membranes and https://doi.org/10.1007/s12034-020-02274-5
nanotechnology [2,10,11]. As the pore size of membranes is greater than the contaminant ions, a pre-treatment proce- dure must be done to increase the overall size of ions. This limits the application of membrane technology in iron remediation. The nanotechnology is rather a new technique and very limited research has been carried out in this area to remove iron from water. However, a recent study by Alimohammadi et al [12] achieved 98.97% total iron removal at pH of 8.2 by the utilization of carbon nanotubes (CNTs). But these nano-materials are difficult to retrieve from the aqueous solutions once they get scattered. Fur- thermore, toxicity issues are arising in the application of CNTs and becoming a major concern in present scenario [13].
Among the various nanoparticles, nZVI has shown remarkable potential to remediate large number of aqueous pollutants. Recently, nZVI has been accounted for effective sequestration of different types of metallic ions such as Cu2?, Pb2?, As3?, Cr6?, Cd2?, As5?, Zn2?, Ba2?and Ni2?
[14–20]. The high surface area and resulting high surface activity of nZVI make it a promising technology for envi- ronmental remediation. Moreover, it is non-toxic and easy to produce [16]. nZVI has a core of iron in its zero-valent state and a surface layer of iron oxides/hydroxides. Studies show that nZVI takes part in heavy metal removal by two mechanisms: sorption and reduction. The core zero-valent iron possesses electron donating power to reduce the heavy metals with standard electrode potential more positive than that of iron. The surface layer of iron oxides/hydroxides offer electrostatic attractions to adsorb heavy metals having standard potentials negative than or close to the iron [20].
This remarkable surface activity of nZVI can be utilized for the iron removal and no such studies have been reported so far. This study aims at evaluating the potential of nZVI on iron remediation by investigating the effects of various influencing operational parameters viz. pH, contact time, iron concentration and nZVI dosage. The results obtained from the experiments are described with different kinetic and adsorption isotherm models. The activity of nZVI and the interaction between different species are assessed by different spectroscopic and microscopic studies.
2. Materials and methods
2.1 Synthesis of nZVI
The liquid-phase reduction technique in the presence of a chelating agent, ethylene diamine tetra acetic acid (EDTA) was used to synthesize nZVI. This prior implemented tech- nique uses sodium borohydride for the reduction of Fe2?[15].
All the chemicals used in this study were of analytical grade purchased from Merck Millipore. In each batch, 0.1 M FeSO47H2O (150 ml) and 0.05 M EDTA (100 ml) were blended in a round bottom flask. The reducing agent (0.75 M NaBH4, 100 ml) was added drop wise into the mixture
prepared. Black particles of nZVI spontaneously appeared in the flask when the first drop of 0.75 M solution of NaBH4was added. After the addition of NaBH4, solution was stirred for 20 min. The black iron nanoparticles were separated by centrifugation. The solid was then washed with 99% absolute ethanol three times. Finally, the synthesized black nanopar- ticles were collected in a glass vial and oven dried at 50°C.
The black colour of the synthesized nZVI particles was per- sistent for 3 months when stored under normal conditions.
2.2 Characterization
The reduction of Fe2?ions to Fe0was ensured by analysing the ultraviolet–visible (UV–Vis) spectrum of nanoparticle solution before and after synthesis. UV–Vis spectropho- tometer (Systronics 2202) was used to analyse the spectrum in the range 200–900 nm.
Scanning electron microscopy/energy dispersive X-ray (SEM/EDX) characterization study was performed using Carlz Zeiss Evo 18 instrument. The solid sample was first applied to a carbon strip attached to a metal disc. Sample surface images were recorded at different magnifications.
The atomic dispersion on the surface of nZVI before and after treatment was mapped by EDX spectrum analysis at randomly chosen areas.
The morphology and surface characteristics of the syn- thesized nZVI was inspected using transmission electron microscopy (TEM). The freshly prepared nZVI was anal- ysed using Jeol/JEM 2100 instrument at 200 kV accelera- tion voltages. Prior to analysis, nZVI was dispersed in ethanol using an ultrasonic bath and a drop of the dispersion was applied to a holey carbon support grid.
A Bruker D8 advance diffractometer which contains a high-power CuKa (k = 1.54 A˚ ) source working at 40 kV/
40 mA was used to record the X-ray diffraction (XRD) patterns. The Fourier transform infrared (FTIR) spectrum was taken by Thermo Nicolet Avtar 370 model spectrom- eter in the range 400–4000 cm-1to detect the characteris- tics of functional groups before and after treatment.
The surface chemistry of synthesized nZVI- and iron- loaded nZVI were examined using X-ray photoelectron spectroscopy (XPS, PHI 5000 VersaProbe II, ULVAC-PHI Inc.) equipped with micro-focused Al-Ka X-ray monochromatic source. High-resolution spectra (narrow scans) as well as survey spectra were recorded. An X-ray source having pass energy of 187.85 eV and power of 50 W was used to record the survey scans. Narrow scans of the major elements were examined at 46.95 eV pass energy.
XPS data were processed using PHI’s Multipak software.
2.3 Batch experimental study
Stock solutions of Fe(II) and Fe(III) ions with 1000 mg l-1 concentration were first prepared by dissolving 280.6 mg of
ammonium ferrous sulphate and 228.2 mg of ammonium ferric sulphate in 500 ml of distilled water, respectively.
These solutions were then diluted to obtain solutions of required concentration. A series of batch studies were conducted to explore the viability of synthesized nZVI particles for iron removal.
This study mainly concentrates on iron removal from simulated surface water containing Fe2?and Fe3?ions. The ratio of ferrous to ferric iron is found to be 2.5:0.5, on an average, in surface waters, even though it shows wide sea- sonal variations [21]. Therefore, the ratio of ferrous ion to ferric ion was fixed as 2.5:0.5 in aqueous solution for the batch study. Ferrous iron solution of 0.42 mg l-1and ferric iron solution of 0.08 mg l-1were prepared and mixed to obtain a sample of the above ratio. The pH of this solution was found to be 6.85. To elucidate the role of contact time, 200 ml of 0.5 mg l-1iron solution at natural solution pH of 6.85 was taken; 1 g l-1of nZVI was added and kept in an orbital shaker
for 1, 2, 3, 6, 12 and 24 h. The experimental data was interpreted by the pseudo-first-order and pseudo-second- order kinetic models. The goodness of kinetic model fit was evaluated by Chi-square (v2) test and the regression coeffi- cient (R2).
A 200 ml iron solution having total iron concentration of 0.5 mg l-1was taken in a series of conical flasks, the pH values were varied from 3–12 and 1 g l-1nZVI was added in each flask. These samples were then agitated for the optimum reaction period (3 h) in order to evaluate the effect of pH.
To investigate the role of adsorbent dosage, 200 ml of 0.5 mg l-1iron solution was taken, the pH was adjusted to the optimum value (pH 10) and added different nZVI dose varying from 0.5–10 g l-1. A series of such conical flasks were agitated for the minimum contact time required to attain equilibrium (3 h). The experimental data was anal- ysed by Langmuir, Freundlich and Dubinin–Radushkevich (D–R) isotherms.
In all batch studies, 50 ml of sample was taken for analysis and allowed 20 min for settling. The supernatant was collected, centrifuged and finally nZVI was magneti- cally separated from treated sample. A control was used in all experiments without adding nZVI to the iron sample solution. The concentration of iron in the clear extracted solution and in control was determined by phenanthroline method [22] using UV–Vis spectrophotometer.
The adsorbed amount and the removal efficiency of nZVI were determined by the equations (1 and 2), respectively [16,23]:
qt¼ðC0CtÞV
m ; ð1Þ
Removal efficiencyð Þ ¼% C0Ct
C0 100; ð2Þ
whereC0andCtare the iron concentrations at the initial and at any timet, respectively,m(g) is the weight of nZVI and V(ml) represents the sample volume.
200 300 400 500 600 700 800 900 -0.5
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5
Absorbance
Wavelength (nm)
Peak pick 425.00 3.219
(a)
200 300 400 500 600 700 800 900 -0.2
-0.1 0.0 0.1 0.2 0.3 0.4 0.5 0.6
Absorbance
Wavelength (nm)
(b)
Peak Pick312.00 0.561
Figure 1. UV–Vis spectrum of (a) Fe2?-EDTA solution before reduction and (b) solution containing nZVI.
Figure 2. SEM image of freshly prepared nZVI.
Figure 3. EDX spectrum of freshly prepared nZVI.
Figure 4. TEM images of freshly prepared nZVI.
3. Results and discussion
3.1 Characterization
The UV–Vis spectroscopic analysis was done to ensure the formation of nanoparticles by comparing the absorption peaks. A shift in the absorption peak from 425 nm (figure1a) to 312 nm (figure1b) confirms the formation of nZVI particles.
To investigate the changes in morphological properties, freshly prepared nZVI was characterized by SEM. The freshly formed iron nanoparticles are in contact with one another and form chain-like aggregates. This chain-like morphology is probably caused by the van der Waals forces and magnetic dipole–dipole interactions (figure2) [24].
The presence of elements C (from EDTA) and B (from NaBH4) can be seen from the EDX spectrum of the freshly prepared nZVI surface in addition to Fe and O (figure3).
The above result is in agreement with the studies of Uzum et al[24] and Zhanget al[25].
Figure4 presents the TEM images of freshly prepared nZVI particles and its chain-like morphology (figure4a and b) justifying SEM results obtained. The prepared nZVI possess a size distribution within the range of 80–99 nm (figure4d) and core-shell structure of these particles could be clearly seen from figure4c. The iron oxide phases and metallic iron can be differentiated from the colour contrast of TEM images. According to the principle of TEM, the components with lower atomic numbers seem lighter than the ones with higher atomic numbers. It can be thus inferred that the average atomic number of surface elements is lower than the average atomic number of the central elements.
Therefore, TEM study verifies the conceptual structure of nZVI, consisting of the metallic iron core and the iron oxides in the shell or surface.
The XRD pattern of freshly prepared nZVI is depicted in figure5. The characteristic diffraction line at 2h = 44.6°in figure5indicates that the prepared iron nanoparticles exist in zero-valent state and persistent without corrosion.
FTIR spectrum for nZVI was recorded in the range 400–4000 cm-1 before and after treatment. As seen from figure6, the absorption peak at 3485 cm-1can be ascribed to O–H stretching vibration of carboxylic groups. A med- ium peak appeared at 1630 cm-1is associated with C=O stretching vibration and the peak at 1383 cm-1reveals C–H bending vibrations. Weak signals observed at 1107 and 617 cm-1are assigned to C–O stretching and C–H bending vibrations, respectively. Moreover, abundant oxygen func- tional groups seen in the FTIR spectrum of freshly prepared nZVI could serve as accessible sites for adsorption. The oxygen atoms in –OH and C–O groups can donate free electrons to intervene with the vacant orbitals of contami- nant ions.
The peaks shown in the XPS surface spectrum of nZVI (figure7a) at around 194, 284.85, 531.66 and 710.10 eV correspond to B, C, O and Fe elements, respectively, consistent with the EDX analysis. The peak for boron is due to the oxidation of the borohydride during synthesis of nZVI. The weak signal displayed at 707 eV corresponds to the binding energy of 2p3/2 of iron (figure7b), reveals the presence of Fe0 in the nanoparticles [26]. The peaks centred at 710 and 724 eV in the survey of Fe 2p core levels (figure7b) represent the binding energies of Fe(2p3/2) and Fe(2p1/2), respec- tively. The shoulder observed at 719 eV results from the overlap of the shake-up satellite of oxidized iron (2p3/2) and zero-valent iron (2p1/2) [10,18]. These prominent peaks are indicative of oxidized iron species which may be present as iron oxides (FexOy) and iron hydroxides (Fe(OH)x) confirming the presence of oxide/hydroxide layer on nZVI surface [27].
0 20 40 60 80 100
In te n s it y
Angle ( 2θ )
44.69 0
Figure 5. XRD pattern of freshly prepared nZVI.
Figure 6. FTIR spectrum of freshly prepared nZVI.
3.2 Batch studies on effect of operational parameters
3.2a Effect of contact time: The effect of contact time on iron removal was investigated at natural solution pH 6.85 for the iron concentration of 0.5 mg l-1 at 1 g l-1 nZVI dosage. The percentage removal of iron species with contact time is presented in figure8. It can be observed that iron is adsorbed rapidly during the first 3 h, beyond which, there is no significant change with further increment. The number of active sites on the nZVI surface is more and iron concentration is high at initial stage. Therefore, these active sites are rapidly interacting with contaminant iron. A total iron removal efficiency of 54% was achieved within 3 h.
After 3 h, the adsorbed amount reached equilibrium due to the slow diffusion of Fe2?/Fe3?ions to the nZVI.
3.2b Effect of initial solution pH: pH has great impacts on the metal chemistry in solution which affect the accessibility of binding sites and in turn affects the metal adsorption [28]. The adsorption ability for target metal ion from aqueous media of 0.5 mg l-1iron concentration was monitored for the optimum reaction time of 3 h at different pH (pH 3 to 12). For all these runs, nZVI dosage was fixed as 1 g l-1. As shown in figure9, iron removal efficacy increased with increase in pH from pH 5 and reached equilibrium at pH of 10 with 57% removal efficiency.
Leaching of iron was also observed in the pH range of 3–4 from nZVI (not shown in figure). Moreover, iron removal was not observed in control, which eliminates the possibility of oxidation. At acidic pH, the competition between the H?ions and the metal ions for the vacant sites results in reduced efficiency [29]. Moreover, the high
1200 1000 800 600 400 200 0
0 1 2 3 4 5
Counts
Binding Energy (ev) C O
Fe O
X104
Atomic % O 46 C 44.7 Fe 8.3 B 1
B (a)
740 730 720 710 700
2000 3000 4000 5000 6000
Counts
Binding Energy (eV)
707 724
719
(b) 710
Figure 7. (a) XPS surface spectrum of freshly prepared nZVI and (b) Fe 2p spectrum of freshly prepared nZVI.
0 3 6 9 12 15 18 21 24 27
0 10 20 30 40 50 60 70 80
Treatment with nZVI Control
Removal Efficiency (%)
Contact time (Hours)
Figure 8. Iron removal efficiency with contact time.
5 6 7 8 9 10 11 12
0 10 20 30 40 50 60 70 80 90 100
pH
RemovalEfficiency (%)
0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60
Final Iron Concentration (mg/L)
Removal efficiency of nZVI Removal efficiency in control Final iron concentration after treatment Final iron concentration in control
Figure 9. Variation of iron removal efficiency and final iron concentration with initial solution pH.
removal efficiency at higher pH can be explained in the context of isoelectric point. The isoelectric point of nZVI is reported as 8.3 in previous studies [30,31]. The oxide surface will act as negatively charged surface at pH higher than the isoelectric point [20]. Therefore, the negatively charged nZVI surface enhances the cation iron adsorption at higher pH.
3.2c Effect of adsorbent dosage: Different dosages (0.5–10 g l-1) of nZVI were applied at optimum pH 10 to 200 ml sample of 0.5 mg l-1 total iron concentration to demonstrate the effect of nZVI dosage on adsorption.
Figure10 shows the effect of adsorbent dosage on iron uptake by nZVI particles. The iron uptake is low at 0.5 g l-1 dosage as the active sites are few at low nZVI amount. However, the iron removal efficiency increases from 49–70% with increase in nZVI dosage and results in constant removal efficacy after 5 g l-1 nZVI dose. Final iron concentration of 0.156 mg l-1 at 5 g l-1 nZVI dose was within the desirable limit of drinking water standards. The dispersion of nZVI particles in aqueous system is best at low adsorbent dosage and all the active sites present on the adsorbent surface are fully exposed which may promote the accessibility of contaminant iron to a large number of active sites. The active sites on the nZVI surface are thus easily saturated, performing high removal efficiency. Higher dosage of nZVI results in lower dispersion which reduces the accessibility of high energy active sites and a larger fraction of low energy active sites being occupied, directing to a reduction in uptake capacity [32]. Besides, the possibility of collision between adsorbent nanoparticles increased by adsorbent dosage, causes particle aggregation, induces an increase in path length of diffusion and a decrease in the total surface area, both resulting in a decrease in contaminant iron adsorption rate [28].
0 2 4 6 8 10
0 10 20 30 40 50 60 70 80
Control
Treatment with nZVI
Removal Efficiency (%)
nZVI dosage (g/L)
Figure 10. Iron removal efficiency with adsorbent dosage.
Table 1. Isotherm model parameters for adsorption of iron on nZVI.
Freundlich model D–R model
kf(l mg-1) n R2 qd(mg g-1) b(mol2J-2) E(kJ mol-1) R2
1.028 1.042 0.9289 21.21 2910-7 1.581 0.9687
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 -1.6
-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2
Log q e
Log Ce y = 0.9591x + 0.0822 R² = 0.9289
(a)
1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 -3.5
-3.0 -2.5 -2.0 -1.5 -1.0 -0.5
ln q e
ε
y = -2E-07x + 3.0547 R² = 0.9687 (b)
2
Figure 11. Adsorption isotherms of iron on nZVI (a) Freundlich isotherm model and (b) D–R isotherm model.
3.3 Adsorption isotherms of iron on nZVI
The adsorption behaviour of total iron on nZVI was mod- elled by Langmuir, Freundlich and Dubinin–Radushkevich (D–R) isotherms. Equations (3–5) represent the linear forms of these isotherms, respectively [16,20,23,33]:
1 qe¼ 1
qmþ 1 qmkL 1
Ce; ð3Þ
logqe¼logkfþ1
nlogCe; ð4Þ
lnqe¼lnqdbe2; ð5Þ
where qe is the equilibrium adsorption capacity (mg g-1), Ce is the equilibrium concentration of iron (mg l-1).
Langmuir constant, kL is related to the free energy of adsorption and qm corresponds to maximum monolayer adsorption capacity (mg g-1).n andkfare the Freundlich isotherm constants signifying the intensity and capacity of the adsorption, respectively. qd in D–R isotherm corre- sponds to maximum coverage (mg g-1),bis D–R isotherm constant related to mean free energy (E), which is in turn given by 1/(2b)0.5. The parameter e in D–R isotherm is calculated according to equation (6) [16,23]:
e¼RTln 1þ 1 Ce
: ð6Þ
The experimental data for different nZVI doses were fitted to the linear forms of isotherm equations. The results show that the Langmuir isotherm (data not shown) yielded poor regression coefficient (R2\0.85) while Freundlich and Dubinin–Radushkevich isotherms provided proper descrip- tion of the data (table1).
According to the Freundlich model, the surface active sites are heterogeneous with extensive affinities. The Table 2. Adsorption kinetic model parameters for iron adsorption on nZVI.
qe,exp(mg g-1)
Pseudo-first-order model Pseudo-second-order model
k1(h-1) qe,cal(mg g-1) R2 k2(h-1) qe,cal(mg g-1) R2
0.280 0.2903 0.124 0.7013 2.476 0.299 0.9931
v2 1.466 0.032
0 5 10 15 20 25
-5.5 -5.0 -4.5 -4.0 -3.5 -3.0 -2.5 -2.0 -1.5
ln
(
q e- q t)
Time (Hours) y = -0.2903x - 2.0859 R² = 0.7013
(a)
0 5 10 15 20 25
0 10 20 30 40 50 60 70 80 90
t/q t
Time (hours) y = 3.3401x + 4.5175 R² = 0.9931
(b)
Figure 12. (a) Pseudo-first-order and (b) pseudo-second-order kinetic model of iron adsorption model.
Figure 13. SEM image of iron-loaded nZVI.
stronger binding sites are filled by the adsorbate species first and as the degree of site occupation increases, the binding force decreases. The Freundlich constants, n andkfcalcu- lated from the plot of lnqeand lnCe(figure11a) establish how favourable an adsorption process and the adsorption distribution are onto the nZVI at equilibrium, respectively.
The value of n[1 indicates that iron ions are favourably adsorbed on nZVI.
The Dubinin–Radushkevich model follows multilayer adsorption behaviour which involves van der Waals forces of attraction and the mean adsorption energy (E) provides information on the chemical and physical adsorption pro- cess. The slope and intercept of the plot ln qe vs. e2 (figure11b) are used to compute D–R isotherm constantsb and qd, respectively. Value of b is inversely related to
concentration change and free energy of adsorption. In physisorption, the values ofb areC7.81 910-9mol2J-2 (value corresponding to 8 kJ mol-1, maximum free energy of physisorption). The magnitude of regression coefficient (R2= 0.9687) confirms that the data is well-fitted with the D–R isotherm. Moreover, the lower value of E(\8 kJ mol-1) suggests that physical sorption have more contribution in this adsorption mechanism [23].
3.4 Adsorption kinetics of iron on nZVI
The evaluation of the adsorption kinetics of contaminant iron on nZVI was done by pseudo-first-order and pseudo- second-order models and goodness of the fit was assessed by Chi-square (v2) test and linear coefficient of determi- nation (R2). Table2 summarizes the parameters of two Figure 14. EDX spectrum of iron-loaded nZVI.
0 20 40 60 80 100
In te n s it y
Angle ( 2θ )
30.2 0
35.6 0 44.6 0
Figure 15. XRD pattern of iron-loaded nZVI.
4000 3500 3000 2500 2000 1500 1000 500 -10
0 10 20 30 40 50 60
% Transmittance
Wave numbers
(
cm-1)
3372 3204
16261392 1120 1020
880
568
Figure 16. FTIR spectrum of iron-loaded nZVI.
kinetic models and figure12a and b shows the linear graphical representation of ln(qe-qt) vs. t and t/qt vs. t, respectively. Among the two models, the maximum adsorption capacity computed from the pseudo-second- order model shows close proximity to the experimental results. In addition, the pseudo-first-order shows lower regression coefficient and data was well-fitted with the pseudo-second-order kinetics withR2= 0.9931. The lower v2value of 0.032 for the pseudo-second-order also recom- mends that iron adsorption on nZVI followed the pseudo- second-order kinetics.
3.5 Mechanism of iron removal by nZVI
The nZVI after treatment of 0.5 mg l-1 iron solution at natural solution pH 6.85 with 5 g l-1dose was subjected to the characterization study to elucidate the mechanism of iron removal from aqueous solution.
As illustrated in SEM image of iron-loaded nZVI (figure13), the sizes of the nZVI particles increased markedly after interacting with Fe2?/Fe3? in the influent solution. Comparing figures2 and 13, more particles are bound to nZVI with coir-like structure, after the treatment, confirming adsorption of Fe2?and Fe3?.
The EDX spectrum was also recorded for Fe2?/Fe3?- loaded nZVI, after the treatment with 0.5 mg l-1 of iron solution, and it is shown in figure14. As per the EDX spectra, the weight percentage of iron on the unreacted nZVI particles is 65.87 and after the treatment it is 85.26, the increase in weight percentage verifying the adsorption of iron on nZVI.
The XRD spectrum of iron-loaded nZVI (figure 15) shows new 2h peaks at 35.6° and 30.2° corresponding to Fe2O3and Fe3O4 [34] confirming the interaction between nZVI and Fe2?/Fe3?. The XRD results cannot distinguish the phases of iron oxide, Fe2O3 and Fe3O4 and therefore
the resulting peaks could stand for either phase or both [14].
The FTIR spectrum of iron-loaded nZVI is shown in figure16. Compared with the FTIR peaks of freshly pre- pared nZVI, the peak at 480 cm-1 disappeared and peak shifting occurred from 3485 to 3370 cm-1. A strong signal appeared at 568 cm-1due to Fe–O stretching vibration. The above changes are illustrations of interaction of contami- nant iron with nZVI. Moreover, the small shifting of peaks in the range of 1626–1020 cm-1 indicates effective adsorption of Fe2?/Fe3? ions on the surface of nZVI by chelating with oxygen containing groups [35].
In the XPS spectra of iron-loaded nZVI (figure17a and b), small new peaks observed around *90 eV correspond to Fe 3s and Fe 3p along with increase in atomic percentage of iron. These observations also indicate the interaction between contaminant iron and nZVI.
Li and Zhang [10] reported that the metal ions having standard reduction potential close to Fe0 are removed by adsorption, while the metal ions with standard reduction potential more positive than Fe0are removed by reduction.
Thus iron removal by nZVI is predominant by adsorption.
4. Conclusions
The present study showed that nZVI can be used as an effective adsorbent for the removal of iron from aqueous solutions. SEM/EDX results demonstrated an increase in atomic percentage of iron after treatment, confirming uptake of contaminant iron from solution by nZVI. The appearance of new peaks and shift in peaks observed in XRD, FTIR and XPS analyses further confirmed that the uptake of iron by nZVI is through the adsorption process.
The performance of batch experiments under various con- ditions indicated that nZVI has good removal ability towards contaminant iron and could bring the effluent iron
740 730 720 710 700
2000 3000 4000 5000 6000
Counts
Binding Energy (eV) (b)
724
719 710
707 (a)
Figure 17. (a) XPS surface spectrum of iron-loaded nZVI and (b) Fe 2p spectrum of iron-loaded nZVI.
concentration to desirable limit of potability. pH is a significant factor affecting the removal efficiency of iron by nZVI. The iron removal is more effective in alkaline pH greater than the isoelectric pH of nZVI (8.3). The experi- mental data was well-fitted with the D–R isotherm with correlation coefficient 0.9687 and maximum adsorption capacity was found to be 21.2 mg g-1. D–R isotherm assumes that the adsorption has a multilayer character, involves van der Waals forces and is applicable for physical adsorption processes. Moreover, mean free energy
\8 kJ mol-1 strongly supported the physical adsorption mechanism for iron uptake by nZVI particles.
Thus the results indicate the feasibility of employing nZVI as an adsorbent for iron remediation from aqueous system. The characterization results show agglomeration of nZVI which may limit its effective utilization. This can be overcome by suitable surface coatings on nZVI or impregnating nZVI on suitable substances. Further modifi- cations are required for the enhancement of iron removal at higher concentration and for the reusability of nZVI.
Acknowledgements
This study was performed in Environmental Engineering Lab of Government Engineering College, Thrissur, Kerala and we are grateful to the lab staff for their help and support. We would like to thank the Sophisticated Test and Instrumentation Centre (STIC, CUSAT), CSIR-National Institute for Interdisciplinary Science and Technology (NIIST), Thiruvananthapuram and Centre for Materials for Electronics Technology (C-MET), Thrissur, for providing facilities for characterization studies. Also, the help and suggestions rendered by Dr K R Dayas, Special Officer Vijnan Sagar Thrissur, is greatly acknowledged. This study was supported by AICTE through the NDF-RPS scheme for the NDF research scholars.
References
[1] Khatri N, Tyagi S and Rawtani D 2017J. Water Process Eng.19291
[2] Ellis D, Bouchard C and Lantagne G 2000Desalination130 255
[3] Abraham N, James J, Banerji T and Menon R 2020 Groundw. Sustain. Dev.10318
[4] Hartmann J, Braulke F, Sinzig U, Wulf G, Maas J H, Konietschke Fet al2013Leuk. Res.37327
[5] Das B, Hazarika P, Saikia G, Kalita H, Goswami D C and Dutta R K 2007J. Hazard. Mater.141834
[6] Dabrowski A, Hubicki Z, Podkoscielny P and Robens E 2004Chemosphere5691
[7] Ghosh D, Solanki H and Purkait M K 2008J. Hazard.
Mater.155135
[8] Kan C, Chen W, Wan M, Phatai P, Wittayakun J and Li K 2012Sustain. Environ. Res.2225
[9] Nemade P D, Kadam A M and Shankar H S 2009Ecol. Eng.
351152
[10] Li X and Zhang W 2007J. Phys. Chem. C11152 [11] Le N L and Nunes S P 2016SM&T71
[12] Alimohammadi V, Sedighi M and Jabbari E 2017Ecol. Eng.
10290
[13] Zhang C, Sui J, Li J, Tang Y and Cai W 2012Chem. Eng. J.
21045
[14] Celebi O, Uzum C, Shahwan T and Erten H N 2007J.
Hazard. Mater.148761
[15] Shi L, Zhang X and Chen Z 2011Water Res.45886 [16] Bopari H K, Joseph M and Carroll D M O 2011J. Hazard.
Mater.186458
[17] Dong H, Zeng G, Zhang C, Liang J, Ahmad K, Xu Pet al 2015J. Environ. Sci. (China)32180
[18] Liang W, Dai C, Zhou X and Zhang Y 2014 PLoS One9 e85686
[19] Mosaferi M, Nemati S, Khataee A, Nasseri S and Ahmad A H 2014J. Environ. Health Sci. Eng.121
[20] Li S, Wang W, Liang F and Zhang W X 2017J. Hazard.
Mater.322163
[21] Subin M P, Anitha C T and Sidhimol P D 2011Nat. Environ.
Pollut. Technol.10583
[22] American Public Health Association (APHA) 2012Standard Methods for the Water and Wastewater, 22nd edn. APHA, Washington, D.C.
[23] Kumar P S, Ramalingam S, Kirupha S D, Murugesan A, Vidhyadevi T and Sivanesan S 2011Chem. Eng. J.167122 [24] Uzum C, Shahwan T, Eroglu A E, Lieberwirth I, Scott T B
and Hallam K R 2008Chem. Eng. J.144213
[25] Zhang X, Shi D, Li X, Zhang Y, Wang J and Fan J 2019 Chemosphere224390
[26] Bhowmick S, Sudipta C, Priyanka M, Wouter V R, Sven V B, Gabriela R Ret al2014Chem.Eng. J.24314 [27] Grosvenor A P, Kobe B A, Biesinger M C and McIntyre N S
2004Surf. Interface Anal.361564
[28] Heidari A, Younesi H and Mehraban Z 2009Chem. Eng. J.
15370
[29] Stumm W and Morgan J J 1996Aquatic chemistry, chemical equilibria and rates in natural waters 3rd edn (New York:
John Wiley & Sons, Inc.)
[30] Sun Y P, Li X Q, Cao J, Zhang W and Wang H P 2006Adv.
Colloid Interface Sci.12047
[31] Karabelli D, Uzum C, Shalhwan T, Eroglu A E, Scott T B, Hallam K Ret al2008Ind. Eng.Chem. Res.474758 [32] Arshadi M, Soleymanzadeh M, Salvacion J W L and Sal-
imiVahid F 2014J. Colloid Interface Sci.426241 [33] Zhang X, Lin S, Lu X Q and Chen Z 2010Chem. Eng. J.163
243
[34] Chen H, Luo H, Lan Y, Dong T, Hu B and Wang Y 2011J.
Hazard. Mater.19244
[35] Yang F, Zhang S, Sun Y, Cheng K, Li J and Tsang D C W 2018Bioresour. Technol.265490
