

Magnetic Nanoparticles and Cellulosic Nanofibers to Remove

Arsenic and Other Heavy Metals from Water
Tania Dey

Brussels University, Belgium. E-mail: taniadey@hotmail.com

 Introduction
 Heavy Metal pollutants in Water
 Magnetic Nanoparticles
 Natural Occurrence
 Synthetic Approaches
... Solution Methods
.... Co-precipitation
.... Microemulsion
.... Polyols
.... High temperature decomposition of precursors
.... Other solution techniques
... Aerosol and Vapor Methods
.... Spray pyrolysis
.... Laser pyrolysis
. Cellulosic Nanofibers
.. Natural Occurrence
.. Processing Techniques
... Mechanical fibrillation
... Electrospinning
... Swelled precursor
... Acid hydrolysis
... By bacteria
. State-of-Art Applications
.. Magnetite (FeO)
.. Maghemite (FeO)
.. Nano zero valent iron (nZVI)
.. Mixed oxides
.. Organic-inorganic nanohybrids
.. Cellulosic biomass
. Nanotechnological Principal applied to Water Purification
.. Nano effect
.. Effect of pH
.. Effect of competing species
.. Effect of temperature
.. Effect of pollutant (adsorbate) concentration
.. Effect of contact time
.. Effect of adsorbent dose


NANOTECHNOLOGY FOR WATER PURIFICATION
. Limitations
. Conclusion
References

. Introduction

One of the most serious problems affecting humanity is the lack of water resources for human con-
sumption. Worldwide . billion people do not have access to clean and safe drinking water and .
billion people lack sanitation []. Every year  million people die of waterborne diseases. This situa-
tion is particularly dramatic in the least developed regions of Asia, Central and South America and
Africa, where new water treatment processes are needed urgently. Problems with water are expected
to grow worse in the coming decades, with water scarcity occurring globally, even in regions currently
considered water-rich. Addressing these problems calls out for a tremendous amount of research to
be conducted to identify new technologies for purifying water at lower cost and with less energy,
while at the same time minimizing the use of chemicals and impact on the environment. This review
chapter highlights some of the state-of-art methodologies of using nanomaterials as adsorbents for
low-cost water treatment.
Ground water contamination by heavy metals is a key environmental concern in areas where the
water supply system draws primarily on groundwater. Several methods of arsenic and chromium re-
moval are already available including oxidation-precipitation, coagulation-precipitation, electrochemi-
cal reduction, adsorption, ion exchange, solvent extraction, nano filtration and reverse osmosis [,].
Adsorption of arsenic and chromium on different sorbents such as iron, iron oxide, iron coated sand,
iron coated activated carbon [], and granular ferric hydroxides [] have also been investigated. How-
ever, their use is limited due to high operation cost, sludge formation, and technical difficulties in
preparation of materials.
Nanoparticles are often associated with greater reactivity due to their large surface area and
unique optical, electronic, magnetic properties in comparison to bulk, originating from quantum con-
finement effect. Magnetic nanoparticles, especially the naturally occurring ones, such as magnetite,
maghemite, zero valent iron, mixed oxides (Co-ferrites, Ni-ferrites etc) as well as cellulosic nanofibers
can be used as an effective adsorbent for water remediation, making the process cost-effective and
sustainable. Being a dispersed system, these adsorbents eliminate the need of high pressure streams
for water purification.

. Heavy Metal Pollutants in Water

Most environmental arsenic problems are the result of mobilization under natural conditions. This
includes mobilization by natural weathering reactions, biological activity, geochemical reactions, vol-
canic emissions and other anthropogenic activities. Soil erosion and leaching contribute to ×
and × g/year of arsenic, respectively, in dissolved and suspended forms in the oceans [].
However, mining activities, combustion of fossil fuels, use of arsenic pesticides, herbicides, and crop
desiccants and use of arsenic additives to livestock feed create additional impacts.
Higher levels of arsenic tend to be found more in ground water sources than in surface water
sources of drinking water. Two forms are common in natural waters: arsenite (AsO−) and arsenate
(AsO−), referred to as arsenic(III) and arsenic(V). Pentavalent (+) or arsenate species are AsO −,
HAsO−, HAsO− while trivalent (+) arsenites include As(OH), As(OH)−, AsOOH− and AsO−.
Pentavalent species predominate and are stable in oxygen rich aerobic environments, whereas triva-
lent arsenites predominate in moderately reducing anaerobic environments such as groundwater [].


 DEY
Arsenic-contaminated water has been a serious problem especially in Vietnam, Bangladesh and
some areas in the world [,]. These areas include Taiwan (. mg/L), Hungary (. mg/L), India
(. mg/L), Mexico (. mg/L), and the southwest United States (. mg/L) [] and to some extent
Argentina, Poland, Japan and New Zealand. Human exposure to arsenic can cause both short and
long term health effects. Short or acute effects can occur within hours or days of exposure. Long or
chronic effects occur over many years. Long term exposure to arsenic has been linked to cancer of
the bladder, lungs, skin, kidneys, nasal passages, liver and prostate, as well as pigmentation changes,
skin thickening (hyperkeratosis), neurological disorders, muscular weakness, loss of appetite, and nau-
sea [-]. Short term exposure to high doses of arsenic can cause other adverse health effects [,
].
The World Health Organization (WHO) has revised the guideline for maximum permissible con-
centration (MPC) value of arsenic in drinking water from  to  ppb i.e. . to . mg/L [].
However, many countries have retained the earlier WHO guideline of  ppb.
Another prominent heavy metal pollutant is water resources is chromium. Chromium, which is
one of the most toxic metals, is mixed into river water and ground water through the electroplating
industries, metal finishing, leather tanning and chrome preparations. In the United States, it is the
second most common inorganic contaminant in waters after lead [].Chromium exists mostly in two
valence states, namely Cr (III) and Cr (VI), out of which the later one is of the great concern due to
its toxicity. It is a US EPA classified group A carcinogen based on its chronic effects [].The mecha-
nistic cytotoxicity of Cr (VI) is not completely understood, however, a large number of studies have
demonstrated that it induces oxidative stress, DNA damage and apoptotic cell death []. Cr (VI) has
been reported to be responsible for lung cancer, chrome ulcer, perforation of nasal septum and brain
and kidney damage []. The major cause of its acute toxicity is its rapid diffusivity through the skin
which enables it to react with biological systems and damage various organs.

. Magnetic Nanoparticles

.. Natural Occurrence

While man strives to synthesize and utilize advanced nanomaterials, Nature’s own laboratory has pro-
duced a range of highly selective nanomaterials, since ages.
Magnetite (FeO) is a ferrimagnetic mineral and a member of the spinel group. Magnetite is the
most magnetic naturally occurring minerals on Earth [].Small grains of magnetite occur in almost all
igneous and metamorphic rocks. Magnetite also occurs in many sedimentary rocks, including banded
iron formations. In many igneous rocks, magnetite-rich and ilmenite-rich grains occur, which were
precipitated together from magma. Magnetite also is produced from peridotites and dunites by ser-
pentinization. Magnetite is sometimes found in large quantities in beach sand. Such black sands (min-
eral sands or iron sands) are found in various places, such as California and the west coast of New
Zealand. Continuous wetting of steel surfaces results in the formation of magnetite as the main con-
stituent of rust formed.
The mineral Hematite (FeO) is known to form pseudomorphs over Magnetite. However, they
differ from Magnetite in that they are only weakly attracted to magnetic field and have a reddish-
brown streak.
Zero Valent Iron (ZVI) is also abundant in nature, from which nanoscale Zero Valent Iron
(nZVI) can be made just by milling. Li et al. has proposed a core-shell model for nZVI particle [],
where the core remains ZVI and the shell, which forms as a result of oxidation reaction, is largely


NANOTECHNOLOGY FOR WATER PURIFICATION
composed of iron oxides and hydroxides. It is the mixed oxide shell (Fe+ and Fe+) that provides
sites for adsorption (Figure ).

Figure : Core-shell model of nZVI particle, ref: []. Reprinted by permission of the publisher, Taylor & Francis Ltd.
http://www.tandf.co.uk/journals

.. Synthetic Approaches

... Solution Methods

Uniform nanoparticles are usually prepared via the homogeneous precipitation reaction, a process
that involves separation between nucleation and the growth of nuclei []. In homogeneous precipita-
tion, a short single burst of nucleation occurs when the concentration of the constituent species
reaches critical supersaturation. The nuclei so obtained are allowed to grow uniformly by diffusion of
solutes from solution to surface until the final size is attained. To achieve monodispersity, these two
stages must be separated, and nucleation should be avoided during the growth of nuclei. This classical
model was first proposed by LaMer and Dinegar [] to explain the mechanism of sulfur colloid for-
mation. However, uniform particles have also been obtained after multiple nucleation events. The
uniformity of the final product in this case is achieved through a self-sharpening growth process
(Ostwald ripening) []. In addition, uniform particles have also been obtained as a result of aggrega-
tion of much smaller subunits rather than continuous growth by diffusion [-]. An artificial separa-
tion between nucleation and growth processes may be achieved by seeding in which foreign particles
are introduced into a solution below critical supersaturation [].

.... Co-precipitation There are mainly two methods for synthesis of spherical magnetite na-
noparticles by this method. In the first, ferrous hydroxide Fe(OH) suspensions are partially oxidized
with different oxidizing agents []. For example, a narrow size distribution of - nm can be ob-
tained using a Fe(II) salt, a base, and a mild oxidant (like nitrate ions) []. The other method involves


 DEY
co-precipitation of a stoichiometric mixture of ferric and ferrous hydroxides in aqueous media, yield-
ing spherical magnetite nanoparticles of quite homogeneous size distribution []. In addition, it has
been shown that by adjusting pH and ionic strength, it is possible to control the mean size of the na-
noparticles over an order of magnitude (from  to  nm) []. The size decreases as the pH and ionic
strength of the medium increase []. Both parameters affect the chemical composition of the surface
and consequently the electrostatic surface charge of the particles.

.... Microemulsion Water-in-oil (w/o) microemulsions (i.e. reverse-micelle solutions) are

transparent, isotropic, thermodynamically stable liquid media. In these systems, fine microdroplets of
the aqueous phase are trapped within assemblies of surfactant molecules dispersed in a continuous oil
phase. The surfactant-stabilized microcavities (typically in the range of  nm) provide a confinement
effect that limits particle nucleation, growth, and agglomeration []. W/o microemulsions have prov-
en to be an adequate, versatile, and simple method to prepare nanosized particles[-] and useful
for in vivo and in vitro applications.
Pileni and coworkers [] prepared nanosized magnetic particles with average sizes from  to 
nm and a standard deviation ranging from . to . using microemulsions. A micellar solution of
ferrous dodecyl sulfate Fe(DS) was used to produce magnetic nanoparticles whose size was con-
trolled by the surfactant concentration and temperature.
A typical transmission electron microscopy (TEM) micrograph of magnetic nanoparticles pro-
duced by this method is shown in Figure a. Magnetite nanoparticles of ~ nm in diameter have been
prepared by controlled hydrolysis with ammonium hydroxide of FeCl and FeCl aqueous solutions
within the reverse-micelle microcavities generated by cosurfactant AOT and heptane as a continuous
oil phase [].
Carpenter and coworkers [] prepared metallic iron particles coated by a thin layer of gold via
the microemulsion method. The gold shell not only protects the iron core but also serves as an easy
means of attaching functionalities using well-known gold–thiol covalent linkage, hence making the
composites lucrative for various applications. The reverse-micelle reaction is carried out using
cetyltrimethylammonium bromide (CTAB) as surfactant, octane as oil phase, and aqueous reactants as
water phase []. The metal particles are formed inside the reverse micelle by reduction of a metal salt
using sodium borohydride. The synthesis technique offered by reverse micelles is sequentially applied,
first to prepare the iron core by reduction of ferrous sulfate using sodium borohydride. After the re-
action was allowed to reach completion, the micelles within the reaction system were enlarged to ac-
commodate the shell using larger micelles obtained from additional borohydride. The gold shell was
formed from hydrogen tetrachloroaurate solution.
Although reverse micelles have been successfully used as nanoreactors for the synthesis of mag-
netic nanoparticles, the main drawbacks associated with this process are () extensive agglomeration,
() poor crystallinity (as the reaction is performed at relatively low temperature), and () low yield.
However, Hyeon and coworkers [] have reported large-scale synthesis of uniformly sized highly
crystalline magnetic nanoparticles (magnetite as well as mixed metal ferrites, where metal = cobalt,
manganese, nickel, and zinc), in microemulsion nanoreactors at high temperature. They have achieved
good control over particle sizes from  to  nm by varying relative concentrations of precursor salt,
surfactant, solvent, and the w ratio (where w = [a polar solvent such as ethanol, methanol, or wa-
ter]/[surfactant]). They have also discussed the temperature and field dependence of magnetic proper-
ties in the same paper. A similar size selection procedure in the range of -  nm for CoFeO has
been reported by Döker et al. []. They observed that particle size of CoFeO increases with increase
in reaction time and decrease in FeSO percentage and AOT (the surfactant) concentration.


NANOTECHNOLOGY FOR WATER PURIFICATION

.... Polyols A very promising technique for the preparation of uniform nanoparticles is the pol-
yol process. Fine metallic particles can be obtained by reduction of dissolved metallic salts and direct
metal precipitation from a solution containing a polyol [, ]. This process was first used to prepare
noble metals such as Ru, Pd, Pt, and Au and others such as Co, Ni, and Cu [, ]. Later the process
was extended to synthesize other materials such as Fe-based alloys [,].
In the polyol process, the liquid polyol acts as a solvent for the metallic precursor, as well as a re-
ducing agent and in some cases as a chelating ligand. The metal glycol can be highly or only slightly
soluble in the polyol. The solution is stirred and heated to a given temperature reaching the boiling
point of the polyol. By controlling the kinetics of precipitation, non-agglomerated nanoparticles of
well-defined size and shape can be obtained. A better control of the average size can be achieved by
seeding the reactive medium with foreign particles (heterogeneous nucleation). In this way, nucleation
and growth steps can be completely separated and uniform particles can be prepared. Iron particles of
~ nm can be obtained by disproportionation of ferrous hydroxide in organic media []. Fe (II)
chloride and sodium hydroxide react with ethylene glycol (EG) or polyethylene glycol, and the precip-
itation occurs at a temperature of –°C.
The author of this book chapter, Dey [, ], has carried out some pioneering work in this area
using a mixture of Fe (II) and Fe (III) chloride salts in the presence of sodium hydroxide in diethylene
glycol (DEG) medium at an elevated temperature (–°C) under argon protection, yielding mag-
netite nanoparticles of very narrow size distribution and a mean diameter of . nm (Figure b). The
underlying mechanism may involve three consecutive steps, such as a metal complexation reaction
between hydrated metal chloride and DEG (where DEG acts as a chelating agent as well as a solvent
and M = Fe+, Fe+, Mn+, Co+, Ni+, and Zn+), followed by hydrolysis to corresponding transition
metal hydroxide and finally dehydration due to heat treatment forming spinel structured metal ferrites.
Capping ligands such as long-chain carboxylic acids (namely, oleic acid) were found to be quite effec-
tive in stabilizing the nanoparticles.
Furthermore, iron alloys can be obtained in a similar manner by co-precipitation of Fe, Ni,
and/or Co in EG or PEG. Monodispersed quasi-spherical and nonagglomerated metallic particles
with ~ nm diameter have been obtained without seeding, while particles between  and  nm
were obtained using Pt as a nucleating agent. Whereas FeCo particles are formed by agglomerates of
Fe and Co primary particles produced over different periods of time, spherical FeNi particles present
good homogeneity due to concomitant Fe and Ni formation and growth by aggregation of nanome-
ter-sized primary particles [].

.... High-temperature decomposition of precursor The decomposition of iron precur-

sors in the presence of hot organic surfactant has yielded markedly improved samples with good size
control, narrow size distribution, and good crystallinity of well-dispersed iron oxide nanoparticles. For
example, Alivisatos and coworkers [] have demonstrated that injecting solutions of FeCup (Cup =
Nnitrosophenylhydroxylamine) in octylamine into long-chain amines at -°C yields nanocrystals
of maghemite. These nanocrystals range from  to  nm in diameter, are crystalline, and are dispersi-
ble in organic solvents. Hyeon and coworkers [] have also been able to prepare monodisperse ma-
ghemite nanoparticles by a nonhydrolytic synthetic method. For example, to prepare maghemite na-
noparticles of  nm diameter (Figure c), Fe(CO) was injected into a solution containing surfactant
and a mild oxidant (trimethylamine oxide).
Recently, Sun and Zeng [] have been able to prepare monodispersed magnetite nanoparticles
with sizes from  to  nm by high-temperature (°C) reaction of iron(III) acetylacetonate with
phenyl ether in the presence of alcohol, oleic acid, and oleylamine. In particular, magnetite nanoparti-


 DEY
cles of size~ nm were obtained by thermal decomposition of the iron precursor, although to obtain
diameters up to  nm seed-mediated growth was required.
Varma and coworkers [] have synthesized iron oxide nanoparticles in three major phases –
namely, α- and γ-FeO and FeO – by a self-sustained solution combustion approach and by using a
range of precursors and oxidizers such as iron nitrate and oxalate, as well as different fuels (e.g., gly-
cine and hydrazine). The product was found to have good crystalline structure and surface area in the
range – m/g, avoiding the need of additional calcination procedures. They showed how the
product composition and properties can be controlled by varying oxidizer as well as fuel.

.... Other solution techniques Sonochemistry-assisted synthesis has also been reported as
an adequate method for the production of magnetite and maghemite nanoparticles [–]. In sono-
chemistry, the acoustic cavitation (i.e., the formation, growth, and implosive collapse of a bubble in an
irradiated liquid) generates a transient localized spot, with an effective temperature of  K and a
nanosecond lifetime []. The cavitation is a quenching process, and hence the composition of the
particles formed is identical to the composition of the vapor in the bubbles, without phase separation.
Electrochemical methods have also been used for the production of maghemite nanoparticles
[]. The electrochemical synthesis of nanoparticles of γ-FeO was performed in an organic medium.
The size was directly controlled by the imposed current density, and the resulting particles were stabi-
lized as a colloidal suspension by use of cationic surfactants. The size distribution of the particles was
narrow, with average sizes varying from  to  nm.

... Aerosol and Vapor Methods

Spray and laser pyrolysis have been shown to be excellent techniques for direct and continuous pro-
duction of well-defined magnetic nanoparticles with exhaustive control on experimental conditions.
Their high production rate can hold a promising future for magnetic nanoparticles useful for in vivo
and in vitro applications. The main difference between spray and laser pyrolysis is in the final state of
the ultrafine particles. In spray pyrolysis, the ultrafine particles are usually aggregated into larger parti-
cles, while in laser pyrolysis, the ultrafine particles are less aggregated due to the shorter reaction time.

.... Spray pyrolysis Spray pyrolysis is a process in which a solid is obtained by spraying a solu-
tion into a series of reactors where the aerosol droplets undergo evaporation of solvent and solute
condenses within the droplet, followed by drying and thermolysis of the precipitated particles at high-
er temperature []. This procedure gives rise to microporous solids, which finally sinter to form
dense particles.
This method represents a convenient procedure for obtaining finely dispersed particles of pre-
dictable shape, size, and variable composition. The resulting powders generally consist of spherical
particles, the final diameter of which can be predetermined from that of the original droplets. The
method offers certain advantages over other more commonly used techniques (such as precipitation
from homogeneous solution) as it is simple, rapid, and continuous. Recently, this method has been
used for the production of mesoporous microspheres [] and phosphorescent nanoparticles [].
Most of the pyrolysis-based processes employed to produce maghemite nanoparticles start with
an Fe+ salt and some organic compound that acts as a reducing agent. It was shown that in this pro-
cedure, Fe+ is partially reduced to a mixture of Fe+ and Fe+ in the presence of organic compounds
with the formation of magnetite, which is finally oxidized to maghemite. In absence of a reducing
agent, hematite is formed instead of maghemite [].


NANOTECHNOLOGY FOR WATER PURIFICATION

Figure : Magnetic nanoparticles prepared in solu-

tion by (a) the microemulsion method (maghemite),
Reprinted with permission from ref: [], copyright
, American Chemical Society (b) the polyol
method (magnetite), Reprinted with permission from
ref: [], copyright at American Scientific Publishers
and (c) high-temperature decomposition of Fe(CO)
(maghemite), Reprinted with permission from ref:
[], copyright , American Chemical Socie-
ty.This entire figure appeared in author’s review
article, ref [].


 DEY
In alcoholic solutions, uniform γ-FeO particles can be prepared with a wide variety of particle
morphology and sizes, ranging from  to  nm, depending on the nature of the iron precursor salt
[]. A detailed description of the device used for this kind of synthesis can be found in the literature
[]. The device essentially consists of an aerosol droplet generator (atomizer, ultrasonic, etc.), a fur-
nace, and a particle recovery system. Dense aggregates of spherical shapes composed of γ-FeO
subunits with mean diameters of  and  nm have been obtained using Fe (III) nitrate and Fe (III)
chloride solutions, respectively. On the other hand, γ-FeO obtained from acetylacetonate solution
resulted in monodispersed particles of about  nm in diameter, while maghemite particles derived
from Fe(II) ammonium citrate appeared as hollow spheres with a mean diameter of  nm. The lat-
ter consisted of small crystallites aggregated to form a shell, the size of which varied between  and
 nm, depending on the heating temperature of the furnace.

.... Laser pyrolysis Since the pioneering work of Cannon and coworkers [] on the continu-
ous production of nanometer-sized powders by laser-induced processes, different powders such as Si,
SiC, SiN, and Si–C–N composite have been prepared under a variety of conditions with sizes rang-
ing from  to  nm [, ]. The method involves heating a flowing mixture of gases with a continu-
ous-wave carbon dioxide laser, which initiates and sustains a chemical reaction. Above certain pres-
sure and laser power, a critical concentration of nuclei is reached in the reaction zone, which leads to
the homogeneous nucleation of particles that are further transported to a filter by an inert gas.
Pure, well-crystallized, and uniform γ-FeO nanoparticles can be obtained in one single step by a
CO laser pyrolysis method. Samples with particles of sizes as small as . and  nm and very narrow
size distribution have been obtained under different experimental conditions [, ]. The device con-
sists of a small reaction zone defined by the overlap of the vertical reactant gas stream and horizontal
laser beam. The reaction zone, being safely separated from chamber walls, offers an ideal environ-
ment for the nucleation of small nanoparticles, with less contamination and narrower size distribution
than that of conventional thermal methods.
To obtain γ-FeO nanoparticles, Fe(CO) was used as a precursor. Due to the fact that this pre-
cursor does not absorb radiation at laser wavelength (. ± . μm), ethylene was used as the ab-
sorbent as well as carrier to transport the carbonyl vapor to the reaction zone. Ethylene does not de-
compose at the energy density used ( Wcm-) but simply absorbs the laser radiation to heat up
Fe(CO), which decomposes into iron and carbon monoxide. In order to obtain iron oxide, air needs
to be introduced into the system, either with Fe(CO) vapor causing oxidation under laser radiation or
mixed with argon.

. Cellulosic Nanofibers

.. Natural Occurrence

Natural fibers are pervasive throughout the world in plants such as grasses, seeds, stalks, and woody
vegetation. They are also referred to as cellulosic fibers. Agricultural and forest wastes are made up of
cellulosic fibers, which can serve as an excellent adsorption substrate to treat polluted water – a con-
cept of ‘using waste to treat waste’. The cost of adsorption material for water treatment can hence be
dramatically reduced, while other advantages include natural abundance, high mechanical strength,
low energy consumption, low density and wide varieties. Coffee husk [], peanut shells [], bagasse
[], rice husk [] and sawdust [] has already been used as an adsorbent for wastewater treatment.


NANOTECHNOLOGY FOR WATER PURIFICATION
Wood fibers, the most abundant biomass resource on earth, are hollow tubes made up of cellu-
lose embedded in a matrix of hemicellulose and lignin. Most of the cell-wall materials are located in
the second layer, which consists of a helically wound framework of microfibrils (Figure ) [].
Reducing the size of cellulosic fibers to its constituent nanofibrils and nanowhiskers may elevate
the overall cost, but may prove to be beneficial, performance wise. However this area is not so well-
explored as far as water treatment application is concerned.

Figure : Microstructure of wood fiber cell wall: S, S and S are the inner, middle and outer layers of the secondary wall, respectively,
Adapted by permission from Macmillan Publishers Ltd. ref: []

.. Processing Techniques

... Mechanical fibrillation

The fibrillation of pulp fiber to obtain nano-scale unit structures, called microfibrillated cellulose, is
obtained through mechanical treatment of pulp fibers – a combination of refining and high pressure
homogenizing processes. The refining process is common in the paper industry, and is done by an
equipment called refiner. In a disk refiner, the dilute fiber suspension is forcefully allowed to go
through a gap between the rotor and stator disks, which have surfaces fitted with bars and grooves,
against which the fibers are subjected to repeated cyclic stresses. This mechanical treatment brings
about irreversible changes in the fibers, increasing their bonding potential by modification of their
morphology and size. In the homogenization process, dilute slurries of cellulose fibers previously
treated by refining, are pumped at high pressure and fed through a spring high pressure loaded valve
assembly. As this valve opens and closes in rapid succession, the fibers are subjected to a large pres-
sure drop with shearing and impact forces. The combination of these forces promotes a high degree
of microfibrillation in the cellulose fibers, resulting in microfibrillated cellulose [].
The refining process is carried out prior to homogenization due to the fact that refining produces
external fibrillation of fibers by gradually peeling off the external cell wall layers (P and S layers) and
exposing the S layer and also causes internal fibrillation that loosens the fiber wall, preparing the pulp
fibers for subsequent homogenization treatment [].


 DEY

... Electrospinning
Electrospinning has been recognized as an efficient technique for the fabrication of polymer nano-
fibers. Various polymers have been successfully electrospun into ultrafine fibers e.g. cellulose acetate.
There are basically three components in the instrument: a high voltage supplier, a capillary tube with a
pipette or needle of small diameter, and a metal collecting screen. In the electrospinning process a
high voltage is used to create an electrically charged jet of polymer solution or melt out of the pipette.
Before reaching the collecting screen, the solution jet evaporates, and is collected as an interconnected
web of small fibers [, ]. One electrode is placed into the spinning solution/melt, the other is at-
tached to the collector. The electric field is applied to the end of the capillary tube that contains the
solution fluid held by its surface tension. This induces a charge on the surface of the liquid. The po-
tential difference depends on the properties of the spinning solution, such as polymer molecular
weight and viscosity. When the distance between the spinneret and the collecting device is short, spun
fibers tend to stick to the collecting device as well as to each other, due to incomplete solvent evapo-
ration. Mutual charge repulsion and contraction of surface charges to the counter electrode cause a
force directly opposite to the surface tension []. As the intensity of the electric field is increased, the
hemispherical surface of the fluid at the tip of the capillary tube elongates to form a conical shape,
known as the Taylor cone []. By further increasing the electric field, a critical value is attained when
the repulsive electrostatic force overcomes the surface tension and the charged jet of the fluid is eject-
ed from the tip of the ‘Taylor Cone’ []. The discharged polymer solution jet undergoes an instability
and elongation process, which allows the jet to become very long and thin. Meanwhile, the solvent
evaporates, leaving behind a charged polymer fiber (Figure ). In case of polymer melts, the same
process is performed, but under vacuum.

Figure : Schematic diagram to show polymer nanofibers by electrospinning, ref: []


... Swelled precursor
Microcrystalline cellulose can be swelled and partly separated to whiskers by chemical and ultra soni-
cation treatments. Dimethyl acetamide with . wt% LiCl solution can be used as a swelling agent.
The microcrystalline cellulose (wt%) in LiCl/dimethyl acetamide is agitated using a magnetic stirrer
for  hours at °C, which causes the microcrystalline cellulose particles to swell. The slightly swelled
particles are then sonicated in an ultrasonic bath for  hours over a period of  days with long inter-
vals between each sonication treatment, to separate the cellulose nano whiskers [].


NANOTECHNOLOGY FOR WATER PURIFICATION

... Acid hydrolysis

Suspensions of nanocrystalline cellulose can be prepared by acid hydrolysis. Hydrolysis is carried out
in sulfuric acid with continuous stirring, immediately followed by -fold dilution with deionized wa-
ter to quench the reaction. The suspension is then centrifuged at  rpm for  min to concentrate
the cellulose and to remove excess aqueous acid. The resultant precipitate should be rinsed, re-
centrifuged, and dialyzed against water for  days until constant neutral pH [].

... By bacteria
Nanofibrils of cellulose can be synthesized by some bacteria. For example, the cellulose produced by
static cultivation of Acetobacter xylinum, sub species BPR, in a fructose/CSL medium at °C
[]. The bacteria is grown in a  ml Erlenmeyer flasks containing  ml of media. In order to re-
move the bacteria and to exchange remaining media, the produced cellulose pellicles are boiled in  M
NaOH at °C for  hour followed by repetitive boiling in deionized water. To prevent drying and to
avoid contamination, the washed cellulose is stored in diluted ethanol, in a refrigerator. The advantage
of using bacterial cellulose lies in its high purity, fine fibrils (high surface area) [], high tensile
strength and water-holding capacity.

. State-of-Art Applications

.. Magnetite

Iron oxides have been reported to have a high affinity for the adsorption of arsenic and arsenate, due
to the ability to form inner-sphere bidentate binuclear complexes with iron oxides [, ]. Magnetite
nanoparticles have highest saturation magnetization of  emu/g among iron oxides. Therefore, mag-
netite nanoparticles can be used to adsorb arsenic ions followed by magnetic decantation.

Maghemite

Tuutijӓrvi et al. [] investigated the suitability of maghemite nanoparticles for As(V) adsorption and
compared the As(V) adsorption efficiency of three different types of maghemite nanoparticle: com-
mercially available (CM), home-made by mechanochemical method (MM) and home-made by sol-gel
process (SM). They observed that the best As(V) adsorption capacity for all adsorbents were achieved
at pH  and decreased gradually with increasing pH. MM possessed the highest As(V) adsorption ca-
pacity of these three maghemites, followed by SM and CM particles, which can be attributed to spe-
cific surface Areas, as MM particles has the highest surface area and CM particles the lowest. The ad-
sorption maximum values based on surface area for all three maghemites were very similar, almost
identical, which indicated that maximum As(V) adsorption capacity is related to surface area. This
observation differs from the previous study [], in which it was indicated that As(V) adsorption ca-
pacity is influenced by nanoparticle size rather than surface area.
Hu et al. [] studied the removal and recovery of Cr(VI) from wastewater using maghemite na-
noparticles. They noted that the adsorption reached equilibrium within  min and was independent
of initial Cr concentration. The maximum adsorption occurred at pH .. The adsorption data were
analyzed and fitted well by Freundlich isotherm. Competition from common coexisting ions were
negligible, which illustrated the selective adsorption of Cr(VI) from wastewater. Regeneration studies
verified that the maghemite nanoparticles, which underwent six successive adsorption-desorption
process, still retained the original metal removal capacity.


 DEY
The adsorption data from Predescu et al. [] indicated that good adsorption capacity for metal
ion removal can be achieved by maghemite nanoparticles coupled with cationic resin (γ-FeO-R-H)
and a higher adsorption tendency for hexavalent chromium in comparison with the other metal ions
were observed.

.. nZVI

nZVI has become one of the most common adsorbents for the rapid removal of As(III) and As(V) in
the subsurface environment [-]. Its application to remove varieties of pollutants has been
demonstrated with halogenated hydrocarbons such as TCE, PCE [-], carbon tetrachloride
[], anions (e.g., NO-, CrO-), heavy metals (e.g. Ni+,Hg+), radio-nuclides [], As(III) [] and
organic compounds such as benzoic acid [].
Kanel et al. [] has characterized pristine nZVI as well as As(V)-treated nZVI materials. Their
HR-TEM study confirmed the core-shell structure of pristine nZVI (~ nm diameter core attributed
to Feo and ~nm thick shell due to iron oxide). Kanel et al. [] has also carried out batch studies
on the removal of As (Figure ) with groundwater samples obtained from Bangladesh and West Ben-
gal, India. The total iron concentration of groundwater from both sources were  mg/L. As(V) was
spiked in both samples to obtain an initial As(V) concentration of  mg/L and its effect was studied
using nZVI. As(V) was removed, with an efficiency reaching %, by ., ., and . g/L nZVI for
the spiked DI water, Bangladesh groundwater and West Bengal groundwater samples, respectively.
They proved that As (V) can be removed by adsorption/precipitation mechanism on nZVI (at neutral
pH) in a relatively short time of only several minutes.

Figure : Sorption of As(V) on nZVI for spiked DI

water with As (V):  mg/L in .MNaCl, Bangladesh
groundwater and West Bengal (India) groundwater sam-
ples; in all cases NZVI: . g/L, pH ,  °C. Reprinted
with permission from ref: [], copyright , American
Chemical Society

.. Mixed oxides

Oxidation of magnetite results in reduction of saturation magnetization. However replacement of

Fe+ in FeO by a small amount of Co+ or Ni+ can improve the oxidation resistance of the com-
pound []. Oxidation resistance is an important factor for arsenic removal under atmospheric con-
ditions. Hai et al. has studied arsenic adsorption ability of Fe-xCoxFeO (Co-ferrites) and Fe-
yNiyFeO (Niferrites) (x, y = , ., ., ., .) nanoparticles []. For Co replacement, adsorption


NANOTECHNOLOGY FOR WATER PURIFICATION
did not change significantly when x < . and reduced slightly when x > .. However, the presence
of Ni improved the adsorption to some extent (Figure ).

Figure : Saturation magnetization of the Co- and Ni-ferrite as a function of Co+ (x)
and Ni+ (y) content, ref: []

.. Organic-inorganic nanohybrids

Literature shows [] magnetic hydroxyapatite, made of hydroxyapatite (HAP) and FeO nanoparti-
cles, can adsorb % phenol in wastewater. Hydroxyapatite is chemically similar to bones and hard
tissues in mammals and is one of the few materials classified as bioactive for biomaterial application.
This unique magnetic sorbent has the capacity to regenerate after reaching adsorption saturation using
ethanol as eluant and external magnetic field as separation unit. The efficiency of adsorption is re-
duced only by % over a six time use period. Besides, Tural [] reported that glucamine modified
magnetic sorbent has the ability to separate and preconcentrate boron.
Carbon nanotubes (CNT) due to their unique structures and exceptional properties have been the
focus of research and development since their discovery in  []. Their thermal and chemical
stabilities have made them attractive adsorbents for organic and inorganic contaminants in water
[,]. Peng et al. [] prepared iron oxide - CNT magnetic composites and used for removal of
Pb(II) and Cu(II) from water. They showed that after adsorption, the adsorbent could be separated
from the medium by a simple magnetic process with >% recovery. While CNT itself is a poor ad-
sorbent for As(V), Poinern et al. [] has reported that CNT-ferrihydrite composite can serve as an
excellent adsorbent in arsenic removal. Their FESEM images show that the composite is made of
regular CNT tubes of length <  um and outer diameter - nm, which is coated with agglomerated
ferrihydrite particles in a bundle-like structure (Figure ) and that surface area of the composite is al-
most double as that of CNT.


 DEY

Figure : FESEM images of (a) CNT and (b) CNT-ferrihydrite composite, ref: []

.. Cellulosic biomass

Mahesh et al. [] has studied the use of cost effective and ecofriendly plant cellulose based adsor-
bents, as an alternative and effective substitute of activated carbon, for the removal of toxic dyes from
waste water, namely the use of sugarcane baggase, pretreated with formaldehyde (referred as raw bag-
gase) and sulphuric acid (referred as chemically activated baggase), for the removal of crystal violet
dye from simulated waste water; the raw baggase showing better adsorption efficiency.
Asandei et al. [] have shown the effectiveness of lead (II) ion removal from aqueous solution
of chitosan being mainly dependent on the initial concentration of lead (II) ions to chitosan dosage
ratio, which should be optimized by a compromise between the removal yield and cost-effectiveness
of the process. Their adsorption data well-fitted Langmuir isotherm.
Igwe et al. [] reported enhanced adsorption capacity of EDTA modified maize husk due to the
chelating effect of ethylenediamine tetra acetic acid (EDTA) while studying Cd(II), Pb(II) and Zn(II)
ion removal form wastewater.
The nonwoven webs of fibers produced from the electrospinning process have high specific sur-
face areas, nano scale pore sizes, high and controllable porosity and extreme flexibility with regard to
the materials used and modification of the surface chemistry of the fibers. Desai et al. reported the
electrospinning difficulties of chitosan/poly (acrylamide) and overcame these by varying the parame-
ters such as polymer concentration and temperature and thereby uniform beadless nanofibers were
achieved. In another work, potential use of electrospun chitosan/poly(ethylene oxide) (PEO) nano-
fibers for heavy metal ion binding, antimicrobial as well as physical separations were clearly examined.
They also proved that filtration efficiency was strongly related to the size of the electrospun fibers and
percentage of the chitosan present on the surface. They have utilized these membranes in binding
hexavalent chromium ions [-].
A company named Elmarco s.r.o. has successfully implemented the nozzle-less electrospining
technology (Nanospider TM) to produce cellulosic nanofibers for air filtration and other applications.

. Nanotechnological Principal applied to Water Purification

The principle behind application of magnetic nanoparticles and cellulosic nanofibers for heavy metal
removal from water, is the theory of adsorption, which can be found in several textbooks of physical


NANOTECHNOLOGY FOR WATER PURIFICATION
chemistry. Basics of surface adsorption has been concisely elucidated in author’s recent study guide
[], which includes different types of adsorption, thermodynamics of adsorption, various adsorp-
tion isotherms – relevant equations, main assumptions and limitations, followed by some applications
and numerical problems.
The profile of adsorption is generally depicted in the form of an adsorption isotherm. The rela-
tionship between extent of adsorption and pressure (if gas) or concentration (if liquid) of adsorbate is
known as adsorption isotherm. Following are the expressions for commonly referred adsorption iso-
therms []:

(Freundlich’s adsorption isotherm) … ()

where q (mg/g) is the saturated absorption, Ce (mg/L) is the equilibrium concentration, Kf is the
equilibrium adsorption coefficients (L/g), and /n is the empirical constant.

(Langmuir adsorption isotherm) … ()

where A and B are adsorption coefficients and A = B. qmax

(BET adsorption isotherm) … ()

where x is the pressure divided by the vapor pressure (p/p) of the adsorbate gas at a given tempera-
ture, v is the STP volume of adsorbed gas, vmon is the STP volume of adsorbated gas required to form
a monolayer and c is the BET constant, which is expressed by:

The adsorption isotherm indicates how the adsorption molecules distribute between the adsorb-
ate and adsorbent when the adsorption process reaches an equilibrium state. The analysis of the iso-
therm data by fitting them to different isotherm models is an important step to find the suitable mod-
el that can be used for design purpose. The two most well-known adsorption isotherms are Langmuir
and Freundlich. Langmuir isotherm assumes monolayer adsorption onto a surface containing a finite
number of adsorption sites of uniform strategies of adsorption with no transmigration of adsorbate in
the plane of surface. Whereas the Freundlich isotherm model assumes heterogeneous surface energies,
in which the energy term in the Langmuir equation varies as a function of the surface coverage.
The regeneration of used sorbent materials is a prime factor to decide the overall cost effective-
ness, which makes desorption studies equally important.

.. Nano effect

The greater reactivity of nanoparticles is associated with increased surface/volume ratio. This princi-
ple is exploited in use of nanomaterials for water treatment or soil remediation.


 DEY
Yean et al. [] has observed that iron spinel oxide i.e magnetite exhibits enhanced adsorption of
arsenic, when the size of the magnetite decreases from  to  and  nm. But when the quantity of
adsorbed arsenic is normalised per surface unit of solid instead of mass unit, magnetite particles of
 and  nm adsorb similar amounts of As (i.e. ~ mmol/m = . As/nm), suggesting similar
adsorption mechanism and reactivity. But, for particles smaller than nm, the adsorption capacity
increases and magnetite of nm adsorbs  times more As per nm (~ mmol/m= . As/nm)
than does  nm-diameter magnetite.

Figure : Scheme representing the crystalline structure of maghemite (γ-FeO) and the atomic arrangement of the iron tetrahedral (FeTd) and
octahedral (FeOh) at the () surface plane as a function of the size of the particles. Reprinted with permission from ref: [], copyright ,
American Chemical Society

This holds true for maghemite nanoparticles as well. The enhanced adsorption per surface unit
suggests a different adsorption mechanism and reveals a particular ‘nano-effect’, the origin of which
can be attributed to two factors []. The first is related to a size-dependent structural modification
of the surface of particles and the decrease in occupancy of the tetrahedral site that leaves vacant,
highly reactive adsorption sites (Figure ). In particular, there is a site localized in a six-membered iron
octahedral ring, which is the only site for which an oxidation of As(III) occurs. The second factor
relates to the thermodynamics of decrease in surface free energy. The high density of As adsorbed
upon saturation of the surface can be related to crystal growth.

.. Effect of pH

The pH of solution determines the concentration distribution of ionic forms of the pollutant. Dixit
and Hering [] showed that arsenite adsorption on iron oxide was independent of pH from  to .


NANOTECHNOLOGY FOR WATER PURIFICATION
However, Raven et al. [] reported that arsenate adsorption on iron oxide decreased with increasing
pH. Fendorf et al. [] showed that arsenic and chromium could form a monodentate complex, a
bidentateebinuclear complex, and a bidentateemononuclear complex with iron oxides, depending on
pH.
Chowdhury et al. [] has investigated the adsorption of arsenic and chromium by - nm
mixed magnetite-maghemite nanoparticles as a function of pH with a contact time of  hour at room
temperature. The maximum arsenic adsorption occurred at pH  with values of . mg/g for arse-
nic(III) and . mg/g for arsenic(V) when the initial concentration was kept at . mg/L for both
arsenic species, while chromium(VI) concentration was . mg/g at pH  with an initial chromium(VI)
concentration of  mg/L. They showed electrostatic attraction is the key mechanism of arsenic and
chromium removal by iron oxide nanoparticles from aqueous solutions. They fitted their experimental
results to Freundlich’s adsorption isotherm and found that the parameters ‘n’ and ‘Kf’ were almost .
and . for arsenic(III),  and . for arsenic(V) and for chromium(VI), . and .. The higher value
for ‘Kf’ indicated higher affinity for arsenic and chromium and the values of ‘n’ lying between  and
 indicated favorable adsorption [].
The effect of pH, in the range -, on adsorption of As(V) by CNT and CNT-ferrihydrite com-
posite was investigated by Poinern et al. []. Rest of the parameters were kept constant i.e.
[As(V)]initial  mg/L, adsorbent concentration g/L, temperature oC and, contact time .h. Figure
 shows that pH variation had little effect on the As(V) adsorption by CNT and adsorption being
quite low (.%), which can be attributed to the lack of suitable functional groups on CNT. However
arsenic adsorption on CNT-ferrihydrite composite was found to be highly pH dependent. As seen in
Figure , the As (V) adsorption is high at pH  and decreases rapidly to pH  after which increase in
pH does not affect the adsorption much. This again indicates that acidic condition offers sufficient
protonated sites on the surface of adsorbent, towards which the negatively charged oxyanions of ar-
senic are attracted.
Kanel et al. [] showed the effect of pH (-) on the adsorption of As (V) ( mgL-) on nZVI
(. g/L) for contact time of  hours. The % sorption of As(T) (total arsenic) at pH - was de-
creased to .% at pH  and to .% at pH  with As(V) addition. This showed that nZVI is ef-
fective in acidic and neutral pH, which is similar to the trends of pH dependency as mentioned earlier
and can be attributed to the ionization of both adsorbates and adsorbents. As(V) has pK, pK, and
pK values of ., ., and ., respectively. In the pH range of -, HAsO- and HAsO- are the
predominant species of As [], and are presumably the major species being adsorbed. As the iso-
electric point (IEP) of nZVI is pH . [], the nZVI surface exhibits a net positive charge at pH
lower than pH . and the adsorption of trace anionic As (V) species is enhanced by electrostatic in-
teraction. However, as the . and  mg/L concentrations of As(V) were added onto . g/L nZVI,
the IEP of nZVI was decreased from . to . and ., respectively, indicating that As(V) might have
become attached on the nZVI corrosion product surface by forming inner sphere complexes
[,,]. They reported that the As(V) adsorption kinetics were rapid and occurred within
minutes following a pseudo-first-order rate expression with observed reaction rate constants (kobs) of
.-. min- at various nZVI concentrations.
Literature shows [] that maximum adsorption capacity of As(V) on  nm magnetite decreases
with increasing pH (Figure ). The surface of magnetite is positively charged at pH values below .
due to the ‘pzc’ (point of zero charge) of magnetite. Thus, stronger As(V) adsorption to magnetite
sample is observed at lower pH values. The reduction in As(V) adsorption at high pH (i.e. pH ) is
attributable to the electrostatic repulsion of the negatively charged As(V) species by the negatively
charged surface sites. Both As(III) and As(V) were found to have great affinities for  nm magnetite
particles, as indicated by their large qmax values.


 DEY

Figure : Effect of initial pH on As(V) adsorption on CNT and CNT-ferrihydrite composite, ref: []

.. Effect of competing species

Arsenic sorption to mineral surface decreases in the presence of competitive anions such as phos-
phate, sulfate, nitrate, and molybdate [-]. Yean et al. [] observed that sorption of both As(III)
and As(V) to nm magnetite nanoparticles in the Lake Houston water sample was lower than that
of in . M NaNO electrolyte. They inferred that Natural Organic Matter (NOM) present in the
Lake Houston sample probably competed with As(III) and As(V) species for the surface sites (hy-
droxyl groups) on magnetite, by a combined mechanism of ligand exchange reactions, hydrogen
bonding and electrostatic interactions, and thus decreased the sorption of As(III) or As(V) to magnet-
ite surface.

Figure : As(V) adsorption on  nm magnetite at pH ., ., and ., Reprinted from Ref: [], with
permission from Cambridge University Press.

As industrial effluent or wastewater is a combination of several ions (e.g. some places in India
having excess fluoride ion in ground water, which can exist together with chromium pollutant from


NANOTECHNOLOGY FOR WATER PURIFICATION
chromo plating or tanning industries), Namdeo et al. [] decided to investigate the effect of sodium
fluoride concentration (in the range  to  mg L-) on Cr (VI) adsorption where the initial concentra-
tion of Cr (VI) was  mg L-. They observed a drastic decrease in uptake of Cr (VI) due to the pres-
ence of F− ions in the adsorption system. This may be explained on the basis of the faster diffusion of
F− ions towards the protonated positively charged surface of magnetite. As the size of F− ions is
much smaller than bulky CrO- ions, its diffusion through the pores of sorbent particles is more rap-
id. As a result the uptake of Cr (VI) decreases sharply with increasing concentrations of F− ions in the
system.
Researchers have found that presence of phosphate can significantly lower the percentage remov-
al of arsenic and chromium by iron oxide particles [, ]. Yet leaching of phosphate fertilizers
from the soil to the water stream does occur in Bangladesh and other countries. Total arsenic or
chromium(VI) and phosphate surface complexation on iron oxide may be one of the causes for the
reduced removal of arsenic or chromium(VI) in the presence of high PO-. At the same time, no sig-
nificant changes were observed in the case of elevated concentrations of Cl-, SO -, and NO- in solu-
tion. Tejedor-Tejedor and Anderson [] reported that phosphate produces three different types of
complexes on iron oxide (i.e. goethite, hematite, etc) surface: protonated [(FeO)(OH)PO], nonproto-
nated bridging bidentate [(FeO)PO] and a nonprotonated monodentate [(FeO)PO]. Stumm and
Morgan [] have reported that the extent of complex formation between Fe(III) and phosphates
(PO- or HPO-) depends on the ratio [OH-]/[PO-] or [HPO-], which in turn makes the phenom-
enon pH dependent.
Similarly for adsorption on nZVI, Kanel et al. found [] that HCO-, SO-, and NO- anions did
not have a negative effect on the As(V) uptake up to a  mM concentration, whereas HSiO and
PO- ( mM) reduced the uptake from .% to ., and .%, respectively, due to competing
ion effect.

.. Effect of temperature

Namdeo et al. [] demonstrated that the process of Cr(VI) uptake by magnetite is pore diffusion
controlled, by plotting log(% sorption) vs. log (time) which yielded straight lines. Owing to the porous
nature of the sorbent, intra-particle diffusion (pore diffusion) is expected. According to Knocke and
Hemphill [], diffusion is an endothermic process and the rate of adsorption will increase with in-
crease in solution temperature when pore diffusion is the rate limiting step. This indeed happened to
be the case for Cr(VI) uptake by magnetite, although for higher temperature the sorption becomes
more dependent on intra-particle diffusion, which would be the rate limiting step.

.. Effect of pollutant (adsorbate) concentration

Researchers have investigated the adsorption isotherm at oC for phenol at low concentrations
(mg/L) and high concentrations (mg/L) on . g FeO/HAP adsorbent []. Their re-
sults show (Figure ) that the isotherms were very well described by the Freundlich equation (eqn. )
at low concentration ( mg/L) for phenol, whereas for high concentrations of phenol (
mg/L), Langmuir equation (eqn. ) fits the best.


 DEY

Figure : Experimental isotherm and fitted models for phenol adsorption on FeO/HAP at high adsorbate concentrations (
mg/L) (A), at low adsorbate concentrations ( mg/L) (B), ref: []. Reprinted with permission from John Wiley & Sons Inc.

Yean et al. [] has studied arsenic adsorption on nm magnetite nanoparticles. The adsorption
isotherm (Figure ) appeared to consist of two Langmuir type isotherms for both As(III) and As(V).
At low concentrations i.e. < μmol/L As(III) or < μmol/L As(V), the adsorption isotherms can
be fitted with simple Langmuir isotherms where maximum adsorption capacities approach 
μmol/g for As(III) and . μmol/g for As(V). This is equivalent to . and . μmol/m for
As(III) and As(V) or . and and . molecules per nm of As(III) and As(V), respectively. But at
high As concentrations i.e. > μmol/L As(III) or > μmol/L As(V), the adsorption data deviates
from the simple Langmuir isotherm and appears to follow a second Langmuir isotherm. For the sec-
ond isotherm, the sorption energy (the ‘B’ term in eqn. ) was assumed to be the same as that of the
first sorption energy to evaluate the second maximum adsorption capacity. The second maximum
adsorption capacities (qmax) for As(III) and As(V) were substantially higher, which may be a result of
the coprecipitation of As with Fe on the surface [].


NANOTECHNOLOGY FOR WATER PURIFICATION

Figure : Plot of the adsorption of (a) As(III) and (b) As(V), both on  nm magnetite at pH ., Reprinted from Ref: [], with
permission from Cambridge University Press.

. Effect of contact time

The contact time between the pollutant and the adsorbent is of significant importance in the
wastewater treatment by adsorption. A rapid uptake of pollutants and establishment of equilibrium in
a short period signifies the efficacy of that adsorbent for its use in wastewater treatment. Figure 
shows the adsorption curves of phenol to iron oxide-hydroxyapatite sorbent at different temperatures
[]. The saturated adsorption was % at oC, raised to % at oC and % at oC. So, high
temperature was in favor of increasing the adsorption capacity, but the impact was not quite large.
Considering the cost, room temperature (oC) will be suitable for this kind of system.


 DEY

Figure : Adsorption curves of phenol with time at different temperatures, ref: []. Reprinted
with permission from John Wiley & Sons Inc.

Chowdhury et al. [] has studied the kinetics of arsenic(V) and arsenic(III) adsorption by vary-
ing the contact time between magnetite-maghemite and metal solution from  to  min using .
g/L adsorbent at initial metal concentrations of  and  mg/L. They showed that a rapid metal uptake
within  min of contact time was followed by a subsequent slower uptake. However the contact time
depends also on pH of solution and initial concentration of pollutant.

.. Effect of adsorbent dose

Moodley et al. [] observed that the removal of nickel by sawdust (cellulosic biomass) adsorbent
decreases with decreasing pH and increasing adsorbent dose. The decrease in adsorption capacity with
increase in dose could be due to the aggregation or overlapping of adsorption sites caused by over-
crowding of the sawdust particles. Such aggregation would lead to decrease in total surface area of the
sorbent and an increase in diffusional path length []. This trend is comparable to that reported by
Ayyappan et al. [] for the removal of Pb(II) from aqueous solution using a low-cost adsorbent.
Moodley et al. [] also found that the adsorption data confirms more to Freundlich model, which
assumes heterogeneity in the surface binding process. This observation is consistent with the hetero-
geneous nature of pine sawdust, which consists of different active sites such as ―OH and ―COOH
groups as revealed by FTIR spectra.
Poinern et al. [] reported the effect of CNT-Ferrihydrite composite dose (in the range  g/L to
 g/L) on As(V) adsorption at pH . The initial As(V) concentration was kept constant at  mg/L
and the contact time used was . h. Figure  shows that the As (V) adsorption increased from %
to  % when the adsorbent dose increased from  g/L to  g/L. The adsorption capacity decreased
from . mg/g to . mg/g. The increase in % As(V) adsorption with the increase in adsorbent dose
may be due to increased available adsorption sites and the decrease in adsorption capacity is mainly
due to unsaturation of adsorption sites through adsorption process.


NANOTECHNOLOGY FOR WATER PURIFICATION

Figure : Effect of CNT-Ferrihydrite adsorbent dose on As(V) adsorption [conditions:  K, pH ., contact time . h,
initial As(V) mg/L], ref: []

. Limitations

There is huge hope that nanotechnological applications and products will lead to a cleaner and health-
ier environment. However, there is also a wide debate about the safety of nanoparticles and their po-
tential impact on environment and biota. Nanotechnology and the environment – is therefore a Ja-
nus-faced relationship.
The use of nanoparticles in environmental applications will inevitably lead to the release of nano-
particles into the environment. Assessing their risks in the environment requires an understanding of
their mobility, bioavailability, toxicity and persistence.
Adsorption methods are considered to be quite effective, because of their treatment stability, ease
of operation and compact facility. Magnetic nanoparticles offer advantage over other adsorbent na-
nomaterials, because they can easily be separated from water using a magnetic field. But in general,
separation, waste formation and poor adsorption capacity remains a challenge in many adsorption-
based nanotechnological water treatment.
Cellulosic nanofibers obtained form wastes can be applied to water treatment, but electrospinning
of natural polymer faces many obstacles due to their high molecular weight, high viscosity at low con-
centration and poor solubility in organic solvents. Many of the naturally occurring polymers cannot be
electrospun due to their limited or poor solubility in organic solvents. However, they can be electro-
spun using toxic solvents such as trifluroaceticacid, which is harmful to the environment. Moreover,
very little is known about the long-term environmental fate of surface engineered nanoabsorbents.

. Conclusion
This chapter was intended to give an overview on major heavy metal pollutants in water, natural oc-
currence and synthetic methods of magnetic nanoparticles and cellulosic nanofibers, the basic princi-
ples and state-of-art applications of how these can be applied to water treatment, various factors af-


 DEY
fecting the process and their limitations. This nanotechnology-based water treatment holds a lot of
promise, but one should bear in mind that the opportunities and risks needs to be carefully weighed
before effective implementation.

References

 World Health Organization (WHO) Meeting the MDG drinking water and sanitation target: A mid–term
assessment of progress. Geneva, .
 J.T. Mayo et al. Sci. Technol. Adv. Mater.  () .
 J. Hu, I.M.C. Lo, G. Chen, Water Sci. Technol.  (), (), .
 B. Petrusevski, J. Boere, S.M. Shahidullah, S.K. Sharma, J.C. Schippers, J. Water SRT-Aqua  () .
 W. Driehaus, M. Jekel, J. Water SRT-Aqua  () .
 World Health Organization (WHO) Guidelines for DrinkingWater Quality. . Recommendations, World
Health Organization, Geneva, .
 N.N. Greenwood, A. Earnshaw, “Chemistry of Elements”, Pergamon Press, Oxford,  (Chapter ).
 http://www.epa.gov/safewater/arsenic
 M. Bissen, F.H. Frimmel, Acta Hydroch. Hydrob.  () .
 F.W. Pontius, K.G. Brown, C.J. Chen, J. Am. Water Works Assoc.  () .
 C.K. Jain, I. Ali, Water Res.  () .
 M.D. Kiping, in: J. Lenihan, W.W. Fletcher (Eds.), “Arsenic, the Chemical Environment, Environment and
Man”, vol. , Glasgow, , pp. -.
 WHO (World Health Organisation), Environmental Health Criteria, : Arsenic, World Health Organisa-
tion, Geneva, .
 B.K. Mandal, K.T. Suzuki, Talanta  () .
 L.C.D. Anderson, K.W. Bruland, Environ. Sci. Technol.  () .
 W.P. Tseng, H.M. Chu, S.W. How, J M. Fong, C.S. Lin, S. Yeh, J. Nat. Cancer Inst.  () .
 B. Wielinga, M. M. Mizuba, C. M. Hansel, S. E .Fendorf. Environ. Sci. Techno.  () .
 E.M. Nkhalambayausi-Chirwa, Yi-Tin Wang. Reactor Performance  () .
 D. Bagchi, S. J. Stohs, B. W. Downs, M. Bagchi, H. G. Preuss. Toxicology  () .
 S. J. Stohs, D. Bagchi, E. Hassoun, M. Bagchi J. Environ. Pathol. Toxicol. Oncol.  () .
 R. J. Harrison, R. E. Dunin-Borkowski, A. Putnis, Proceedings of the National Academy of Sciences  ()
.
 X.-Q. Li, D. W. Elliott, W.-X. Zhang, “Zero-Valent Iron Nanoparticles for Abatement of Environmental
Pollutants: Materials and Engineering Aspects” Critical Reviews in Solid State and Materials Sciences  ()
.
 T. Sugimoto, “Fine Particles: Synthesis, Characterization and Mechanism of Growth,” New York: Marcel
Dekker, .
 V. K. LaMer, R. H. Dinegar, J. Am. Chem. Soc.  () .
 C. J. J. Den Ouden, R. W. Thompson, J. Coll. Interf. Sci.  () .
 M. Ocana, R. Rodriguez-Clemente, C. J. Serna, Adv. Mater.  () .
 M. P. Morales, T. González-Carreño, C. J. Serna, J. Mater. Reson.  () .
 T. Sugimoto, E. Matijevic, J. Coll. Interf. Sci.  () .
 R. Massart, V. Cabuil, J. Chem. Phys.  () .
 J. P. Jolivet, “Metal Oxide Chemistry and Synthesis: From Solutions to Solid State,” New York: Wiley, .
 M. P. Pileni, J. Phys. Chem.  () .
 I. Lisiecki, M. P. Pileni, J. Am. Chem. Soc.  () .
 K. Zhang, C. H. Chew, G. Q. Xu, J. Wang, L. M. Gan, Langmuir () .
 A. J. Zarur, J. Y. Ying, Nature  () .
 P. Tartaj, L. C. De Jonghe, J. Mater. Chem.  () .
 P. Tartaj, J. Tartaj, Chem. Mater.  () .
 M. P. Pileni, Nat. Mater.  () .


NANOTECHNOLOGY FOR WATER PURIFICATION
 N. Feltin, M. P. Pileni, Langmuir  () .
 M. A. López-Quintela, J. Rivas, J. Coll. Interf. Sci.  () .
 E. E. Carpenter, J. Magn. Magn. Mater.  () .
 M. Boutonnet, J. Kizling, P. Stenius, Coll. Surf. A,  () .
 Y. Lee, J. Lee, C. J. Bae, J.-G. Park, H-J. Noh, J.-H. Park, T. Hyeon, Adv. Funct. Mater.  () .
 O. Döker, E. Bayraktar, Ü. Mehmetoglu, A. Calimli, Rev. Adv.Mater. Sci.  () .
 E. Matijevic (Ed.), “Fine Particles,” MRS Bull.  ()  (special issue).
 G. Viau, F. Ravel, O. Acher, F. Fievet-Vicent, F. Fievet, J. Appl. Phys.  () .
 F. Fievet, J. P. Lagier, B. Blin, B. Beaudoin, M. Figlarz, Solid State Ionics / () .
 G. Viau, F. Fievet-Vicent, F. Fievet, J. Mater. Chem.  () .
 G. Viau, F. Fievet-Vicent, F. Fievet, Solid State Ionics  () .
 F. Fievet, J. P. Lagier, B. Blin, B. Beaudoin, M. Figlarz, Solid State Ionics / () .
 T. Dey, ‘Polymer-coated Magnetic Nanoparticles: Surface Modification and End-functionalization’ J. Na-
nosci. Nanotechnol.  () .
 T. Dey, C. J. O’Connor, NSTI Nanotech.  () .
 J. Rockenberger, E. C. Scher, A. P. Alivisatos, J. Am. Chem. Soc.  () .
 T. Hyeon, S. S. Lee, J. Park, Y. Chung, H. B. Na, J. Am. Chem. Soc.  () .
 S. Shen, H. Zeng, J. Am. Chem. Soc.  () .
 K. Deshpande, A. Mukasyan, A. Varma, Chem. Mater.  () .
 X. Cao, Y. Koltypin, G. Katabi, R. Prozorov, I. Felner, A. Gedanken, J. Mater. Res.  () .
 K. V. P. M. Shafi, A. Ulman, X. Yan, N.-L. Yang, C. Estournés, H. White, M. Rafailovich, Langmuir 
() .
 K. V. P. M. Shafi, A. Ulman, A. Dyal, X. Yan, N.-L. Yang, C. Estournés, L. Fournes, A. Wattiaux, H. White,
M. Rafailovich, Chem. Mater.  () .
 K. S. Suslick, Science  () .
 C. Pascal, J. L. Pascal, F. Favier, M. L. Elidrissi-Moubtassim, C. Payen, Chem. Mater.  () .
 G. L. Messing, S. Zhang, G. V. Jayanthi, J. Am. Ceram. Soc.  () .
 Y. Lu, H. Fan, A. Stump, T. L. Ward, T. Rhieker, C. J. Brinker, Nature  () .
 B. Xia, I. W. Lenggoro, K. Okuyama, Adv. Mater.  () .
 C. Pecharroman, T. González-Carreño, J. E. Iglesias, Phys. Chem. Min.  () .
 T. González-Carreño, M. P. Morales, M. Garcia, C. J. Serna, Mater. Lett.  () .
 T. González-Carreño, A. Mifsud, C. J. Serna, J. M. Palacios, Mater. Chem. Phys.  () .
 W. R. Cannon, S. C. Danforth, J. H. Flint, J. S. Haggerty, R. A. Marra, J. Am. Ceram. Soc.  () .
 M. Cauchetier, O. Croix, N. Herlin, M. Luce, J. Am. Ceram. Soc.  () .
 S. Veintemillas-Verdaguer, M. P. Morales, C. J. Serna, Mater. Lett.  () .
 S. Veintemillas-Verdaguer, M. P. Morales, C. J. Serna, Appl. Organomet. Chem.  () .
 U. Kumar, Sci. Res. Essay  () .
 W. Wafwoyo, C. W. Seo, W. E. Marshall, J. Chem. Technol. Biotechnol.  () .
 S. S. Azhar, A. G. Liew, D. Suhardy, K. F. Hafiz, M. D. I. Hatim, Am. J. Appl. Sci.  () .
 U. Kumar, M. Bandyopadhyay, Bioresour. Technol.  () .
 M. E. Argun, S. Dursun, C. Ozdemir, M. Karatas, J. Hazard. Mater.  () .
 D. H. Page, F. El-Hosseiny, K. Winkler, Nature,  () .
 F. W. Herrick, R. L. Casebier, J. K. Hamilton, K. R. Sandberg, Journal of Applied Polymer Science,  () .
 L. Wagberg, “Wood material science”. Finnish-Swedish research programme, –, Year Book -
().
 J. M. Deitzel, J. D. Kleinmeyer, J. K. Hirvonen, N. C. Beck Tan, Polymer,  () –.
 H. Fong, D. H. Reneker, Electrospinning and formation of nano-fibers. in “Structure formation in poly-
meric fibers” (eds.: Salem D. R., Sussman M. V.) Hanser, Munich, - ().
 X. Fang, D. H. Reneker, Journal of Macromolecular Science,  () .
 G. I. Taylor, Proceedings of the Royal Society: Series A,  () .
 S. Kamel, eXPRESS Polymer Letters () () .
 K. Oksman, A. P. Mathew, D. Bondeson, I. Kvien, Composites Science and Technology,  () .
 E. D. Cranston, D. G. Gray, Biomacromolecules,  () .


 DEY
 S. Albert, G. S. Mittal, Food Research International,  () .
 R. Ek, C. Gustafsson, A. Nutt, T. Iversen, C. Nyström, Journal of Molecular Recognition,  () .
 S. Fendorf, M.J. Eick, P. Grossl, Environ. Sci. Technol.  () .
 A. Manceau, Geochim. Cosmochim .Acta  () .
 T. Tuutijärvi, J. Lu, M. Sillanpää, G. Chen, Journal of Hazardous Materials  () .
 S. Yean, L. Cong, C. T. Yavuz, J. T. Mayo, W. W. Yu, A. T. Kan, V. L. Colvin, M. B. Tomson, ‘Effect of
magnetite particle size on adsorption and desorption of arsenite and arsenate’ J. Matter. Res  () .
 J. Hu, G. Chen, I. M. Lo, Water. Res.  () .
 A. Predescu, A. Nicolae, U.P.B. Sci. Bull. B,  () .
 B. A. Manning, M. Hunt, C. Amrhein, J. A. Yarmoff, Environ. Sci. Technol.  () .
 C. Su, R. W. Puls, Environ. Sci. Technol.  () .
 C. Su, R. W. Puls, Environ. Sci. Technol.  () .
 J. Farrell, J. Wang, P. O’Day, M. Coklin, Environ. Sci. Technol.  () .
 O. X. Leupin, S. J. Hug, Water Res.  () .
 J. A. Lackovic, N. P. Nikolaids, G. M. Dobbs, Environ. Eng. Sci.  () .
 S. Bang, M. D. Johnson, G. P. Korfiatis, X. Meng, Water Res.  () .
 C. B. Wang, W. Zhang, Environ. Sci. Technol.  (a) .
 H.-L. Lien, W. Zhang, J. Environ. Eng.  () .
 B. Schrick, J. L. Blough, A. D. Jones, T. E. Mallouk, Chem. Mater.  () .
 G. V. Lowry, K. M. Johnson, Environ. Sci. Technol.  () .
 J. T. Nurmi, P. G. Tratnyek, V. Sarathy, D. R. Baer, J. E. Amonette, K. Pecher, C. Wang, J. C. Linehan, D.
W. Matson, R. L. Penn, M. D. Driessen, Environ. Sci. Technol.  () .
 W. X. Zhang, J. Nanopart. Res.  () .
 S. R. Kanel, B. Manning, L. Charlet, H. Choi, Environ. Sci. Technol.  () .
 S. H. Joo, A. J. Feitz, D. L. Sedlak, T. D. Waite, Environ. Sci. Tech.  () .
 S.J. Kanel, J-M. Greneche, H. Choi, Environ. Sci. Technol.  () .
 N.H. Hai, N.D. Phu, N. Chau, H.D. Chinh, L.H. Hoang, D.L. Leslie-Pelecky, J. Korean Phys. Soc.  ()
.
 N.H. Hai, N.D. Phu, VNU Journal of Science, Mathematics - Physics  () .
 X. Wang, Clean: Soil, Air, Water  () . DOI: ./clen.
 B. Tural, Clean: Soil, Air, Water  () .
 S. Iijima, Nature,  () .
 Y. H. Li, S. Wang, J. Wei, X. Zhang, C. Xu, Z. Luan, D. Wu, B. Wei, Chemical Physics Letters,  () .
 X. Peng, Y. H. Li, Z. Luan, Z. Di, H. Wang, B. Tian, Z. Jia, Chemical Physics Letters,  () .
 X. Peng, Z. Luan, Z. Di, Z. Zhang, C. Zhu, Carbon,  () .
 G.E.J. Poinern, D. Parsonage, T.B. Issa, M.K. Ghosh, E. Paling, P. Singh, International Journal of Engineering,
Science and Technology  () .
 S. Mahesh, G. Vijay Kumar, P. Agrawal, Journal of Environmental Biology  () .
 D. Asandei, L. Bulgariu, E. Bobu, Cellulose Chem. Technol.,  () .
 J. C. Igwe, A. A. Abia, Electronic Journal of Biotechnology,  () .
 K. Desai, K. Kit, Polymer  () .
 K. Desai, K. Kit, J. J. Li, P. M. Davidson, S. Zivanovic, H. Meyer, Polymer  () .
 K. Desai, K. Kit, J. Li, S. Zivanovic, Biomacromolecules  () .
 T. Dey, “The Basics of Surface Adsorption” () Brain Mass Inc: Ontario,
http://brainmass.com/ebooks/BasicsSurfaceAdsorption
 T. Dey, “Synthesis and Biomedical Application of Polymer-coated Magnetic Nanoparticles” in Encyclopedia
of Nanoscience and Nanotechnology, - volume set, H. S. Nalwa (Ed.),  () .
 M. Auffan, J. Rose, O. Proux, D. Borschneck, A. Masion, P. Chaurand, J-L. Hazemann, C. Chaneac, J-P.
Jolivet, M.R. Wiesner, A. Van Geen, J-Y. Bottero, Langmuir  () .
 S. Dixit, J. G. Hering, Environ. Sci. Technol.  () .
 K. P. Raven, A. Jain, R. H. Loeppert, Environ. Sci. Technol.  () .
 S. Fendorf, M. J. Eick, P. Grossl, Environ. Sci. Technol.  () .
 S.R. Chowdhury, E.K. Yanful, Journal of Environmental Management  () .


NANOTECHNOLOGY FOR WATER PURIFICATION
 G. McKay, H. S. Blair, J. R. Garden, J. Appl. Polym. Sci.  () .
 B. A. Manning, M. Hunt, C. Amrhein, J. A. Yarmoff, Environ. Sci. Technol.  () .
 E. O. Kartinen, C. J. Martin, Desalination  () .
 J. Goldberg, J. Colloid Interface Sci.  () .
 A. Jain, R.H. Loeppert, J. Environ. Qual.  () .
 B.P. Jackson, W.P. Miller, Soil Sci. Soc. Am. J.  () .
 C.M. Su, R.W. Puls, Environ. Sci. Technol.  () .
 M. Namdeo, S.K. Bajpai, Agricultural and Food Chemistry  () .
 W. Zhang, P. Singh, E. Paling, S. Delides, Minerals Eng.  () .
 B. A. Manning, S. Goldberg, Soil Sci. Soc. Am. J.  () .
 M. I. Tejedor-Tejedor, M. A. Anderson, Langmuir  () .
 W. Stumm, J. Morgan, “Aquatic Chemistry: an Introduction Emphasizing Chemical Equilibria in Natural
Waters”, vol. . () Wiley, New York, pp. -.
 W. R. Knocke, L. H. Hemphill, Water Res.  () .
 K. Moodley, R. Singh, E.T. Musapatika, M.S. Onyango, A. Ochieng, Water SA.  () .
 A. Shukla, Y. Zhang, P. Dubey, J. L. Margrave, S. S. Shukla, J. Hazard. Mater. B () .
 R. Ayyappan, A. C. Sophia, K. Swaminathan, S. Sandhya, Process Biochem.  () .

