Nanoparticle Aggregation: Challenges to Understanding Transport

TECHNICAL REPORTS

1909

Unique forms of manufactured nanomaterials, nanoparticles, and their suspensions are rapidly being created by manipulating properties such as shape, size, structure, and chemical composition and through incorporation of surface coatings. Although these properties make nanomaterial development interesting for new applications, they also challenge the ability of colloid science to understand nanoparticle aggregation in the environment and the subsequent eff ects on nanomaterial transport and reactivity. Th is review briefl y covers aggregation theory focusing on Derjaguin-Landau-Verwey-Overbeak (DLVO)-based models most commonly used to describe the thermodynamic interactions between two particles in a suspension. A discussion of the challenges to DLVO posed by the properties of nanomaterials follows, along with examples from the literature. Examples from the literature highlighting the importance of aggregation eff ects on transport and reactivity and risk of nanoparticles in the environment are discussed.

Nanoparticle Aggregation: Challenges to Understanding Transport and Reactivity

in the Environment

Ernest M. Hotze, Tanapon Phenrat, and Gregory V. Lowry* Carnegie Mellon University

As of the submission date of this review, approximately

4700 articles on nanoparticle (NP)-related topics have been

published in the calendar year 2009 alone (Th ompson Reuters,

2009). Many of these papers discuss novel particles, and applica-

tions of nanomaterials along with their use in consumer products

are expanding rapidly (Project on Emerging Nanotechnologies,

2009). For example, nanomaterials are used for environmental

clean-up and biomedical applications (Alivisatos, 2001; Liu et al.,

2005). Th is expansion in applications, and the incorporation of

“nano” ingredients into products that may be released during the

life cycle of those products, will increase the occurrence of manu-

factured nanomaterials in the environment. For example, the

leaching of silver NPs used as antimicrobial agents in cloth was

recently reported (Benn and Westerhoff , 2008). Unintentionally

or intentionally, manufactured NPs might come into contact with

biological receptors. For this reason, the risk of nanomaterials to

humans and to the environment garners attention from the scien-

tifi c community, governmental agencies, and public stakeholders.

Nanoparticless in engineered or environmental systems could

be thought of as dispersions of primary particles. However, pri-

mary NPs tend to aggregate into clusters up to several microns

in size (Brant et al., 2007; Chen and Elimelech, 2007; Jiang et

al., 2009; Limbach et al., 2005; Phenrat et al., 2008). Th erefore,

NP aggregation plays an important role in determining reactivity,

toxicity, fate, transport, and risk in the environment. Th e critical

role of aggregation in environmental implications had occasion-

ally been ignored (Baveye and Laba, 2008), but many recent stud-

ies have begun applying colloid science principles, based around

Derjaguin-Landau-Verwey-Overbeak (DLVO) theory, to under-

stand NP aggregation under various conditions. Manufactured

NPs challenge the limits of colloid science, however, due to their

small size, variable shape, structure, composition, and poten-

tial presence of adsorbed or grafted organic macromolecules.

Laboratory aggregation studies, which are covered in detail in this

review, have demonstrated that these challenges, acting simulta-

neously, will ultimately aff ect how NPs behave in the environ-

ment. Additionally, particles released to the environment undergo

Abbreviations: CT, carbon tetrachloride; Df, fractal dimension; DLVO, Derjaguin-Landau-

Verwey-Overbeak; EDL, electrostatic double layer; MWCNT, multi-walled carbon

nanotubes; NOM, natural organic matter; NP, nanoparticle; NZVI, nano zero valent iron;

PZC, point of zero charge; ROS, reactive oxygen species; Velas

, elastic-steric potential; VOSM

,

osmotic potential; XDLVO, extended Derjaguin-Landau-Verwey-Overbeak.

Center for Environmental Implications of NanoTechnology (CEINT) and Deps. of Civil

& Environmental Engineering, Carnegie Mellon Univ., Pittsburgh, PA 15213-3890.

Assigned to Associate Editor Dionysios Dionysiou.

Copyright © 2010 by the American Society of Agronomy, Crop Science

Society of America, and Soil Science Society of America. All rights

reserved. No part of this periodical may be reproduced or transmitted

in any form or by any means, electronic or mechanical, including pho-

tocopying, recording, or any information storage and retrieval system,

without permission in writing from the publisher.

J. Environ. Qual. 39:1909–1924 (2010)

doi:10.2134/jeq2009.0462

Published online 20 May 2010.

Received 18 Nov. 2009.

*Corresponding author ([email protected]).

© ASA, CSSA, SSSA

5585 Guilford Rd., Madison, WI 53711 USA

SPECIAL SUBMISSIONS

1910 Journal of Environmental Quality • Volume 39 • November–December 2010

chemical and physical transformations and encounter a multi-

tude of solution conditions important for aggregation, includ-

ing solution pH, dissolved ions, naturally occurring organic

matter, clays, and biocolloids.

Th is review focuses on three questions: (i) Can basic colloid

science predict the behavior of nanomaterials in the environ-

ment? We provide a brief overview of colloid science. Several

excellent resources are available for additional information

on colloid science (Evans and Wennerstrom, 1999; Hiemenz

and Rajagopalan, 1997; Buffl e et al., 1998; Elimelech and

O’Melia, 1990; O’Melia, 1980). (ii) Where are challenges to

existing theory posed by manufactured nanomaterials? We

discuss the intrinsic NP properties that raise these challenges.

(iii) How will these challenges aff ect studies of fate and trans-

port, reactivity, and risk to the environment? We describe

several case studies examining environmental implications of

nanomaterial aggregation.

Colloid Science and AggregationColloid science is defi ned as a branch of chemistry dealing with

colloids, heterogeneous systems consisting of a mechanical

mixture of dispersion (colloids) and dispersant. When particles

are dispersed, sizes range between 1 nm and 1 μm in a continu-

ous medium. Above this size range, particles begin to sediment

out of suspension. Several theories to explain aggregation of

submicrometer spherical particles have been developed and

experimentally studied for decades, the most common being

DLVO (Derjaguin and Landau, 1941; Verwey et al., 1948)

(an alternative approach is statistical thermodynamics; see

Hermansson [1999] and Koponen [2009]). By defi nition, the

size range of colloidal particles overlaps the size range of manu-

factured nanomaterials (i.e., <100 nm). Th erefore, theories in

colloid science should be applicable to manufactured nanoma-

terials, but, due to the novelty of nanomaterial properties, sev-

eral studies have found that this is not always the case (He et

al., 2008; Kozan et al., 2008; Phenrat et al., 2008). However,

colloid science is fundamental to understanding and develop-

ing theories for nanomaterials systems. Th is section, therefore,

provides a very brief review of the types of aggregation and

DLVO theory.

Types of Aggregation and Aggregate StructureAggregation of colloidal particles occurs when physical pro-

cesses bring particle surfaces in contact with each other and

short-range thermodynamic interactions allow for particle–

particle attachment to occur. For particles <100 nm in size,

Brownian diff usion controls the long-range forces between

individual nanoparticles, causing collisions between particles.

When contact occurs, it can result in attachment or repul-

sion. Short-range thermodynamic interactions that control

this can be understood in the context of DLVO theory. Th ere

are two types of aggregation relevant to manufactured NPs in

the environment: homoaggregation and heteroaggregation.

Homoaggregation refers to aggregation of two similar par-

ticles (i.e., NP–NP attachment). Th is is observed in homo-

geneous suspensions of particles that are typically studied in

the laboratory to make useful correlations with DLVO theory.

Heteroaggregation refers to aggregation of dissimilar particles

(e.g., NP–clay particle attachment). Real environmental sys-

tems contain natural particles (e.g., clay particles) in numbers

far greater than the number of manufactured NPs, so heteroag-

gregation is an important fate process for manufactured NPs.

Th e probability that two particles attach is given by a “stick-

ing coeffi cient,” also known as the attachment effi ciency (α)

(i.e., when α = 1, sticking occurs 100% of the time; when α =

0.5, sticking occurs 50% of the time). Particles attaching at fi rst

contact (α = 1) form large dendritic aggregates, whereas par-

ticles sticking only after several collisions (α < 1) form denser

and less dendritic aggregates (Buffl e and Leppard, 1995). Th e

concept of fractal dimension is used to describe aggregate struc-

ture. Fractal dimension (Df) can range from 1 (a line) to 3 (a

sphere). Th ese are limits that come from Euclidean dimension,

where 1 is a straight line (e.g., x1), 2 is a fl at surface (e.g., x2),

and 3 is a volume (e.g., x3). Fractal dimensions are those that

lie between Euclidean dimensions. Ideal fractal structures are

self-similar throughout. Homoaggregation under controlled

laboratory conditions produces aggregates of reasonably pre-

dictable fractal dimension (Barbot et al., 2010; Chakraborti

et al., 2003; Hansen et al., 1999), whereas heteroaggregation

typically forms natural fractals (statistically self-similar over a

limited range of length scales), making aggregation state more

diffi cult to predict. Th e physical dimensions of the aggregates

formed can aff ect the reactive surface area, reactivity, bioavail-

ability, and toxicity. Aggregate structure must therefore be

determined and considered when interpreting fate, transport,

and toxicity data.

Derjaguin-Landau-Verwey-Overbeak and Extended

Derjaguin-Landau-Verwey-OverbeakAccording to the classical DLVO theory of aggregation, the

sum of attractive and repulsive forces determines attachment

(Derjaguin and Landau, 1941; Verwey et al., 1948). Th is

theory maintains that only two forces dominate interactions

between particles: van der Waals (vdW) attractive and electro-

static double layer (EDL) forces. Table 1 lists an expression for

the vdW attraction experienced by two spheres in suspension

(Stumm and Morgan, 1996). Expressions for vdW commonly

calculate the attraction between two spherical particles; how-

ever, due to unique NP shapes and compositions, expanded

theoretical approaches may be needed (Gatica et al., 2005).

An EDL has a charge that refl ects not only the surface that it

surrounds but also the solution chemistry of the surrounding

media. Ionic strength, which is a measure of the amount of

ions present, controls the extent of the radius of the diff use

layer from the surface. Low ionic strength means the EDL ion

cloud extends far out from the particle; high ionic strength

conditions compress the EDL. Table 1 lists an expression for

the EDL experienced by two charged spheres in suspension

(Stumm and Morgan, 1996).

Classical DLVO simplifi es thermodynamic surface inter-

actions and predicts the probability of two particles stick-

ing together by simply summing van der Waals and electric

double-layer potentials to determine if forces are net attractive

(−VT) or net repulsive (+V

T). For example, in Fig. 1, van der

Waals and electric double-layer potentials are plotted as a func-

tion of separation distance between the particles. Summing

Hotze et al.: Nanoparticle Aggregation: Environmental Transport and Reactivity 1911

these curves (VT) demonstrates that particles can have a net

attraction in a primary or secondary minimum (well). Particles

in the primary well are considered to be irreversibly aggre-

gated, whereas particles in the secondary well are reversibly

aggregated (i.e., re-entrainment is possible if shear forces are

exerted) (Hahn et al., 2004). Figure 1 also shows something

Table 1. Derjaguin-Landau-Verwey-Overbeak and extended Derjaguin-Landau-Verwey-Overbeak forces in aggregation.

Force X/DLVO† Equations‡§ Origin References

van der Waals attraction

DLVO

( ) ( )( )( )

2 2VDW

2 2

42 2ln

6 4 2 2

s a sV A a a

kT kT s a s a s a s

⎛ ⎞+⎜ ⎟= − + +⎜ ⎟+ + +⎝ ⎠

interactions of electrons in particles

Stumm and Morgan (1996)

Electrostatic double layer

DLVO ( )( )

( )2 2

2EDL 64tanh exp

2 4

sdaV n kT ze

kT s a kT

−κ − δ+ δ ⎡ ⎤π Ψ⎛ ⎞= ⎜ ⎟⎢ ⎥κ + ⎝ ⎠⎣ ⎦1

22 2

rs 0å å

N

i ii

e z n

kT

⎡ ⎤⎢ ⎥⎢ ⎥κ =⎢ ⎥⎢ ⎥⎣ ⎦

∑

charged surfaces Delgado et al. (2007), Stumm and Morgan (1996)

Magnetic attraction

XDLVO 2 3o S

M 3

8

9 2

M aV

s

a

− πμ=⎛ ⎞+⎜ ⎟⎝ ⎠

aligning electron spins

Phenrat et al. (2007)

Hydrophobic (Lewis acid-base)

XDLVO

0

0Vexp

surface area

ABABs

s sG

−⎛ ⎞= Δ ⎜ ⎟λ⎝ ⎠

entropic penalty of separating hydrogen bonds in water

Hoek and Agarwal (2006), Wu et al. (1999)

Osmotic repulsion

XDLVOOSM2 0

Vd s

kT

⎧ ⎫≤ ⇒ =⎨ ⎬⎩ ⎭

22OSM

1

4 12

2 2p

V a sd s d d

kT v

⎧ ⎫π⎪ ⎪⎛ ⎞⎛ ⎞≤ < ⇒ = φ − χ −⎨ ⎬⎜ ⎟⎜ ⎟⎝ ⎠ ⎝ ⎠⎪ ⎪⎩ ⎭

2 2OSM

1

4 1 1ln

2 2 4p

V a s ss d d

kT v d d

⎧ ⎫⎛ ⎞π⎪ ⎪⎛ ⎞ ⎛ ⎞< ⇒ = φ − χ − −⎨ ⎬⎜ ⎟⎜ ⎟ ⎜ ⎟⎝ ⎠ ⎝ ⎠⎪ ⎪⎝ ⎠⎩ ⎭

concentration of ions between two particles (Fig. 4)

Fritz et al. (2002), Ortega-Vinuesa et al. (1996), Phenrat et al. (2008), Romero-Cano et al. (2001)

Elastic-steric repulsion

XDLVOelas 0

Vd s

kT

⎧ ⎫≤ ⇒ =⎨ ⎬⎩ ⎭

2

2elas32

ln2

36 ln 3 1

2

p pW

s dV a s sd s d

kT M d d

s d s

d

⎧ ⎧ ⎫⎡ ⎤−⎛ ⎞ ⎛ ⎞π⎪ ⎪ ⎪⎢ ⎥> ⇒ = φ ρ⎨ ⎨ ⎬⎜ ⎟ ⎜ ⎟⎢ ⎥⎝ ⎠⎝ ⎠⎪ ⎪ ⎪⎣ ⎦⎩ ⎭⎩

⎫−⎛ ⎞ ⎪⎛ ⎞− + + ⎬⎜ ⎟ ⎜ ⎟⎝ ⎠⎪⎝ ⎠ ⎭

molecules on particle surfaces resist loss of entropy due to compaction

(Fritz et al., 2002; Ortega-Vinuesa et al., 1996; Phenrat et al., 2008; Romero-Cano et al., 2001)

Bridging attraction

XDLVO NA surface molecules bridge to other particles

Chen and Elimelech (2007), Domingos et al. (2009)

† DLVO, Derjaguin-Landau-Verwey-Overbeak; EDLVO, extended Derjaguin-Landau-Verwey-Overbeak.

‡ A, Hamaker Constant; a, average particle radius; χ, Flory-Huggins solvency parameter; d, thickness of “brush” polymer layer (~δ); 0

ABsGΔ :free energy of

acid base interaction between particles at distance s0; δ, Stern layer thickness; ε, relative permittivity of water (=78.5 C V−1 m−1); ε

o, permittivity of a vacuum

(=8.854 × 10−12); e, elementary charge of an electron; κ, inverse Debye length; k, Boltzmann constant; λ, decay length for acid–base interactions (=0.6

nm); Ms, saturation magnetization; μ

o, vacuum permeability; M

W, molecular weight of polyelectrolyte; n, number concentration of ion pairs; ρ

p, density

of polymer; Ψd diff use potential (~zeta potential); s, distance between interacting surfaces; s

0, minimum equilibrium separation distance (=0.158 nm); T,

absolute temperature; φp, volume fraction of polymer in “brush”; v

1, volume of solvent molecule; z, electrolyte valence,

§ Refer to referenced publications for details of equation derivations and constraints.


that is commonly the case with manufactured NPs having an

organic coating: Classical DLVO forces alone are not suffi cient

to accurately predict aggregation behavior. Steric repulsion

forces (VT + V

ELAS) resulting from adsorbed polymer or poly-

electrolyte coatings or natural organic matter (NOM) makes it

so these particles may only have a net attraction in a second-

ary minimum. Coated NPs may therefore aggregate reversibly

(Graf et al., 2009), and aggregation reversibility has signifi cant

consequences regarding the fate, transport, bioavailability, and

eff ects of NPs in the environment. Th ese additional forces are

collectively known as extended DLVO (XDLVO).

In addition to steric repulsion forces (VT + V

ELAS) due to NP

coatings, other XDLVO forces have been invoked to match

experimental data for various types of aggregating primary

particles (Chen and Elimelech, 2007; Fritz et al., 2002; Hoek

and Agarwal, 2006; Ortega-Vinuesa et al., 1996; Phenrat et

al., 2007; Phenrat et al., 2008; Romero-Cano et al., 2001; Wu

et al., 1999). Th is means that additional short-range forces are

operating on an equivalent magnitude as vdW attraction and

EDL. Th ese forces can include bridging (Chen and Elimelech,

2007), osmotic (Fritz et al., 2002; Ortega-Vinuesa et al., 1996;

Phenrat et al., 2008; Romero-Cano et al., 2001), steric (Fritz

et al., 2002; Ortega-Vinuesa et al., 1996; Phenrat et al., 2008;

Romero-Cano et al., 2001), hydrophobic Lewis acid-base (Wu

et al., 1999) (Hoek and Agarwal, 2006), and magnetic forces

(Phenrat et al., 2007). Table 1 summarizes these forces. Th e rel-

evant force type depends on the system tested, and frequently

more than one XDLVO force is important. Moreover, forces

are not completely independent of one another. In the case of

nanomaterials, understanding aggregation using a DLVO- or

XDLVO-based theory presents many challenges.

Nanoparticle Challenges to Derjaguin-

Landau-Verwey-OverbeakColloid science demonstrates that several forces with diff er-

ent natures and origins govern the kinetics and extent of par-

ticle aggregation. Th is section elaborates how manufactured

NPs present a combination of challenges to two predominant

theories in colloid science, DLVO and XDLVO.

Among these challenges are shape, size, structure,

composition, and adsorbed or grafted organic

macromolecules (e.g., coatings or NOM) (Fig.

2). Although particles not classifi ed as “nano”

(i.e., <100 nm) also present some of these chal-

lenges, manufactured NPs are often designed

with the intention of amplifying these properties.

Examining some early studies demonstrating these

challenges can inform our discussion in the previ-

ous section on implications of NP aggregation on

their transport and reactivity in the environment.

Particle SizeTh e small size of most NPs confl icts with a funda-

mental assumption of DLVO. Equations describ-

ing EDL are mathematically incorrect if this

assumption does not hold (Delgado et al., 2007).

Th is is because when a particle reaches a small

enough size, its surface curvature is too substantial

to assume it is fl at (i.e., when the assumption of

κ*a >>1 no longer applies). Th erefore, this is one of the pri-

mary challenges when considering NP aggregation in a DLVO

framework. Aside from this, vdW and EDL forces (Table 1)

scale similarly with particle size (Evans and Wennerstrom,

1999; Hiemenz and Rajagopalan, 1997; Israelachvili, 1991),

so moving from microscale to nanoscale particles should theo-

retically aff ect a change on vdW and EDL forces to a similar

degree. However, as a particle decreases in size, a greater per-

centage of its atoms exist on the surface. Electronic structure,

surface charge behavior, and surface reactivity can be altered as

a result. Recent studies suggest that redistribution and chang-

ing coordination environment of charge and atoms at the sur-

faces of NPs strongly infl uences their reactivity (Hochella et al.,

2008). He et al. (2008) reported that as the particle size became

smaller, the particles became more susceptible to surface charge

titration. Th ey found an inverse correlation between particle

size and the measured point of zero charge (PZC); 12, 32, and

65 nm hematite particles had PZCs of 7.8, 8.2, and 8.8 pH

units, respectively (He et al., 2008). Th ese hematite NPs were

synthesized using the same method so they would be identi-

cal in every aspect except for size. However, for these particles,

at pH 7 the smaller particles are less charged than larger par-

ticles, leading to lower EDL repulsion forces and subsequently

greater aggregation tendency. Indeed—given the same ionic

strength—smaller hematites have faster aggregation rates than

larger ones (He et al., 2008). Th is suggests that size polydisper-

sity in manufactured NPs aff ects their aggregation behavior.

Polydispersion is the reality in large-scale applications. For

example, reactive nano-scale zerovalent iron (NZVI) particles

used for environmental remediation are highly polydisperse

(Liu et al., 2005; Phenrat et al., 2007; Phenrat et al., 2009a).

Unlike van der Waals attraction, long-range magnetic attrac-

tion between two NZVI particles increases with particle radius

to the sixth power (Table 1). Polydispersity and particle size

therefore greatly aff ected the aggregation of NZVI having par-

ticle sizes ranging from 15 to 260 nm, with aggregation being

proportional to the fraction of large particles in the dispersion

(Phenrat et al., 2009a).

Fig. 1. van der Waals force (dashed line), electrostatic double layer (EDL) force (gray line), total Derjaguin-Landau-Verwey-Overbeak (DLVO) forces, and elastic force plotted together to fi nd total potential as a function of separation distance. XDLVO, extended Derjaguin-Landau-Verwey-Overbeak. V/KT, potential energy divided by Boltzmann’s constant and absolute temperature.


Another example of a size eff ects is “halo” stabilization,

where nano sized materials can control electrostatic repulsion

between two approaching micrometer-sized particles, hinder-

ing their fl occulation by keeping them separated beyond the

range of DLVO attractive forces (Fig. 3). Tohver et al. (2001)

found that adding 6-nm zirconia NPs to a dispersion of unsta-

ble submicron silica particles (0.6 μm in diameter) stabilized

the mixture (Tohver et al., 2001). Th e same phenomenon was

observed for bimodal magnetorheological fl uids consisting of

micrometer sized magnetite particles (1.45 μm in diameter)

and magnetite NPs (8 nm diameter) (Viota et al., 2007). Size

and resultant polydispersity, therefore, present a large challenge

to understanding how NP dispersions behave because these

parameters can enhance or diminish aggregation.

Chemical CompositionColloid science captures the eff ect of chemical composition on

aggregation through the Hamaker constant, which governs van

der Waals attraction; saturation magnetization, which control

magnetic attraction; hydrophobicity, which controls hydro-

phobic interaction (Wu et al., 1999); and surface charge, which

governs electrostatic double layer (EDL) interaction (Table 1)

(Evans and Wennerstrom, 1999).

Particles with a high Hamaker constant have greater aggre-

gation tendency compared with particles with a low Hamaker

constant at the same solution and surface chemistry. Because

the origin of van der Waals attraction is permanent or induced

dipoles, a bulk material with electronic or molecular structure

that favors generation of permanent or induced dipoles nor-

mally has a large Hamaker constant value. For example, the

Hamaker constants of gold, silver, and polystyrene are 45.3,

39.8, and approximately 9.8 × 10−20 J, respectively (Hiemenz

and Rajagopalan, 1997). van der Waals attraction forces are

approximately fi ve times stronger for gold particles in compari-

son with polystyrene (see equation in Table 1). In addition to

controlling van der Waals attraction, some materials can origi-

nate forces that challenge the DLVO framework, including

magnetic and hydrophobic interactions that promote aggrega-

tion. For example, ferromagnetic and ferrimagnetic materials,

such as zerovalent iron and magnetite, can cause long-range

magnetic attraction without an applied magnetic fi eld (Table

1) (Elena et al., 2009; Phenrat et al., 2007). Th is magnetic

force contributes to abnormally rapid aggregation. Moreover,

carbon-based nanomaterials, such as C60

or CNTs, have hydro-

phobic surfaces in aqueous environments. Interaction of all-

carbon surfaces with water molecules involves a signifi cant

entropic penalty driving an aggregation process that might

Fig. 2. Derjaguin-Landau-Verwey-Overbeak (DLVO) frames particle aggregation by modeling two solid spheres approaching in suspension. The sum of van der Waals (VDW) force and electrostatic double layer (EDL) force determines if the particles attach or repulse. Nanoparticles present challenges to DLVO frameworks because of shape, structure, composition, size, and macromolecule considerations. Not drawn to scale.


be understood through the use of XDLVO frameworks that

model hydrophobic Lewis acid–base interactions (Wu et al.,

1999). Th ese types of XDLVO frameworks have yet to be

developed for carbon-based nanomaterials.

Unique NP composition and morphology (e.g., core-shell

NPs) and environmental transformations of NPs will chal-

lenge DLVO explanations of aggregation. Particles with a high

surface potential have lower aggregation tendencies compared

with particles with a low surface potential given the same

solution chemistry. Nanoparticle chemical composition alters

surface potential by dissociation or ion adsorption by surface

atoms (Evans and Wennerstrom, 1999; Israelachvili, 1991).

Surface charge of uncoated NPs is therefore governed in part

by the type of atoms at the surface of the particles. However,

the core of core-shell particles, which are common, can infl u-

ence the charge of the surface atoms. For example, uncoated

Fe0/Fe3O

4 core-shell NPs have a zeta potential near −32 mV (in

1 mmol L−1 NaHCO3 at pH 7.5), but as the Fe0 core of the par-

ticles oxidizes to magnetite, the surface charge becomes more

like that of magnetite (Phenrat et al., 2007). Gold NPs (syn-

thesized via citrate reduction) have a charge of around −55 mV

(Kim et al., 2009), presumably due to citrate acid adsorbed on

the surface. Multiwall carbon nanotubes (MWCNTs) treated

with a mixture of sulfuric acid and nitric acid (3:1) together

with ultrasonication have zeta potential around −55 mV (Kim

and Sigmund, 2004) due to carboxylation of the surface of the

MWCNTs (Smith et al., 2009). Brant et al., Cheng et al., and

Chen et al. have shown that stable fullerene clusters with nega-

tive charge of approximately −40 to −80 mV (from pH 2 to

12) were formed from mixing fullerene clusters in water for a

period of several weeks (Brant et al., 2005; Chen and Elimelech,

2009; Cheng et al., 2004). Th e nature of these charges is under

investigation. Th ese examples demonstrate that core-shell

structures, the presence of coatings, and chemical transforma-

tions impart charges to NP surfaces. Th ese charge sources, or

combinations thereof, are not accounted for in colloid science.

Understanding the fate of NPs in the environment therefore

requires a fundamental comprehension of the types of trans-

formations that the NPs may undergo during manufacturing,

under commercial use, and after release into the environment

(as discussed later in this review).

Complex Crystal StructuresComplex NP crystal structures can present

challenges to colloid science because changes

in the crystal structure can alter Hamaker con-

stants and surface charges in ways not explain-

able by theory. Additionally, defects in the

crystal lattice can alter surface charge (Evans

and Wennerstrom, 1999; Israelachvili, 1991).

Th e crystal structure of materials with the same

atomic composition can play a large role in

determining surface charge. Nano-sized TiO2,

for example, has three diff erent polymorphs:

rutile, anatase, and brookite, each with a unique

crystal structure. Th e zeta potential is approxi-

mately −20 mV (French et al., 2009) when

anatase and brookite structures are predomi-

nant and is approximately −35 mV (Lebrette et

al., 2004) when the rutile phase is predominant

(both measured at pH 7.5). Th is is consistent with fi ndings

that the dielectric constant can be altered by annealing pure

anatase-phase TiO2 to the rutile phase (Kim et al., 2005). Th e

Hamaker constant is a function of dielectric constant (Tabor

and Winterton, 1969); therefore, the Hamaker constant also

can depend on the crystal structure of NPs. Th is is important

to consider for NP systems containing mixtures of phases or

where crystal structures are altered for applications, such as

photocatalysis with TiO2.

ShapeIn DLVO modeling, one of the primary assumptions is that

particles are spherical. Th is assumption is reasonable for ideal

latex particles and some ideal colloidal systems. Manufactured

NPs, however, come in a variety of nonspherical shapes.

Triangles, icosahedrons, ellipses, rhombohedrons, spindle,

rod, and tube shapes are possible (AgNPs [Moon et al., 2009],

Fe2O

3NPs [Wang et al., 2009], and Ag

2S [Ma et al., 2007]),

thereby complicating traditional DLVO and XDLVO.

Both vdW and EDL forces are aff ected by changes in

shape. Several researchers have investigated these changes

(Bhattacharjee and Elimelech, 1997; Montgomery et al., 2000;

Vold, 1954). At a separation distance smaller than the mean

diameter (∼v1/3, where v is the volume of particles), the attrac-

tion between anisometric particles, such as plates, rectangu-

lar rods, and cylinders, is larger than for spherical particles of

equal volume because a greater number of atoms are in close

proximity (Vold, 1954). Th e attraction between spheres, rods

(and cylinders), and platelets varies as s−1, s−2, and s−3, respec-

tively, where s is separation distance (Vold, 1954). For cylindri-

cal particles, parallel orientation is energetically favored over a

perpendicular orientation at small distances. For rectangular

rods or platelets, an orientation with the largest faces oppo-

site each other is the most favorable energetically. Electrostatic

double-layer forces of cylindrical particles are theoretically a

function of particle orientation (Vold, 1954). Th ese results

imply that shape can theoretically control aggregation under

some favored orientations. For example, Kozan et al. (2008)

studied the aggregation of tungsten trioxide (WO3) nanowire

suspensions of diff erent types (uneven, single, or bundled in

Fig. 3. Without nanoparticles present, microscale particles aggregate because net forces are attractive. Nanoparticles (NPs) create a “halo” eff ect that increases electrostatic double layer repulsive forces and causes net forces between microscale particles to be unattractive.


diameter) and dimensions (diameter of 40–200 nm and nomi-

nal lengths of 2, 4, 6, and 10 μm) in various polar solvents

and compared the aggregate morphology of the nanowires with

that of spherical WO3 NPs (40 nm in diameter) (Kozan et al.,

2008). Th e aggregates of spherical WO3 NPs were more com-

pact and more spherical in shape (Df ~ 2.6 and R

g ~3.8 μm)

than aggregates of nanowires with diameter of 200 nm (Df ~

2.3 and Rg ~2.2 μm). Moreover, aggregation of the nanow-

ires for 7 d in a quiescent environment resulted in no signifi -

cant Rg increase but an increase in D

f (Kozan et al., 2008).

Th is reorganization is related to the theoretical prediction that

anisometric particles are likely to aggregate under a secondary

minimum (Vold, 1954). Th e degree of secondary minimum

aggregation is dependent on shape and is most favorable for

platelets, less favorable for rods and cylinders, and least favor-

able for spherical NPs (Vold, 1954). Loose aggregates formed

in the secondary minimum might gradually restructure into

denser aggregates in the primary minimum. Nanoparticless

designed with a myriad of unique shapes (Cai and Sandhage,

2005; Kawano and Imai, 2008; Sanchez-Iglesias et al., 2009)

therefore challenge colloid science to understand their aggrega-

tion behavior in the environment because of the model com-

plexity of understanding these shapes in a DLVO framework.

Although nonconventional theories, such as surface element

integration (Bhattacharjee and Elimelech, 1997), may account

for interfacial forces in irregular shapes (e.g., ellipsoids), these

theories are diffi cult to apply to the unique (e.g., helical) and

diverse types of nanomaterial shapes.

Surfactants and Macromolecular Surface CoatingsNanomaterials are commonly manufactured with surface

coatings (e.g., surfactants, polymers, and polyelectrolytes) to

enhance dispersion stability or to provide specifi c functionality

(He et al., 2007; Hezinger et al., 2008; Hydutsky et al., 2007;

Mayya et al., 2003; Phenrat et al., 2008; Saleh et al., 2008;

Saleh et al., 2005; Zhang et al., 2002). Table 2 summarizes

some examples of surface-coated NPs and their applications.

Adsorbed or covalently bound surfactants prevent aggregation

and enhance dispersion stability of NPs by increasing surface

charge and electrostatic repulsion or by reducing interfacial

energy between particle and solvent (Rosen, 2004). Coulombic

attractions (Rosen, 2004; Vaisman et al., 2006; Wang et al.,

2004), hydrophobic interactions (Vaisman et al., 2006), and

surfactant concentration (Vaisman et al., 2006) aff ect the

adsorbed surfactant mass and layer conformation and hence

the ability of a surfactant to stabilize a NP against aggrega-

tion. Table 3 summarizes the properties of surfactants that

aff ect their ability to stabilize NPs against aggregation. Low-

molecular-weight surfactants reversibly adsorb on particle sur-

faces (Rosen, 2004), so, without a strong interaction between

the surfactant and NP surface, the surfactant may desorb or be

displaced by higher-molecular-weight natural polymers such as

NOM when the particles are released into the environment.

Th erefore, changes in the stability of surfactant-coated NPs are

expected on release to the environment.

Polymers are organic macromolecules consisting of repeti-

tions of smaller monomer chemical units (Fleer et al., 1993).

Table 2. Examples of surface coated nanoparticles and their applications.

Particle type Surface modifi er Purpose References

Quantum dots amphiphilic alkyl-modifi ed polyacrylic acid

alter particle hydrophilicity and modify quantum dot interactions with other biological compounds

Luccardini et al. (2006)

Titanium dioxide polyacrylic acid stabilize dispersions and modify surface chemistry allowing attachment of an antibody for a substrate selective photocatalytic system

Kanehira et al. (2008)

Fe0/Fe-oxide poly(styrene sulfonate), polyaspartate, carboxymethyl cellulose, or triblock copolymers

inhibition of aggregation, decreasing adhesion to solid surfaces, increasing mobility in the subsurface, and targeting specifi c pollutants

He et al. (2009), Kanel et al. (2008), Phenrat et al. (2008), Phenrat et al. (2009c), Phenrat et al. (2009a), Saleh et al. (2005)

Carbon nanotubes various kinds of anionic, cationic, nonionic surfactants and polymers

individually stabilized, single-walled carbon nanotubes

Moore et al. (2003)

Silver daxad 19 (sodium salt of a high-molecular-weight naphthalene sulfonate formaldehyde condensate)

stabilize dispersions silver nanoparticles Sondi et al. (2003)

Table 3. Examples of surfactant properties and their relationship to particle stabilization.

Property Findings References

Surfactant chain length

A linear relationship exists between the surfactant chain length and the logarithm of the dispersion concentration (cdc), where cdc is defi ned as the lowest concentration of a surfactant necessary to disperse hydrophobic particles.

Dederichs et al. (2009)

Molecular weight Stabilization of CNT† with nonionic surfactants increased with molecular weight (e.g., Brij 78 [MW = 1198 g mole−1] stabilized 4.3% of CNT, while Brij 700 [MW = 4670 g mole−1] could stabilize 6.4% of CNT)

Moore et al. (2003)

Type of ionic head groups

At similar molecular weight, cationic surfactants such as dodecyltrimethylammonium bromide (MW = 308 g mole−1) and cetyltrimethylammonium bromide (MW = 364 g mole−1) were slightly more eff ective in stabilizing CNT than anionic sodium dodecylbenzenesulfonate (MW = 348 g mole−1).

Moore et al. (2003)

Self-assembly structure

Sodium dodecyl sulfate, anionic surfactant, forms cylindrical micelles as helices or double helices and hemicellar structures on CNT. Random adsorption of surfactants on CNT hypothesized to enhance stabilization of the CNT dispersions.

O’Connell et al. (2002), Richard et al. (2003), Yurekli et al. (2004)

† CNT, carbon nanotube.


Th ey can be synthetic (e.g., polyethylene glycol) or naturally

occurring (e.g., biomacromolecules, such as proteins and poly-

saccharides) (Brewer et al., 2005; Jain et al., 2005; Liu et al.,

2006). Th ey also are surface active agents but are generally of

higher molecular weight than surfactants discussed previously.

Unlike surfactants, the repeating monomer units can adsorb

to NPs at multiple places along the polymer chain, making

polymer adsorption stronger than surfactant adsorption and

relatively irreversible. For example, Kim et al. (2009) modi-

fi ed NZVI by physisorption of various types of strong and

weak polyelectrolytes and found that <30% by mass of poly-

mer desorbed from the NPs over a period of 4 mo (Kim et al.,

2009). Polyelectrolytes are charged polymers. Th eir charge is

derived from ionizable groups built into the polymer structure

(e.g., from -COO−, -SO4−, or -SO

3− groups) (Fleer et al., 1993).

Adsorbed polymers and polyelectrolytes provide electrosteric

repulsions that prevent NP aggregation. Th ese repulsive forces

have osmotic (VOSM

) and elastic-steric (Velas

) components due

to their charge and large molecular weight, respectively (Fig.

4). Th e magnitude of the electrosteric repulsion is governed

by the surface excess (adsorbed mass of polymers per unit sur-

face area of a NP), polymer molecular weight, polymer den-

sity, adsorbed layer thickness, and solution composition (see

expressions in Table 1 and examples in Tables 4 and 5) (Fritz

et al., 2002; Ortega-Vinuesa et al., 1996; Phenrat et al., 2008;

Romero-Cano et al., 2001).

Surfactant–NP and polymer–NP interactions are extremely

complex and often kinetically controlled rather than being

in thermodynamic equilibrium. Th e science describing or

predicting adsorbed mass and adsorbed layer conformation

is immature, and the adsorbed mass and layer conformation

depends on NP properties and solution conditions (pH, ionic

strength, ionic composition). Th erefore, predicting the eff ects

of adsorbed surfactants, polymers, or polyelectrolytes on NP

behavior in the environment is challenging.

Eff ect of Nanoparticle Aggregation on

Applications and Implications

for the EnvironmentUnderstanding NP aggregation under well controlled solution

conditions in the laboratory is diffi cult because NP properties

present challenges for colloid science. However, ultimately we

would like to comprehend the fate, transport, reactivity, and

risks associated with manufactured NPs released into the envi-

ronment. How do these intrinsic challenges in understanding

aggregation aff ect our understanding of NP fate, transport, and

eff ects in more complicated environmental settings? Figure 5

summarizes some of the environmental interactions of NPs

that aff ect or are aff ected by aggregation and are discussed next.

Eff ect of Solution Chemical Composition on

Aggregation: pH and Ionic SolutespH and dissolved ionic solutes play crucial roles on aggregation

of NPs in engineered and natural systems. Th ese parameters can

be controlled in laboratory settings but may vary spatially and

temporally in natural systems. Spatial heterogeneity of miner-

als in the subsurface alters the concentration of ionic species

present in diff erent systems (e.g., groundwater versus surface

water). Variation in aquatic environment pH stems from the

same phenomenon. A better understanding of how pH and dis-

solved ionic solutes aff ect the behavior of organic macromol-

ecule–coated NPs in the environment is needed. Surface charge

titration and EDL screening are the two primary ways that

pH and ionic solutes promote NP aggregation. Most NP sur-

faces have surface functional groups (e.g., hydroxide and oxide

groups) that are titratable by H+ or OH−. Excess of H+ (low

pH) normally results in a positively charged particle surface,

Fig. 4. Osmotic plus elastic repulsive forces (VOSM

+ Velas

) of polyelectro-lytes are dependent on surface layer thickness.

Table 4. Examples of polymer properties and their relationship to particle stabilization.

Polymer property Findings References

Adsorbed layer thickness and surface excess Correlate to dispersion stability of polymer modifi ed Fe0–Fe oxide nanoparticles.

Phenrat et al. (2008)

Molecular weight Dispersion stability of carbon nanotubes increased as the molecular weight of Pluronic increased.

Moore et al. (2003)

Condition of polymer adsorption: heat treatment Adsorption of polyacrylic acid on TiO2 enhanced at 150°C

versus 20°C. This resulted in the enhanced dispersion stability.Kanehira et al. (2008)

Condition of polymer adsorption: solvent property More polyacrylic acid adsorbed on TiO2 in dimethylformamide

than in water. Kanehira et al. (2008)


Table 5. Examples of particle stabilization and destabilization by natural organic matter and biomolecules.

System Findings References

TiO2 (positively charged)

at pH 2 to 6 with 1 mg L−1 FA†Disaggregation of TiO

2, resulting in a smaller particle size (3 nm) than the

original bare TiO2 obtained from the manufacturer (10 nm).

Domingos et al. (2009)

TiO2 (negatively charged)

pH > PZC with 1 mg L−1 FAFlocculation of TiO

2 due to bridging FA was observed. Aggregation decreased

with increasing ionic concentration. High ionic concentration screened the negative charges of FA and TiO

2, resulting in greater FA adsorption onto TiO

2

promoting stabilization, not bridging.

Domingos et al. (2009)

C60

or CNT with HA (CaCl2

concentration <40 mmol L−1)Humic acid as low as 1 mg L−1 increased dispersion stability of C

60 up to 5 mg L−1.

Humic acid concentrations increased the dispersion stability of CNTs from 5 to 300 mg L−1.

Chappell et al. (2009), Chen and Elimelech (2007)

C60

with HA (CaCl2 concentration

>40 mmol L−1)Unadsorbed HA aggregated through intermolecular bridging via calcium complexation. Aggregates subsequently bridged C

60, resulting in fl occulation.

Chen and Elimelech (2007)

AuNPs with BSA Gold nanoparticles coated with BSA were stable as colloids at the PZC of bare AuNPs, suggesting steric stabilization due to adsorbed BSA layers.

Brewer et al. (2005)

AuNPs with lysozymes Enhanced aggregation of lysozyme–AuNP assemblies at physiological pH at very low lysozyme concentrations (16 nmol L−1). Au nanoparticles were found embedded in the lysozyme matrix.

Zhang et al. (2009)

† AuNP, gold nanoparticle; BSA, bovine serum albumin; CNT, carbon nanotubes; FA, fulvic acid; HA, humic acid; PZC, point of zero charge.

Fig. 5. Hypothesized eff ects of the environment on nanoparticle aggregation and how aggregation state could alter nanoparticle (NP) environ-mental interactions. (A) pH and ions. Increasing acidity can lead to charge neutralization at the point of zero charge. Increasing ionic strength reduces the size of the electrostatic double layer and repulsive forces. Certain ions (e.g., Ca2+) can bridge functional groups on the surface of nanoparticles. (B) Heteroaggregation. Macromolecules present in the environment coat nanomaterials leading to a steric eff ect. Larger mac-romolecules trap NPs in a mesh or gel causing stabilization or destabilization. Nanoparticles also associate with other particles such as clays or biocolloids. Nanoparticles fl ow through porous media depending on their aggregation state. (C) Biological interactions. Cells activate phagocy-tosis mechanisms depending on if a particle is aggregated or not. Nanoparticles containing metals aggregate at the surface of organisms and release metal ions at varying rates. (D) Transformations. Oxidants or sunlight degrade particle coatings or the particles themselves. Aggregated photoactive NPs only have the outside particles active.


whereas excess of OH− (high pH) normally yields a negatively

charged surface. For each particular system, the pH at which

the H+ and OH− concentration causes suspended particles to

have a neutral charge is called the PZC. As pH moves toward

this point, the EDL repulsion decreases, and aggregation is pro-

moted by vdW attraction (Fig. 5A). Charge reversal (i.e., from

negatively charged to positively charged particles) is also possi-

ble. Th is change would dramatically aff ect NP behavior because

most surfaces in the environment are negatively charged at near

neutral pH. As for the ionic species eff ect, high ionic concen-

tration decreases the Debye length (κ−1) (see Table 1 for calcu-

lation of EDL using DLVO), resulting in the decrease of the

extent of EDL repulsion (Fig. 5A). Th e ionic concentration

where the repulsive energy barrier is completely screened and

rapid aggregation occurs is known as the critical coagulation

concentration. Table 6 summarizes factors governing nanoma-

terial critical coagulation concentration and dispersion sensitiv-

ity toward electrolytes. In many cases, surface titration and EDL

screening eff ects occur in NP dispersions as they occur for large

colloidal particles. However, there are notable exceptions where

the pHPZC

depends on the NP size. Table 7 summarizes param-

eters altering nanoparticle PZC.

Heteroaggregation and DepositionTh e environment is a complex system with multiple primary

molecules and particles as well as solution compositions.

Straightforward DLVO theory is diffi cult to apply to these sys-

tems, and typically XDLVO forces are brought into models to

improve results (Wu et al., 1999). At best, DLVO + XDLVO

provides qualitative information about aggregation and deposi-

tion, but this information is useful for understanding the infl u-

ence of various parameters on aggregation and deposition.

Nanoparticle Interaction with Naturally Occurring Organic Matter,

Biomolecules, and Biomacromolecules

As with synthetic surface coatings used on NPs (e.g., polyelec-

trolytes), naturally occurring organic macromolecules (e.g.,

NOM) in the environment can signifi cantly alter the aggre-

gation behavior of NPs. Even without fully defi ned structure,

due to its macromolecular nature, NOM is expected to prevent

aggregation and deposition presumably due to electrosteric sta-

bilization (Amirbahman and Olson, 1993; Amirbahman and

Olson, 1995). Natural organic matter consists of mainly fulvic

and humic substances. Fulvic compounds typically make up 40

to 80% of NOM compounds, and, due to their small size (∼1

kDa), they can coat the surfaces of particles. Fulvic compounds

alter the surface charge and have been shown to stabilize with

an eff ect similar to surfactants (Domingos et al., 2009). In fact,

NOM may stabilize MWCNTs better than surfactants com-

monly used for laboratory suspensions of MWCNTs (Hyung et

al., 2007). However, enhanced dispersion stability due to elec-

trosteric forces and destabilization due to bridging have been

experimentally observed, and it is not often possible to predict

these interactions. Natural organic matter attaches to the sur-

face of particles in a variety of ways. For example, irreversible

adsorption onto the surfaces of iron oxide via ligand exchange

between carboxyl/hydroxyl functional groups of humic acid

and iron oxide surfaces (Gu et al., 1994) or hydrophobic inter-

actions with carbon based nanomaterials (Hyung et al., 2007)

have been reported. Whether or not attachment results in stabi-

Table 6. Factors governing nanomaterial critical coagulation concentration and dispersion sensitivity toward electrolyte concentration (ionic strength).

Parameter Findings References

Type of ionic species

Regardless of particle type, CCC† for divalent ions (e.g., CaCl2 or MgCl

2)

is lower than for monovalent ions (e.g., NaCl). Chen and Elimelech (2007), Evans and

Wennerstrom (1999), Saleh et al. (2008)

Particle size Values of CCC increase with increasing particle size (e.g., CCC values of Fe2O

3 with

diameter of 12, 32, and 65 nm have CCCs of 45, 54, and 70 mmol L−1 NaCl, respectively).He et al. (2008)

Type of particles The CCC of carbon nanomaterials such as fullerene appears to be higher than that of metal oxide such as Fe

2O

3 [e.g., CCC of C

60 [d

p = 60 nm] is around 120 mmol L−1 NaCl,

whereas that of Fe2O

3 [d

p = 65 nm] is 70 mmol L−1 NaCl). This is presumably due to

stronger van der Waals attractions of Fe2O

3 in comparison to C

60, making the stability

of metal oxide nanoparticles more sensitive to screening eff ects.

Chen and Elimelech (2007), He et al. (2008)

Particle shape The CCC of carbon nanotubes is lower than C60

particles. Van der Waals attraction is greater and EDL repulsion is lower for rod-like shapes than for spherical shapes.

Chen and Elimelech (2007), Saleh et al. (2008)

Surface modifi er Due to steric repulsions polymeric surface modifi cation decreases the sensitivity of aggregation induced by ionic species. NZVI (d

p = 162 nm) modifi ed by guar-gum

prevented aggregation at ionic concentrations up to 0.5 mmol L−1. TiO2 modifi ed

by PAA prevented aggregation in ionic concentrations up to 0.5 mol L−1 NaCl.

Kanehira et al. (2008)

† CCC, critical coagulation concentration; EDL, electrostatic double layer; NZVI, nano zero valent iron; PAA, polyacrylic acid.

Table 7. Parameters altering nanoparticle point of zero charge.

Parameter Findings References

Chemical composition PZCs of metal NPs,† metal oxides, and carbon nanomaterials have a wide range and need to be measured for each material and preparation method.

Brant et al. (2005), Kim et al. (2009), Kosmulski (2009), Sirk et al. 2009

Surface modifi cation Surface modifi cation using anionic polyelectrolytes shifts PZC toward lower values (e.g., ~3 pH units for Fe0/Fe-oxide nanoparticles).

Kim et al. (2009), Sirk et al. (2009)

Particle size PZCs for hematite nanoparticles of diameter 12, 32, and 65 nm are ~7.5, 7.8, and 8.5, respectively.

He et al. (2009)

Particle transformation Aged Fe0/Fe oxide nanoparticles have lower PZC than fresh NPs by ~3 pH units, presumably due to surface oxidation and hydrolysis.

Kim et al. (2009)

† NP, nanoparticle; PZC, point of zero charge.


lization or destabilization depends on several factors, including

type of NOM, NOM concentration, and solution chemistry

(Chen and Elimelech, 2007; Chen et al., 2006; Diegoli et al.,

2008; Domingos et al., 2009). Stabilization usually results from

NOM forming a charged stabilizing layer on the outside of the

particle. Destabilization, on the other hand, results from par-

ticles being bridged by larger NOM molecules, such as rigid

biopolymers (Fig. 5B, left). Th is occurs for relatively large-sized

NOM (~10–100 kDa) and can dominate interactions between

nanoparticles (Buffl e et al., 1998). Large-molecular-weight bio-

molecules and biomacromolecules, including proteins, poly-

peptides, and amino acids (McKeon and Love, 2008; Moreau

et al., 2007; Vamunu et al., 2008), aff ect aggregation of NPs in

aquatic environments similarly. Table 5 summarizes reports of

NP stabilization and destabilization aff orded by the presence of

NOM and biomacromolecules reported in literature.

Nanoparticle Interaction with Inorganic and Biocolloids

Nanoparticles released to the environment can interact with

inorganic colloids (e.g., clays, aluminoscilicates and iron oxy-

hydroxides) and biocolloids (e.g., bacteria), altering the disper-

sion stability of engineered NPs via heteroaggregation (Fig.

5B). Heteroaggregation with bacteria can prevent aggregation

or promote disaggregation of NPs. Cerium oxide NPs were

reported to attach the surface of bacteria at the maximum

adsorption capacity of 15 mg of CeO2 m−2 due to electrostatic

attraction (Th ill et al., 2006). Holden et al. (2009c) showed

that TiO2 NPs were disaggregated in the presence of micro-

organisms (unpublished data). Heteroaggregation of NZVI

and bacteria also increases dispersion stability when com-

pared with dispersions containing no bacteria (Li et al., 2010).

Heteroaggregation with bacteria, NOM, or mobile colloids

(e.g., clay particles) could enhance NP stability, but heteroag-

gregation to large enough particles could also destabilize NP

dispersions. A complete understanding of the impacts of het-

eroaggregation on NP behavior requires more experimentation.

Eff ect of Aggregation on Deposition and Sedimentation

Aggregation of NPs aff ects their fate and transport in ground

and surface waters (Fig. 6). In the subsurface, aggregation aff ects

deposition processes; in surface waters, aggregation drives sedi-

mentation processes. Homoaggregation of NPs mostly leads to

Fig. 6. Nanoparticles are hypothesized to transport diff erently in the environment depending on their homo- or heteroaggregation state. Homoaggregated particles are more likely to attach to surfaces and be removed in porous media. Homoaggregation can also lead to the forma-tion of aggregates that are large enough to be settled out of surface waters by gravitational forces. Heteroaggregation may lead to surface particle coatings that enhance transport in porous media. Heteroaggregation can also lead to stabilization or destabilization eff ects in surface waters (e.g., with bacteria or natural organic matter).


high deposition rates in porous media because of increased col-

lisions between NP clusters and aquifer materials via sedimenta-

tion and interception as opposed to relying only on Brownian

motion. For example, NZVI injected into the subsurface at high

particle concentrations (1–10 g L−1) aggregate rapidly, and their

mobility in the subsurface suff ers as a result (Kanel et al., 2008;

Phenrat et al., 2007; Saleh et al., 2007) (i.e., they do not trans-

port far from their injection point).

Th ere are numerous studies that suggest prevention of

aggregation improves transport in porous media. Polymeric

surface modifi cation has been used to decrease NZVI aggrega-

tion and therefore deposition (Kanel et al., 2008; Kim et al.,

2009; Phenrat et al., 2009c; Phenrat et al., 2008; Saleh et al.,

2008). He et al. (2009) and Tiraferri and Sethi (2009) observed

enhanced mobility of NZVI modifi ed with adsorbed carboxy-

methyl cellulose and guar gum (He et al., 2009; Tiraferri and

Sethi, 2009). Phenrat et al. (2009a) demonstrated that NZVI

transport was limited specifi cally by NP aggregation during

transport in porous media by comparing the transport of non-

aggregating hematite NPs and aggregating NZVI of the same

size and surface chemistry. When NP deposition is interpreted

by classical fi ltration theory (i.e., assuming no aggregation

during particle transport in porous media), the eff ect of aggre-

gation is neglected and can lead to error in the reported attach-

ment coeffi cient. Th is makes comparisons between diff erent

experiments and systems diffi cult.

In contrast to homoaggregation, heteroaggregation with

mobile, natural colloidal particles (e.g., clay) and with natu-

ral particles that have low density (e.g., microorganisms) can

enhance transport of NPs in porous media by decreasing the

particle collision rate. For example, the presence of 20 mg

L−1 NOM enhanced NZVI transportability in porous media

(Johnson et al., 2009). Clay particles were also used as NZVI

supports to enhance NZVI mobility (Hydutsky et al., 2007),

which indicates that heteroaggregation between clay and NZVI

enhances NZVI mobility.

In surface waters, homoaggregation can enhance sedimen-

tation if aggregates or agglomerates reach a size where gravity

forces acting on the particles becomes signifi cant compared

with Brownian diff usion (O’Melia, 1980). Nanoparticles with

low dispersion stability therefore tend to accumulate in the

sediment near their source (Fig. 6). Heteroaggregation, with

low-density natural colloids, may facilitate stabilization or dis-

aggregation of NPs, which would increase their residence times

in water bodies (Fig. 6). Stabilizing or destabilizing NPs against

aggregation can have far-reaching eff ects. For example, cerium

oxide NPs stabilized with surfactant remained dispersed as col-

loids and were not eff ectively removed by an activated sludge

system (Limbach et al., 2008). Additionally, the removal effi -

ciency of functionalized versus nonfunctionalized fullerene

NPs by alum coagulation was compared. Increased steric sta-

bilization by surface functionalization prevented aggregation

and attraction with alum fl ocs (Hyung and Kim, 2009). In

contrast, surface coatings, such as Tween 20 on SiO2 NPs,

have been shown to enhance their fl occulation and removal

during water treatment, whereas uncoated SiO2 NPs were not

removed (Jarvie et al., 2009). Th e ability to predict the aggre-

gation eff ects of adsorbed or grafted organic macromolecular

coatings on NP surfaces is not possible with current DLVO

or XDLVO theory and is an important area of environmental

nanotechnology research.

Eff ect of Aggregation on Nanoparticle Reactivity

and Eff ect of Reactivity on Aggregation StateAggregation can alter NP reactivity, and NP aggregation state

can be altered by chemical or photochemical environmen-

tal processes (Fig. 5D). Several types of NPs are photoactive

(e.g., C60

, quantum Dots, and TiO2) and can produce reactive

oxygen species (ROS). A relationship between Df and photoac-

tivity indicates that the structure of the aggregates formed has a

role in determining ROS production and subsequent reactivity.

Fullerene (C60

) is a photoactive carbon-based class of NP that

forms crystalline aggregates in water with high Df (i.e., closely

packed aggregates). Th is aggregation promoted triplet–triplet

annihilation and self-quenching due to close contact between

cages and severely reduced the quantum yield of C60

(Hotze

et al., unpublished data). Fullerol (hydroxylated fullerene),

which has a lower theoretical quantum yield than C60

produced

greater amounts of ROS than C60

under identical solution con-

ditions because fullerol formed aggregates with lower Df (less

closely packed fractal aggregates), which decreased the amount

of triplet–triplet annihilation and increased ROS yield relative

to C60

(Hotze et al., unpublished data).

Aggregation has been shown to decrease the rate of dissolu-

tion of lead sulfi de (PbS) NPs (Liu et al., 2008). Th e dissolu-

tion rate of 240 nm aggregates of PbS NPs (4.7 × 10−10 mol m−2

s−1) was an order of magnitude lower than for monodisperse

14-nm primary PbS NPs (4.4 × 10−9 mol m−2 s−1) and also

lower than for micrometer-sized particles of the same materi-

als. In addition to aggregation decreasing available surface area

for dissolution, they attributed the retarded dissolution to inhi-

bition in the highly confi ned space between the particles in

the aggregates. In confi ned space, water molecules arrange dif-

ferently than in the bulk, increasing viscosity at the interface.

Enhanced viscosity in confi ned space decreases the diff usion

rate of water to the surface and therefore the dissolution rate.

Th ey also hypothesized that the overlap of the diff usion layer

between two adjacent particles in an aggregate can reduce the

concentration gradient (i.e., the driving force for dissolution)

from NP surfaces to bulk solution.

Another example of aggregation decreasing NP reactivity

is the dechlorination of carbon tetrachloride (CT) by 9-nm-

diameter magnetite NPs (Vikesland et al., 2007). Th e CT

dechlorination rate decreased as the degree of aggregation

of the magnetite NPs increased. By promoting aggregation

through addition of diff erent concentrations of monovalent

or divalent cations, an inverse relationship between aggre-

gate size (measured by DLS) and the CT dechlorination rate

was established.

Although aggregation can aff ect NP reactivity, NP reactiv-

ity can also aff ect aggregation. Clusters of C60

also undergo

photochemical oxidation and subsequent disaggregation in

water (Hou and Jafvert, 2009; Lee et al., 2009). For exam-

ple, Hou and Jafvert (2009) reported that the photochemical

transformation of aqueous C60

clusters in sunlight decreased

mean aggregate size from 500 to 160 nm and produced

unidentifi ed soluble reaction products after 65 d of exposure


to natural sunlight (Hou and Jafvert, 2009). Lee et al. (2009)

observed similar transformations under UV light exposure

(Lee et al., 2009). Photochemical transformations of C60

also took place during heteroaggregation with NOM (Li et

al., 2009) and enhanced the dispersion stability of NOM-

coated C60

in comparison to nonilluminated samples (Li et

al., 2009). Enhanced dispersion appeared to be due to a sur-

face erosion or dissolution-recrystallization process catalyzed

by sunlight, rather than a simple breakage of C60

from clus-

ters (Li et al., 2009). Th erefore, NPs designed to be reactive

under photoilluminated conditions (e.g., C60

or TiO2) could

have their aggregation properties altered by long-term sun-

light exposure.

Chemical oxidation could act in a similar manner by alter-

ing surface coatings or the underlying NP so that the particles

gain or lose stability against aggregation (Fig. 5D). For exam-

ple, ozone reacts with C60

(Fortner et al., 2007) and CNTs,

breaking the carbon cage structure and altering the stability of

the NPs against aggregation in aqueous suspension. Th is ozo-

nation process has been demonstrated to render the oxidized

C60

more photoactive than its unoxidized parent aggregates

(Cho et al., 2009). Additionally, surfactants and coatings are

susceptible to these same types of oxidation processes possi-

bly decreasing the stability of NPs in the environment due to

loss of the stabilizing coating. A better understanding of the

biological and abiotic reactions aff ecting NP stability against

aggregation is needed to predict their fate in the environment.

Eff ects of Aggregation on Biological Impacts

from NanoparticlesAggregation of NPs can alter their biological eff ects by aff ecting

ion release from the surface, if and where particles collect inside

an organism, and the reactive surface area of NPs (Fig. 5C).

Calculation of metal fl ux at cell surfaces is important for deter-

mining the bioavailability and ecotoxicity of nanomaterials. Th is

is especially true in the case of silver NPs where silver ion release

plays an important role in the inhibition of bacterial growth

(Taurozzi et al., 2008; Ju-Nam and Lead, 2008). Ion release

may also prove to be important for determining the impacts of

quantum dots and iron NPs, among others. Heteroaggregation

can confound understanding of metal fl ux because the metal

can also bind to surfaces and internal sites of naturally occur-

ring colloids of various sizes and compositions (Buffl e et al.,

2007). Solution chemistry (e.g., ionic strength and the amount

and type of NOM) controls the fractal dimension of NP aggre-

gates, with lower Df (1.3–2.0) formed in solutions with high

NOM, and higher Df (2.3–2.7) formed in solutions without

signifi cant NOM present (Zhang et al., 2007). Aggregate struc-

ture plays an important role in how metal ions diff use inside of

the fl ocs where the kinetics of metal association and dissociation

from binding sites is slowed by orders of magnitude compared

with rates expected for free ions in solution (Buffl e et al., 2007).

Understanding mechanisms by which aggregate structure aff ects

the dissolution and release of ions and resulting ecotoxicity (e.g.,

from silver NPs) will improve insights into the potential impacts

of such particles in the environment.

Aggregation alters the NP size distribution, in turn aff ect-

ing the fate of NPs in biological systems as well as their toxic-

ity. For example, 20-nm NPs were found to deposit mostly in

the alveolar region, whereas 5- to 10-nm particles deposited

in tracheobronchial region, and particles smaller than 10 nm

accumulate mostly in the upper respiratory tract (Heyder et

al., 1986; Oberdorster et al., 2007; Swift et al., 1994). Th is

indicates that aggregation can alter exposure locations or

exposure pathways. Nanoparticle aggregation was also shown

to aff ect the mode of cellular uptake of NPs and subsequent

biological responses. For example, phagocytosis by macro-

phages and giant cells is a mechanism used by cells to clear

particles of a few microns or less (Oberdorster et al., 2007).

Phenrat et al. (2009b) found that aggregation aff ected uptake

of bare and polymer-modifi ed NZVI particles by mammalian

glial cells. Although the primary particle sizes of bare NZVI

particles (20–100 nm) are smaller than the optimal size range

(1–2 μm) for phagocytosis, the NZVI loosely aggregated to

micrometer-sized clusters in the physiological buff er, trigger-

ing phagocytosis and subsequent uptake. Biological eff ects,

therefore, vary depending on whether or not a particle is

aggregated, among other confounding factors (e.g., endotox-

ins, adsorption of biomacromolecules, etc.). Given this fact,

understanding aggregation in biological systems is essential

for elucidating mechanisms of toxicity.

ConclusionsTh e ability to manipulate atoms allows scientists to create

NPs with signifi cantly diff erent reactivity than their micro-

sized analogs (He et al., 2008; Kanehira et al., 2008).

Reports on NP behavior and eff ects to date indicate that

the aggregation behavior of the NPs will signifi cantly aff ect

their fate, transport, reactivity, bioavailability, and eff ects in

environmental systems. Fortunately, the science of aggrega-

tion is well developed. Unfortunately, the translation of that

science into the regime of nanomaterials remains a challenge

due their unique physical and chemical properties (e.g., size,

shape, composition, and reactivity). Th e presence of organic

macromolecular coatings on NPs including those engineered

as part of the NP, or those that are acquired in the environ-

ment (e.g., adsorption of NOM), further complicate analysis

of NP behavior by traditional colloid science. Nanomaterial

aggregation studies, therefore, aim to take into account

how these types of novel surface properties will alter what

is already understood about aggregation science. Ultimately,

this will guide scientists’ grasp of how nanomaterials and

NPs will occur in and aff ect environmental systems. Much

work remains to gain a full understanding of NP aggregation

and how to apply these principles to create better applica-

tions and lower impacts of nanomaterials.

AcknowledgmentsTh is material is based on work supported by the National Science

Foundation (NSF) and the Environmental Protection Agency (EPA)

under NSF Cooperative Agreement EF-0830093, Center for the

Environmental Implications of NanoTechnology (CEINT). Any

opinions, fi ndings, conclusions or recommendations expressed in this

material are those of the author(s) and do not necessarily refl ect the

views of the NSF or the EPA. Th is work has not been subjected to

EPA review, and no offi cial endorsement should be inferred.


ReferencesAlivisatos, A. 2001. Less is more in medicine: Sophisticated forms of nanotech-

nology will fi nd some of their fi rst real-world applications in biomedical research, disease diagnosis and, possibly, therapy. Sci. Am. 285:66–73.

Amirbahman, A., and T.M. Olson. 1993. Transport of humic matter-coated hematite in packed beds. Environ. Sci. Technol. 27:2807–2813.

Amirbahman, A., and T.M. Olson. 1995. Th e role of surface conformations in the deposition kinetics of humic matter-coated colloids in porous media. Colloids Surf. A Physicochem. Eng. Asp. 95:249–259.

Barbot, E., P. Dussouillez, J.-Y. Bottero, and P. Moulin. 2010. Coagulation of bentonite suspension by polyelectrolytes or ferric chloride: Floc breakage and reformation. Chem. Eng. J. 156:83–91.

Baveye, P., and M. Laba. 2008. Aggregation and toxicology of titanium dioxide nanoparticles. Environ. Health Perspect. 116:A152–A152.

Benn, T.M., and P. Westerhoff . 2008. Nanoparticle silver released into wa-ter from commercially available sock fabrics. Environ. Sci. Technol. 42:4133–4139.

Bhattacharjee, S., and M. Elimelech. 1997. Surface element integration: A novel technique for evaluation of DLVO interaction between a particle and a fl at plate. J. Colloid Interface Sci. 193:273–285.

Brant, J., H. Lecoanet, M. Hotze, and M. Wiesner. 2005. Comparison of electrokinetic properties of colloidal fullerenes (nC

60) formed using two

procedures. Environ. Sci. Technol. 39:6343–6351.

Brant, J.A., J. Labille, C.O. Robichaud, and M. Wiesner. 2007. Fullerol cluster formation in aqueous solutions: Implications for environmental release. J. Colloid Interface Sci. 314:281–288.

Brewer, S.H., W.R. Glomm, M.C. Johnson, M.K. Knag, and S. Franzen. 2005. Probing BSA binding to citrate-coated gold nanoparticles and sur-faces. Langmuir 21:9303–9307.

Buffl e, J., and G. Leppard. 1995. Characterization of aquatic colloids and mac-romolecules: 1. Structure and behavior of colloidal material. Environ. Sci. Technol. 29:2169–2175.

Buffl e, J., K.J. Wilkinson, S. Stoll, M. Filella, and S. Zhang. 1998. A gener-alized description of aquatic colloidal interactions: Th e three-colloidal component approach. Environ. Sci. Technol. 32:2887–2899.

Buffl e, J., Z. Zhang, and K. Startchev. 2007. Metal fl ux and dynamic spe-ciation at (bio)interfaces. Part I: Critical evaluation and compilation of physicochemical parameters for complexes with simple ligands and ful-vic/humic substances. Environ. Sci. Technol. 41:7609–7620.

Cai, Y., and K.H. Sandhage. 2005. Zn2SiO4-coated microparticles with bio-logically-controlled 3D shapes. Phys Status Solidi A 202:R105–R107.

Chakraborti, R., K. Gardner, J. Atkinson, and J. Van Benschoten. 2003. Chang-es in fractal dimension during aggregation. Water Res. 37:873–883.

Chappell, M.A., A.J. George, K.M. Dontsova, B.E. Porter, C.L. Price, P. Zhou, E. Morikawa, A.J. Kennedy, and J.A. Steevens. 2009. Surfactive stabiliza-tion of multi-walled carbon nanotube dispersion with dissolved humic substances. Environ. Pollut. 157:1081–1087.

Chen, K., and M. Elimelech. 2007. Infl uence of humic acid on the aggrega-tion kinetics of fullerene (C

60) nanoparticles in monovalent and divalent

electrolyte solutions. J. Colloid Interface Sci. 309:126–134.

Chen, K.L., and M. Elimelech. 2009. Relating colloidal stability of fullerene (C

60) nanoparticles to nanoparticle charge and electrokinetic properties.

Environ. Sci. Technol. 43:7270–7276.

Chen, K.-L., S.E. Mylon, and M. Elimelech. 2006. Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent elec-trolyte. Environ. Sci. Technol. 40:1516–1523.

Cheng, X.K., A.T. Kan, and M.B. Tomson. 2004. Naphthalene adsorption and desorption from aqueous C

60 fullerene. J. Chem. Eng. Data 49:675–683.

Cho, M., J.D. Fortner, J.B. Hughes, and J.-H. Kim. 2009. Escherichia coli inactivation by water-soluble, ozonated C

60 derivative: Kinetics and

mechanisms. Environ. Sci. Technol. 43:7410–7415.

Dederichs, T., M. Möller, and O. Weichold. 2009. Colloidal stability of hy-drophobic nanoparticles in ionic surfactant solutions: Defi nition of the critical dispersion concentration. Langmuir 25:2007–2012.

Delgado, A.V., F. Gonzalez-Caballero, R.J. Hunter, L.K. Koopal, and J. Lykl-ema. 2007. Measurement and interpretation of electrokinetic phenom-ena. J. Colloid Interface Sci. 309:194–224.

Derjaguin, B.V., and L. Landau. 1941. Th eory of the stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solutions of electrolytes. Acta Phys. Chim. URSS 14:633–662.

Diegoli, S., A.L. Manciulea, S. Begum, I.P. Jones, J.R. Lead, and J.A. Preece. 2008. Interaction between manufactured gold nanoparticles and natu-rally occurring organic macromolecules. Sci. Total Environ. 402:51–61.

Domingos, R.F., N. Tufenkji, and K.J. Wilkinson. 2009. Aggregation of tita-nium dioxide nanoparticles: Role of a fulvic acid. Environ. Sci. Technol.

43:1282–1286.

Elena, D.V., M. Coisson, C. Appino, F. Vinai, and R. Sethi. 2009. Magnetic characterization and interaction modeling of zerovalent iron nanopar-ticles for the remediation of contaminated aquifers. J. Nanosci. Nano-technol. 9:3210–3218.

Elimelech, M., and C. O’Melia. 1990. Eff ect of particle size on collision ef-fi ciency in the deposition of Brownian particles with electrostatic energy barriers. Langmuir 6:1153–1163.

Evans, D.F., and H. Wennerstrom. 1999. Th e colloidal domain: Where phys-ics, chemistry, biology, and technology meet. Wiley-VCH, New York.

Fleer, G.J., M.A. Cohen Stuart, J.M.H.M. Scheutjens, T. Cosgrove, and B. Vincent. 1993. Polymers at interfaces. Chapman & Hall, London, UK.

Fortner, J.D., D.-I. Kim, A.M. Boyd, J.C. Falkner, S. Moran, V.L. Colvin, J.B. Hughes, and J.-H. Kim. 2007. Reaction of water-stable C

60 aggregates

with ozone. Environ. Sci. Technol. 41:7497–7502.

French, R.A., A.R. Jacobson, B. Kim, S.L. Isley, R.L. Penn, and P.C. Baveye. 2009. Infl uence of ionic strength, pH, and cation valence on aggrega-tion kinetics of titanium dioxide nanoparticles. Environ. Sci. Technol. 43:1354–1359.

Fritz, G., V. Schadler, N. Willenbacher, and N. Wagner. 2002. Electrosteric stabilization of colloidal dispersions. Langmuir 18:6381–6390.

Gatica, S., M. Cole, and D. Velegol. 2005. Designing van der Waals forces between nanocolloids. Nano Lett. 5:169–173.

Graf, P., A. Mantion, A. Foelske, A. Shkilnyy, A. Masic, A.E. Th uenemann, and A. Taubert. 2009. Peptide-coated silver nanoparticles: Synthesis, surface chemistry, and pH-triggered, reversible assembly into particle as-semblies. Chem. Eur. J. 15:5831–5844.

Gu, B., J. Schmitt, Z. Chen, L. Liang, and J.F. McCarthy. 1994. Adsorption and desorption of natural organic matter on iron oxide: Mechanisms and models. Environ. Sci. Technol. 28:38–46.

Hahn, M., D. Abadzic, and C. O’Melia. 2004. Aquasols: On the role of sec-ondary minima. Environ. Sci. Technol. 38:5915–5924.

Hansen, P., M. Malmsten, B. Bergenstahl, and L. Bergstrom. 1999. Ortho-kinetic aggregation in two dimensions of monodisperse and bidisperse colloidal systems. J. Colloid Interface Sci. 220:269–280.

He, F., D. Zhao, J. Liu, and C.B. Roberts. 2007. Stabilization of Fe-Pd nanoparticles with sodium carboxymethyl cellulose for enhanced trans-port and dechlorination of trichloroethylene in soil and groundwater. Ind. Eng. Chem. Res. 46:29–34.

He, F., M. Zhang, T. Qian, and D. Zhao. 2009. Transport of carboxymethyl cellulose stabilized iron nanoparticles in porous media: Column experi-ments and modeling. J. Colloid Interface Sci. 334:96–102.

He, Y.T., J. Wan, and T. Tokunaga. 2008. Kinetic stability of hematite nanoparticles: Th e eff ect of particle sizes. J. Nanopart. Res. 10:321–332.

Hermansson, M. 1999. Th e DLVO theory in microbial adhesion. Colloids Surf. B Biointerfaces 14:105–119.

Heyder, J., J. Gebhart, G. Rudolf, and C. Schiller. 1986. Deposition of par-ticles in the human respiratory tract in the size range 0, 005–15 μm. J. Aerosol Sci. 17:811–825.

Hezinger, A.F.E., J. Tessmar, and A. Göpferich. 2008. Polymer coating of quantum dots: A powerful tool toward diagnostics and sensorics. Eur. J. Pharm. Biopharm. 68:138–152.

Hiemenz, P.C., and R. Rajagopalan. 1997. Principles of colloid and surface chemistry. CRC, New York.

Hochella, M.F., S.K. Lower, P.A. Maurice, R.L. Penn, N. Sahai, D.L. Sparks, and B.S. Twining. 2008. Nanominerals, mineral nanoparticles, and earth systems. Science 21:1631–1635.

Hoek, E., and G. Agarwal. 2006. Extended DLVO interactions between spher-ical particles and rough surfaces. J. Colloid Interface Sci. 298:50–58.

Hotze, E.M., J. Labille, P. Alvarez, and M.R. Wiesner. 2008. Mechanisms of photochemistry and reactive oxygen production by fullerene suspensions in water. Environ. Sci. Technol. 42:4175–4180.

Hou, W.-C., and C.T. Jafvert. 2009. Photochemical transformation of aqueous C

60 clusters in sunlight. Environ. Sci. Technol. 43:362–367.

Hydutsky, B.W., E.J. Mack, B.B. Beckerman, J.M. Skluzacek, and T.E. Mal-louk. 2007. Optimization of nano- and microiron transport through sand columns using polyelectrolyte mixtures. Environ. Sci. Technol. 41:6418–6424.

Hyung, H., J. Fortner, J. Hughes, and J. Kim. 2007. Natural organic matter stabilizes carbon nanotubes in the aqueous phase. Environ. Sci. Technol. 41:179–184.

Hyung, H., and J.-H. Kim. 2009. Dispersion of C60

in natural water and re-moval by conventional drinking water treatment processes. Water Res. 43:2463–2470.

Israelachvili, J.N. 1991. Intermolecular & surface forces. Academic Press,


New York.

Jain, T.K., M.A. Morales, S.K. Sahoo, D.L. Leslie-Pelecky, and V. Labhaset-war. 2005. Iron oxide nanoparticles for sustained delivery of anticancer agents. Mol. Pharm. 2:194–205.

Jarvie, H., H. Al-Obaidi, S. King, M. Bowes, M. Lawrence, A. Drake, M. Green, P. Dobson. 2009. Fate of silica nanoparticles in simulated pri-mary wastewater treatment. Environ. Sci. Technol. 43:8622–8628.

Jiang, J., G. Oberdorster, and P. Biswas. 2009. Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxico-logical studies. J. Nanopart. Res. 11:77–89.

Johnson, R.L., G.O. Johnson, J.T. Nurmi, and P.G. Tratnyek. 2009. Natural organic matter enhanced mobility of nano zerovalent iron. Environ. Sci. Technol. 43:5455–5460.

Ju-Nam, Y., and J.R. Lead. 2008. Manufactured nanoparticles: An overview of their chemistry, interactions and potential environmental implications. Sci. Total Environ. 400:396–414.

Kanehira, K., T. Banzai, C. Ogino, N. Shimizu, Y. Kubota, and S. Sonezaki. 2008. Properties of TiO

2–polyacrylic acid dispersions with potential for

molecular recognition. Colloids Surf. B Biointerfaces 64:10–15.

Kanel, S.R., R.R. Goswami, T.P. Clement, M.O. Barnett, and D. Zhao. 2008. Two dimensional transport characteristics of surface stabilized zero-valent iron nanoparticles in porous media. Environ. Sci. Technol. 42:896–900.

Kawano, T., and I. Hiroaki. 2008. A simple preparation technique for shape-controlled zinc oxide nanoparticles: Formation of narrow size-distrib-uted nanorods using seeds in aqueous solutions. Colloid Surface A 319:130–135.

Kim, B., and W.M. Sigmund. 2004. Functionalized multiwall carbon nano-tube/gold nanoparticle composites. Langmuir 20:8239–8242.

Kim, H.-J., T. Phenrat, R.D. Tilton, and G.V. Lowry. 2009. Fe0 nanoparticles remain mobile in porous media after aging due to slow desorption of polymeric surface modifi ers. Environ. Sci. Technol. 43:3824–3830.

Kim, J., D. Kim, H. Jung, and K. Hong. 2005. Infl uence of anatase-rutile phase transformation on dielectric properties of sol-gel derived TiO

2 thin

fi lms. Jpn. J. Appl. Phys. 1:6148–6151.

Koponen, I.T. 2009. Cluster growth poised on the edge of break-up, II: From reaction kinetics to thermodynamics. Physica A 388:2659–2665.

Kosmulski, M. 2009. pH-dependent surface charging and points of zero charge. IV. Update and new approach. J. Colloid Interface Sci. 337:439–448.

Kozan, M., J. Th angala, R. Bogale, M.P. Menguc, and M.K. Sunkara. 2008. In-situ characterization of dispersion stability of WO

3 nanoparticles and

nanowires. J. Nanopart. Res. 10:599–612.

Lebrette, S., C. Pagnoux, and P. Abélard. 2004. Stability of aqueous TiO2 sus-

pensions: Infl uence of ethanol. J. Colloid Interface Sci. 280:400–408.

Lee, J., M. Cho, J.D. Fortner, J.B. Hughes, and J.-H. Kim. 2009. Transforma-tion of aggregated C

60 in the aqueous phase by UV irradiation. Environ.

Sci. Technol. 43:4878–4883.

Li, Q., B. Xie, Y.S. Hwang, and Y. Xu. 2009. Kinetics of C60

fullerene disper-sion in water enhanced by natural organic matter and sunlight. Environ. Sci. Technol. 43:3574–3579.

Li, Z., K. Greden, P. Alvarez, K. Gregory, and G.V. Lowry. 2010. Adsorbed polymer and NOM limits adhesion and toxicity of nano scale zero-valent iron (NZVI) to E. coli. Environ. Sci. Technol. (in press).

Limbach, L.K., R. Bereiter, E. Muller, R. Krebs, R. Glli, and W.J. Stark. 2008. Removal of oxide nanoparticles in a model wastewater treatment plant: Infl uence of agglomeration and surfactants on clearing effi ciency. Envi-ron. Sci. Technol. 42:5828–5833.

Limbach, L.K., Y.C. Li, R.N. Grass, T.J. Brunner, M.A. Hintermann, M. Muller, D. Gunther, and W.J. Stark. 2005. Oxide nanoparticle uptake in human lung fi broblasts: Eff ects of particle size, agglomeration, and diff usion at low concentrations. Environ. Sci. Technol. 39:9370–9376.

Liu, J., D.M. Aruguete, J.R. Jinschek, J.D. Rimstidt, and M.F. Hochella. 2008. Th e non-oxidative dissolution of galena nanocrystals: Insights into mineral dissolution rates as a function of grain size, shape, and aggrega-tion state. Geochim. Cosmochim. Acta 72:5984–5996.

Liu, Y., S.A. Majetich, R.D. Tilton, D.S. Sholl, and G.V. Lowry. 2005. TCE dechlorination rates, pathways, and effi ciency of nanoscale iron particles with diff erent properties. Environ. Sci. Technol. 39:1338–1345.

Liu, Z., W. Cai, L. He, N. Nakayama, and K. Chen. 2006. In vivo biodistribu-tion and highly effi cient tumour targeting of carbon nanotubes in mice. Nature Nanotechnol. 2:47–52.

Luccardini, C., C. Tribet, F. Vial, V. Marchi-Artzner, and M. Dahan. 2006. Size, charge, and interactions with giant lipid vesicles of quantum dots coated with an amphiphilic macromolecule. Langmuir 22:2304–2310.

Ma, D., X. Hu, H. Zhou, J. Zhang, and Y. Qian. 2007. Shape-controlled

synthesis and formation mechanism of nanoparticles-assembled Ag2S

nanorods and nanotubes. J. Cryst. Growth 304:163–168.

Mayya, K.S., B. Schoeler, and F. Caruso. 2003. Preparation and organization of nanoscale polyelectrolyte-coated gold nanoparticles. Adv. Funct. Ma-ter. 13:183–188.

McKeon, K.D., and B.J. Love. 2008. Th e presence of adsorbed proteins on particles increases aggregated particle sedimentation, as measured by a light scattering technique. J. Adhes. 84:664–674.

Montgomery, S., M. Franchek, and V. Goldschmidt. 2000. Analytical disper-sion force calculations for nontraditional geometries. J. Colloid Interface Sci. 227:567–584.

Moon, S.Y., T. Kusunose, and T. Sekino. 2009. CTAB-assisted synthesis of size- and shape-controlled gold nanoparticles in SDS aqueous solution. Mater. Lett. 63:2038–2040.

Moore, V.C., M.S. Strano, E.H. Haroz, R.H. Hauge, R.E. Smalley, J. Schmidt, and Y. Talmon. 2003. Individually suspended single-walled carbon nanotubes in various surfactants. Nano Lett. 3:1379–1382.

Moreau, J.W., P.K. Weber, M.C. Martin, B. Gilbert, I.D. Hutcheon, and J.F. Banfi eld. 2007. Extracellular proteins limit the dispersal of biogenic nanoparticles. Science 316:1600–1603.

O’Connell, M.J., S.M. Bachilo, C.B. Huff man, V.C. Moore, M.S. Strano, E.H. Haroz, K.L. Rialon, P.J. Boul, W.H. Noon, C. Kittrell, J.P. Ma, R.H. Hauge, R.B. Weisman, and R.E. Smalley. 2002. Band gap fl uorescence from individual single-walled carbon nanotubes. Science 297:593–596.

O’Melia, C.R. 1980. ES&T features. Aquasols: Th e behavior of small particles in aquatic systems. Environ. Sci. Technol. 14:1052–1060.

Oberdorster, G., V. Stone, and K. Donaldson. 2007. Toxicology of nanopar-ticles: A historical perspective. Nanotoxicology 1:2–25.

Ortega-Vinuesa, J., A. Martín-Rodríguez, and R. Hidalgo-Alvarez. 1996. Col-loidal stability of polymer colloids with diff erent interfacial properties: Mechanisms. J. Colloid Interface Sci. 184:259–267.

Phenrat, T., H.-J. Kim, F. Fagerlund, T. Illangasekare, R.D. Tilton, and G.V. Lowry. 2009a. Particle size distribution, concentration, and magnetic at-traction aff ects transport of polymer-modifi ed Fe0 nanoparticles in sand columns. Environ. Sci. Technol. 43:5079–5085.

Phenrat, T., T. Long, G. Lowry, and B. Veronesi. 2009b. Partial oxidation (“ag-ing”) and surface modifi cation decrease the toxicity of nanosized zerova-lent iron. Environ. Sci. Technol. 43:195–200.

Phenrat, T., Y. Liu, R. Tilton, and G. Lowry. 2009c. Adsorbed polyelectrolyte coatings decrease Fe0 nanoparticle reactivity with TCE in water: Con-ceptual model and mechanisms. Environ. Sci. Technol. 43:1507–1514.

Phenrat, T., N. Saleh, K. Sirk, H. Kim, R. Tilton, and G. Lowry. 2008. Stabilization of aqueous nanoscale zerovalent iron dispersions by an-ionic polyelectrolytes: Adsorbed anionic polyelectrolyte layer proper-ties and their eff ect on aggregation and sedimentation. J. Nanopart. Res. 10:795–814.

Phenrat, T., N. Saleh, K. Sirk, R.D. Tilton, and G.V. Lowry. 2007. Aggrega-tion and sedimentation of aqueous nanoscale zerovalent iron dispersions. Environ. Sci. Technol. 41:284–290.

Project on Emerging Nanotechnologies. 2009. Nanotechnology consumer products inventory. Available at http://www.nanotechproject.org (veri-fi ed 22 Apr. 2010). Project on Emerging Nanotechnologies, Washing-ton, DC.

Richard, C., F. Balavoine, P. Schultz, T.W. Ebbesen, and C. Mioskowski. 2003. Supramolecular self-assembly of lipid derivatives on carbon nanotubes. Science 300:775–778.

Romero-Cano, M., A. Martin-Rodriguez, and F. de las Nieves. 2001. Elec-trosteric stabilization of polymer colloids with diff erent functionality. Langmuir 17:3505–3511.

Rosen, M.J. 2004. Surfactants and interfacial phenomena. 3rd ed. Wiley-In-terscience, New York.

Saleh, N., H.-J. Kim, T. Phenrat, K. Matyjaszewski, R.D. Tilton, and G.V. Lowry. 2008. Ionic strength and composition aff ect the mobility of surface-modifi ed Fe0 nanoparticles in water-saturated sand columns. En-viron. Sci. Technol. 42:3349–3355.

Saleh, N., T. Phenrat, K. Sirk, B. Dufour, J. Ok, T. Sarbu, K. Matyiaszewski, R.D. Tilton, and G.V. Lowry. 2005. Adsorbed triblock copolymers de-liver reactive iron nanoparticles to the oil/water interface. Nano Lett. 5:2489–2494.

Saleh, N., K. Sirk, Y. Liu, T. Phenrat, B. Dufour, K. Matyjaszewski, R.D. Til-ton, and G.V. Lowry. 2007. Surface modifi cations enhance nanoiron transport and DNAPL targeting in saturated porous media. Environ. Eng. Sci. 24:45–57.

Sanchez-Iglesias, A., M. Grzelczak, B. Rodriguez-Gonzalez, R.A. Alvarez-Puebla, L.M. Liz-Marzan, and N.A. Kotov. 2009. Gold colloids with unconventional angled shapes. Langmuir 25:11431–11435.


Sirk, K.M., N.B. Saleh, T. Phenrat, H.-J. Kim, B. Dufour, J. Ok, P.L. Golas, K. Matyjaszewski, G.V. Lowry, and R.D. Tilton. 2009. Eff ect of adsorbed polyelectrolytes on nanoscale zero valent iron particle attachment to soil surface models. Environ. Sci. Technol. 43:3803–3808.

Smith, B., K. Wepasnick, K.E. Schrote, H.-H. Cho, W.P. Ball, and D.H. Fair-brother. 2009. Infl uence of surface oxides on the colloidal stability of multi-walled carbon nanotubes: A structure-property relationship. Lang-muir 25:9767–9776.

Sondi, I., D.V. Goia, and E. Matijevic. 2003. Preparation of highly concen-trated stable dispersions of uniform silver nanoparticles. J. Colloid Inter-face Sci. 260:75–81 .

Stumm, W., and J.J. Morgan. 1996. Aquatic chemistry: Chemical equilibria and rates in natural waters. 3rd ed. John Wiley & Sons, New York.

Swift, D., Y. Cheng, Y. Su, and H. Yeh. 1994. Ultrafi ne aerosol deposition in the human nasal and oral passages. Ann. Occup. Hyg. 38:77–81.

Tabor, D., and R. Winterton. 1969. Th e direct measurement of normal and retarded van der Waals forces. Proceed. R. Soc. London A. 312:435–450.

Taurozzi, J.S., H. Arul, V.Z. Bosak, A.F. Burban, T.C. Voice, M.L. Bruen-ing, and V.V. Tarabara. 2008. Eff ect of fi ller incorporation route on the properties of polysulfone-silver nanocomposite membranes of diff erent porosities. J. Membr. Sci. 325:58–68.

Th ill, A., O. Zeyons, O. Spalla, F. Chauvat, J. Rose, M. Auff an, and A.M. Flank. 2006. Cytotoxicity of CeO

2 nanoparticles for Escherichia coli:

Physico-chemical insight of the cytotoxicity mechanism. Environ. Sci. Technol. 40:6151–6156.

Th ompson Reuters. 2009. Web of knowledge. Available at http://WebofKnowl-edge.com (verifi ed 22 Apr. 2009). Th ompson Reuters, Philadelphia, PA.

Tiraferri, A., and R. Sethi. 2009. Enhanced transport of zerovalent iron nanoparticles in saturated porous media by guar gum. J. Nanopart. Res. 11:635–645.

Tohver, V., J.E. Smay, A. Braem, P.V. Braun, and J.A. Lewis. 2001. Nanopar-ticle halos: A new colloid stabilization mechanism. Proc. Natl. Acad. Sci. USA 98:8950–8954.

Vaisman, L., H.D. Wagner, and G. Marom. 2006. Th e role of surfac-tants in dispersion of carbon nanotubes. Adv. Colloid Interface Sci. 128–130:37–46.

Vamunu, C.I., P.J. Hol, Z.E. Allouni, S. Elsayed, and N.R. Gjerdet. 2008. Formation of potential titanium antigens based on protein binding to

titanium dioxide nanoparticles. Int. J. Nanomedicine 3:69–74.

Verwey, E., J. Overbeek, and K. van Nes. 1948. Th e theory of the stability of liophobic colloids: Th e interaction of sol particles having an electric double layer. Elsevier, Amsterdam.

Vikesland, P.J., A.M. Heathcock, R.L. Rebodos, and K.E. Makus. 2007. Par-ticle size and aggregation eff ects on magnetite reactivity toward carbon tetrachloride. Environ. Sci. Technol. 41:5277–5283.

Viota, J.L., F. González-Caballero, J.D.G. Durán, and A.V. Delgado. 2007. Study of the colloidal stability of concentrated bimodal magnetic fl uids. J. Colloid Interface Sci. 309:135–139.

Vold, M.J. 1954. Van der Waals’ attraction between anisometric particles. J. Colloid Sci. 9:451–459.

Wang, G.-H., W.-C. Li, K.-M. Jia, B. Spliethoff , F. Schüth, and A.-H. Lu. 2009. Shape and size controlled �-Fe

2O

3 nanoparticles as supports for

gold-catalysts: Synthesis and infl uence of support shape and size on cata-lytic performance. Appl. Catal. A 364:42–47.

Wang, W., B. Gu, L. Liang, and W.A. Hamilton. 2004. Adsorption and structural arrangement of cetyltrimethylammonium cations at the silica nanoparticle-water interface. J. Phys. Chem. B 108:17477–17483.

Wu, W., R.F. Giese, and C.J. van Oss. 1999. Stability versus fl occulation of particle suspensions in water-correlation with the extended DLVO ap-proach for aqueous systems, compared with classical DLVO theory. Col-loids Surf. B Biointerfaces 14:47–55.

Yurekli, K., C.A. Mitchell, and R. Krishnamoorti. 2004. Small-angle neutron scattering from surfactant-assisted aqueous dispersions of carbon nano-tubes. J. Am. Chem. Soc. 126:9902–9903.

Zhang, D., O. Neumann, H. Wang, V.M. Yuwono, A. Barhoumi, M. Per-ham, J.D. Hartgerink, P. Wittung-Stafshede, and N.J. Halas. 2009. Gold nanoparticles can induce the formation of protein-based aggregates at physiological pH. Nano Lett. 9:666–671.

Zhang, Y., N. Kohler, and M. Zhang. 2002. Surface modifi cation of super-paramagnetic magnetite nanoparticles and their intracellular uptake. Biomaterials 23:1553–1561.

Zhang, Z., J. Buffl e, and D. Alemani. 2007. Metal fl ux and dynamic speciation at (bio)interfaces. Part II: Evaluation and compilation of physicochemi-cal parameters for complexes with particles and aggregates. Environ. Sci. Technol. 41:7621–7631.

Documents

Nanoparticle Aggregation: Challenges to Understanding Transport