United States
Environmental Protection
mI m m Agency
Characterizing Concentrations
and Size Distributions of
Metal-Containing Nanoparticles
in Waste Water
APM 272
RESEARCH AND DEVELOPMENT
-------
-------
EPA/600/R-10/117
October 2010
www.epa.gov
Characterizing Concentrations
and Size Distributions of
Metal-Containing Nanoparticles
in Waste Water
APM 272
Edward M. Heithmar3
Spiros A. Pergantisb
aU.S. Environmental Protection Agency
National Exposure Research Laboratory
Environmental Sciences Division
944 E. Harmon Ave.
Las Vegas, NV 89119
bNRC Senior Research Fellow
U.S. Environmental Protection Agency
National Exposure Research Laboratory
Environmental Sciences Division
944 E. Harmon Ave.
Las Vegas, NV 89119
Although this work was reviewed by EPA and approved for publication, it may not necessarily reflect official
Agency policy. Mention of trade names and commercial products does not constitute endorsement or
recommendation for use.
U.S. Environmental Protection Agency
Office of Research and Development
Washington, DC 20460
-------
-------
Abstract
Nanomaterials containing metals are finding increasing use in consumer, industrial, and
medical products, and they are subsequently being released into the environment. Methods for
detecting, quantifying, and characterizing these materials in complex matrices are critical for the
eventual understanding of their implications to environmental quality and human health. This
report describes recent progress in the development of new metrology tools. Single particle -
inductively coupled plasma mass spectrometry (SP-ICPMS) is used to analyze complex aqueous
samples. SP-ICPMS simultaneously quantifies the concentration of nanoparticles containing an
analyte metal and measures the metal mass in individual nanoparticles. The accuracy of SP-
ICPMS is assessed over a range of nanoparticle sizes. The utility of the technique in a number of
applications is examined. Nanoparticles containing the analyte element can be measured
accurately and precisely in the presence of a 10,000-fold greater concentration of other
nanoparticles. SP-ICPMS is used to assess transformation of nanoparticles induced by changes
in ionic strength. A screening-level assessment of metal-containing nanoparticles in urban run-
off using SP-ICPMS is demonstrated, and preliminary studies of coupling SP-ICPMS with on-
line nanoparticle size separation methods are presented.
11
-------
Notice
The United States Environmental Protection Agency's (EPA) Office of Research and
Development performed and funded the research described here. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
111
-------
Table of Contents
Abstract ii
Notice iii
Table of Contents iv
Abbreviations and symbols v
Acknowledgments vi
Introduction 1
Uses of engineered nanomaterials 1
Environmental implications of increased use of engineered nanomaterials 2
Rationale for the development of metrology tools for metal-containing ENMs 3
Methods for characterizing engineered nanomaterials 4
Theory of SP-ICPMS 7
Experimental 9
Materials 9
Instrumentation 9
Methods 9
Results and Discussion 10
Assessment of SP-ICPMS accuracy 10
Precision and selectivity of SP-ICPMS 11
Nanoparticle transformation study 11
Screening urban runoff using SP-ICPMS 12
Coupling SP-ICPMS with separation methods 13
Conclusions and Future Work 14
References 16
Tables 22
Figure Captions 26
iv
-------
Abbreviations and symbols
AFM Atomic force microscopy
ca Analyte concentration in aqueous sample (g mL"1)
CAPS 3-(cyclohexylamino)propanesulfonic acid
CDF Cumulative distribution function
CE Capillary electrophoresis
cp Particle concentration in aqueous sample (mL"1)
DLS Dynamic light scattering
EDS Energy dispersive X-ray spectrometry
EHS Environmental health and safety
sn Nebulization transport efficiency (dimensionless)
ENM Engineered nanomaterials
EPA Environmental Protection Agency
ESD Environmental Sciences Division
FFF Field flow fractionation
Flow-FFF Flow field flow fractionation
HDC Hydrodynamic chromatography
ICPMS Inductively coupled plasma mass spectrometry
IRZ Initial radiation zone
m;Lp Analyte element mass in the particle (g)
NAZ Normal analytical zone
n. Number of analyte ions detected (unitless number)
NOM Natural organic matter
NTA Nanoparticle tracking analysis
qu Ionized analyte flux (s"1)
qp Particle flux (s"1)
qs Sample uptake rate (mL s"1)
rg Radius of gyration
rh Hydrodynamic radius
SDS Sodium dodecyl sulfate
Sed-FFF Sedimentation field flow fractionation
SEM Scanning electron microscopy
SLS Static light scattering
SP-ICPMS Single particle - inductively coupled plasma mass spectrometry
td Dwell time (s)
TEM Transmission electron microscopy
tp Plume pulse transit time (s)
v
-------
Acknowledgments
I am grateful to PerkinElmer for providing the DRCe inductively coupled plasma mass
spectrometer used for some of this research. Charlita Rosal of EPA developed and performed
the capillary electrophoresis separation shown in this report.
vi
-------
Introduction
The National Nanotechnology Initiative defines nanotechnology as "the understanding and
control of matter at dimensions between approximately 1 and 100 nanometers"(Council 2010),
and nanomaterials are defined as materials with at least one characteristic dimension in this
range. Nanomaterials can be natural (e.g., humic and fulvic acids) (Thurman and Malcolm
1981), incidental to human activity (e.g., diesel emissions, welding fumes) (Peters, Elzey et al.
2009), or engineered. Engineered nanomaterials (ENMs) exhibit optical, electrical, and chemical
characteristics different from either their bulk or dissolved forms. Because of this, they are the
source of the so called nanotechnology revolution. Most ENMs can be divided into two general
classes, depending on whether they are carbon-based (e.g., carbon nanotubes and fullerenes) or
metal-containing (e.g., Ag, Ti02, CeC>2, Fe).
Uses of engineered nanomaterials.
The use of ENMs in consumer, industrial, and agricultural products, as well as in
environmental technology is rapidly increasing (Ponder, Darab et al. 2001; Chaudhry, Scotter et
al. 2008; Nanotechnologies 2010). This is largely due to the unique optical, electrical, and
chemical properties of nano-scale particles. Often, the benefit of using nanomaterials stems from
the increased surface area per unit mass of material, which increases with the inverse of the
diameter in the case of spherical particles. This results in faster rates of chemical reactions such
as oxidative catalysis that occur at surfaces. Sometimes the benefit of nanomaterials arises from
the quantum nature of energy states at the nanometer scale, as in the wavelength tuning of the
fluorescence of quantum dots (Michalet, Pinaud et al. 2005). In the health sciences, the ability of
ENMs to bind to cell walls can be utilized for drug delivery (Lai, Trewyn et al. 2003; Zhang,
Pornpattananangkul et al. 2010).
The Project on Emerging Nanotechnologies maintains a database of consumer products that
manufacturers claim contain ENMs (Nanotechnologies 2010). Currently the database contains
over 1000 products. Consumer products containing nano-scale silver dominate current ENM
usage (Table 1). Note that of the carbon-based ENMs, nearly % of the products are durable
goods that incorporate carbon nanotubes in the construction material. The nanotubes are bound
in this matrix for the lifetime of the product. Conversely, the metal-containing ENMs are most
often used in dispersive applications, where they are intentionally released from the product or
where incidental release is substantial. For example, fabrics containing nano-scale silver release
silver during washing at varying rates, depending on the type of fabric and the washing
conditions (Benn and Westerhoff 2008; Geranio, Heuberger et al. 2009). Of the products using
carbon-based ENMs only 7 very low-volume uses of fullerenes are documented. Although there
is only one product listed using cerium oxide, it is as a diesel fuel additive that potentially could
entail high-volume dispersive use.
1
-------
Environmental implications of increased use of engineered
nanomaterials
The unusual properties that result in unique benefits from the use of ENMs have also elicited
intense interest in the environmental behavior of these materials. The environmental transport,
transformation, fate, exposure potential, and effects of ENMs cannot be predicted by the
behavior of either the corresponding dissolved or neat chemicals. In addition, the metrics that
affect all these environmental properties are different and more complex than those of
conventional pollutants. The number of funded research projects on the environmental health
and safety (EHS) implications of ENMs (Figure 1) increased dramatically from 1998 to 2007
(Nanotechnologies 2010). There has been some decline in newly initiated projects in the past
two years (although data for 2010 are incomplete); nonetheless, research into EHS of ENMs
remains intense. The level of research activity is roughly evenly apportioned to carbon-based
and metal containing ENMs. For the former, there are currently about 54 investigations of
nanotubes and 29 of fullerenes. Of the metal-containing ENMs, most investigations involve
titanium dioxide and iron, followed by quantum dots, silver, and then the remaining metal
oxides. The interest in carbon nanotubes stems from the known toxicity of these materials that is
mediated by the particle aspect ratio. Research focus on the metal-containing ENMs is prompted
by their high likelihood to be released from products. This leads to a greater potential loading of
the metal-containing ENMs in environmental compartments.
An essential capability for any successful research into the environmental behavior of ENMs
is a set of metrology tools for the several metrics of the materials that control their release
transport, transformation, fate, and effects. These tools would also be required for determining
the occurrence and distribution of ENMs in the environment, and they would permit the temporal
trends in these to be assessed. A previous report by the Environmental Sciences Division (ESD)
described the importance of characterization of metal-containing ENMs in the exposure research
of these potential environmental stressors (Heithmar 2009). It discussed the several methods
currently available for the characterization of various exposure metrics and why the current
methods are inadequate. The report introduced single particle - inductively coupled plasma
mass spectrometry (SP-ICPMS) as a method for both screening-level and selective determination
of metal-containing ENM dispersions. The potential of SP-ICPMS to become the first practical
analytical method for characterization of metal-containing ENMs in environmental matrices was
demonstrated.
This report describes recent progress in the development of SP-ICPMS, which can
simultaneously quantify the concentration of nanoparticles containing an analyte metal and
measure the metal mass in individual nanoparticles. The report briefly reviews the current
metrology methods discussed in the previous report and introduces two not covered there. The
introduction concludes with a conceptual description of SP-ICPMS, with a real analytical
example, and the equations and associated assumptions to calibrate the method. The
experiments discussed in this report first assess the accuracy of SP-ICPMS over a range of
nanoparticle sizes. The utility of the technique in a number of applications is then examined.
2
-------
Nanoparticles containing the analyte element can be measured accurately and precisely in the
presence of a 10,000-fold greater concentration of other nanoparticles. SP-ICPMS is used to
assess transformation of nanoparticles induced by changes in ionic strength. Limitations in
sampling and sample handling of real environmental water samples for nanoparticle
measurements are demonstrated. A screening-level assessment of metal-containing
nanoparticles in urban run-off using SP-ICPMS is demonstrated. Finally, preliminary studies of
coupling SP-ICPMS with on-line nanoparticle size separation methods are presented.
Rationale for the development of metrology tools for metal-containing
ENMs
A 2009 workshop of approximately 50 scientists involved in EHS research of ENMs was
sponsored by the International Council on Nanotechnology (ICON). The charge of the workshop
was to identify and rank the key research priorities for informed decisions on developing ENMs
with minimal potential adverse environmental impacts. The workshop report (Alvarez, Colvin et
al. 2009) identified fourteen research priorities, which were ranked according to two criteria: (1)
the importance of each issue in advancing our ability to design ENMs responsibly, and (2) the
current development of scientific understanding of the issue. Tools to detect, measure and
characterize ENMs ranked highest in importance. This area was also determined to be the most
poorly developed. Consequently, development of metrology tools, especially for real
environmental matrices, was the highest ranked research priority.
The first reason for the critical need for research to develop metrology tools is their
central role in every area of EHS research. These areas include (1) measuring the environmental
occurrence and distribution of ENMs, (2) determining temporal trends in environmental loads,
(3) laboratory studies of transport, transformation, and fate, and (3) toxicity studies. Ultimately,
exposure models must be developed for ENMs (Mueller and Nowack 2008), and these will
require measuring releases for source terms, as well as measuring ENM distributions in the
environment to verify the models.
Another reason for developing ENM metrology tools is the complex set of metrics. For
conventional pollutants, mass concentration is often the only relevant metric for determining
exposure. In the case of ENMs, exposure potential is affected by particle size. Depending on the
exposure model, measurement of size distributions in terms of particle mass, volume, or
equivalent hydrodynamic radius may be most relevant (Hassellov, Readman et al. 2008).
Specific surface area (area/mass ratio) is important to the catalytic activity of ENMs such as
TiC>2, and it can influence toxicity (Schulte, Geraci et al. 2008). Surface charge and the related
property of C, potential can determine the degree ENMs tend to aggregate (Kim, Lee et al. 2008).
As previously discussed, particle shape, characterized by aspect ratio, can dramatically affect
toxicity of carbon nanotubes (Magrez, Kasas et al. 2006; Takagi, Hirose et al. 2008).
A third factor fueling the high interest in developing ENM metrology tools is the current
lack of many practical methods. There are a number of techniques for measuring concentrations
and size distributions of ENMs in pure suspensions used as starting materials in laboratory
3
-------
studies. However, very few methods exist for detection, quantification, and characterization of
suspensions after they are introduced into test systems. Such techniques are critical, because the
ENM can transform in the test system (Alvarez, Colvin et al. 2009). Going one step further in
complexity, virtually no practical methods have been published for measuring concentration and
size distributions of ENMs in real environmental samples (Handy, von der Kammer et al. 2008).
There are two reasons for this situation. First, concentrations of ENMs in the real environment
are very low (Benn and Westerhoff 2008). Compounding this low concentration is the presence
of natural colloids that interfere (Klaine, Alvarez et al. 2008), especially in the case of metal-
containing ENMs. Detection and quantification of metals usually involves measuring total
element concentration, and natural colloids often include minerals that contain the same element
as the ENM of interest. In addition, analyte elements can also adsorb on natural organic matter
(NOM) in colloids.
Advances in understanding the environmental behavior of metal-containing ENMs can
only be made if detection, quantification, and characterization methods are developed, especially
for the concentration and size distribution metrics of the nanomaterials. In the short term, these
methods must be effective at least in laboratory test systems, and at least screening-level
methods must be available to detect the probable presence of metal-containing ENMs in real
environmental samples. Ultimately, selective determinative methods that greatly minimize false
positives in real-world samples must be developed. The development of characterization
methods for metrics other than nanoparticle concentration and size distribution may never be
attainable for these systems.
Methods for characterizing engineered nanomaterials
Methods that are currently available to measure several common ENM characterization
metrics (Table 2, adapted from Heithmar 2009a) generally work well in simple matrices. One
general class of characterization techniques is single particle imaging. The most common of
these methods are scanning electron microscopy (SEM) or transmission electron microscopy
(TEM) (Lin and Yang 2005; Pyrz and Buttrey 2008). In well controlled systems, atomic force
microscopy (AFM) can be advantageous, because interaction forces between ENMs and
substrates can be studied (Ebenstein, Nahum et al. 2002). When combined with X-ray emission
spectroscopy, usually in the energy dispersive mode (EDS), SEM or TEM can definitively
identify ENMs in a simple sample matrix. However, particle concentration must be generally at
least 109 mL"1 with current sampling methods, making the use of this method for
environmentally relevant concentrations of ENMs infeasible. Representative sampling of the test
material is generally not attainable, so quantification is impossible. In environmental samples,
distinguishing metal-containing ENMs from colloids containing the same metal is difficult
(Tiede, Hassellov et al. 2009). Finally, sampling methods can introduce artifacts, especially in
the size distribution of ENMs (Tang, Wu et al. 2009).
An alternative to single-particle imaging for characterizing nanoparticle concentration
and size distribution is the general class of ensemble measurement techniques, in which a signal
produced by a sample population of particles is measured. Several light scattering methods are
4
-------
among the ensemble approaches, and they can provide representative size distribution
characterization at relatively low concentrations (ca. 107 -108mL"'), Dynamic light scattering
(DLS), also called photon correlation spectroscopy, measures the decay of the autocorrelation of
laser light scattering to calculate the hydrodynamic radius (rh) distribution (Filella, Zhang et al.
1997). DLS combined with applying an oscillating electric field also can measure C, potential,
which is related to the surface charge of an ENM. Static light scattering (SLS), which measures
the continuous scatter signal at various scattering angles, provides the complementary size
characteristic of radius of gyration (rg) (Kammer, Baborowski et al. 2005). The ratio rg/rh can
provide an estimation of ENM shape (Schurtenberger, Newmen et al. 1993). Unfortunately, both
these ensemble laser light scattering methods are prone to errors in widely polydisperse
suspensions, because scatter signal is exponentially related to size. Nanoparticle tracking
analysis (NTA) is a single particle light scattering technique that does not suffer this limitation.
None of the light scattering techniques are effective in complex media with multiple particle
types, because they do not measure particle elemental composition.
Field flow fractionation (FFF) has become popular for physical separation of
nanoparticles of different sizes. It can be used with an ensemble detector, such as inductively
coupled plasma mass spectrometry (ICPMS), that can provide elemental concentration for each
size fraction. Therefore, the FFF-ICPMS combination is applicable in complex systems. FFF is
actually a general technique with several implementations. All the FFF methods use the
application of a laminar flow of eluent through a narrow channel combined with an orthogonal
force field (hence the term field flow). In most implementations of FFF, the orthogonal force
field results in bigger particles having an average position closer to the channel wall than smaller
particles, resulting in a lower average velocity for the bigger particles, and a resulting size
separation at the channel outlet. The two commonly used FFF techniques for ENM
characterization are flow field (flow-FFF) (Stolpe, Hassellov et al. 2005; Lesher, Ranville et al.
2009) or, to a lesser extent, a sedimentation (gravity) field (sed-FFF) (Taylor, Garbarino et al.
1992; Hassellov, Lyven et al. 1999). Flow-FFF separates nanoparticle sizes by hydrodynamic
radius, while sed-FFF separates by buoyant mass. Flow-FFF combined with ICPMS detection
has recently been used to study the adsorption of uranium ions on iron oxide nanoparticles
(Lesher, Ranville et al. 2009).
Another technique for physically separating nanoparticle size fractions that has recently
been applied to ENMs is hydrodynamic chromatography (HDC) (Tiede, Hassellov et al. 2009).
HDC can be coupled on-line with ICPMS to provide an ensemble characterization method
(Tiede, Boxall et al. 2010). In HDC, the sample is passed through a column of inert material
such as silica with a mobile phase. The inter-particle channels in the stationary phase approach
the size of the nanoparticles, so that nanoparticles with larger hydrodynamic radius experience a
higher average velocity of the laminar flow of mobile phase than smaller particles. Thus, larger
particles elute before smaller ones. HDC potentially has less likelihood for artifacts due to
interactions with the stationary phase than FFF methods, which use semi-permeable polymer
membranes. On the other hand, FFF provides separations of higher size resolution.
Capillary electrophoresis (CE) has recently been applied to separation of engineered
nanomaterials (Liu and Wei 2004; Liu, Lin et al. 2005). CE has the advantage of relatively fast
5
-------
analysis compare with FFF and HDC, and very good theoretical resolving power. It separates
nanoparticles based on electrophoretic mobility (Petersen and Ballou 1992; Huff and Mclntire
1994; Quang, Petersen et al. 1996). Therefore, nanoparticle charge and size both affect retention
time. This can lead to complications in data interpretation.
Ensemble methods based on size separation coupled on-line with elemental detection
(i.e., hyphenated methods) provide elemental information in addition to size distribution. This
level of specificity makes them effective tools in moderately complex systems like laboratory
test media. However, they can only give screening-level characterization of real environmental
samples. They provide the information on total analyte element content associated with different
size particles. Therefore, they cannot distinguish a large concentration of nanoparticles, each
containing a small fraction of analyte element, from a lower concentration of the same size
particles, each with a high analyte element fraction. For example, a large background of natural
organic matter of 20 nm hydrodynamic radius with adsorbed silver ions cannot be distinguished
from a low concentration of 20 nm silver nanoparticles.
The lack of any practical methods for detecting, quantifying, and characterizing metal-
containing ENMs in environmental media has resulted in the adaptation of a single particle
technique from aerosol analysis and colloid chemistry. This method, as modified in our
laboratory for aqueous suspensions of metal-containing ENMs, is called single particle-
inductively coupled plasma mass spectrometry. A related technique was originally developed to
characterize airborne particulates (Nomizu, Nakashima et al. 1992; Nomizu, Hayashi et al.
2002). That technique was later adapted for characterizing zirconia colloids (Degueldre,
Favarger et al. 2004), thorium oxide (Degueldre and Favarger 2004), and gold particles
(Degueldre, Favarger et al. 2006). Recently, the technique was applied to characterize silver
nanoparticles in municipal waste water (Monserud, Lesher et al. 2009), and preliminary results
of the technique coupled with flow-FFF have been presented (Hassellov 2009). ESD presented a
study of the performance of SP-ICPMS and how ICPMS acquisition and plasma parameters
control performance (Heithmar 2009).
SP-ICPMS relies on the fact that metal-containing nanoparticles entering an ICPMS
plasma produce intense ion plumes of the metal isotopes in very short time periods (< 1 ms). If
the ICPMS signal is monitored very fast (<10 ms data points), any moderate background from
dissolved analyte metal, or plasma matrix ions (Lam and Horlick 1990), diminishes to an average
count of less than 1 ion detected per data point. The ion plume pulses from the nanoparticles are
easily distinguished. The frequency of the pulses is proportional to the nanoparticle
concentration in the sample, and the number of ions detected in each pulse is proportional to the
analyte mass in the nanoparticle creating the pulse.
SP-ICPMS is a single particle rather than an ensemble method, so it analyzes individual
nanoparticles (Table 3). SP-ICPMS does not measure particle size, but rather analyte mass for
each particle and the nanoparticle concentration. The hyphenated methods do not give
information on individual nanoparticles, but rather on each size fraction of nanoparticles as a
group. One other difference between SP-ICPMS and the hyphenated methods is that the former
measurement is made on a time scale of several seconds to a minute depending on the
6
-------
concentration, while the latter typically require 10 - 30 minutes. It can be seen that SP-ICPMS
and the hyphenated approaches are complementary, and preliminary results of coupling the two
will be discussed in this report.
Theory of SP-ICPMS
A detailed description of ICPMS processes in the conventional mode as well as in the SP
mode was given in our previous report (Heithmar 2009). This report gives a conceptual
description of the techniques with real analytical results to demonstrate the concepts.
In conventional ICPMS, an aerosol of droplets is produced from the sample usually using some
form of pneumatic nebulizer (Taylor 2001). The larger aerosol droplets are removed by
collisions within a spray chamber before a tertiary aerosol of droplets enters the plasma. The
flux of droplets is typically greater than 106 s"1. Each droplet contains a few hundred to a few
thousand analyte atoms. The efficiency of the sampling process from aqueous sample to aerosol
flux in the plasma is quantified by the nebulization transport efficiency, sn, which is generally
between 2% and 30%, depending on the nebulizer and spray chamber used. The aerosol is
evaporated in the preheating zone of the plasma, within the load coil region. The salt residues
containing the analye begin to vaporize in this region. The salts are atomized in the initial
radiation zone (IRZ), which extends from the load coil several mm (Koirtyohann, Jones et al.
1980). Ionization occurs in the normal analytical zone (NAZ) (Thomas 2004). The
vaporization, atomization, and ionization processes are usually >80% efficient for most elements
(O'Connor and Evans 1999).
In conventional ICPMS, the above processes result in a fairly constant flux of analyte ions
reaching the detector, due to the large number of droplets and the small number of analyte atoms
dissolved in each droplet. The resulting ICPMS signal for 30 pg/mL dissolved gold with a short
measurement window (known as the dwell time, td) of 10 ms is very low and somewhat noisy
(Figure 2a). Therefore, a longer measurement window is usually used in conventional ICPMS
(Figure 2b). The signal for the 30 pg/mL dissolved gold is roughly constant at about 80 s"1.
When the same 30 ng of gold in 50 nm nanoparticles are suspended in each mL of sample,
very few aerosol droplets contain any gold, and a few contain ca. 4 x 106 gold atoms (one
nanoparticle). The resulting signal (Figure 3a) for the conventional long td (2 s in this example)
is somewhat noisier than for dissolved gold, which is expected by the episodic nature of the gold
flux. The average signal intensity is the same as for the same concentration of dissolved gold
(Figure 2b). Each gold nanoparticle produces a plume of millions of ions that enter the mass
spectrometer interface over a period of less than 500 |is (Gray, Olesik et al. 2009; Heithmar
2009). In SP-ICPMS, this discrete nature of the ion signal is exploited by using very short dwell
times. With td = 10 ms, each gold nanoparticle ion plume reaching the detector contains about
30 ions (the rest of the ions entering the interface are lost in the spectrometer), producing a
corresponding pulse (Figure 3b). The number of ions detected in each pulse is directly
proportional to the gold mass in the corresponding single nanoparticle. The frequency of pulses
is proportional to the concentration of nanoparticles in the sample (in this case, ca. 2 x 104 mL"1).
7
-------
This theory of SP-ICPMS that results in an easily calibrated signal relies on two
assumptions. First, every nanoparticle that reaches the plasma is detected. This also depends on
a sufficiently long residence time, so the ion plume expands enough to substantially fill the cross
section of the central channel of the plasma. If so, equation 1 is valid.
(1) qp/cp = qs8n,
where qp = flux of particles detected in plasma (s"1), cp = concentration of nanoparticles
containing the detected metal in the sample (mL"1), qs = sample uptake rate (mL s"1), and sn =
nebulization efficiency (dimensionless). Note that qs and sn are properties of the ICPMS
instrument conditions, and independent of the element. Therefore, equation 1 can be used to
calculate their product using any type of nanoparticle suspension of known cp.
The second assumption of the SP-ICPMS theory is that ICPMS sensitivity is constant for
an analyte, irrespective of whether it is dissolved or contained in nanoparticles. Again, this
requires that the residence time in the plasma to be sufficiently long. If so, equation 2 is valid.
(2) ma.p [qs l'-n Ca / C|i.a | 11 l p k Hi p-
where ma p = mass of analyte element in a single nanoparticle (g), n, p = number of ions of analyte
element detected in the corresponding plume (number of ions detected in a single SP-ICPMS
pulse - see Figure 3 b), ca = the analyte concentration in a dissolved standard of the analyte (g
mL"1), qLa = ion flux measured for the dissolved standard (s"1). For each analyte element,
calibration of the element mass in individual particles (calculation of the response factor k)
requires only the qs sn product from Equation 1 and analysis of a known concentration of
dissolved analyte element using a conventional ICPMS standard).
Equation 2 provides calibration of nanoparticle element mass. The calibration of
nanoparticle concentration is provided by rearrangement of equation 1 for any unknown
nanoparticle suspension, once qs sn has been determined:
(3) cp qp / qs sn.
8
-------
Experimental
Most of the materials, instrumentation and methods used in the present study are found in
our previous report (Heithmar 2009). Only those added or modified are described here.
Materials
Sodium dodecyl sulfate (SDS), Triton X-100, and 3-(cyclohexylamino)propanesulfonic
acid (CAPS) were obtained from Sigma-Aldrich, St. Louis, MO. Calcium chloride was obtained
from EMD Chemicals, Darmstadt, Germany. Syringe filters (0.45 [j.m and 5.0 [j,m nylon) were
obtained from Sterlitech, Kent, WA. Urban runoff water for SP-ICPMS experiments were
obtained from Las Vegas Wash, Las Vegas, NV (36° 6.826' N, 115° 8.741' W, ca. 2 km east of
Las Vegas Blvd.).
Instrumentation
ICPMS analyses were performed on both the DRCe (PerkinElmer, Waltham, MA)
described previously and on a 7500ce (Agilent, Santa Clara, CA), with an upgraded 7500cx lens
system. HDC-SP-ICPMS experiments were conducted with a hydrodynamic chromatography
column (5-300 nm size range, Polymer Laboratories, Shropshire, UK) with either an Agilent
1200 HPLC pump or an 100DM syringe pump (ISCO, Lincoln, NE). CE-ICPMS experiments
were conducted with a MDQ capillary electrophoresis instrument (Beckman-Coulter, Brea, CA).
Methods
Because of limitations in software, SP-ICPMS measurements on the 7500ce with dwell
times less than 10 ms had to be acquired by entering multiple elements, so the total integration
time was at least 10 ms.
Silver nanoparticle transformation studies were conducted on either 500 ng/mL (4.5 x 108
mL"1 particle concentration) or 0.25 ng/mL (2.2 x 105 mL"1) silver nanoparticle suspensions in
water. The samples were spiked to final concentration of 200 mM CaCh. Subsamples were
taken at designated times after spiking and diluted to 1 x 104 mL"1 for SP-ICPMS analysis.
Urban runoff water (10 mL) was collected in polypropylene syringes and filtered on-site
with 5.0 [^m syringe filters. One sample was spiked with gold and silver nanoparticle
suspensions prior to filtering. Samples were analyzed by SP-ICPMS within 36 hours, usually
within three hours.
HDC separations were conducted using 10 mM SDS buffer optimized for the
nanomaterial. Triton X-100 (0.1%) was added in some experiments. Eluent flow was 1.0
9
-------
mL/min. The column eluent was split 60/40 with a tee and a PEEK capillary tube to waste in
order to match the optimum uptake rate of the nebulizer (400 |iL/min).
The 30 and 50 nm gold nanoparticle CE separation was conducted in 75 |im ID x 80 cm
fused capillary with 40 mM SDS and 10 mM CAPS, pH 7.3. The SP-ICPMS dwell time was 10
ms.
Results and Discussion
Assessment of SP-ICPMS accuracy
The validity of the SP-ICPMS theory as described in this report was evaluated by
applying it to a series of gold nanoparticles (30, 50, 80, and 200 nm nominal diameter). A
suspension of 30 nm gold at 2 x 104 mL'1 was analyzed by SP-ICPMS with a 10 ms dwell time to
calculate qs snfrom Equation 1, and a 1.0 ng/mL dissolved gold ICPMS standard was analyzed to
calculate k in Equation 2. This was then applied to the SP-ICPMS responses (mean nLp of
suspensions of the four nanoparticle standards). The resulting ma.p values were used to calculate
the theoretical diameters of corresponding spherical gold nanoparticles. The linear least square
fit of the calculated diameters from this procedure against the nominal values (Figure 4) is:
dcalculated = 1-00 dnominal + 0.73; (r2 = 0.9989).
The agreement between the SP-ICPMS calculated diameters and the nominal values
demonstrates that the SP-ICPMS theory is valid for gold nanoparticles up to at least 200 nm. At
some point, larger particles will not vaporize, atomize and ionize with the same efficiency as
dissolved analyte, or their ion plume will not be collected with the same efficiency. At that
point, calculated m^ values will be negatively biased, so quantitative analysis by SP-ICPMS
should be evaluated with standards of the largest nanoparticles expected. When nanoparticle
standards of well characterized size distribution are not available for a given material, SP-ICPMS
should be considered a semi-quantitative technique for measuring particle elemental mass.
There are published studies of the behavior of ion populations in the interface region of
the ICPMS mass spectrometer (Macedone, Gammon et al. 2001; Mills, Macedone et al. 2006;
Farnsworth, Spencer et al. 2009) and of ion clouds in the plasma produced by single particles
(Hobbs and Olesik 1993; Hobbs and Olesik 1997; Gray, Olesik et al. 2009). These studies
present the vaporization and ionization, as well as the transmission of ions through the interface,
as complex processes influence by many factors. They might be used to hypothesize that the
assumptions made by the SP-ICPMS theory are suspect. However, the results of our
investigation demonstrate that the assumptions are valid, at least for gold nanoparticles up to 200
nm diameter.
10
-------
Precision and selectivity of SP-ICPMS
The precision of SP-ICPMS was assessed by replicate analyses of 50 nm gold
nanoparticle suspensions (4 x 103 mL"1; 5.0 pg/mL). No significant differences in the slopes or
positions of the cumulative distribution functions (CDFs) of the detected ion pulses (Fgure 5) are
detectable. The relative standard deviation of 2.4% for the mean ion counts reflects good
precision of the calculated mean particle element mass (Table 4). The relative standard deviation
of the pulse count (11%) is consistent with Poisson statistics for a mean of 101, the best precision
attainable for the nanoparticle concentration metric.
The selectivity of SP-ICPMS in complex matrices was evaluated by spiking the
suspension analyzed in the last two replicates with 60 nm silver nanoparticles at a concentration
of 5 x 107 mL"1. Neither the mean particle gold mass nor the gold nanoparticle concentration
metric were affected by a greater than 104 ratio of silver to gold nanoparticles (Table 4).
Nanoparticle transformation study
There have been several studies of changing size distributions of metal-containing
nanoparticles as produced by changes in matrix (Kapoor, Lawless et al. 1994; Peukert,
Schwarzer et al. 2005; Shim and Gupta 2007; Baalousha 2009). Increased ionic strength is one
factor that often induces greater degree of aggregation, usually attributed to a collapse of the
double layer and a decrease of the associated electrostatic repulsion (Domingos, Tufenkji et al.
2009; Gilbert, Ono et al. 2009). SP-ICPMS is a potentially powerful tool for studying
nanoparticle transformation processes. Because of the speed of SP-ICPMS analyses, fast
transformation processes can be followed. The sensitivity of SP-ICPMS could allow the first
studies of transformations at environmentally relevant concentrations.
SP-ICPMS was used to analyze suspension of 60 nm silver over several minutes after the
ionic strength was increased by addition of CaCl2 to a final concentration of 200 mM. The
original nanoparticle concentration was similar to those used in most transformation studies to
date (4.5 x 108 mL"1 or 500 ng/mL). The resulting CDFs at 0 (unspiked), 8, and 14 minutes after
CaCh addition shift progressively to the right (higher ion counts per pulse) with time (Figure 6).
This reflects the expected aggregation to larger particle clusters.
A similar experiment was conducted at a concentration of silver nanoparticles that might
be found in a wastewater effluent (2.2 x 105 mL"1 or 0.25 ng/mL) (Benn and Westerhoff 2008).
The transformation was followed over two days, reflecting the slow transformation kinetics
expected at low concentration. The CDFs at 0, 1, and 49 hours indicate that at least 85% of the
dwell times containing any ion signal produced only 1 or 2 ion counts (Figure 7). This result
indicates either dissolved silver or silver nanoparticles of less than about 30 nm diameter. Silver
nanoparticles in this experiment appear to dissolve at environmentally relevant concentrations.
11
-------
The portion of the CDFs with particle mass rank >0.85 shift progressively to the left over time,
indicating that the dominant process is a continued dissolution, rather than aggregation. The
small number of large nanoparticles (ion counts >200 per particle) does increase over time,
indicating slight competing aggregation. This very preliminary study is the first of silver
nanoparticle size transformations at environmentally relevant concentrations. It is made possible
by the high particle concentration sensitivity of SP-ICPMS.
Screening urban runoff using SP-ICPMS
SP-ICPMS was applied to a small screening-level assessment of metal-containing
nanoparticles in a surface water consisting of urban runoff (Las Vegas Wash). Water samples
were to be filtered for two reasons. First, it was hoped that a size cutoff of about 0.45 |im could
be achieved that would produce data that reflected true nanoparticle content. Second, unfiltered
water could contain larger particles (>5 |im) that could potentially clog the nebulizer. The effect
of filtration on nanoparticle recovery was studied.
A reagent water suspension of 50 nm diameter gold nanoparticles (1 x 104 mL"1) before
and after filtration through a 0.45 |im nylon filter (Figure 8a and 8b, respectively) was analyzed
by SP-ICPMS. Although the nominal cut-off of the filter is 10 times the diameter of the
nanoparticles, the filter retains virtually all the nanoparticles. The CDFs of the unfiltered (Figure
8c) and filtered (Figure 8d) suspensions show that only a few smaller nanoparticles (< 30% of
the original mean particle mass) were recovered in the filtrate. The poor filtrate recovery may
indicate adsorption of the nanoparticles on the membrane material.
Recovery studies with reagent water suspensions of 50 nm gold and 60 nm silver showed
that nearly 100% recovery was obtained in the filtrate using 5.0 |im cut-off nylon filters, so Las
Vegas Wash water samples were filtered in the field with these filters. One sample was spiked
before filtration with 50 nm gold and 60 nm silver nanoparticle suspensions. Even accounting
for a 10:1 dilution of silver nanoparticles in the spiked sample, the silver SP-ICPMS analysis of
the spiking suspension (5 x 104 mL"1 - Figure 9a) and the 5.0-|im filtered spiked Las Vegas
Wash water sample (5 x 103 mL"1 - Figure 9b) demonstrate poor recovery. In addition, the
CDFs of the spiking suspension (Figure 9c) and the 5.0-|im filtered spiked sample (Figure 9d)
show a large bias toward small nanoparticles in the filtered sample. The poor recovery in the
real water samples could be due to either adsorption of nanoparticles on larger suspended solids
in the spiked sample or a layer of solids on the filter membrane blocking the passage of
nanoparticles. An investigation of the mechanism of the poor nanoparticle recovery and of
possible approaches to ameliorate the problem is needed. For the preliminary demonstration
study reported here, 5.0-|im filtered samples were analyzed to protect the nebulizer, despite the
poor recovery. No attempt at correcting results for recovery was made.
Silver-bearing particles are present in the Las Vegas Wash water (Figure 10). The pulse
heights indicate a Ag mass equivalent to about 50-nm diameter metallic Ag particles. The
particle concentration is approximately 200 mL"1, not correcting for the low recovery. By
12
-------
contrast, the recovered gold-bearing particle concentration may be on the order of 10 mL'1
(Figure 10).
Approximately 400 particles per mL of titanium-bearing particles above an equivalent
Ti02 diameter of about 60 nm are present in Las Vegas Wash water samples (Figure 11). The
largest particle is equivalent to about 400 nm Ti02.
The SP-ICPMS analysis of Las Vegas Wash water for the lanthanoid elements and
yttrium (Figure 12) show the presence of cerium-bearing particles. It is likely that the source is
natural, given the presence of the other lanthanoids and yttrium which tend to occur with cerium
in minerals.
SP-ICPMS analysis of Las Vegas Wash water at m/z 57, which is the isotope commonly
used for iron determination, was conducted at a 10 ms dwell time (Figure 13, upper trace). The
continuous signal is largely due to ArOH+ plasma background. Significant pulses above the
background are hard to distinguish. SP-ICPMS analysis with a 3 ms dwell time demonstrates the
improved signal-to-noise in SP-ICPMS as dwell time decreases (Figure 13, lower trace). In SP-
ICPMS, background is proportional to dwell time, while signal is independent of dwell time, as
long as the latter is much longer than the plume pulse transit time, tp (Heithmar 2009a). The
background signal in Figure 13 is decreased proportionally to the 3/10 dwell time ratio. The
background is apparently dominated by flicker noise rather than shot noise, because the
background fluctuation appears to decrease nearly as much as the mean background, rather than
the square-root relationship produced by shot noise. Significant pulses above the 3 ms
background are obvious, indicating iron-bearing particles.
Coupling SP-ICPMS with separation methods
A preliminary investigation of the coupling of SP-ICPMS with HDC separations was
conducted. Initial results indicated significant background interference in the SP-ICPMS of
gold. SP-ICPMS analysis of SDS eluent from the Agilent reciprocating piston pump (no HDC
column) detected numerous background pulses that precluded analysis of HDC chromatograms
(Figure 14). It was considered unlikely that the background pulses were due to gold. Tantalum,
with a single isotope at m/z 181, is known to form oxide polyatomic ions (m/z 197). The
presence of tantalum-containing particles in the SDS eluent was confirmed by the SP-ICPMS
analysis at m/z 181 (Figure 14 b). The intensities of the m/z 197 background peaks to the m/z
181 peaks indicate a roughly 1-2% TaO+ formation relative to the Ta+. This is consistent with
expected values. Because the tantalum appeared to come from the HPLC pump, the
reciprocating piston pump was replaced with a syringe pump. It was hoped that the much slower
moving parts of the latter would produce fewer tantalum-bearing particles. The resulting m/z
197 and m/z 181 SP-ICPMS backgrounds revealed only two background peaks greater than 5
ions appear at m/z 197 over a 12 minute period (Figure 15). This level of background is
acceptable. Interestingly, the two peaks observed are likely gold, since none of the peaks at m/z
181 are large enough to have produced a pulse of these magnitudes, assuming 2% TiO+/Ti+.
13
-------
A suspension of 60 nm gold nanoparticles (50 pg/mL - 2.3 x 104 mL"1) was analyzed by
HDC coupled to the ICPMS. The m/z 197 chromatogram with 1 s integration time typical for
conventional hyphenated analysis includes a poorly defined peak around 760 s, indicating the
elution of 60 nm gold (Figure 16a). The small peak at about 830 s is near the elution volume of
dissolved gold. The chromatogram of an identical sample at the SP-ICPMS dwell time of 10 ms
(Figure 16b) exhibited a group of individual pulses with a mean ion count of about 40-50 at 760
s, consistent with a SP-ICPMS particle gold mass equivalent to 60 nm gold. Thus, the retention
time of the HDC peak provides information that the particle diameter is about 60 nm, while the
SP-ICPMS signal confirms that the individual particles contain a high fraction of gold. This
complementary information would eliminate the possibility in a real analysis that the signal at a
60 nm retention is due to nanoparticles with adsorbed gold or minerals with low gold content.
The HDC-ICPMS chromatogram (conventional 1 s integration windows) of a mixture of
60 nm gold nanoparticles (50 pg/mL - 2.3 x 104 mL"1) and 30 nm gold nanoparticles (25 pg/mL
- 9.2 x 104 mL"1) contains peaks around 770 s (50 nm gold) and 785 s (30 nm gold) (Figure 17a).
There is no complementary information in the chromatogram to confirm that composition. The
HDC-SP-ICPMS chromatogram (10 ms dwell time) includes individual pulses around 770 s
producing about 40-50 ions each, while those around 785 s produce about 6-7 ions (Figure 17b).
This is in excellent agreement with the ratio predicted by their nominal diameters of about 8:1.
One issue with HDC-ICPMS in either the conventional mode or the SP mode is the presence of
occasional spurious single particles (see for example the peaks at about 820, 840, and 850
seconds). The cause of these peaks is currently unclear. However, we suspect that they result
from contamination from earlier chromatographic runs using larger Au nanoparticles at much
higher concentrations ([j,g/mL).
A preliminary experiment coupling CE with SP-ICPMS was conducted with a suspension
of 30 nm gold (4 x 108 mL"1; retention time ca. 13 minutes) and 50 nm gold (9 x 107 mL"1 ;
retention time ca. 15 minutes). The 30 nm particle concentration in this experiment resulted in a
substantial fraction of dwell times with multiple nanoparticles, so the pulse ion-count ratio was
less than the expected 5:1 (Figure 18). Nevertheless, the 30 nm and 50 nm particles appear to be
nearly completely resolved.
Conclusions and Future Work
The experiments described in this report indicate the promise of SP-ICPMS as a viable
metrology tool for metal-containing ENMs. The theory of the technique is valid for gold
nanoparticles up to at least 200 nm diameter (83 femtogram mass). Particle element mass and
particle concentration measurements are precise and unaffected by large excesses of non-analyte
particles.
SP-ICPMS allows the study of nanoparticle size transformation at an environmentally
relevant concentration. The particle concentration sensitivity of other methods have previously
limited transformation studies to concentrations several orders of magnitude higher. The results
presented in this report indicate that, while aggregation is significant at high nanoparticle
14
-------
concentration (2.2 x 105 mL"1), dissolution is the dominant transformation process at a more
environmentally relevant concentration (4.5 x 108 mL"1).
SP-ICPMS is a convenient stand-alone method for screening urban runoff water for
particles containing substantial mass of a selected analyte element. The method does not
respond to high concentrations of particles with low element mass per particle, which complicate
analyses by hyphenated methods.
Sample handling artifacts, such as low recoveries of filtered samples, must be addressed
to allow reliable screening-level assessments of water containing suspended solids. Alternative
methods for removing large particles will be investigated. These methods include filtering with
large-area metal sieves and centrifugation.
The coupling of separation methods such as HDC or CE with SP-ICPMS provides more
selective analysis than either the separation technique with conventional ICPMS or SP-ICPMS
alone. Additional coupled techniques will be investigated. Coupling flow-FFF with SP-ICPMS
is particularly promising, given the high resolution of that method compared to HDC. Other
emerging techniques such as microfluidic oscillating electric field-FFF may be investigated in
collaboration with others.
The main limitation of the SP-ICPMS is its limited nanoparticle concentration dynamic
range using commercial ICPMS instruments. This limitation may be largely overcome using a
commercial instrument introduced this year that greatly reduces the dead time between
individual dwell times, allowing shorter dwell times that increase the upper limit of the dynamic
range. Implementation of SP-ICPMS on that instrument is planned soon.
15
-------
References
Alvarez, P. J. J., V. Colvin, et al. (2009). "Research Priorities to Advance Eco-Responsible
Nanotechnology." ACS Nano 3(7): 1616-1619.
Baalousha, M. (2009). "Aggregation and disaggregation of iron oxide nanoparticles: Influence of
particle concentration, pH and natural organic matter." Science of The Total Environment
407(6): 2093-2101.
Benn, T. M. and P. Westerhoff (2008). "Nanoparticle Silver Released into Water from
Commercially Available Sock Fabrics." Environmental Science & Technology 42: 4133-4139.
Chaudhry, Q., M. Scotter, et al. (2008). "Applications and implications of nanotechnologies for
the food sector." Food Additives & Contaminants: Part A: Chemistry. Analysis. Control
Exposure & Risk Assessment 25(3): 241 - 258.
Council, National Science and Technology (2010). The National Nanotechnology Initiative -
Research and development leading to a revolution in technology and industry.
Degueldre, C. and P. Y. Favarger (2004). "Thorium colloid analysis by single particle
inductively coupled plasma-mass spectrometry." Talanta 62(5): 1051-1054.
Degueldre, C., P. Y. Favarger, et al. (2004). "Zirconia colloid analysis by single particle
inductively coupled plasma-mass spectrometry." Analvtica Chimica Acta 518(1-2): 137-142.
Degueldre, C., P. Y. Favarger, et al. (2006). "Gold colloid analysis by inductively coupled
plasma-mass spectrometry in a single particle mode." Analvtica Chimica Acta 555(2): 263-268.
Domingos, R. F., N. Tufenkji, et al. (2009). "Aggregation of Titanium Dioxide Nanoparticles:
Role of a Fulvic Acid." Environmental Science & Technology 43(5): 1282-1286.
Ebenstein, Y., E. Nahum, et al. (2002). "Tapping Mode Atomic Force Microscopy for
Nanoparticle Sizing: Tipa"'Sample Interaction Effects." Nano Letters 2(9): 945-950.
Farnsworth, P. B., R. L. Spencer, et al. (2009). "A comparison of ion and atom behavior in the
first stage of an inductively coupled plasma mass spectrometer vacuum interface: Evidence of
the effect of an ambipolar electric field." Spectrochimica Acta Part B: Atomic Spectroscopy
64(9): 905-910.
Filella, M., J. Zhang, et al. (1997). "Analytical applications of photon correlation spectroscopy
for size distribution measurements of natural colloidal suspensions: capabilities and limitations."
Colloids and Surfaces A: Physicochemical and Engineering Aspects 120(1-3): 27-46.
16
-------
Geranio, L., M. Heuberger, et al. (2009). "The Behavior of Silver Nanotextiles during Washing."
Environmental Science & Technology 43(21): 8113-8118.
Gilbert, B., R. K. Ono, et al. (2009). "The effects of nanoparticle aggregation processes on
aggregate structure and metal uptake." Journal of Colloid and Interface Science 339(2): 285-295.
Gray, P. J., J. W. Olesik, et al. (2009). Particle Size Effects on Vaporization in Laser Ablation
Inductively Coupled Plasma Mass Spectrometry. Pittsburgh Conference on Analytical Chemistry
and Applied Spectroscopy, Chicago, IL.
Handy, R., F. von der Kammer, et al. (2008). "The ecotoxicology and chemistry of manufactured
nanoparticles." Ecotoxicology 17(4): 287-314.
Hassellov, M. (2009). Detection and Characterization of engineered nanoparticle in the
enviroment using Field-Flow Fractionation coupled to ICP-MS in single nanoparticle detection
mode. Goteborg Setae Europe.
Hassellov, M., B. Lyven, et al. (1999). "Sedimentation Field-Flow Fractionation Coupled Online
to Inductively Coupled Plasma Mass SpectrometryNew Possibilities for Studies of Trace Metal
Adsorption onto Natural Colloids." Environmental Science & Technology 33(24): 4528-4531.
Hassellov, M., J. Readman, et al. (2008). "Nanoparticle analysis and characterization
methodologies in environmental risk assessment of engineered nanoparticles." Ecotoxicology
17(5): 344-361.
Heithmar, E. M. (2009a). Characterization of metal-containing nanoparticles in water, U.S.
Environmental Protection Agency.
Heithmar, E. (2009b). "Characterizing Metal-based Nanoparticles in Surface Water by Single-
Particle ICPMS." 237th ACS National Meeting, Salt Lake City, Utah.
Heithmar, E. (2009c). "Single particle ICPMS for characterizing metal-based nanoparticles and
monitoring transformation processes in surface water." SET AC Europe Meeting, Gothenburg,
Sweden.
Hobbs, S. E. and J. W. Olesik (1993). "The effect of desolvating droplets and vaporizing
particles on ionization and excitation in Ar inductively coupled plasmas." Spectrochimica Acta
PartB: Atomic Spectroscopy 48(6-7): 817-833.
Hobbs, S. E. and J. W. Olesik (1997). "The influence of incompletely desolvated droplets and
vaporizing particles on chemical matrix effects in inductively coupled plasma spectrometry:
time-gated optical emission and laser-induced fluorescence measurements." Spectrochimica Acta
PartB: Atomic Spectroscopy 52(3): 353-367.
17
-------
Huff, B. V. and G. L. Mclntire (1994). "Determination of the electrophoretic mobility of
polystyrene particles by capillary electrophoresis." Journal of Microcolumn Separations 6(6):
591-594.
Kammer, F. v. d., M. Baborowski, et al. (2005). "Field-flow fractionation coupled to multi-angle
laser light scattering detectors: Applicability and analytical benefits for the analysis of
environmental colloids." Analvtica Chimica Acta 552(1-2): 166-174.
Kapoor, S., D. Lawless, et al. (1994). "Reduction and Aggregation of Silver Ions in Aqueous
Gelatin Solutions." Langmuir 10(9): 3018-3022.
Kim, T., C.-H. Lee, et al. (2008). "Kinetics of gold nanoparticle aggregation: Experiments and
modeling." Journal of Colloid and Interface Science 318(2): 238-243.
Klaine, S. J., P. J. J. Alvarez, et al. (2008). "NANOMATERIALS IN THE ENVIRONMENT:
BEHAVIOR, FATE, BIOAVAILABILITY, AND EFFECTS." Environmental Toxicology &
Chemistry 27: 1825-1851.
Koirtyohann, S. R., J. S. Jones, et al. (1980). "Nomenclature system for the low-power argon
inductively coupled plasma." Analytical Chemistry 52(12): 1965-1966.
Lai, C.-Y., B. G. Trewyn, et al. (2003). "A Mesoporous Silica Nanosphere-Based Carrier System
with Chemically Removable CdS Nanoparticle Caps for Stimuli-Responsive Controlled Release
of Neurotransmitters and Drug Molecules." Journal of the American Chemical Society 125(15):
4451-4459.
Lam, J. W. H. and G. Horlick (1990). "A Comparison of Argon and Mixed Gas Plasmas for
Inductively Coupled Plasma-mass Spectrometry." Spectrochimica Acta Part B: Atomic
Spectroscopy 45.
Lesher, E. K., J. F. Ranville, et al. (2009). "Analysis of pH Dependent Uranium(VI) Sorption to
Nanoparticulate Hematite by Flow Field-Flow Fractionation - Inductively Coupled Plasma Mass
Spectrometry." Environmental Science & Technology 43(14): 5403-5409.
Lin, W.-C. and M.-C. Yang (2005). "Novel Silver/Poly(vinyl alcohol) Nanocomposites for
Surface-Enhanced Raman Scattering-Active Substrates." Macromolecular Rapid
Communications 26(24): 1942-1947.
Liu, F.-K., Y.-Y. Lin, et al. (2005). "Highly efficient approach for characterizing nanometer-
sized gold particles by capillary electrophoresis." Analvtica Chimica Acta 528(2): 249-254.
Liu, F.-K. and G.-T. Wei (2004). "Adding sodium dodecylsulfate to the running electrolyte
enhances the separation of gold nanoparticles by capillary electrophoresis." Analvtica Chimica
Acta 510(1): 77-83.
18
-------
Macedone, J. H., D. J. Gammon, et al. (2001). "Factors affecting analyte transport through the
sampling orifice of an inductively coupled plasma mass spectrometer." Spectrochimica Acta Part
B: Atomic Spectroscopy 56(9): 1687-1695.
Magrez, A., S. Kasas, et al. (2006). "Cellular Toxicity of Carbon-Based Nanomaterials." Nano
Letters 6(6): 1121-1125.
Michalet, X., F. F. Pinaud, et al. (2005). "Quantum Dots for Live Cells, in Vivo Imaging, and
Diagnostics." Science 307(5709): 538-544.
Mills, A. A., J. H. Macedone, et al. (2006). "High resolution imaging of barium ions and atoms
near the sampling cone of an inductively coupled plasma mass spectrometer." Spectrochimica
Acta PartB: Atomic Spectroscopy 61(9): 1039-1049.
Monserud, J. H., E. K. Lesher, et al. (2009). Real Time Single Particle-inductively Coupled
Plamsa-mass Spectrometry for Detection and Characterization of Nanoparticles. 237th ACS
National Meeting, Salt Lake City, Utah.
Mueller, N. C. and B. Nowack (2008). "Exposure Modeling of Engineered Nanoparticles in the
Environment." Environmental Science & Technology 42: 4447-4453.
Nanotechnologies, The Proect on Emerging (2010). "Environment, Health and Safety Research
Inventory." from http://www.nanotechproiect.org/inventories/ehs/.
Nanotechnologies, The Proect on Emerging (2010). "Nanotechnology Consumer Products
Inventory." from http://www.nanotechproiect.org/inventories/consumer/.
Nomizu, T., H. Hayashi, et al. (2002). "Determination of zinc in individual airborne particles by
inductively coupled plasma mass spectrometry with digital signal processing." Journal of
Analytical Atomic Spectrometry 17(6): 592-595.
Nomizu, T., H. Nakashima, et al. (1992). "Simultaneous Measurement of the Elemental Content
and Size of Airborne Particles by Inductively Coupled Plasma Emission Spectrometry Combined
with the Laser Light-Scattering Method." Analytical Scienes 8(4): 527-531.
O'Connor, G. and E. H. Evans (1999). Fundamental aspects of ICP-MS. Inductively Coupled
Plasma Spectrometry and its Applications. S. J. Hill. Sheffield, Sheffield Academic Press: 119-
144.
Peters, T. M., S. Elzey, et al. (2009). "Airborne Monitoring to Distinguish Engineered
Nanomaterials from Incidental Particles for Environmental Health and Safety." Journal of
Occupational and Environmental Hygiene 6(2): 73-81.
19
-------
Petersen, S. L. and N. E. Ballou (1992). "Effects of capillary temperature control and
electrophoretic heterogeneity on parameters characterizing separations of particles by capillary
zone electrophoresis." Analytical Chemistry 64(15): 1676-1681.
Peukert, W., H.-C. Schwarzer, et al. (2005). "Control of aggregation in production and handling
of nanoparticles." Chemical Engineering and Processing 44(2): 245-252.
Ponder, S. M., J. G. Darab, et al. (2001). "Surface Chemistry and Electrochemistry of Supported
Zerovalent Iron Nanoparticles in the Remediation of Aqueous Metal Contaminants." Chemistry
of Materials 13(2): 479-486.
Pyrz, W. D. and D. J. Buttrey (2008). "Particle Size Determination Using TEM: A Discussion of
Image Acquisition and Analysis for the Novice Microscopist." Langmuir 24(20): 11350-11360.
Quang, C., S. L. Petersen, et al. (1996). "Characterization and separation of inorganic fine
particles by capillary electrophoresis with an indifferent electrolyte system." Journal of
Chromatography A 732(2): 377-384.
Schulte, P., C. Geraci, et al. (2008). "Occupational Risk Management of Engineered
Nanoparticles." Journal of Occupational & Environmental Hygiene 5: 239-249.
Schurtenberger, P., M. E. Newmen, et al. (1993). "Environmental Particles." J. Buffle. H.P. van
Leeuwen (Eds) 2: 37.
Shim, J.-Y. and V. K. Gupta (2007). "Reversible aggregation of gold nanoparticles induced by
pH dependent conformational transitions of a self-assembled polypeptide." Journal of Colloid
and Interface Science 316(2): 977-983.
Stolpe, B., M. Hassellov, et al. (2005). "High resolution ICPMS as an on-line detector for flow
field-flow fractionation; multi-element determination of colloidal size distributions in a natural
water sample." Analvtica Chimica Acta 535(1-2): 109-121.
Takagi, A., A. Hirose, et al. (2008). "Induction of mesathelioma in p53+/- mouse by
intraperitioneal application of multi-wall carbon nanotube." Journal of Toxicol Science 33: 105-
116.
Tang, Z., L. Wu, et al. (2009). "Size fractionation and characterization of nanocolloidal particles
in soils." Environmental Geochemistry and Health 31(1): 1-10.
Taylor, H. E. (2001). Inductively Coupled Plasma-Mass Spectrometry: Practices and
Techniques. San Diego, Academic Press.
Taylor, H. E., J. R. Garbarino, et al. (1992). "Inductively coupled plasma-mass spectrometry as
an element-specific detector for field-flow fractionation particle separation." Analytical
Chemistry 64(18): 2036-2041.
20
-------
Thomas, R. (2004). Practical Guide to ICP-MS. New York, Marcel Dekker, Inc.
Thurman, E. M. and R. L. Malcolm (1981). "Preparative isolation of aquatic humic substances."
Environmental Science & Technology 15(4): 463-466.
Tiede, K., A. B. A. Boxall, et al. (2010). "Application of hydrodynamic chromatography-ICP-
MS to investigate the fate of silver nanoparticles in activated sludge." Journal of Analytical
Atomic Spectrometry 25(7): 1149-1154.
Tiede, K., M. Hassellov, et al. (2009). "Considerations for environmental fate and ecotoxicity
testing to support environmental risk assessments for engineered nanoparticles." Journal of
Chromatography A 1216(3): 503-509.
Zhang, L., D. Pornpattananangkul, et al. (2010). "Development of Nanoparticles for
Antimicrobial Drug Delivery." Current Medicinal Chemistry 17: 585-594.
21
-------
Tables
Table 1 Engineered nanomaterials uses in consumer products.
Nanomaterial
Number of products
Common uses (with selected examples)
Silver
121
Nutritional supplements; personal care
products (tooth brushes, tooth paste);
biocide coating on membranes (gloves,
masks, condoms); cothing (socks,
underwear, baby apparel); toys and
pacifiers; dispersed biocide (washing
machines)
Titanium dioxide
32
Resin coatings; personal care products
(skin cream, hair dryers); nutritional
supplements; catalysts (self-cleaning
coatings)
Zinc oxide
30
Sun block
Carbon nanotubes
19
Material construction (sporting goods,
automobile and aircraft parts)
Gold
15
Nutritional supplements, personal care
products (skin cream); catalysts.
Fullerenes
7
Cosmetics
Cerium oxide
1
Fuel additive
22
-------
Table 2 Common ENM characterization metrics and selected characterization methods.
Characterization metric
Measurement methods
Size distribution - diameter
SEM, TEM, AFM, DLS, SLS, NT A, flow-
FFF, sed-FFF, CE, HDC
Size distribution - mass
Sed-FFF
Specific surface area
adsorption isotherm
Surface charge
C potential by DLS
Shape
SEM, TEM, AFM, SLS with DLS or FFF
23
-------
Table 3 Comparison of SP-ICPMS with hyphenated ensemble methods.
SP-ICPMS
Hyphenated ensemble methods
Measures elemental mass of nanoparticles, not
nanoparticle size distribution.
Separate nanoparticles into size fractions
diameter, and determine total metal
concentration associated with each particle size
fraction.
Provides concentration of metal-based
nanoparticles, and mass of metal in each
particle.
Cannot differentiate large concentration of
particles containing little analyte from low
concentration with high analyte content.
Therefore, no information is provided on
number or characteristics of metal-based
particles.
Fast analysis (e.g., < 60 seconds)
Lengthy analysis (e.g., 20 - 30 minutes)
24
-------
Table 4. Statistics of replicate SP-ICPMS analyses of 50 nm gold nanoparticles
(4 x 103 mL1).
Replicate
Mean number of ions per particle
Number of particles
1
25.6
94
2
25
105
3
26.4
87
41
26.3
101
51
25.9
115
Grand mean
25.8
101
% RSD
2.22
11
1 Replicates 4 and 5 were analyzed in presence of 60 nm silver particles at 5 x 107 mL"1.
2 Corresponds to < 0.8% RSD in mean particle diameter.
25
-------
Figure Captions
Figure 1. Number of environmental health and safety (EHS) research projects initiated since
1998.
Figure 2. Time resolved ICPMS signal for dissolved gold (30 pg/mL) at (a) 10 ms dwell time
and (b) 2 s dwell time.
Figure 3. Time resolved ICPMS signal for 50 nm gold nanoparticle suspension (30 pg/mL) at
(a) 2 s dwell time and (b) 10 ms dwell time.
Figure 4. Correlation of calculated gold nanoparticle diameter from the particle elemental mass
determined by SP-ICPMS vs. the nominal diameter reported by the manufacturer.
Figure 5. CDF of five replicate SP-ICPMS analyses of 50 nm gold nanoparticle suspensions at 4
x 103 mL"1. Replicates 4 and 5 also contained 60 nm silver nanoparticles at 5 x 103 mL"1.
Figure 6. Change in CDF of 60 nm silver nanoparticles (500 ng/mL) over time after spiking with
200 mM CaCl2.
Figure 7. Change in CDF of 60 nm silver nanoparticles (0.25 ng/mL) over time after spiking
with 200 mM CaC^.
Figure 8. SP-ICPMS of 50 nm gold nanoparticles (1.0 x 104 mL-1) - (a) time-resolved signal of
unfiltered suspension, (b) signal of filtrate suspension - 0.45 [j,m Nylon filter, (c) CDF of
unfiltered suspension, (b) CDF of filtrate suspension - 0.45 [j,m Nylon filter.
Figure 9. SP-ICPMS of 60 nm silver ENMs - (a) time-resolved signal of 5xl04 mL"1 spiking
solution, (b) signal of Las Vegas Wash water spiked to 5xl03 mL"1 and then passed through 5.0
|im filter, (c) CDF of 5xl04 mL"1 spiking solution, (d) CDF of spiked and filtered Las Vegas
Wash sample.
Figure 10. Time resolved SP-ICPMS signal of silver and gold in 5.0-|im filtered Las Vegas
Wash water.
Figure 11. Time resolved SP-ICPMS signal of titanium in 5.0-|im filtered Las Vegas Wash
water.
Figure 12. Time resolved SP-ICPMS signal of lanthanoid elements and yttrium in 5.0-|im
filtered Las Vegas Wash water.
Figure 13. Time resolved SP-ICPMS signal of m/z 57 in 5.0-|im filtered Las Vegas Wash water
at two dwell times.
26
-------
Figure 14. Time resolved SP-ICPMS signal of eluent from liquid chromatograph reciprocating
pump - (a) m/z 197 and (b) m/z 181.
Figure 15. Time resolved SP-ICPMS signal of eluent from piston pump - (a) m/z 197 and (b)
m/z 181.
Figure 16. HDC-ICPMS chromatograms of 60 nm gold nanoparticles (50 pg/mL) - (a) 1 s dwell
time and (b) 10 ms dwell time.
Figure 17. HDC-ICPMS chromatograms of 60 nm gold nanoparticles (50 pg/mL) and 30 nm
gold nanoparticles (25 pg/mL) mixture - (a) 1 s dwell time and (b) 10 ms dwell time.
Figure 18. CE-SP-ICPMS electropherogram of 30 nm and 50 nm gold nanoparticles.
27
-------
600 n
500
1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010
Year initiated
-------
a
ICPMS (dissolved)
10 ms dwell time
¦c
0)
O
a>
0)
¦o
c
o
1
b
ICPMS (dissolved) 1
2 s dwell time ^
¦o
a>
•*->
o
-------
a
250 n
ICPMS (nanoparticles) 200
2 s dwell time
¦c
0)
o 150
0)
-------
250
CD
0
E
03
b
200
E
c
CO
CL
O
§5 150
tn
w
CC
E
03
-*—>
C
0
E
a)
o
E
o
100
84
50
48.6
30.4
0
0
50 100
Nominal diameter (nm)
150
200
-------
0 10 20 30 40 50 60 70 80
ions detected
~ Rep. 1 ¦ Rep. 2 Rep. 3 Rep. 4 x Rep. 5
-------
100 200 300 400 500 600
ions detected
~ Omin. ¦ 8 min. 14 minutes
-------
c
CC
L_
w
CO
CC
0.6
s 0.4
03
0.2
0
0 100 200 300 400 500 600
Ions detected
~ Ohr ¦ 1 hr 49 hr
-------
90
c
o
20
10
0
20 25 30
time (s)
(D)
..I j . 1.1 ., ... ..
10
15
20 25 30
time (s)
35
40
45
50
ions detected
-------
"O
(D
-i—¦ *
O
CD
200
180
160
I 140
120
100
^ 80 "
60 "
£=
O 40
20
0
(a)
150
time (s)
200
180
160
140
120
100
80
60
40
20
0
(b)
Ll—i L
¦¦ia .b.>L,.L.h
150
time (s)
O 50 10O 150 200 250
ions detected
(d)
1
0.9
0.8
£=
0.7
m
O (=>
(/j
(/;
TO
O f>
fc
0
0.4
O
O 3
(T5
Q_
Q_
0.2
0.1
O
O 50 1 OO 150 200 250
ions detected
-------
90
80
70
60
50
40
30
20
10
0
Ag107 Au 197
1 Jl L|J"U -J Jl ¦L'1^
X'L Wll-ll. I* ' I • ~L.-tJt.IH. . ILIlJ¦ J Ijl- LlJl
50
100
150
time (s)
200
250
300
-------
1600 n
1400
1200
1000
800
600
400
200
jJ. L
50
100
150
jjl jJi IliiU-ll - 111 i i I Li i
200 250 300 350 400
time (s)
-------
350 n
11 i
I.
50
100
Ce140
Nd142 Y89
-------
6000 -n
5000
4000
a3
I 3000
CD
0
3
1 2000
in
c
o
1000
0
0 100 200 300 400
time (s)
10 ms dwell time 3 ms dwell time
-------
a
m/z 197
120
100
80
8 60
40
20
llIL
Ui
100
200
300
b 25000
m/z181 20000
c/5 15000
° 10000
5000
LJjJ
¦ -1 I - I-
La,
100
200
300
jJjlLlliL.J]
400 500
time (s)
600
700
800
- -t.-.ii .-i i.-J,..]!,- Ij . I.—... 1...^.ii i J^_.
400 500 600 700 800
time (s)
-------
a
120 n
m/z 197
100
<2 80
C
§ 60
40
20
g ¦ ¦ • ¦ L — *- I ¦¦ ¦ |.IL ... I. .. - ¦ ¦! .
0 100 200 300
b 120
m/z 181 100
<0 80
¦l—'
8 60
40
20
0
HM
Ji—J *¦
100
lLLX
200
300
¦¦L- ¦¦
400 500
time (s)
600
700
800
-U-lLjij
400
500
600
700
800
time (s)
-------
a
1 s
400 i
350 -
300 -
c/5 250 -
4—>
§ 200 -
o
150 -
100 -
50 -
0 -W
750
„
770 790 810 830
time (s)
850
870
890
10 ms 80
750 770 790 810 830 850
time (s)
870
890
-------
a
1s
o
o
1400
1200
1000
800
600
400
200
o 4—
600
650
700
b
10 ms
o
o
200
180
160
140
120
100
80
60
40
20
0 -W-
600
"
650
700
JIL
750
time (s)
800
850
900
- J* -
750
time (s)
800
850
900
-------
30 & 50 nm Au @ 10-ms integration time
140
120
100
80
60
40
20
0
0
5
10
15
20
25
30
minutes
-------
-------
&EPA
United States
Environmental Protection
Agency
Please make all necessary changes on the below label, detach
or copy and return to the address in the upper left hand corner.
If you do not wish to receive these reports CHECK HERE
detach, or copy this cover, and return to the address in the
upper left hand corner.
PRESORTED STANDARD
POSTAGE & FEES PAID
EPA PERMIT No. G-35
Office of Research
and Development (8101R)
Washington, DC 20460
Official Business
Penalty for Private Use
$300
EPA/600/R-10/117
October 2010
www.epa.gov
Recycled/Recyclable
Printed with vegetabie-based ink on
paper that contains a minimum of
50% post-consumer fiber content
processed chiorine free
------- |