ArticlePDF Available

Industrial Application of Nanomaterials – Chances and Risks

Authors:
Future Technologies Volume 54 Industrial application of nanomaterials -- chances and risks
With the support of the European Commission
Technological Analysis
Industrial application
of nanomaterials --
chances and risks
Industrial application of nanomaterials
- chances and risks
Technology analysis
Wolfgang Luther (ed.)
Published by:
Future Technologies Division
of VDI Technologiezentrum GmbH
Graf-Recke-Str. 84
40239 Düsseldorf
Germany
This technological analysis arose in connection with the project “Risk Assessment in
Production and Use of Nanoparticles with Development of Preventive Measures and
Practice Codes“ (”Nanosafe”). The Project was funded by the European Community
under the “Competitive and Sustainable Growth” Programme (Contract-No. G1MA-
CT-2002-00020, project coordinator Dr. Rüdiger Nass). The publication of the report
was supported by the German Federal Ministry of Education and Research (BMBF)
within the project “Innovation accompanying measures nanotechnology”, (FKZ BM1,
project director Dr. Dr. Axel Zweck).
The following experts contributed to the report as authors, informants or advisers:
Dr. Rüdiger Nass, Mr. Robert Campbell, Ms. Ulrike Dellwo (Nanogate Technologies
GmbH, Germany),
Mr. Frédéric Schuster, Mr. François Tenegal (Commissariat à l'Energie Atomique,
CEA, France)
Ms. Marke Kallio, Dr. Pertti Lintunen, (VTT Processes Advanced Materials, Finland)
Dr. Oleg Salata (University of Oxford, United Kingdom)
Dr. Maja Remškar, Dr. Marko Zumer (Jožef Stefan Institute, Ljubljana, Slovenia)
Prof. Peter Hoet (Katholieke Universiteit Leuven, Belgium)
Dr. Irene Brüske-Hohlfeld (GSF-Forschungszentrum für Umwelt und Gesundheit,
GmbH, GSF, Germany)
Dr. Sarah Lipscomb (Oxonica Ltd., United Kingdom)
Dr. Wolfgang Luther, Dr. Norbert Malanowski, Dr. Dr. Axel Zweck (Future
Technologies Division, VDI-Technologiezentrum GmbH, Germany)
Contact: Dr. Wolfgang Luther (luther@vdi.de)
Future Technologies No. 54
Düsseldorf, August 2004
ISSN 1436-5928
The authors are responsible for the content. All rights reserved except those agreed by
contract. No part of this publication may be translated or reproduced in any form or by
any means without prior permission of the authors.
Front Page: left above: microelectronic clean room facility, left below: flourescent cadmium telluride
nanoparticles (source: University of Hamburg), right above: TEM image of agglomerated silicon carbide
nanoparticles (source : CEA), right below: macrophage intaking ultrafine particles (source: GSF)
Future Technologies Division
of VDI Technologiezentrum GmbH
Graf-Recke-Straße 84
40239 Düsseldorf, Germany
The VDI Technologiezentrum GmbH is an associated company of the Association of
Engineers (VDI) under contract to and with the support of The Federal Ministry of
Education and Research (BMBF).
Foreword
Nanotechnology is seen as one of the most relevant technologies for the 21st century.
This opinion on nanotechnology is derived from its economic potential on new or
optimised products as well as on the expected contributions for minimising ecological
stress and consumption of resources. Nanotechnology does not only mean to make
technology one step smaller, from micro- to nanotechnology. Nanotechnology rather
prepares the way for handling and using quantum effects. Showing complete new
characteristics and behaviours, nanomaterials are opening new product innovations e.g.
for protection against sun, biochip makers or copy protection.
On the other hand there is an upcoming discussion about the potential risks of
nanotechnology. Like in every early state of discussion on risks, fears, arguments and
speculations are merging. Visionary aspects such as nanobots or grey-goo and questions
about health and environmental implications of nanomaterials are named with the same
breath like well known existing problems of nanoparticulate carbon emissions by car
diesel engines. This discussion has given rise to a demand for a moratorium on
nanotechnology by some NGOs.
Nevertheless we have to expect a large diffusion of nanotechnology based or
nanotechnology related products and production processes in the coming years. The
European Commission intend to support the gathering of scientific data in order to
analyse chances and risks of nanotechnology as basis for matter-of-fact oriented public
discussion. The intended result is seen in a reliable database and extensive safety for
decision making in the sense of protection and regulation to a minimum extend.
The objectives of this report are to assemble available information from public and
private sources on chances but also possible hazards involving industrial nanoparticle
production, to evaluate the risks for workers, consumers and the environment, and to
give recommendations for setting up regulatory measures and codes of good practice.
Dr. Dr. Axel Zweck
Head of Future Technologies Division, VDI Technologiezentrum GmbH
Table of Content
1 INTRODUCTION 1
1.1 Objectives of the report 3
1.2 Methods 3
2 CLASSIFICATION AND PROPERTIES 5
2.1 Classification of nanomaterials 5
2.2 Properties of nanomaterials 7
2.3 Characteristics of nanoparticulate materials 9
3 INDUSTRIAL APPLICATIONS AND MARKET POTENTIALS 11
3.1 Nanoparticles 11
3.2 Nanocomposites 14
4 PRODUCTION METHODS 17
4.1 Top-down approaches 17
4.2 Bottom-up approaches 21
4.3 Stabilisation and functionalisation of nanoparticles 26
5 CHARACTERISATION AND DETECTION TECHNIQUES 29
5.1 Atomic structure and chemical composition 29
5.2 Determination of size, shape and surface area 32
5.3 Determination of nanoparticles in aerosols 37
5.4 Determination of nanoparticles in biological tissue 41
6 RISK ASSESSMENT 43
6.1 Potential particle release 44
6.2 Exposure assessment 55
6.3 Toxicological assessment 58
6.4 Toxicological testing 74
6.5 Preliminary scheme for risk assessment 75
7 RISK MANAGEMENT 77
7.1 Preventive measures at the work place 77
7.2 Preventive measures for the environment 81
7.3 Standardisation and regulation activities 83
8 CONCLUSION AND RECOMMENDATIONS 91
8.1 Key findings of the report 91
8.2 Policy options 93
9 REFERENCES 95
LIST OF ABBREVIATIONS 111
LIST OF EXPERTS AND INTERNETLINKS 112
1
1 INTRODUCTION
Atoms and molecules are the essential building blocks of all things. The
manner in which things are “constructed” with these building blocks is
vitally important to their properties and how they interact.
Nanotechnology refers to the manipulation or selfassembly of individual
atoms, molecules, or molecular clusters into structures to create materials
and devices with new or vastly different properties. Nanotechnology is
developing new ways to manufacture things. Since the late 90`s,
nanotechnology has shot into the limelight as a new field with
tremendous promise. The potential beneficial impact of nanotechnology
on society has been compared with that of silicon and plastics. This new,
“small” way of manipulating materials has already led to new research
areas and the development of new products, which are available
commercially.
Nanostructured materials play a key role in most of the nanotechnology
based innovations. By tailoring the structure of materials at the
nanoscale, it is possible to engineer novel materials that have entirely
new properties. With only a reduction of size and no change in substance,
fundamental characteristics such as electrical conductivity, colour,
strength, and melting point – the properties we usually consider constant
for a given material – can all change. Therefore nanomaterials show
promising application potentials in a variety of industrial branches such
as chemistry, electronics, medicine, automotive, cosmetics or the food
sector. Nanotechnology here holds the promise for producing better
goods with less input of energy and /or materials, developing specific
drug delivery systems and lab-on-a-chip based diagnostics for a minimal
invasive medicine, improving information and communication through
smaller and more powerful electronic devices, etc. An optimistic view on
these developments predicts that a new „industrial revolution“ will take
place in the following decades.
In recent times however, an increasing number of sceptical voices
concerning nanotechnology can be heard in the public. Beside the
discussion of risks of visionary developments like “nanobots” and
“nanoassemblers” (Drexler 1986 and 1991, Joy 2000) most critics focus
on potential health and environmental risks of nanomaterials. This can be
illustrated with several articles in newspapers and top scientific journals
(e.g. Service 2003, Malakoff 2003) discussing potential negative effects
and risks of nanoparticle applications. Although, not very much has been
published concerning specific nanomaterials, the potential health and
environmental risks of nanoparticles respective ultrafine particles (UFP)
with aerodynamic diameters < 100 nm, has gained public attention in the
last years. In this discussion the terms ‘ultrafine particles’ - used in
aerosol and epidemiology terminology - and ‘nanoparticles’ are often
used interchangeably.
Nanotechnology is a
key technology for
the 21
st
Century
Material properties
can be tailored at
the nanoscale
Public discussion on
potential health
and environmental
risks of
nanoparticles
Terminology of
nanoparticles and
nanomaterials is
often ambiguous
2 Industrial application of nanomaterials – chances and risks
One of the sharpest critics of industrial nanoparticle applications is the
Canadian-based non-governmental organisation ETC Group, which
called for an immediate moratorium on commercial production of new
nanomaterials and for a transparent global process for evaluating the
socio-economic, health and environmental implications of
nanotechnology (ETC 2002). The fear of risk associated with
nanoparticle use is mainly caused by limited scientific knowledge about
potential side effects of nanoparticle in the human body and the
environment due to their special properties. Conventional compounds
normally considered harmless might prove to be dangerous on a
nanometer scale. For example nanoparticles can penetrate into body cells
and even break through biological barriers (such as the blood-brain
barrier).
Call for a
moratorium on
commercial
production of
nanomaterials
Epedimiological
studies show an
association
between number
concentration of
ultrafine particles
in polluted air and
health risks
Few data available
for physiological
effects of
nanoparticles
Epidemiological studies have consistently shown an association between
particulate air pollution and health, not only in exacerbations of illness in
people with respiratory disease but also in rising numbers of deaths from
cardiovascular and respiratory disease among older people. It has been
proposed that the adverse health effect of particulate air pollution was
mainly associated with the number concentrations of ultrafine particles
(Oberdörster et al. 1994, Seaton et al. 1995) rather than the mass
concentrations of coarser particle fractions. These epidemiological
studies were conducted in the environmental context with traffic and
industrial combustion processes being the main source of particulate
matter in ambient air. So far, no epidemiological studies are available to
describe the work place situation in regard to the production of
nanoparticles. However, there have been some studies showing that
nanoparticles, after deposition in the lungs, largely escape alveolar
macrophage surveillance and gain access to the pulmonary interstitium
with greater inflammatory effect than larger particles (Oberdörster 2001).
From occupational medicine it has been known for decades that particles
deposited in the alveolar region of the lungs can lead to the development
of chronic diffuse interstitial lung disease like silicosis and asbestosis.
Recent findings from animal studies suggest a fast translocation of
nanoparticles from pulmonary and gastrointestinal epithelium into the
systemic circulation (Frampton 2001, Nemmar et al. 2002, Oberdörster et
al. 2002) Also, there is some evidence that carbon nanoparticles can
directly enter the brain via the respiratory nasal mucosa and the olfactory
bulb (Calderon-Garciduenas et al. 2002, Oberdörster 2004). All these
properties make the epidemiologically observed association of inhaled
nanoparticles and adverse health effects biologically plausible. However,
without hard data it is impossible to know what physiological effects will
occur.
Although nanotechnology in most fields is still at an experimental stage,
the next few years will probably see a dramatic increase in the industrial
generation and use of nanoparticles (Mazzola 2003, Paull 2003).
Introduction 3
Therefore impact of these materials on worker safety, consumer
protection, public health and the environment will have to be considered
carefully by legislation and regulation authorities.
1.1 Objectives of the report
The objectives of this report are to assemble available information from
public and private sources on chances but also possible hazards involving
industrial nanoparticle production, to evaluate the risks to workers,
consumers and the environment, and to give recommendations for setting
up regulatory measures and codes of good practice to obviate any danger.
The report gives information on characteristics of nanoparticles (size,
shape, types, etc.), production methods, industrial application fields,
characterisation and detection methods as well as a risk assessment
including potential particle release and exposure, toxicological aspects
and protective measures.
It has to be noted that this report focuses on the assessment of the
production, handling and treatment, and use of nanoparticles in industrial
processes and products, as well as in consumer products. Risks of other
kind of ultrafine or nanoparticles e.g. from vehicle or power plant
emissions will not be dealt with, although some of the knowledge and
information acquired may be relevant.
1.2 Methods
The report summarises information from scientific literature, project
studies within the partner organisations, environmental, health and
worker protection associations as well as national and European
legislation. Literature databases, proceedings of relevant workshops and
conferences as well as internet searches and expert interviews were used
as information sources.
The authors realise that in view of the broad scope and a very early stage
of the discussion this report is rather a working-document that should be
criticised and discussed to come to a better understanding of the topic. It
has also to be mentioned, that many information concerning the
development of nanomaterial based products are kept confidental by the
involved companies, so this report may not represent the state-of-the-art
in some areas.
4
5
2 CLASSIFICATION AND PROPERTIES
2.1 Classification of nanomaterials
All conventional materials like metals, semiconductors, glass, ceramic or
polymers can in principle be obtained with a nanoscale dimension. The
spectrum of nanomaterials ranges from inorganic or organic, crystalline
or amorphous particles, which can be found as single particles,
aggregates, powders or dispersed in a matrix, over colloids, suspensions
and emulsions, nanolayers and –films, up to the class of fullerenes and
their derivates. Also supramolecular structures such as dendrimers,
micelles or liposomes belong to the field of nanomaterials. Generally
there are different approaches for a classification of nanomaterials, some
of which are summarised in table 1.
Broad range of
different
nanomaterial
classes
Classification
approaches of
nanomaterials
Classification examples
Dimension
3 dimensions < 100nm
2 dimensions < 100nm
1 dimension < 100nm
particles, quantum dots, hollow spheres, etc.
tubes, fibers, wires, platelets, etc.
films, coatings, multilayer, etc.
Phase composition
single-phase solids
multi-phase solids
multi-phase systems
crystalline, amorphous particles and layers, etc.
matrix composites, coated particles, etc.
colloids, aerogels, ferrofluids, etc.
Manufacturing process
gas phase reaction
liquid phase reaction
mechanical procedures
flame synthesis, condensation, CVD, etc.
sol-gel, precipitation, hydrothermal processing, etc.
ball milling, plastic deformation, etc.
Table 1: Classification of nanomaterials with regard to different parameters
The main classes of nanoscale structures can be summarised as follows:
2.1.1 Nanoparticles
Nanoparticles are constituted of several tens or hundreds of atoms or
molecules and can have a variety of sizes and morphologies (amorphous,
crystalline, spherical, needles, etc.). Some kind of nanoparticles are
already available commercially in the form of dry powders or liquid
dispersions. The latter is obtained by combining nanoparticles with an
aqueous or organic liquid to form a suspension or paste. It may be
necessary to use chemical additives (surfactants, dispersants) to obtain a
uniform and stable dispersion of particles. With further processing steps,
nanostructured powders and dispersions can be used to fabricate
coatings, components or devices that may or may not retain the
nanostructure of the particulate raw materials. Industrial scale production
of some types of nanoparticulate materials like carbon black, polymer
dispersions or micronised drugs has been established for a long time.
6 Industrial application of nanomaterials – chances and risks
Another commercially important class of nanoparticulate materials are
metal oxide nanopowders, such as silica (SiO2), titania (TiO2), alumina
(Al
2
O
3
) or iron oxide (Fe
3
O
4
, Fe
2
O
3
). But also other nanoparticulate
substances like compound semiconductors (e.g. cadmium telluride,
CdTe, or gallium arsenide, GaAs) metals (especially precious metals
such as Ag, Au) and alloys are finding increasing product application.
Metal oxide
nanopowders have
already broad
commercial
applications
Carbon nanotubes
are expected to
have a big market
potential in the
future
Beside that, the range of macromolecular chemistry with molecule sizes
in the range of up to a few tens of nanometers is often referred to as
nanotechnology. Molecules of special interest that fall within the range of
nanotechnology are fullerenes or dendrimers (tree-like molecules with
defined cavities), which may find application for example as drug
carriers in medicine.
2.1.2 Nanowires and -tubes
Linear nanostructures such as nanowires, nanotubes or nanorods can be
generated from different material classes e.g. metals, semiconductors or
carbon by means of several production techniques. As one of the most
promising linear nanostructures carbon nanotubes can be mentioned,
which can occur in a variety of modifications (e.g. single- or multi-
walled, filled or surface modified). Carbon nanotubes are expected to
find a broad field of application in nanoelectronics (logics, data storage
or wiring, as well as cold electron sources for flat panel displays and
microwave amplifiers) and also as fillers for nanocomposites for
materials with special properties. At present carbon nanotubes can be
produced by CVD methods on a several tons per year scale and the gram
quantities are already available commercially.
2.1.3 Nanolayers
Nanolayers are one of the most important topic within the range of
nanotechnology. Through nanoscale engineering of surfaces and layers a
vast range of functionalities and new physical effects (e.g.
magnetoelectronic or optical) can be achieved. Furthermore a nanoscale
design of surfaces and layers is often necessary to optimise the interfaces
between different material classes (e.g. compound semiconductors on
silicon wafers) and to obtain the desired special properties. Some
application ranges of nanolayers and coatings are summarised in table 2.
Classification and properties 7
Surface Properties Application examples
Mechanical properties (e.g.
tribology, hardness, scratch-
resistance)
Wear protection of machinery and
equipment, mechanical protection of soft
materials (polymers, wood, textiles, etc.)
Wetting properties (e.g.
antiadhesive, hydrophobic,
hydrophilic)
Antigraffiti, antifouling, Lotus-effect,
self-cleaning surface for textiles and
ceramics, etc.
Thermal and chemical
properties (e.g. heat
resistance and insulation,
corrosion resistance)
Corrosion protection for machinery and
equipment, heat resistance for turbines
and engines, thermal insulation
equipment and building materials, etc.
Biological properties
(biocompatibility, anti-
infective)
Biocompatible implants, abacterial
medical tools and wound dressings, etc.
Electronical and magnetic
properties (e.g. magneto-
resistance, dielectric)
Ultrathin dielectrics for field-effect
transistors, magnetoresistive sensors and
data memory, etc.
Optical properties (e.g. anti-
reflection, photo- and
electrochromatic)
Photo- and electrochromic windows,
antireflective screens and solar cells, etc.
A vast range of
functionalities and
new physical
effects can be
achieved by
nanoscale
engineering of
surfaces
Table 2: Tunable properties by nanoscale surface design and their application potentials
2.1.4 Nanopores
Materials with defined pore-sizes in the nanometer range are of special
interest for a broad range of industrial applications because of their
outstanding properties with regard to thermal insulation, controllable
material separation and release and their applicability as templates or
fillers for chemistry and catalysis. One example of nanoporous material
is aerogel, which is produced by sol-gel chemistry. A broad range of
potential applications of these materials include catalysis, thermal
insulation, electrode materials, environmental filters and membranes as
well as controlled release drug carriers.
2.2 Properties of nanomaterials
The physical and chemical properties of nanostructured materials (such
as optical absorption and fluorescence, melting point, catalytic activity,
magnetism, electric and thermal conductivity, etc.) typically differ
significantly from the corresponding coarser bulk material. A broad
range of material properties can be selectively adjusted by structuring at
the nanoscale (see table 3).
8 Industrial application of nanomaterials – chances and risks
Properties Examples
Catalytic Better catalytic efficiency through higher surface-to-volume
ratio
Electrical Increased electrical conductivity in ceramics and magnetic
nanocomposites, increased electric resistance in metals
Magnetic Increased magnetic coercivity up to a critical grain size,
superparamagnetic behaviour
Mechanical Improved hardness and toughness of metals and alloys,
ductility and superplasticity of ceramic
Optical Spectral shift of optical absorbtion and fluorescence properties,
increased quantum efficiency of semiconductor crystals
Sterical Increased selectivity, hollow spheres for specific drug
transportation and controlled release
Biological Increased permeability through biological barriers
(membranes, blood-brain barrier, etc.), improved biocom-
patibility
Table 3: Adjustable properties of nanomaterials
These special properties of nanomaterials are mainly due to quantum size
confinement in nanoclusters and an extremely large surface-to-volume
ratio relative to bulk materials and therefore a high percentage of
atoms/molecules lying at reactive boundary surfaces. For example in a
particle with 10 nm diameter only approx. 20 per cent of all atoms are
forming the surface, whereas in a particle of 1 nm diameter this figure
can reach more than 90 per cent. The increase in the surface to volume
ratio results in the increase of the paricle surface energy, which leads to
e.g. a decreasing melting point or an increased sintering activity. It has
been stated that large specific surface area of particles may significantly
raise the level of otherwise kinetically or thermodynamically
unfavourable reactions (Jefferson 2000). Even gold (Au), which is a very
stable material, becomes reactive when the particle size is small enough
(Haruta 2003).
Special properties
of nanomaterials
are due to
quantum effects
and a large
surface-to-volume
ratio
With precise control of the size of the particles their characteristics can
be adjusted in certain borders. Though it is usually difficult to maintain
these desired characteristics beyond the different manufacturing
processes to the final product, because loose nano-powders tend to grow
to larger particles and/or firmly connected agglomerates already at room
temperature and thus loosing there nano-specific characterisitcs.
Therefore it is necessary to select or develop suitable production
processes and further refining/treatment processes (e.g. coating of
nanoparticles) to prevent or attentuate agglomeration and grain growth
during generation, processing and use of nanomaterials (see also chapter
4.3).
Classification and properties 9
2.3 Characteristics of nanoparticulate materials
Relevant
parameters for the
characterisation of
nanoparticulate
materials
In this report we focus on nanoparticulate materials which have structure
sizes smaller than 100 nm in at least two dimensions. These
nanoparticulate materials can have various shapes and structures such as
spherical, needle-like, tubes, platelets, etc. Chemical composition is
another important parameter for the characterisation of nanoparticulate
materials, which comprise nearly all substance classes e.g. metals/ metal
oxides, polymers, compounds as well as biomolecules. Under ambient
conditions nanoparticles tend to stick together and form aggregates and
agglomerates. These aggregates/ agglomerates have various forms, from
dendritic structure to chain or spherical structures with sizes normally in
the micrometer range. The properties of nanoparticles can be
significantly altered by surface modification. For example, nanoparticles
are often stabilised with coatings or molecule adducts to prevent
agglomeration. For the characterisation of nanoparticulate materials it is
further important in which medium the nanoparticles are dispersed e.g. in
gaeous, liquid or solid phase. The following figure summarises relevant
parameters for the characterisation of nanoparticulate materials.
Dispersion in
• gases (aerosols)
• liquids (e.g. gels, ferrofluids)
• solids (e.g. matrix materials)
Surface modification
• untreated (as obtained in production process)
• coated (e.g. conjugates, polymeric films)
• core/shell particles (e.g. spheres, capsules)
Shape/Structure
• spheres
• needles
• platelets
• tubes
Chem. composition
• metals/ metal oxides
• polymers, carbon
• semiconductors
• biomolecules
• compounds ...
Nanoparticulate
Materials
Ultrafine
Aerosols
Quantum
dots
Nano-
particles
Nano-
tubes
Nano-
capsules
Origin
• natural
• unintentionally released
• manufactured („old“, „new“)
Aggregation state
• single particles
• aggregates
• agglomerates
Figure 1: Characterisation parameters of nanoparticulate materials (source: VDI-TZ)
Some examples of different types of nanoparticulate materials are
presented in the following figure.
10 Industrial application of nanomaterials – chances and risks
Nanostructured Al
2
O
3
-Ni composite
powder. (Keskinen 2003)
Nickel nanoparticles. (Groza et al.
2003)
Needle-like crystals Ag-(NbS
4
)
x
I
(Remskar 2002)
Multiwalled carbon nanofibres
(MWCNF) grown on substrate
(Meyyappan et al. 2003)
100 nm
SiC nano-structures (source: CEA) Fe-nanoparticles stabilised with polyvi-
nyl alcohol, Scale bar = 20 nm (Pardoe
et al. 2001)
Figure 2: Electron microscopy images showing structure and shape of different
nanoparticulate materials
11
3 INDUSTRIAL APPLICATIONS AND MARKET
POTENTIALS
The production of nanomaterial based products involves several
manufacturing steps. It usually starts with the production of nanoscaled
particles from precursors or bulk materials, goes to master batches or
dispersions which can be intergrated into commercial products to make
semi-manufactured products and ends in products over a wide range of
applications. The processing of nanoparticles depends on the basic
formulation, solid as nanopowders or liquid as dispersions. Nanopowders
can be used as fillers for different materials such as varnish, paint,
plastics, etc. or they can be used as educts e.g. for the production of
ceramics. Liquid nanodispersions can be integrated into other liquid
systems such as paints or can be used to create new composites with new
properties. The following figure shows a typical value chain of
nanoparticulate material based products.
Production steps of nanomaterial based products
precursor preparation
semi
manufactured
products
products
bulk materials
and precursors
e.g. metal organic
compounds,
alcoholates etc.
nano-dispersions,
master batch,
compounds and
composites,
nanopowders
e.g. foils, coated
components,
abrasives, paints,
adhesives,
membranes etc.
products with
custom-designed
properties e.g.
textiles, windows,
ceramics, etc.
Figure 3: Production steps and value chain of nanomaterial based products
The following chapters summarise existing as well as potential
applications of different types of nanoparticulate materials.
3.1 Nanoparticles
3.1.1 Metal oxides/metals
Metal oxides, in particular silica (SiO2), titania (TiO2), alumina (Al
2
O
3
),
iron oxide (Fe
3
O
4
, Fe
2
O
3
) at present occupy the first position in terms of
economic importance within the range of inorganic nanoparticles. Also
of increasing importance are mixed oxides, such indium-tin oxide (ITO)
and antimony-tin oxide (ATO), silicates (aluminum and zirconium
silicates) and titanates (e.g. barium titanate). While silica and iron oxide
nanoparticles have a commercial history spanning half a century or more,
other nanocrystalline metal oxides have entered the marketplace more
12 Industrial application of nanomaterials – chances and risks
recently. Main applications fields of metal oxid nanoparticles are electro-
nics, pharmacy/medicine, cosmetics as well as chemistry and catalysis.
In the range of cosmetics the most economic relevant application are
nanoparticle-based sunscreens. Here nanoparticulate titania and zinc
oxide are used as UV light absorbing components, which are transparent
due to their small size and provide an effective protection. One marketing
advantage of inorganic particles is the ability to provide broad-spectrum
protection in a non-irritating sunscreen product. Certain organic active
agents, including avobenzone, which provides full UVA shielding, can
cause skin irritation. As a result, TiO
2
and ZnO are finding increasing
application in sensitive skin and baby products and daily-wear skin
lotions. One concern regarding the use of metal oxide nanoparticles, is
that upon absorption of UV radiation, they release free radicals, which
can damage DNA, and thus maybe prove to be carcinogenic. Therefore,
suppliers of nanoparticles generally offer the particles with coatings,
which cause the free radicals to recombine before entering the skin.
However, recent concern about the fate of the particles when applied to
the skin, as they possibly can penetrate much deeper than microparticles
(see chapter 6.3), may complicate the use of organic and inorganic
nanoparticles in cosmetics. Applications of nanoparticles in medicine are
e.g. markers for biological screening tests (e.g. gold or semiconductor
particles), contrast agents for magnetic resonance imaging (MRI) as well
as antimicrobic coatings and composite materials for abacterial surfaces
and medical devices (Salata 2004).
In the field of catalysis the biggest market volume can be assigned to
porous catalysts support for car exhaust catalysts. Nanoporous alumina
here serves as supporting material for noble metal catalysts, which were
finely dispersed on to the substrate. Nanoparticles will also find
increasing applications as catalysts in PEM fuel cells and hydrogen
reformers. The Business Communication Company (BCC) estimates the
world market volume of metal oxide and metal nanoparticles at 750 mill.
EURO in 2005. Table 4 gives an overview on applications in different
industrial branches.
Electronic, optoelectronic
magnetic applications
Biomedical, pharmaceu-
tical cosmetic applications
Energy, catalytic
structural applications
Chemical–mechanical
polishing
Electroconductive
coatings
Magnetic fluid seals and
recording media
Multilayer capacitors
Optical fibers
Phosphors
Quantum optical devices
Antimicrobials
Biodetection and
labeling
Biomagnetic
separations
Drug delivery
MRI contrast agents
Orthopedics/implants
Sunscreens
Thermal spray coatings
Automotive
catalyst
Membranes
Fuel cells
Photocatalysts
Propellants
Scratch-resistant
coatings
Structural ceramics
Solar cells
Table 4: Current and emerging applications of nanoparticles (source: Rittner 2002)
Industrial applications and market potentials 13
3.1.1.1 Carbon
Nanostructured carbon comprise long established mass produced
materials like carbon black as well as relatively new compounds like
fullerenes and carbon nanotubes (CNT). At present, conventional ma-
terials like carbon black are clearly dominating the world market with a
sales volume of about 5 billion EURO (SRI 2002). Carbon black consists
of chainlike aggregates of carbon nanoparticles, which have an average
primary particles size of a few nanometers and are mainly used as fillers
for rubber in car tyres or pigments in toners for photocopiers.
Fig. 4: Different modifi-
cations of carbon nano-
tubes, (single-walled,
multi-walled, filled with
metal atoms, etc.)
For CNT, which can occour single- or multiwalled, a big market
potential is forecasted due to their outstanding properties, e.g. extremely
high tensile strength (theoretically approx. 100 times stronger than steel)
and excellent thermal and electric conductance (CMP 2003). The main
barrier to a broad economic use of carbon nanotubes, e.g. in sensor
technology, electronics (CNT based connects and transistors), composite
materials (e.g. electrically conductive polymers) or flat screens (electron
emitters in field emission displays) is due to the high price of approx. 150
EURO per gram for single wall CNT (Loefken and Mayr 2003). The
high price reflects the early undeveloped stages of industrial production
and purification. While the present market potential of CNT lies within
the range of some million EURO, very optimistic prognoses forecast a
world market size of 1 billion EURO already for the year 2006 (Fecht et
al. 2003). However, these predictions will strongly depend on whether a
cheap production of carbon nanotubes on an industrial scale can be
implemented and significant performance gains in comparison with
conventional products can be achieved.
3.1.2 Nanoclays
Nanoclays as fillers
for polymers
Fig. 5: SEM Image
showing the morpholo-
gy of nanoclay particles
(source: IRC London)
Nanostructured organically modified layer silicates (nanoclays) have
been used for some time as fillers in polymers for improving barrier
characteristics (e.g. gas tightness), as flame-retardant and also as
mechanical reinforcement. Although some products are already on the
market, problems during the manufacturing process as well as the
relatively high price and only moderate performance gains impair a broad
economic application of these materials. Up to the year 2006 the world
market size for nano-layer silicates is estimated at 21 million EURO (SRI
2002).
3.1.3 Organic nanoparticles
Organic nanoparticles with economic relevance can be classified as
follows (Horn und Rieger 2002):
Polymer nanoparticles/-dispersions
14 Industrial application of nanomaterials – chances and risks
Micronised drugs and chemicals (vitamines, pigments and
pharmaceuticals)
Macro molecules (e.g. dendrimers)
Micells, liposomes
Nanostructured
vitamines, pigments
and drugs with an
improved
effectiveness
Currently only micronised drugs, vitamins and polymer dispersions have
a significant economic contribution. Through micronisation of organic
compounds such as vitamines, pigments and pharmaceuticals, which
often have a low solubility in water and require special formulation
procedures when applicated in aqueous solution, the increased surface-to-
volume ratio improves the water solubility significantly and thus
optimises the physiological (in pharmacy, cosmetic, crop protection,
nutrition) or technological effectiveness (e.g. in lacquers and printing
inks). Such nanoparticles can be made by mechanical milling or
precipitation and/or condensation of colloidal solutions. The world
market potential for organic nanoparticles (in particular vitamines) has
been estimated to approx. 1 billion EURO in the year 2002 (Ebenau
2002).
A still larger market with approx. 15 billion EURO in the year 2002
exhibit aqueous polymer dispersions (Distler 2002). These material class
is long established in industry but can be optimised by application of
modern nanotechnological procedures, e.g. increasing the solid content
due to a controlled particle size distribution, selective surface
modification of the polymers or the production of nanocomposites by
mixing with organic or inorganic fillers. Such polymer dispersions offer
broad application fields, e.g. as binders in colors and lacquers, adhesives
for labels and tapes or as coating systems for textiles, wood or leather.
Beside that, aqueous polymer dispersions are more environmentally
benign than products, which are based on organic solvents.
Organic macromolecules such as dendrimers and hyperbranched
polymers (e.g. on polyurethane basis) are used in the niche markets but
might have a promising future (Bruchmann 2002). Application potentials
of dendritic molecules can be seen for example as supports for catalysts
or pharmaceutical active substances (Drug Delivery) or as cross-linking
materials for scratch-proof autolacquers or printing inks. The world-wide
market potential of dendrimers is estimated at 5-15 million EURO in the
year 2006 (SRI 2002).
3.2 Nanocomposites
Nanoparticles and –fibers are often used as reinforcement for other
material classes such as polymers, ceramic or metals to yield nano-
composites with special properties.
Industrial applications and market potentials 15
3.2.1 Polymer nanocomposites
Polymer nanocomposites comprises block copolymers as well as polymer
materials, which are doped with ceramic, silicates, metal or also
semiconductor nanoparticles. The incorporation of nanoparticles into the
polymer matrix serves the improvement of material properties e.g.
(thermo-)mechanical and electronic characteristics. The following
examples can be mentioned:
Nanoparticle filled
polymers with
improved
mechanical and
electric properties
Nanoclay doped polymers for improvement of barrier properties (e.g.
gas tightness), as flame retardant or mechanical reinforcement
Nanoparticle doped epoxies as insulation for electric car cables or for
improved resins in coils
Electric conductive polymers, e.g. doped with carbon black or
henceforth with carbon nanotubes, for applications as electrostatic
shielding of electronic devices, etc.
Nanoparticle doped (e.g. silver) polymers with antimicrobic
properties for applications in medicine and hygiene
In the medium-term a strong market growth is expected for the world
market of polymer nanocomposites from 13 mill. EURO in 2001 up to
250 mill. EURO in 2006 (SRI 2002).
3.2.2 Metal matrix composites
By reinforcement of metals with ceramic fibers, in particular silicon
carbide, but also alumium oxide or aluminum nitride, their thermo-
mechanical properties can be improved significantly. Such metal matrix
composites (MMC), e.g. SiC in aluminum alloys or TiN in Ti/Al alloys,
possess due to their high heat resistance, hardness, thermal conductivity,
controllable thermal expansion and low density, a high potential for
structural applications in aerospace or the automotive sector.
3.2.3 Ceramic nanocomposites
Within ceramic nanomaterials a special focus lies on the production of
controlled micro/nano-structured grain sizes, the production of gradient
materials as well as application of nanostructured coatings and surface
functionalisation. One objective is the improvement of thermomechanical
properties, fracture toughness and formability ("super-plasticity") of this
brittle material class. In addition, the sintering temperatures and the
consolidation time of ceramic materials can be reduced by applying
nanopowders, which saves not only money but also allows new
manufacturing techniques like coprocessing of ceramics and metals.
Ceramic nanopowders meanwhile can be manufactured with high
chemical purity and adjustable powder grain size. Both gas or liquid
phase processes are used for the production of ceramic nanopowders, for
non-oxidic powders (e.g. Si
3
N
4
, SiC, TiCN) preferentially gas phase
processes and for oxidic powders (e.g. Al
2
O
3
, SiO
2
) also sol gel
16 Industrial application of nanomaterials – chances and risks
procedures. A further relevant topic are nanostructured gradient
materials, in which the gradient can be adjusted both regarding
thermomechanical or chemical properties. These materials could be used
for example in the production of photonic structures in optical data
communication or in the production of micromechanical and
microelectronic components with a high degree of miniaturisation.
3.2.4 Application fields
The following table gives an overview on potential markets, market
segments and products based on nanoparticulate materials.
Automotive industry
lightweight
construction
painting (fillers, base
coat, clear coat)
catalysts
tires (fillers)
sensors
Coatings for wind-
screen and car bodies
Chemical industry
fillers for paint systems
coating systems based
on nanocomposites
impregnation of papers
switchable adhesives
magnetic fluids
Engineering
wear protection for
tools and machines
(anti blocking coatings,
scratch resistant
coatings on plastic
parts, etc.)
lubricant-free bearings
Electronic industry
data memory (MRAM,
GMR-HD)
displays (OLED, FED)
laser diodes
glass fibres
optical switches
filters (IR-blocking)
conductive, antistatic
coatings
Construction
construction materials
thermal insulation
flame retardants
surface-functionalised
building materials for
wood, floors, stone,
facades, tiles, roof
tiles, etc.
facade coatings
groove mortar
Medicine
drug delivery systems
active agents
contrast medium
medical rapid tests
prostheses and
implants
antimicrobial agents
and coatings
agents in cancer
therapy
Textile/fabrics/non-
wovens
surface-processed
textiles
smart clothes
Energy
fuel cells
solar cells
batteries
capacitors
Cosmetics
sun protection
lipsticks
skin creams
tooth paste
Food and drinks
package materials
storage life sensors
additives
clarification of fruit
juices
Household
ceramic coatings for
irons
odors catalyst
cleaner for glass,
ceramic, floor,
windows
Sports /outdoor
ski wax
antifogging of
glasses/goggles
antifouling coatings
for ships/boats
reinforced tennis
rackets and balls
Table 5: Overview on applications of nanomaterial based products in different areas
17
4 PRODUCTION METHODS
Two basic
approaches to
produce nano-
materials: „top-
down“ and
„bottom-up“
There are two general ways available to produce nanomaterials (Moriarty
2001, Schmid et al. 1999) as shown in the following figure. The first way
is to start with a bulk material and then break it into smaller pieces using
mechanical, chemical or other form of energy (top-down). An opposite
approach is to synthesise the material from atomic or molecular species
via chemical reactions, allowing for the precursor particles to grow in
size (bottom-up). Both approaches can be done in either gas, liquid,
supercritical fluids, solid states, or in vacuum (Mayo 1993, 1996). Most
of the manufacturers are interested in the ability to control: a) particle
size b) particle shape c) size distribution d) particle composition e)
degree of particle agglomeration.
ENERGY
BUL
K
PARTICLES
ATOMS OR
MOLECULES
ENERGY
Figure 6: Two basic approaches to nanomaterials fabrication: top-down (shown here
from left to the right) and bottom-up (from right to the left)
4.1 Top-down approaches
Methods to produce nanoparticles from bulk materials include high-
energy ball milling, mechano-chemical processing, etching, electro-
explosion, sonication, sputtering and laser-ablation. These processes are
done in an inert atmosphere or in vacuum. Immediately after processing
nanoparticles are very reactive and can easily form agglomerates. If a
reactive gas is present some additional reactions may occur. This can be
used to coat nanoparticles with a material that would prevent further
interaction with other particles or the environment. In the following a
more detailed description of the basic nanomaterials manufacturing
techniques from bulk to nano are presented below.
18 Industrial application of nanomaterials – chances and risks
4.1.1 Mechanical milling
Fig. 8: A worker is
placing a milling vial
into planetary type ball
mill. Starting powder
and milling balls are
inside tighly closed
milling vial filled with
inert gas (source: VTT)
Mechanical milling is a process which is routinely used in powder
metallurgy and mineral processing industries. In this process, mixtures of
elemental or prealloyed powders are subjected to grinding under
protective atmosphere in equipment capable of high-energy compressive
impact forces such as attrition or shaker mills.
A variety of ball mills have been developed for different purposes
including tumber mills, attrition mills, shaker mills, vibratory mills,
planetary mills, etc. Powders with typical particle diameters of about 50
µm are placed together with a number of hardened steel or tungsten
carbide (WC) coated balls in a sealed container which is shaken or
violently agitated. Since the kinetic energy of the balls is a function of
their mass and velocity, dense materials are preferable to ceramic balls.
During the continuous severe plastic deformation associated with high-
energy mechanical attrition, a continuous refinement of the internal
structure of the powder particles to nanometer scales occurs.
Figure 7: Schematic diagram showing the different forms of impact which might occur
during high-energy ball milling (Zhang 2003)
When a single phase elemental powder or intermetallic compound
powder is milled, the grain size of the powder particles continues to
decrease until it reaches a minimum level – in the range of 3–25 nm. For
some intermetallic compounds, the powder becomes amorphous beyond
this point. For intrinsic brittle powders, such as silicon powder or carbide
and oxide powders, the reduction of the grain size is a natural outcome of
the transgranular fracturing and cold welding, and the minimum grain
size is determined by the minimum grain size which does not allow
nucleation and propagation of cracks within grains. No study has been
seen which attempts to theoretically determine this minimum grain size.
Very important advantage of the mechanical milling process is that the
processing temperature is low, so the newly formed grains grow very
slowly.
Production methods 19
Mechanical attrition methods allow the preparation of alloys and
composites which can not be synthesised via conventional casting routes,
e.g. uniform dispersions of ceramic particles in a metallic matrix and
alloys of metals with quite different melting points with the goal of
improved strength and corrosion resistance. Mechanical attrition has also
gained a lot of attention as a nonequilibrium process resulting in solid-
state alloying beyond the equilibrium solubility limit and the formation
of amorphous or nanostructured materials for a broad range of alloys,
intermetallics, ceramics and composites (Edelstein and Cammarata
1996).
High-energy mechanical milling is a very effective process for
synthesizing metal–ceramic composite powders as it allows incorporation
of the metal and the ceramic phases into each powder particle, as shown
schematically in the figure below.
Figure 9: Schematic diagram showing the formation of composite powder after high-
energy mechanical milling (Zhang 2003)
As previously mentioned high-energy mechanical milling can be used to
produce nanopowders. There are two routes for producing nanopowders
using mechanical milling: (a) milling a single phase powder and
controlling the balance point between fracturing and cold welding, so that
particles larger than 100 nm will not be excessively cold welded; and (b)
producing nanopowders using mechanochemical processes.
Mechanochemical Processing (MCP) is a novel, cost effective method of
manufacturing a wide range of nanopowders. MCP can most simply be
described as the use of a conventional ball mill as a low temperature
chemical reactor. It is important to realise that the ball mill is not being
used as a simple grinding tool. Instead, the ball mill increases the
reaction kinetics in the reacting powder mixture as a result of the intimate
mixing and refinement of the grain structure to the nanometer scale,
allowing the reaction to occur during the actual milling. Chemical
reactions, which normally require high temperatures, are thus activated
during milling. This is the key element of the MCP technology.
20 Industrial application of nanomaterials – chances and risks
To produce nanoparticles of a specific material, a suitable precursor is
chosen. Often a particular product can be produced from a range of
precursors allowing the process to be optimised to use industry standard
precursors to reduce cost. Oxides, carbonates, sulphates, chlorides,
fluorides, hydroxides or other compounds are all candidates for use as the
precursor material. The chosen precursor is then milled with an
appropriate reactant. The resulting product phase is formed as individual
single nanometer sized grains in a by-product matrix. After milling a low
temperature heat treatment is often used to ensure the reaction is
complete before the by-product is removed, leaving the pure, non-
agglomerated nanopowder, which consists of dispersed nano-sized
particles of 1-1000 nm in diameter (Froes et al. 2001).
One simple example is described as follows: The process starts by high-
energy milling a mixture of FeCl
3
powder and Na pieces. The milling
induces a reaction between FeCl
3
and Na, forming Fe nanoparticles
mixed with NaCl. The NaCl can be easily leached out from the powder
by using water, and Fe nanopowder is produced (Ding et al. 1995).
4.1.2 Etching (chemical)
A combination of lithographically defined patterning with etching is a
basis of microelectronics. Regular arrays of the nanometer-sized
structures can be produced on a planar substrate. Unmasked
electrochemical or photo-electrochemical etching can be used to produce
regular arrays of shapes within nanometer range. For example, layers of
porous silicon are formed by electrochemically etching the crystalline
silicon wafers, employing a mixture of hydrofluoric acid and ethanol as
an electrolyte. Another example is porous alumina.
4.1.3 Electro-explosion (thermal/chemical)
Electro-explosion involves providing a very high current over a very
short time through thin metallic wires, in either an inert or reactive gas,
such that extraordinary temperatures are achieved. The wire is converted
into a plasma state, but the plasma is contained and is in fact compressed
by the very high fields produced during the pulse. The very high currents
heat the wire to 20.000 – 30.000 degrees, and at these temperatures the
resistivity of the metal becomes virtually infinite, terminating the flow of
current. At that point the electromagnetic field disappears and
superheated metal plasma expands with supersonic velocity creating a
shock wave in the ionised gas surrounding the wire. The extremely fast
cooling (10
6
to 10
8
deg/sec) rate provides ideal conditions for
stabilisation of different metastable structures.
The process of electro-explosion of wire has prepared metallic powders
of approximately 100 nanometers, where an electric power impulse is
applied to the wire under argon pressure. The resulting powders have
Production methods 21
greater chemical and metallurgical reactivity as compared to other
powders. They also have internal strain and surface energies that are
released as the powders go through a transformation from their active as-
produced state to form sub-micron spheres. When turned into pellets and
heated to their transition temperatures, which is ordinarily well below
their melting points, they will release heat to cause the compacts to "self-
sinter". Their reactivity allows alloying to occur at substantially reduced
temperatures. Examples include a mixture of electro-exploded aluminium
and amorphous boron, which react to form aluminium diboride by
igniting the pellet with an electric wire, and where a pellet of electro-
exploded copper and zinc will react at 200° C to form brass directly
(Argonide 2004).
4.1.4 Sputtering (kinetic)
The impact of an atom or ion on a surface produces sputtering from the
surface as a result of the momentum transfer from the incoming particle.
Unlike many other vapour phase techniques there is no melting of the
material. Sputtering is done at low pressure on a cold substrate.
4.1.5 Laser ablation (thermal)
A broad range of
nanoparticulate
materials can be
obtained by laser
ablation
In laser ablation, pulsed light from an excimer laser is focused onto a
solid target inside a vacuum chamber to "boil off" a plume of energetic
atoms of the target material (Ullmann et al. 2002). A substrate positioned
to intercept the plume will receive a thin film deposit of the target
material. This phenomenon was first observed with a ruby laser in the
mid-1960s. Because this process then contaminated the films made with
particles, little use was found for such "dirty" samples.
Laser ablation method has the following advantages for the fabrication of
nanomaterials:
a) the fabrication parameters can be easily changed in a wide range b)
nanoparticles are naturally produced in a laser ablation plume so that the
production rate is relatively high c) virtually all materials can be
evaporated by laser ablation. A modification of this technique includes
laser ablation of microparticles (LAM), which helps to reduce size
dispersion
1
.
4.2 Bottom-up approaches
Methods to produce nanoparticles from atoms are chemical processes
based on transformations in solution e.g. sol-gel processing, chemical
vapour deposition (CVD), plasma or flame spraying synthesis, laser
pyrolysis, atomic or molecular condensation. These chemical processes
1
http://www.ph.utexas.edu/~laser/laser_stuff/nano1/Nano-Web.html
22 Industrial application of nanomaterials – chances and risks
rely on the availability of appropriate “metal-organic” molecules as
precursors. Sol-gel processing differs from other chemical processes due
to its relatively low processing temperature. This makes the sol-gel
process cost-effective and versatile. In spraying processes the flow of
reactants (gas, liquid in form of aerosols or mixtures of both) is
introduced to high-energy flame produced for example by plasma
spraying equipment or carbon dioxide laser. The reactants decompose
and particles are formed in a flame by homogeneous nucleation and
growth. Rapid cooling results in formation of nanoscale particles.
These are chemical processes to materials based on transformations in
solution such as sol-gel processing, hydro or solvo thermal syntheses,
Metal Organic Decomposition (MOD), or in the vapour phase chemical
vapour deposition (CVD). Most chemical routes rely on the availability
of appropriate “metal-organic” molecules as precursors. Among the
various precursors of metal oxides namely metal b-diketonates and metal
carboxylates, metal alkoxides are the most versatile. They are available
for nearly all elements and cost-effective synthesis from cheap feedstock
have been developed for some.
Two general ways are available to control the formation and growth of
the nanoparticles. One is called arrested precipitation and depends either
on exhaustion of one of the reactants or on the introduction of the
chemical that would block the reaction. Another method relies on a
physical restriction of the volume available for the growth of the
individual nanoparticles by using templates.
4.2.1 Sol-gel
Sol-gel process is a
long established
method for
producing
nanopowders
The sol gel technique is a long established industrial process for the
generation of colloidal nanoparticles from liquid phase, that has been
further developed in last years for the production of advanced
nanomaterials and coatings (e.g. Yu. 2001, Fendler 2001, Meisel 1997).
Sol-gel-processes are well adapted for oxide nanoparticles and
composites nanopowders synthesis. The main advantages of sol-gel
techniques for the preparation of materials are low temperature of
processing, versatility, flexible rheology allowing easy shaping and
embedding. They offer unique opportunities for access to organic-
inorganic materials. The most commonly used precursors of oxides are
alkoxides due to their commercial availability and to the high liability of
the M-OR bond allowing facile tailoring in situ during processing.
2
2
http://www.solgel.com/articles/jun02/preintro.asp
Production methods 23
Figure 10: System model for nanocomposites produced by sol-gel (source: Fraunhofer
IST).
4.2.2 Aerosol based processes
Aerosol based processes are a common method for the industrial
production of nanoparticles (e.g. Gurav 1993, Kammler 2001, Pratsinis
1998). Aerosols can be defined as solid or liquid particles in a gas phase,
where the particles can range from molecules up to 100 µm in size.
Aerosols were used in industrial manufacturing long before the basic
science and engineering of the aerosols were understood. For example,
carbon black particles used in pigments and reinforced car tires are
produced by hydrocarbon combustion; titania pigment for use in paints
and plastics is made by oxidation of titanium tetrachloride; fumed silica
and titania formed from respective tetrachlorides by flame pyrolysis;
optical fibres are manufactured by similar process (Kodas and Hampden-
Smith 1999).
Traditionally spraying is used either to dry wet materials or to deposit
coatings. Spaying of the precursor chemicals onto a heated surface or
into the hot atmosphere results in precursor pyrolysis and formation of
the particles. For example, a room temperature electro-spraying process
was developed at Oxford University to produce nanoparticles of
compound semiconductors and some metals. In particular, CdS
nanoparticles were produced by generating aerosol micro-droplets
containing Cd salt in the atmosphere containing hydrogen sulphide.
4.2.3 Chemical vapour deposition
CVD consists in activating a chemical reaction between the substrate
surface and a gaseous precursor. Activation can be achieved either with
24 Industrial application of nanomaterials – chances and risks
temperature (Thermal CVD) or with a plasma (PECVD : Plasma
Enhanced Chemical Vapour Deposition). The main advantage is the non-
directive aspect of this technology. Plasma allows to decrease
significantly the process temperature compared to the thermal CVD
process. CVD is widely used to produce carbon nanotubes (Meyyappan
et al. 2003).
4.2.4 Atomic or molecular condensation
This method is used mainly for metal containing nanoparticles. A bulk
material is heated in vacuum to produce a stream of vaporised and
atomised matter, which is directed to a chamber containing either inert or
reactive gas atmosphere. Rapid cooling of the metal atoms due to their
collision with the gas molecules results in the condensation and
formation of nanoparticles. If a reactive gas like oxygen is used then
metal oxide nanoparticles are produced.
The theory of gas-phase condensation for the production of metal nano-
powders is well known, having been first reported in 1930
3
. Gas-phase
condensation uses a vacuum chamber that consists of a heating element,
the metal to be made into nano-powder, powder collection equipment and
vacuum hardware.
Aggregation Nanoparticles
N
ucleation Modification
0.1...10 µm
10...100 nm
+ Reactive gas
+ Heat
W-platen
°
Metal vapor
Flow
0.1...1 m/s
Carrier
gas
Filter deposition
Figure 11: Principle of Inert Gas Condensation method for producing nanoparticulate
material (source: FHG-IFAM, Bremen)
The process utilises a gas, which is typically inert, at pressures high
enough to promote particle formation, but low enough to allow the
production of spherical particles. Metal is introduced onto a heated
element and is rapidly melted. The metal is quickly taken to temperatures
far above the melting point, but less than the boiling point, so that an
adequate vapour pressure is achieved. Gas is continuously introduced
into the chamber and removed by the pumps, so the gas flow moves the
evaporated metal away from the hot element. As the gas cools the metal
vapour, nanometer-sized particles form. These particles are liquid since
they are still too hot to be solid. The liquid particles collide and coalesce
in a controlled environment so that the particles grow to specification,
remaining spherical and with smooth surfaces. As the liquid particles are
3
A.H. Pfund, Phys. Rev. (1930)1434
Production methods 25
further cooled under control, they become solid and grow no longer. At
this point the nanoparticles are very reactive, so they are coated with a
material that prevents further interaction with other particles
(agglomeration) or with other materials.
4.2.5 Supercritical fluid synthesis
Methods using supercritical fluids are also powerful for the synthesis of
nanoparticles. For these methods, the properties of a supercritical fluid
(fluid forced into supercritical state by regulating its temperature and its
pressure) are used to form nanoparticles by a rapid expansion of a
supercritical solution. Supercritical fluid method is currently developed at
the pilot scale in a continuous process.
4.2.6 Spinning
An emerging technology for the manufacture of thin polymer fibers is
based on the principle of spinning dilute polymer solutions in a high-
voltage electric field. Electro spinning is a process by which a suspended
drop of polymer is charged with thousands of volts. At a characteristic
voltage the droplet forms a Taylor cone, and a fine jet of polymer
releases from the surface in response to the tensile forces generated by
interaction of an applied electric field with the electrical charge carried
by the jet. This produces a bundle of polymer fibers. The jet can be
directed to a grounded surface and collected as a continuous web of
fibers ranging in size from a few µm’s to less than 100 nm.
4
4.2.7 Use of templates
Any material containing regular nano-sized pores or voids can be used as
a template to form nanoparticles. Examples of such templates include
porous alumina, zeolites, di-block co-polymers, dendrimers, proteins and
other molecules. The template does not have to be a 3D object. Artificial
templates can be created on a plane surface or a gas-liquid interface by
forming self-assembled monolayers (Huczko 2000).
4
http://www.polymers.dk/research/posters/ElectrospinningSUH.pdf
26 Industrial application of nanomaterials – chances and risks
4.2.8 Self-assembly
Nanoparticles of a wide range of materials- including a variety of organic
and biological compounds, but also inorganic oxides, metals, and
semiconductors- can be processed using chemical self-assembly
techniques (Meier 2000, Zhang 2002, Shimizu 2003, Shimomura 2000,
Tomalia 1999, Fendler 2001). These techniques exploit selective
attachment of molecules to specific surfaces, biomolecular recognition
and selfordering principles (e.g. the preferential docking of DNA strands
with complementary base pairs) as well as well-developed chemistry for
attaching molecules onto clusters and substrates (e.g. thiol (-SH) end
groups) and other techniques like reverse micelle, sonochemical, and
photochemical synthesis to realise 1-D, 2-D and 3-D self-assembled
nanostructures. The molecular building blocks act as parts of a jigsaw
puzzle that join together in a perfect order without an obvious driving
force present. Long-term and visionary nanotechnological conceptions
however go far beyond these first approaches. This applies in particular
to the development of biomimetic materials with the ability of self
organisation, self healing and self replication by means of molecular
nanotechnology. One objective here is the combination of synthetic and
biological materials, architectures and systems, respectively, the
imitation of biological processes for technological applications. This field
of nanobiotechnology is at present still in the state of basic research, but
is regarded as one of the most promising research fields for the future
(European Commission 2001).
Self Assembly of
nanomaterials using
biomolecular
recognition and
selfordering
principles
Fig. 12: Titania particle
coated with a nanometer
thick silica layer, the
inset shows the particle
morphology, (source:
American Chemical
Society)
4.3 Stabilisation and functionalisation of
nanoparticles
Due to their high reactivity nanoparticles have a high tendency to build
aggregates resp. agglomerates, which could lead to a loss of the desired
properties. Therefore it is often necessary to stabilise the nanoparticles
with additional treatments. The commercial success or failure of
nanoparticles in a particular application usually depends upon the ability
to prepare stable dispersions in water or organic fluids with controlled
rheology. In turn, the ability to prepare stable nanoparticle dispersions
with controlled rheology is enabled by tailoring nanoparticle coatings.
On the other hand coating nanoparticles with another material of
nanoscale thickness is a simple way to alter the surface properties of
nanoparticles. Core-shell structured nanoparticles have been shown to
display advanced optical, mechanical and magnetic properties.
One common method to stabilise or modify the reactivity of the
nanoparticle is the encapsulation with a molecular or polymeric layer. A
thin polymeric shell enables compatibility of the particles with a wide
variety of fluids, resins and polymers (Bourgeat-Lami 2002). In this way,
the nanoparticles retain their original chemical and physical properties,
but the coating can be tailored for wide variety of applications and
Production methods 27
environments, ranging from extremely non-polar (hydrophobic) to very
polar systems (Gerfin et al. 1997). One example is the Discrete Particle
Encapsulation (DPE) method patented by Nanophase Technologies
Corporation
5
.
Collecting nano-
particles in liquid
suspension to pre-
vent agglomeration
Another way to ensure the stability of the collected nanoparticle powders
against agglomeration, sintering, and compositional changes is to collect
the nanoparticles in a liquid suspension. For semiconducting particles,
stabilisation of the liquid suspension has been demonstrated by the
addition of polar solvent; surfactant molecules have been used to stabilise
the liquid suspension of metallic nanoparticles (Sailor and Lee 1997).
Alternatively, inert silica encapsulation of nanoparticles by a gas-phase
reaction and by oxidation in colloidal solution has been shown to be
effective for metallic nanoparticles (Mulvaney et al. 2000). For carbon
nanotubes which are usually generated as mixtures of solid morphologies
that are mechanically entangled or that self-associate into aggregates it is
often necessary to disperse the CNT in fluid suspensions to obtain a
regular orientation in the composite material resulting in unique
mechanical or electrical characteristics. Milling, ultrasonication, high
shear flow, elongational flow, functionalisation, and surfactant and
dispersant systems are used to affect the morphologies of carbon
nanotubes and their interactions in the fluid phase (Hilding et al. 2003).
Nanoparticles dispersed in aqueous solutions also tend to build
aggregates due to attractive van der Waals forces. By altering the
dispersing conditions repulsive forces can be introduced between the
particles to prevent the aggregation. There are two general ways of
stabilising nanoparticles in aequous solutions. Firstly by adjusting the pH
of the system the nanoparticle surface charge can be manipulated in such
way that an electrical double layer is generated around the particle.
Overlap of two double layers on different nanoparticles causes repulsion
and hence stabilisation. The magnitude of this repulsive force can be
measured via the zeta potential. The second method involves the
adsorption of polymers onto the nanoparticles in such way that the
particles are physically prevented from coming close enough for the van
der Waals attractive force to dominate. This is termed steric stabilisation.
A combination of these two mechanisms is called electrosteric
stabilisation and occurs when polyelectrolytes are adsorbed on the
nanoparticle surface (Caruso 2001).
5
www.nanophase.com
28
29
5 CHARACTERISATION AND DETECTION TECHNIQUES
One essential prerequisite for the development, manufacturing and
commercialisation of nanomaterials is the availability of techniques,
which allow the characterisation of their physical, chemical and
biological properties on a nanoscale level. Powerful analytical detection
and characterisation methods are also the basis of a risk assessement of
nanomaterials to investigate how nanomaterials behave under different
chemical and physical conditions, how they move and distribute in
different environmental compartments like water, soil and air and how
they interact with the biosphere and the human organism.
Meanwhile there is a considerable arsenal of detection and
characterisation methods for nanomaterials. These methods are normally
used in research laboratories for the study of nanomaterial properties.
However, most of them are not suitable for the realisation of systematic
on-line measurements for safety analyses (i.e detection in a continuous
mode in industrial environment). For example, microscopy methods as
well as X-rays spectroscopies are very powerful methods for the
determination of nanoparticles characteristics but their use requires large
instruments, UHV (Ultra-High Vacuum) and extensive sample
preparation. Moreover, they are not adapted to continuous analysis for
safety purposes. In the following sections the specification and
limitations of the main methods used for nanomaterial characterisation
will be briefly summarised.
Most detection
methods are not
suitable for a
continous online
monitoring of
nanoparticles
5.1 Atomic structure and chemical composition
The following paragraph presents some methods for the determination of
atomic structure and chemical composition of solid or liquid
nanomaterials. Though the techniques presented below are not
specifically used for nanomaterials, they can provide valuable
information on nanoscale material properties, which can differ
significantly from the bulk properties.
5.1.1 Spectroscopic methods
Spectroscopic methods such as vibrational, nuclear magnetic resonance,
X-ray and UV spectroscopies have been extensively used for the
characterisation of nanomaterials. The following paragraphes summarise
some examples.
5.1.1.1 Vibrational spectroscopies
Vibrational spectroscopies comprise Fourier Transform Infrared (FTIR)
spectroscopy and Raman Scattering (RS). These two methods are used to
investigate vibrational structure of molecules or solids. FTIR is well
adapted for organic compounds and is extensively used for the
30 Industrial application of nanomaterials – chances and risks
characterisation of carbon nanoparticles for the detection of fullerenes or
Polycyclic Aromatic Hydrogenated species (PAH). Both methods can be
performed on dry powders or on liquid suspensions. For FTIR, the
absorption spectra can be deduced from transmission measurements
through a KBr pellet with entrapped nanoparticles or directly on
nanoparticles in a reflection mode measurement (DRIFT).
2000 1500 1000 500
0,1
0,2
C262
C254
C261
Absorbance (a. u.)
W avenumber (cm
-1
)
Figure 13: FTIR spectra of different laser-synthesised carbon black samples with
varying fullerene content, fullerene signatures are indicated by the arrows (Tenegal et
al. 2003)
FTIR can be also used to determine the the crystallisation and grain sizes
in ceramic powders e.g. Si/C/N composites, where the spectra of the
nanostructured powders differ significantly from the coarser bulk
material (Dez et al. 2002).
5.1.1.2 Nuclear magnetic resonance
High resolution liquid and solid state NMR is another tool that has been
widely adapted for the characterisation of nanomaterials. To be
mentioned here are the characterisation of zeolites (Zhang et al. 1999),
the investigation of nanoscale effects such as hydrogen-bonding and
transfer (Limbach 2002) or the surface properties and chemistry of
nanolayer systems (Liu et al 2003).
5.1.1.3 X-ray and UV spectroscopies
X-ray and UV spectroscopies are used to investigate the electronic
structure of materials to deduce their atomic structure. Core levels,
valence and conduction band are probed using an X-ray or a UV
excitation. X-ray Photoemission Spectroscopy (XPS) or Electron
Spectroscopy for Chemical Analysis (ESCA) refers to the photoemission
of electrons produced by a monochromatic X-ray or UV beam (Briggs
Characterisation and detection techniques 31
1983). XPS spectrometers measure the kinetic energies of the electrons.
Due to the limited mean free path of the electrons in matter, only few
nanometric layers are investigated. Since binding energies are highly
sensitive to chemical bonding, a map of the bonding configuration is
obtained for surface layers. With the photoemission cross-sections,
chemical compositions of the surface material can be calculated and
compared to bulk chemical compositions. Surface values can differ
significantly as shown in the following table for Si/C/N/O nanopowders
(Gheorghiu et al. 1997).
Si C N O
Chem. anal. (bulk)
40 % 21 % 37 % 2 %
XPS (surface)
31.6 % 32.6 % 28.2 % 7.6 %
Table 6: XPS chemical composition. O and C atoms are concentrated at the surface of
the nanopowders
Another X-ray method used to investigate the conduction band of
materials is X-ray Absorption Spectroscopy (XAS), which involves
Extended X-ray Absorption Fine Structure (EXAFS) and X-ray
Absorption Near Edge Structure (XANES). The principle is based on the
absorption of a monochromatic X-ray beam by a core shell electron of
selected atomic species inside a sample. By changing the energy of the
incident beam around the binding energy, modulation of the absorption
cross section are observed and interpreted by an interference
phenomenon between the wave associated with the emitted electron and
the scattered waves emitted by the neighbouring atoms. XAS methods
are selective and well adapted for samples with low crystallinity. They
are local order techniques useful to follow the early stages of the
crystallisation of amorphous nanoparticles. The absorption spectra can be
measured by recording either the attenuation of an incident beam
(transmission), the electron yield or the fluorescence yield. In the two last
cases, modulations of the yield with the energy of the incident beam are
that of the absorption cross section. For the electron yield, the
information is more sensitive to surface for the same reasons as for XPS.
Measurements using the fluorescence yield are efficient to probe the
local environment of atomic species in diluted samples. They can be
performed on dry powders or on liquid suspensions. However, an
important drawback is that performing XAS measurements require
Synchrotron Radiation Facilities (SRF).
5.1.2 X-ray and neutron diffraction
To characterise nanoparticles atomic structure at larger scales, diffraction
techniques appear as the most powerful methods. They are based on the
32 Industrial application of nanomaterials – chances and risks
diffraction of an incident beam (X-ray, neutrons) by reticular planes of
the crystalline phases inside a sample. The beam (X-ray or neutrons) is
diffracted at specific angular positions with respect to the incident beam
depending on the phases of the sample. When crystal size is reduced
toward nanometric scale, then a broadening of diffraction peaks is
observed and the width of the peak is directly correlated to the size of the
nanocrystalline domains (Debye-Scherrer relation).
XRD and neutron diffraction are complementary methods used to obtain
a contrast effect on the diffracted beam. Indeed, diffracted intensities are
modulated by weighting factors, which differ between X-ray and neutron
due to the difference in the nature of interaction.
Intensity (a.u)
Figure 14: XRD pattern of a Si/C/N nanopowder. The broad peaks correspond to the
crystalline β-SiC phase. The Debye Scherrer analysis gives a crystal size of 5 nm β-SiC
crystals are mixed with an amorphous phase (source: CEA)
If the crystal size continues to decrease, peaks become broader until they
transform into a smooth oscillation (for amorphous structures), which can
be measured only with a sufficient bright beam of X-rays. Then, the
modulations give rise to a total radial distribution function by a Fourier
transformation. In this last case, the method is called Wide Angle X-ray
Scattering (WAXS) and is performed only on a SRF as for XAS. Neutron
diffraction requires also neutron facilities. At small angles (SAXS and
neutron), particles size can be obtained. XRD, WAXS, SAXS and
neutrons can be performed on dry nanopowders or on liquid suspensions.
5.2 Determination of size, shape and surface area
5.2.1 Electron microscopies
Electron microscopies are methods of choice to investigate particles size,
shape and structure and also agglomerates. They regroup two techniques:
Scanning Electron Microscopy (SEM) or Transmission Electron
Microscopy (TEM).
Characterisation and detection techniques 33
In SEM experiments, electrons emitted from a filament are reflected by
the sample and images are formed using either secondary electrons or
backscattered electrons. However, in the case of SEM, a field emission
microscope (FE-SEM) is necessary to investigate the nanometric scale
(electrons are emitted from a field-emission gun). FE microscopes could
reach resolutions of the order of 1 nm using a cold cathode. If they are
equipped with an Energy Dispersive Spectrometer (EDS), chemical
composition can be obtained. Then, size distribution, shape and chemical
composition of nanoparticles can be investigated by FE-SEM.
Fig. 15: FE-SEM image
of cold-compacted
Si/C/N nanopowders
(source: CEA)
100 nm
In TEM experiments, electrons pass through the sample and the
transmitted beam is used to build the images. As for FE-SEM, shape and
size distribution can be obtained by TEM. Size distribution (with low
statistic) can be obtained by counting the number of particles as a
function of the size on the micrograph. At lower magnifications, the way
in which nanoparticles are connected can be observed. Then, qualitative
information about the agglomerates structure is deduced from
observations.
100
nm
0
10
20
30 40 50 60
(%
)
Diameter
(
n
m
)
SiCN
Figure 16: left: shape determination of SiC nanostructures, right: size distribution of
Si/C/N nanoparticles derived from TEM Measurements (source: CEA)
The TEM resolution is below 1 nm for the High Resolution Microscopes
(HRTEM). High resolution is performed to look at crystal quality and
interfaces. EDS can be used for chemical composition determination.
STEM microscopes are field emission gun scanning/transmission
electron microscopes. The STEM combines the features of both TEM
and SEM. Analysis can be performed in transmission mode or in
scanning mode. In scanning mode, a high brightness source produces a
focused beam with high current density and small diameter for EDS
microanalysis. Spatial resolution for microanalysis (about 2 nm for thin
specimens) is much better than it is for microanalysis in SEM for bulk
samples (about 0.5-3 microns). With STEM, EELS (Electron Energy
Loss Spectroscopy) can be performed. This method allows measurements
of the concentration profile of nanoparticles.
34 Industrial application of nanomaterials – chances and risks
Figure 17: Concentration profile obtained by EELS for one nanometric grain of one
Si/C/N nanopowder (Monthioux et al. 2000, data from N. Lebrun, LPS, Paris, France)
SEM and TEM are performed on dry powders. For SEM, environmental
microscopes are available to perform analysis on wet samples e.g. in
biology and medicine. Electron microscopies are powerful methods to
investigate size distribution, shape, chemical composition but also phases
analysis (nature and repartition) but require an extensive sample
preparation.
5.2.2 BET and pycnometry
Specific surface area and density of nanoparticles are obtained using the
Brunauer Emmet Teller (BET) method and helium pycnometry. BET is
based on the measurement of the adsorption isotherm of an inert gas (N
2
)
at the surface of the particles. The surface determination is performed for
an adsorbed volume corresponding to a monomolecular adsorption.
Helium pycnometry is based on the variation of the helium pressure (in a
calibrated cell) produced by a variation of volume. Nanoparticles are pre-
compacted into small pellets and put into the cell. Helium pycnometry is
a measurement of the true bulk density of the particles if they do not
contain closed pores. BET and He pycnometry are performed on dry
powders. By coupling the results obtained by these two methods, an
average grain size can be calculated if the particles are isolated, spherical
with a monodisperse size distribution. The value of d given by BET and
helium pycnometry can be compared to the value obtained with electron
microsocopies. If the diameters are equal, then grains do not contain any
open porosity. If the value of the model is quite smaller than that
observed by microscopy, then particles contain open porosity. BET and
helium pycnometry are performed on outgassed dry powders.
Characterisation and detection techniques 35
5.2.3 Epiphaniometer
The epiphaniometer (Gäggeler et. al. 1989) is an instrument developed at
the Paul Scherrer Institute in Switzerland that measures the surface
concentration of aerosol particles in both the nuclei and accumulation
mode size ranges. The epiphaniometer is most sensitive to particles in the
accumulation mode, but Gäggeler et al. (1989) reported successfully
measuring silver particles between 20 and 90 nm that were agglomerates
formed from smaller primary silver particles. The maximum
concentrations the epiphaniometer can handle is not reported, but it is
capable of measuring low atmospheric particle concentrations found in
remote locations. In an epiphaniometer, aerosol is passed through a
charging chamber where lead isotopes created from a decaying actinium
source are attached to the particle surfaces. The particles are transported
through a capillary to a collecting filter. The epiphaniometer uses a
surface barrier detector to measure the level of radioactivity of the
particles collected on the filter. The amount of radioactivity is
proportional to the particle’s Fuchs surface area and follows Fuchs theory
of attachment of radioactive isotopes. Because of the short half-life of the
lead isotopes, the filter does not become saturated and essentially real-
time radioactivity measurements can be made. Although not clearly
stated in the studies, the surface barrier detector measures radioactivity
and must be related to the size of particles being sampled to get a
measure of surface area of the particles. A pre-classifier, such as a DMA,
may be used before the epiphaniometer to allow a determined range of
particles to enter the instrument.
5.2.4 Laser granulometries and Zeta potential
Laser granulometries are statistical methods for the determination of
quantitative particles size distributions. These methods are based on the
diffraction/scattering of a laser beam by particles in stable suspensions.
The first method is based on laser diffraction. The diffraction pattern
(width of the ring and intensity) is directly connected to the particles size.
However, sizes lower than λ/20 are not observable by this method.
Practically, only particles with sizes higher than 80 nm can be charac-
terised using laser diffraction. Quantification of the size distribution
using the Mie theory can be performed if refractive indexes (particles and
solution) are known.
The second method more adapted to ultrafine particles is based on photon
correlation (PCS). This method measures at selected angles the variation
of the scattered intensities (due to the Brownian motion of particles) as a
function of time
6
. An autocorrelation function giving changes of intensity
as a function of time is measured and size distribution is extracted. As for
6
A. Rawle, “PCS in 30 minutes”, info-brochure Malvern Instruments Ltd., U.K.
36 Industrial application of nanomaterials – chances and risks
laser diffraction, PCS requires the prior knowledge of solution viscosity
and refractive index. PCS is highly sensitive to the presence of
agglomerates. If agglomerates are present, the autocorrelation function
will be dominated by their signal. The sensitivity of the detection is
highly dependent on the size of the particles/agglomerates. Sizes as low
as 1 nm can be characterised by PCS. Both methods (laser diffraction and
PCS) are well adapted for diluted or ultra-diluted stable suspensions.
They work for dense, spherical low-absorbing particles.
Laser granulometries are frequently coupled with Zeta potential
analysers. Zeta potential is a measurement of charges carried by the
particles in suspensions. The principle of the commonly used Zeta
potential analyser is based on electrophoresis. Zeta potential
measurements are used to characterise stability of the suspensions using
electrostatic repulsion.
5.2.5 Elliptically polarised light scattering
A new method based on laser light scattering was recently developed to
investigate size distribution, shape distribution but also agglomerates
structure and size distribution (Pinar 2003). Unlike laser granulometries,
the incident laser beam is elliptically polarised and modifications of the
polarisation state due to sample are measured at specific angular
positions. Polarisation analysis gives complementary information about
shape and agglomerate structure.
Fig. 19: Agglomerates
size distribution for
SnO
2
particles with a
fractal dimension of 2.1
Figure 18: Size distribution is obtained for particles with different aspect ratio
The method determines the fractal dimension (D
f
) of the agglomerates.
This parameter is a function of the agglomerates structure (linear, porous,
compact).
Characterisation and detection techniques 37
Figure 20: Different types of agglomerate structures
In the field of nanomaterial research, the way in which nanoparticles are
agglomerated is important since many properties depend on this
parameter. Agglomerates size can also be characterised by this method
with an accessible range between 50 nm and 2 µm.
5.3 Determination of nanoparticles in aerosols
Most critical for assessing exposure of people and workers is the
determination of nanoparticles in air. Whilst mass is the current metric
for measuring exposure to the coarse aerosol fractions, there is evidence
to suggest that it may not be so for the ultrafine particle fraction.
Relevant metrics for nanoparticle exposures may be the number
concentration, size distribution, surface area or morphology.
Mass is not a
suitable metric for
measuring ultrafine
particle exposure
Only few devices
are capable of
measuring number,
size-distribution or
surface area of
ultrafine particles
Technologies to measure these metrics for nanoparticles are not readily
available, particularly in a form which may be used to measure personal
exposure on a routine basis. Personal exposure measurements are
preferred for workplace measurements as they give better exposure
estimates and therefore lead to better risk assessment and risk
management. Methods available for determining the number concen-
trations of airborne nanoparticles include the CPC (Condensation Particle
Counter), which counts particle number, the SMPS (Scanning Mobility
Particle Sizer and ELPI (Electrical Low Pressure Impactor) which count
particle number and give number-weighted particle size distribution
information. Both the SMPS and ELPI are relatively large and
cumbersome. Although the CPC is much smaller, it is not a personal
device and the lack of size discrimination is a major limitation. The
situation with the measurement of particle concentration in terms of the
surface area metric is even poorer. One instrument that has been
specifically developed to measure the surface area of airborne
nanoparticles is the epiphaniometer. This detects the radiation arising
from radon isotopes attached to aerosol particles when exposed to a
38 Industrial application of nanomaterials – chances and risks
radioactive lead source inside the instrument. It gives near real time
information on the total Fuchs surface area of collected particles in the
range 10 to 1000 nm. However, the instrument is large, with a
registerable radioactive source and has never been fully developed from
the experimental stage. Whilst rough estimates of surface area of airborne
particles can also be obtained from number size distributions (as outlined
above), there is currently no instrument (personal or static) suitable to
measure the surface area-weighted particle exposures of workers.
At the present time, few techniques exist for the detection of ultrafine
airborne particles in aerosols derived mainly from the automotive
industry (OICA 2002) or from the biomass combustion (Johansson
2002). Methods for for real-time particle measurements to be mentioned
here are:
DLPI (Dekati Low Pressure Impactor)
ELPI (Electrical Low Pressure Impactor)
SMPS (Scanning Mobility Particle Sizer)
Specificities of the analyser give the following detection limits: DLPI
and ELPI detect particles with size as low as 7 nm whereas SMPS can
detect particles with diameter as low as 3 nm
7
,
8
. Methods by impaction
(DLPI and ELPI) are based on inertial size classification (see figure
below). The device have two co-linear plates of which one has a small
nozzle in it. The sample aerosol passes through this nozzle at high speed
and makes a sharp turn with the flow between the plates. Particles with
sufficient inertia cannot follow the flow and impact on the second plate,
particles with small enough inertia remain in the flow and are impacted at
a subsequent stage.
Figure 21: Principle of DLPI and ELPI analyser (source: www.dekati.com).
7
http://www.dekati.com/dlpi.shtml, http://www.dekati.com/elpi.shtml
8
http://www.tsi.com/particle/downloads/brochures/3936.pdf
Characterisation and detection techniques 39
The cut diameter for one impactor is defined as the size of particles
collected with 50 % efficiency. Cascade impactor, as shown on the figure
above, consist of several successive impactor stages with decreasing cut
diameters. A DLPI impactor has 13 successive impactors. For DLPI, a
gravimetric analysis is performed whereas for the ELPI, particles are
preliminary charged before they enter the impactor and an electric
analysis is performed resulting in a fast response time (Kerkinen et al.
1991, 1992). SMPS method allows number size distributions (from 3 to
1000 nm) and number concentrations measurements (Wang and Flagan
1990). The aerosol first passes through a single-stage, inertial impactor.
This serves to remove large particles outside the measurement range that
may contribute to data inversion errors caused by multiple charging.
Then, particles of the aerosol are charged and enter the DMA
(Differential Mobility Analyser). At this point, particles are separated
according to their electrical mobility, by using their deviation in an
electric field produced by a charged rod. Only particles within a narrow
range of electrical mobility have the correct trajectory to pass through an
open slit near the DMA exit. At the exit of the DMA, particles enter a
Condensation Particle Counter (CPC). By changing the voltage of the rod
inside the DMA, the entire size distribution can be measured.
Figure 22: Principle of the DMA (Differential Mobility Analyser) of a SMPS analyser
(source: www.tsl.com)
The CPC (or CNC) instrument is used to count the particles emitted from
the DMA. The size-selected nanoparticles are flowing through a zone
saturated by n-butanol vapours and subsequently cooled to cause the
condensation of vapours at their surface (Harrison et al. 2000). Then
particles are growing to the order of 10 µm in diameter at which they are
efficient light scatterers. The particles can be counted as they pass
through a light beam (low number densities – single count mode) or by
40 Industrial application of nanomaterials – chances and risks
detecting the light scattered in a sensing zone (high number densities –
photometric mode). Typically, size ranges between 3 and 20 nm can be
characterised by CPC/CNC.
Figure 23: Principle of the Condensation Particle Counter (source: www.tsl.com)
Key benefits of the SMPS analyser are:
fast results (about 60 s or less)
high-resolution data, broad size range (3 to 1000 nm)
wide concentration range (1 to 10
8
particles per cubic centimeter)
simple control of operations.
Need for standar-
disation of meas-
urement and samp-
ling procedures
As it has been shown by a round robin test, SMPS is a valid tool to
measure size distribution and number concentrations of nanoparticles but
only with uniform instrument parameters (Dahmann et al. 2001).
Therefore there is an urgent need for the standardisation of measurement
and sampling procedures and conditions. Furthermore the emergence of
new sources of nanoparticle releases in industrial environments requires
the setting up of well-adapted techniques for the detection of new types
of nanoparticles. For analysers, the following specifications would be
desirable:
portable instruments
adapted to aerosols and liquids
continuous mode instruments
real-time diagnostics
personal exposure measurement systems
Among the existing characterisation methods, candidates for an online
measurement system for nanoparticles in work place atmospheres could
be based on light scattering methods (see chapter 5.2.4 to 5.2.5), which
have specifications close to those given above and which can be also
adapted to achieve detection in liquid effluents. For a detailed
Characterisation and detection techniques 41
characterisation of size and morphology of nanoparticles only off-line
measurement techniques such as Scanning Transmission Electron
Microscopy (STEM), High Resolution Transmission Electron
Microscopy (HRTEM) or Scanning Probe Microscopy techniques
(SNOM, AFM) are available.
Table 7 gives an overview on characterisation parameters and respective
measurement techniques for assessing nanoparticle exposure in aerosols.
Parameter Measurement Techniques
Number Concentration
Condensation Particle Counter (CPC)
Particle Number and number-
weighted particles size
distribution
Scanning Mobility Particle Sizer
(SMPS)
Electrical Low Pressure Impactor (ELPI)
Submicron particle surface
area
Epiphaniometer
Diffusion Charger
Size, morphology and surface
properties
(for collected particles)
STEM
HRTEM
SNOM
AFM
Table 7: Parameters and measurement techniques for assessing nanoparticle exposures
in the atmosphere (Harrison et al. 2000, Maynard 2000)
Beside the lack of personal exposure measurement systems also
measurement standards for a reliable and comparable nanoparticle
determination are presently not available. To perform an evaluation of
the existing detection techniques, nanoparticles produced in industrial or
pre-industrial environment must be completely characterised from their
atomic structure to their agglomeration using a combination of the above
mentioned complementary methods. The acquisition of detailed reference
data as well as intercomparisons and round robin tests will be necessary
to assess the reliability and the limitations of the applied detection
techniques. Moreover, the understanding of the interaction mechanisms
of nanoparticles with their environment requires a precise knowledge of
the characteristics of the nanoparticles produced from pre-industrial and
industrial processes.
5.4 Determination of nanoparticles in biological
tissue
Most of the above mentioned detection techniques for nanoparticles are
restricted to measurements in gaseous phase or solid phase. A critical
point for assessing effects of nanoparticle exposure on living organisms
and the environment are therefore measurement techniques which are
capable of analysing liquid and/or biological samples. This is important
because biological samples often require complex sample preparation
steps before nanoanalytical methods such as electron microscopy can be
42 Industrial application of nanomaterials – chances and risks
applied. Methods for nanoparticle detection in biological samples to be
mentioned here are Cryogenic Transmission Electron Microscopy (Cryo-
TEM) and Scanning Transmission Ion Microscopy (STIM). These
methods have been applied for example to assess, if nanoparticles used in
cosmetics (sunscreens, etc.) can penetrate into the human skin and
consequently may cause systemic effects (see chapter 6.3).
Cryo-TEM and STIM
for characteri-
sation of nanopar-
ticulate materials
in liquids and
biological tissue
The Transmission Electron Microscope (TEM) that has been widely used
in research in the fields of materials science and technology has now
become capable of observing specimens at atomic resolution and is
making valuable contributions to research and development of industrial
products. In recent years, with the progress of molecular biology, the
TEM has begun to be used in analyses of biological macromolecules
such as proteins (enzymes) and viruses on the molecular level. Also, it is
being applied to elucidating life phenomena, including those on the
molecular level, as well as to development and improvement of
industrial, pharmaceutical, and agricultural products. Also, recently,
strong demands for TEM observation of three-dimensional structures of
biological macromolecules in the hydrated state (the native state in living
bodies) at atomic resolution have been made.
However, there are two difficult problems in using the TEM for these
purposes. First, the path of the electron beam must be in vacuum and,
therefore, the specimen to be observed must also be kept in vacuum.
Second, damage of the specimen due to electron-beam irradiation is large
and must be reduced considerably. A method for overcoming these
difficulties and for observing the specimen at atomic resolution while
keeping it in the hydrated state is the so-called ice embedding method for
preparing a frozen specimen. Beiersdorf used the cryo-TEM method for
demonstrating the non-penetration of titania nanoparticles into the human
dermis (Pflücker et al. 2001).
Based on the idea similar to the Scanning Transmission Electron
Microscopy the Scanning Transmission Ion Microscopy (STIM) was
developed in the early 80‘s which uses an ion beam instead of an electron
beam. The main advantage of STIM is the greater penetration depth that
allows the analysis of much more thicker objects. This technique makes
imaging as well as mass normalisation possible at resolution down to 100
nm (and less in future). STIM has been applied e.g. by the university of
Leipzig for the investigation of percutaneous uptake of ultrafine TiO
2
particles (Menzel et al. 2004).
43
6 RISK ASSESSMENT
Risk assessment in general comprises several components including
hazard identification
hazard characterisation
exposure assessment
risk calculation
One the basis of a reliable risk assessment measures for risk management
have to be undertaken comprising preventive measures, standardisation
and regulation activities which are elaborated in chapter 7. The following
figure gives an overview of different aspects and components, which
have to be taken into account for the assessment and management of
risks associated with industrial nanoparticle production and use.
Particle Characteristics
• Aspect-ratio
• Diameter (particle/aggregate)
• Surface area/ properties
• Water solubility
• Chemical composition
Emission
• Production volume
• Material flows
• Potential
particle release
(production, use, disposal)
Health effects
Humans
• Experimental animals
Environmental effects
• Persistence
• Biomagnification
• Long range transport
1 Hazard identification 2 Hazard characterization
Epidemiological Studies
• Workers
• Consumers
• Exposed population
In vivo studies
• acute/chronic
• different species
In vitro studies
• Human/ animal, different cell types
• Models (lung, skin, systemic effects)
Exposure routes
• Inhalation, dermal, ingestion
Environmental monitoring
• Biological uptake
Occupational monitoring
• Personal exposure
3 Exposure assessment
4 Risk calculation
Susceptibility extrapolation
models
• high dose low dose
• animal human
Threshold value calculation
• Intake, immission concentration,
maximum workplace concentration
5 Risk Management
Preventive Measures
• Personal protection equipment
• Modification of processes
Standardization
• Measurement techniques
• Toxicological assessment
Regulation
• Exposure/ immission standards
• Production standards/restrictions
Figure 24: Components and aspects of risk assessment and management associated with
industrial nanoparticle production and use.
It should be pointed out, that many of the above mentioned aspects
concerning nanoparticles have not been investigated yet and are still
unknown. The following chapter tries to give an overview on existing
information with regard to the different steps of risk analyses. For the
risk assessement it is useful to distinguish different types of
nanoparticulate materials, whether they are dispersed in gaseous, liquid
or solid phase, whether they occur as single-particles or as agglomerates
or whether they are untreated or surface modified (see chapter 2).
Many aspects
concerning risk
assessment of
nanoparticles are
still unknown
Most critical with regard to potential health and environmental risk are
nanoparticles dispersed in air (aerosols), because of their mobility and the
possible intake into the human body via the lungs which represents the
most critical exposure route for humans. On the other side nanoparticles
dispersed in a solid matrix are much less likely to raise concerns because
of their immobilisation.
44 Industrial application of nanomaterials – chances and risks
As mentioned above aerosol nanoparticles tend to build larger
agglomerates with sizes in the µm-range. So the question arises whether
particle-size dependent phenomena like cell barrier crossing, etc. are still
valid for engineered nanomaterials. To answer this question investiga-
tions have to be performed, e.g. if agglomerates can deagglomerate in the
lung liquid or other biological liquids.
Another point which has to be considered in the risk assessment of
nanoparticles is the fact that natural aerosols also contain particles with
sizes between 100 µm to 10 nm and even smaller (Primmermann 2000).
The amount of nanoparticles is nearly the same in urban and rural areas
with as much as 10
6
to 10
8
particles/liter air. Whereas, in rural areas the
particles are mainly soil-derived and bioaerosols, in urban areas the
particle are mainly composed of Man-made materials (combustion and
mechanical abrasion, etc.) and bioagents. Hence one might expect that
over millions of years e.g. the lungs had to adapt their natural exposed
tissue to function and fulfil their work even in the present of 10
6
-10
8
nanoparticles/liter air and 10
2
particles/liter with a size between 100 nm
to 10 µm.
Natural aerosols
also contain ultra-
fine particles in
high concentrations
To sum up it has to be kept in mind when assessing risks of engineered
nanomaterials that:
At present combustion processes from traffic and energy generation
as well as mechanical abrasion processes contribute much more to
anthropogenic nanoparticle emissions than industrial nanoparticle
production
Industrial nanoparticulate materials usually build aggregates with
sizes in the µm-range
Also natural aerosols contain huge amounts of particles with sizes <
100 nm
Nevertheless due to fact that the next few years will probably see a
dramatic increase in the industrial generation and use of nanoparticles
and entirely new substance classes like carbon nanotubes are released
into the environment, a careful risk assessment of engineered
nanomaterials is obviously necessary.
6.1 Potential particle release
The objective of this part is to make an inventory of possible sources of
potential particle release in nanoparticle production processes during the
whole life cycle from nanoparticle generation to end products and finally
disposal. Due to a variety of different production methods for
nanoparticles, the process conditions vary widely and thus in principle
the risk of a potential particle release has to be considered separately for
each different process. However, most of the processes like plasma and
laser deposition as well as aerosol process are usually performed in
Risk assessment 45
evacuated or at least closed reaction chambers. Therefore, exposure to
nanoparticles is more likely to happen after the manufacturing process
itself except in the case of unexpected failure during the processing (e.g.
chamber failures causing leakages).
6.1.1 Nanoparticle production
Processes working at high pressure (supercritical fluid for example) or
with high energy mechanical forces (mechanical synthesis), particle
release could occur in the case of failure of sealing of the reactor or the
mills. Then large quantities of nanopowder could be released in a short
time into the atmosphere. For laser processes (laser pyrolysis/ ablation),
breaking of reactor laser windows (windows on the optical path of the
laser beam) is a possible source of release.
Potential particle
release in case of
failure of reactor
sealings or collec-
ting apparatus
Moreover, when sealing is broken, reactive mixtures can be put in
contact with air an in some case, violent chemical reactions can occur.
For example when silane is used for the synthesis of silicon based
nanopowders, the accidental contact with air provokes a spontaneous
very exothermic reaction with oxygen and flame can appear inside the
process unit or in the close environment.
Failure of collecting apparatus are also important sources of potential
release during the processes. The collecting apparatus must be able to
stop the nanoparticles and to evacuate effluents produced from the
processes. If collecting apparatus is designed for the recovery of
nanoparticles in a dry form (using filters for example), failure of filters
efficiency could be a source of potential release towards the evacuation
system (pumping unit for example). In order to avoid this kind of release,
gas and aerosols treatments units must be connected at the exit of the
plants to prevent release in atmosphere.
6.1.1.1 Mechanical processes
In mechanical milling processes raw material powders are usually
crushed together with process control agents (PCA) under inert gas
atmosphere. PCA can be in solid or liquid form. Risk arises when inert
gas atmosphere is removed from the milling vial. Fine particles are very
reactive at this stage. If the liquid PCA is used during the milling
process, the nanoparticles stay stable in suspension until it is dried. After
drying the suspension, release of nanoparticles to the surrounding
atmosphere may occur.
6.1.1.2 Vacuum processes
In vacuum processes nanoparticles are formed inside the chamber and
then collected on a substrate. These type of methods are e.g. CVD, PVD.
The manufacturing process itself is safe, but opening the chamber may
cause release of nanoparticles.
46 Industrial application of nanomaterials – chances and risks
6.1.1.3 Spraying methods
Different kind of spray methods are used to product nanoparticles.
Typical examples of these methods are plasma spray synthesis, flame
spray and laser pyrolysis. Liquid precursor/ fuel mixture is feeded into
the flame. Nanoparticles synthesised in the flame are collected as powder
e.g. in an electrostatic precipitator, baghouse filter or as a deposit on a
substrate. Plasma spray synthesis has been used even in open atmosphere
to produce nanoparticles. Flow velocity of nanoparticles in this process is
high and collecting all of the produced nanoparticles is a highly
demanding task. The manufacturing method is quite simple, unexpensive
and also suitable for mass-production, but efficient and safe particle
collecting system is required.
6.1.1.4 Sol-gel processes
Sol-gel processes are chemical methods based on hydrolysis or
condensation reactions. They are well adapted for oxide nanoparticles
synthesis. During the sol-gel processing nanoparticles are precipitated
from solution. By controlling the amount of reactants or by using the
chemical that blocks the reaction, precipitation can be arrested so that
nanosize particles are formed. If the precipitated nanoparticles can stay in
the solution, there will be no risk for release into the atmosphere.
However, drying of this solution and collecting of dry nanoparticles will
again arise risk for nanoparticle release.
6.1.2 Collection of nanoparticles
Risks are increasing during the collect of nanoparticles particularly in a
dry form. When opening collecting apparatus or reactors, nanoparticles
can be released and travel in air due to their high volatility. In gaseous
atmosphere the behaviour of dry nanoparticles is primarily determined by
the balance between attractive and lift forces. Gravity force has no
noticeable effect on nanoparticles. Therefore nanoparticles in gaseous
atmosphere will not settle down easily and may stay in the air as
impurities for a long time causing health risks via inhalation. Drastic
effects can be observed for metallic or non-oxide nanoparticles due to the
high pyrophoricity of dispersed nanopowders.
6.1.2.1 Dust explosion
When handling small particels there may arise risk for dust explosion,
especially in the case of metal powders. During various manufacturing
processes, dust or dust clouds may be generated. Once dust has formed
into the proper mixture with air, it can be ignited by energy from various
internal or external sources. Figure 25 summarises the general conditions
necessary for a dust explosion and/or fire to occur.
Risk assessment 47
Figure 25: Summary of conditions required for dust explosion with metal powders
(Dahn et al. 2000)
Many chemical and physical material properties, various atmosphere
conditions, the type and magnitude of energy of the ignition source
determine whether a dust cloud ignites and how intense the explosion
output is.
6.1.2.2 Ventilation
During collection of solid nanopowders special care must be taken with
regard to ventilation at the working place. Air streams could disperse
nanopowders to form aerosols. This could occur if fume cupboards are in
the proximity of pieces covered with nanopowders (parts of the reactors
or of the collecting apparatus, filters).
48 Industrial application of nanomaterials – chances and risks
Figure 26: Nanopowders deposit in a small laser-pyrolysis reactor (source: CEA). The
white arrow shows the deposit of nanopowders on the wall of a laser pyrolysis reactor
6.1.3 Cleaning operations
Nanoparticle release can also occur during cleaning operations of
reactors, after the disassembling, when nanoparticles have to be removed