Fire and Explosion Properties PDF

Health and Safety
Executive
Fire and explosion properties

of nanopowders
Prepared by the Health and Safety Laboratory

for the Health and Safety Executive 2010
RR782
Research Report
Health and Safety
Executive

of nanopowders
P Holbrow, M Wall, E Sanderson, D Bennett, W Rattigan,

R Bettis & D Gregory
Health and Safety Laboratory
Harpur Hill
Buxton
Derbyshire
SK17 9JN
Nanotechnology is a rapidly expanding technology in which existing and novel materials are engineered at
the nanoscale, typically in the range of 1 to 100 nanometres. Engineered nanomaterials include uniquely
manufactured products with unique shapes and enhanced physical and chemical properties, compared
with conventional materials of the same composition. There is currently little available information on the
explosion risks of these materials. The Health and Safety Executive therefore commissioned this project
to investigate the potential fire and explosion hazards associated with nanopowders. Test equipment
and procedures were developed to assess the key properties of a selected number of nanopowders.
A specialised 2 litre test vessel was developed to determine the explosion characteristics and modified
standard test apparatus was used to measure the minimum ignition energy of nanopowders. Resistivity
and electrostatic charging characteristics were assessed using specially designed test apparatus. Key
information including KSt, Pmax and MIE values were obtained for a range of metal and carbon nanopowders.
Generally, the explosibility (maximum explosion pressure, rates of pressure rise and equivalent KSt) of
nanopowders were found to be broadly similar to conventional micron-scale powders. However, the
minimum ignition energies of some nanopowders were found to be lower than the equivalent material at
micron-scale. It was demonstrated that with increasing relative humidity the resistivity of most nanopowders
decreases. There was also a tendency for nanopowders to have higher resistivity values than conventional
micron-scale powders. All the powders produced electrostatic charge. Generally, the charge developed by
nanopowders was comparable with the micron-scale powders.
This report and the work it describes were funded by the Health and Safety Executive (HSE). Its contents,
including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily
reflect HSE policy.
HSE Books
© Crown copyright 2010
First published 2010
All rights reserved. No part of this publication may be

reproduced, stored in a retrieval system, or transmitted
in any form or by any means (electronic, mechanical,
photocopying, recording or otherwise) without the prior
written permission of the copyright owner.
Applications for reproduction should be made in writing to:

Licensing Division, Her Majesty’s Stationery Office,
St Clements House, 2-16 Colegate, Norwich NR3 1BQ
or by e-mail to hmsolicensing@cabinet-office.x.gsi.gov.uk
ii
CONTENTS
1 INTRODUCTION......................................................................................... 1
2 LEGAL BACKGROUND............................................................................. 2
3 EXPLOSION SEVERITY............................................................................. 3
3.1 Standard test ........................................................................................... 3
3.2 Test vessel .............................................................................................. 4
3.3 Control and Instrumentation .................................................................... 7
3.4 Test procedure ........................................................................................ 8
4 MINIMUM IGNITION ENERGY ................................................................. 10

4.1 Test apparatus....................................................................................... 10
5 FIRE HAZARDS ....................................................................................... 13

5.1 Fire Hazards in nanopowders................................................................ 13
5.2 Practical Assessment of Fire Properties ................................................ 13
6 ELECTROSTATIC PROPERTIES ............................................................ 15

6.1 Electrostatic charge ............................................................................... 15
6.2 Charging test ......................................................................................... 16
6.3 Resistivity test........................................................................................ 17
7 TEST MATERIALS ................................................................................... 19
8 RESULTS ................................................................................................. 22
8.1 Commissioning tests – explosion severity ............................................. 22
8.2 Explosion severity of nanopowders ....................................................... 28
8.3 Commissioning MIE test results ............................................................ 34
8.4 MIE test results – nanopowders ............................................................ 35
8.5 Electrostatics ......................................................................................... 37
9 DISCUSSION............................................................................................ 54
9.1 Explosion test Equipment ...................................................................... 54
9.2 Explosion severity and MIE ................................................................... 54
9.3 Electrostatic issues ................................................................................ 58
10 CONCLUSIONS .................................................................................... 61
11 APPENDICES ....................................................................................... 63
12 REFERENCES ...................................................................................... 68
iii
iv
EXECUTIVE SUMMARY
Objectives
Nanotechnology is a rapidly expanding technology in which existing and novel materials are
engineered at the nanoscale. However, there is currently little available information on the fire
and explosion risks of nanoparticles. The objective of this project is to characterize the fire and
explosion properties of a range of commercially available nanopowders.
Main Findings
1. A unique facility has been developed specifically to allow the safe handling and testing
of nanopowders. The equipment comprises:
i) 2 litre explosion test apparatus for measuring the rate of pressure rise and
the maximum explosion pressure.
ii) Minimum ignition energy test apparatus. The minimum ignition energy of
nanopowders was measured using a modified Kuhner MIKE3 test apparatus
modified to allow nanopowders to be safely handled within the test
apparatus.
iii) Sealed systems that allow the safe handling of oxidized and non-oxidized
nanopowders under inert atmospheres.
2. Explosion properties of a range of nanopowders have been characterised including

aluminium, iron, zinc, copper and several carbon nanomaterials. Generally, the
explosibility (maximum explosion pressure, rates of pressure rise and equivalent KSt) of
nanopowders are broadly similar to conventional micron-scale powders. The minimum
ignition energies of some nanopowders have been lower than the equivalent material at
micron-scale. This indicates that the nanopowders may be more susceptible to ignition
but once ignited the explosion violence is no more severe than micron-scale powders.
3. Electrostatic properties of nanopowders have been considered. Dust layer resistivity

measurements and charge tests have been made on a range of nanopowders.
Recommendations
To build on the data already obtained, and develop further expertise and knowledge, further
nanopowders should be obtained and characterised.
v
vi
1 INTRODUCTION
Nanotechnology is a rapidly expanding technology in which existing and novel materials are
engineered at the nanoscale. Engineered nanomaterials include uniquely manufactured products
with unique shapes and enhanced physical and chemical properties, compared with conventional
materials of the same composition.
The Health and Safety Executive has commissioned a project to investigate the potential
hazards associated with nanopowders. The general method of approach is to use standard, and
modified, fire and explosion characterization methods to assess the key properties of a selected
number of nanopowders.
As part of this project, a review (HSL report number XS/08/119) has been carried out to update
a review done by HSL in 2003 and to ensure that all relevant new material in this fast changing
area is incorporated.
There is currently little available information on the explosion risks of nanoparticles. The little
data that does exist is contradictory and suggests the explosion violence is less than that of
larger particles, which goes against the expectation that smaller particles burn faster than larger
ones due to the increased surface area.
Fine powders are known to be an explosion risk, particularly organic and metallic powders.
For fine particulates, there is standard test equipment available to determine the explosion and
ignition characteristics for powders, but these typically require kilogram quantities of powder.
Also, the powder is dispersed in the standard apparatus using compressed air. With nano-
powders, their large surface to volume ratio means that many are spontaneously flammable on
contact with air, or surface oxidation alters their properties. Hence equipment that avoids
oxidation until the point of ignition has been developed.
This project report describes the design, operation of test apparatus and results obtained relating
to the explosion properties of nanopowders. Key information including explosion severity,
minimum ignition energy and electrostatic properties is reported.
1
2 LEGAL BACKGROUND
The major legal regulations that could apply to nanopowders include the following:
DSEAR Regulations
It is the duty of employers to comply with the Dangerous Substances and Explosive
Atmosphere Regulations 2002 (DSEAR) which seek to eliminate, reduce and control the fire
and explosion risks from dangerous substances. Flammable nanopowders capable of fueling a
dust explosion fall within the definition of a dangerous substance. Regulation 5 of these
regulations requires that, where a dangerous substance is present at a workplace, the employer
should make a suitable and sufficient assessment of the risks to their employees. Those risks
should, where possible, be eliminated. Where it is not reasonably practicable to do this they
should be reduced and controlled (Regulation 6). Measuring the fire explosion characteristics of
nanopowders is an essential step in assessing the risk.
CHIP Regulations
The CHIP (Chemical Hazard Information and Packaging for Supply) Regulations 2009 have
implications for nanopowders. Dangerous substances or preparations cannot be supplied unless
they have been classified in accordance with the regulations. Classification requires a
knowledge of the toxic and fire and explosion properties of the substance. In addition, suppliers
are required to provide information about the hazards of the substances and to package and
label them for safe transport.
REACH Regulations
REACH is a new European Union regulation concerning the Registration, Evaluation,

Authorisation and restriction of CHemicals. It came into force on 1st June 2007 and replaces a
number of European Directives and Regulations with a single system. A major part of REACH
is the requirement for manufacturers or importers of substances to register them with a central
European Chemicals Agency (ECHA). A registration package will be supported by a standard
set of data on that substance. The amount of data required is proportionate to the amount of
substance manufactured or supplied. The data requirement includes the fire and explosion
properties of the substance.
2
3 EXPLOSION SEVERITY
3.1 STANDARD TEST
The procedure for measuring the explosion severity of dust/air mixtures is described in a
European standard available as BS EN 14034-1 (2004) and BS EN 14034-2 (2006). The
standard test vessel for these determinations is the 1 m3 vessel.
The peak maximum explosion pressure, Pmax, and the peak maximum rate of pressure rise,
(dP/dt)max, are measured in this standard test procedure. The Pmax and (dP/dt)max are the highest
values generated in an enclosed dust explosion. These characteristics are measured in a standard
test at the optimum dust concentration and are obtained by testing the dust over a wide range of
dust concentrations. The two peaks normally occur at different dust concentrations. The
(dP/dt)max is used to calculate a dust specific explosibility characteristic called the KSt value.
The KSt is given by:
KSt = (dP/dt)maxV1/3 (2.1)
Where (dP/dt)max is the peak maximum rate of pressure rise (bar/s) and V is the total internal
volume of the test vessel (m3). The units of KSt are bar m/s.
Equation (2.1) is well known as the “cubic law” (Barton (2002)). The KSt is considered to be a
constant for any dust, independent of vessel size and equation (2.1) acts as a simple scaling law.
The KSt value is derived only from measurements in either a 20 litre sphere vessel or a 1 m3
vessel and if any other vessel is used to measure the KSt it must be calibrated against the 1 m3 or
20 litre standard test vessels.
The ISO standard also allows the use of alternative vessels provided they give comparable
results. The criterion for demonstrating conformity is given in the standard BS EN 14034-2
(2006); the standard has been prepared by CEN Technical Committee 305. This gives
alternative test equipment procedures. It states that the maximum rate of pressure rise of dust
clouds can be determined using alternative test equipment and/or test procedures. When using
an alternative it shall be shown that at least for the following dusts that the method yields results
within specific deviations. The dusts shall include at least two metal powders, two natural
organic powders, two synthetic organic powders and two coal dusts. In the standard 1 m3 vessel
the KSt value is equal to the maximum rate of pressure rise.
The standard test apparatus typically requires kilogram quantities of powder and therefore a
smaller scale test apparatus is required for nanopowders. Also, the powder is dispersed in the
standard apparatus using compressed air. With nanopowders, their large surface to volume ratio
means that many are spontaneously flammable on contact with air, or surface oxidation alters
their properties. Hence equipment that avoids oxidation until the point of ignition is required.
No commercial equipment is available to satisfy these requirements, so a specially designed test

apparatus, to measure the explosion characteristics of nanopowders, has been developed in this
project.
3
3.2 TEST VESSEL
An explosion test vessel with a volume of approximately 2 litres has been designed and
manufactured specifically for testing nanopowders (Figure 1 - 3). The construction quality was
specified to be similar to the existing standard 20 litre sphere apparatus at HSL but with
specially treated internal surfaces.
The vessel is constructed from stainless steel and designed for a maximum working pressure of
20 bar. It is fully certified and tested to British Standards and in accordance with the Pressure
Systems Directive PD 5500: 2005.
The shape of the internal test chamber is spherical since this will allow the dust to be dispersed
as a homogeneous dust cloud and thus allow the burning material to expand uniformly towards
the vessel wall. The internal volume of the spherical test chamber is 2 litres with an internal
diameter of 156 mm.
A twin-skinned wall to facilitate water-cooling as used in the 20 litre sphere apparatus was
considered. This was rejected on the grounds that much of the vessel body will be congested
with flanges and ports and the cooling would be ineffective. The vessel is therefore constructed
with a single skin wall. The whole vessel assembly is capable of electrical isolation.
Ports
The test vessel incorporates a number of ports and flanged connections:
1. The main access into the vessel is via the top of the vessel. This access is sealed using a
flanged plate that incorporates two holes to accept the ignition electrodes. The opening
is sized to enable hand-access for cleaning and for installing the dispersion nozzle.
2. Two diametrically opposed flanges tapped with M14 threads for Kistler pressure
transducers.
3. A port incorporating a glass viewing window with a diameter of 40 mm.
4. A dust injection port is located at the bottom of the vessel.
5. A pressure release valve and an evacuation port.
Dust injection
The dust is introduced into the vessel from an external dust container via a dust injection valve
and solenoid system mounted on an intermediate flanged adapter at the base of the vessel and a
dispersion nozzle.
Dispersion nozzle
A pepper-pot nozzle incorporating a rebound plate was selected for the test programme. The
pepper-pot design with a circular deflector is designed to encourage a homogeneous dust
distribution and is located into the base of the test chamber.
Dust container
A special dust container, designed for a working pressure of 20 bar, is attached to the dust
injector. The container has a volume of approximately 0.06 litre. A ¼ inch BSP port is used for
4
the attachment of a pressure hose and pressure gauge. The container was designed to allow
remote loading of nanopowder under an inert atmosphere within a glovebox.
Ignition
Ignition of the dust cloud is achieved using an ignition source attached to two central electrodes.
There are a number of possible ignition sources including: chemical igniters, electric fuse heads,
spark and hot wire. A voltage is applied to the ignition source from the control system at the
preset time delay following dust injection into the vessel.
Figure 1: Vessel general arrangement
5
Figure 2: Test vessel
Figure 3: Test vessel
The test vessel was located within a fume cupboard, that incorporated a HEPA filter system and
a sink fitted with waste filters for cleaning the contaminated components. Located outside the
fume cupboard and connected via a pass-through port were: vacuum pump, high pressure air
supply, control system and data acquisition system and a glovebox fitted with an integral HEPA
6
filter. To minimise operator contact with nanopowders, the materials were handled, weighed
and loaded into the enclosed dust injection chamber within the glovebox (Figure 4). The
glovebox also allowed the handling of nanopowders under an argon atmosphere, enabling
powders to be transferred from the glovebox to the test apparatus under argon thus preventing
premature oxidation of powders.
Figure 4: Glovebox
3.3 CONTROL AND INSTRUMENTATION
The control and instrumentation system is interfaced to a PC via custom control circuitry and
commercial USB data acquisition and control hardware. The PC executes software providing
control, data logging and analysis functions. The fume cupboard is fitted with safety interlocks
to help ensure operator safety. The software has been developed using the National Instruments
LabVIEW development environment. It is installed for use on the Microsoft Windows XP
platform.
In brief the test operates as follows. The sphere is sealed and evacuated. The powder under
investigation is released into the sphere to form a cloud, driven by the in-rush of air from
external pressure. The cloud is ignited, and the pressure rise in the sphere logged. This data is
then analysed to determine the maximum rate of pressure rise and the peak pressure.
The software controls the nanopowder release and ignition, logs the pressure data from the
enclosure, and will analyse the data. Safety interlock status is monitored to prevent risk to
personnel from unplanned attempts at powder release or ignition.
The PC has two free USB 2.0 ports to allow interfacing to the USB data acquisition devices.
The software provides appropriate interface signals via the interface hardware to control the
apparatus.
The software accepts appropriate interface signals to provide feedback for the control process
and to log the required test data from two independent piezoelectric pressure sensors. The
pressure transducers measure the pressure difference, not absolute pressure.
7
The software calculates dP/dt (the rate of pressure rise) and Pmax (maximum explosion over-
pressure) following each test. These are calculated for pressure sensor 1, for pressure sensor 2,
and for the averaged readings of the two sensors. Kistler (type 701A) pressure transducers are
used in the test apparatus and the details are shown in Table 1.
Table 1: Pressure transducers

Laboratory Channel Serial number Transducer Charge
instrument location amplifier
reference sensitivity
PT140 1 128026 L.H. port 152.1
2 128027 R.H. port 154.1
3.4 TEST PROCEDURE
The following is a summary of the general operational procedure.
Sample preparation
1. Remove container with nanopowder from the storage cupboard. Place in a secondary
transport container to protect the nanopowder container from accidental breakage. Transport to
the test facility.
2. Remove the nanopowder container from the transport container a place in the glove box. The
balance will be located in the glovebox.
3. Weigh the required mass of nanopowder by carefully scooping the powder and placing on a
stainless steel dish on the balance. Close the nanopowder container.
4. Pour the nanopowder into the dust injection vessel and seal the vessel. Transfer the vessel to
the fume cupboard, fit to the dust injection vessel and connect the supply lines.
5. Non-oxidised nanopowder is handled under an inert atmosphere.
Firing
1. Prepare the ignition source.
2. If a chemical igniter is used, connect to the electrodes inside the vessel and check the circuit
continuity using an approved ohmmeter tester.
3. If spark electrodes are used, ensure that the gap is set and the correct discharge energy is
selected.
4. Replace the vessel lid and ensure a full complement of screws are fitted and are tightened.
5. Close the vessel outlet valve, open the evacuation valve and open the dust injection isolation
valve.
8
6. Partially evacuate the explosion vessel to the appropriate pressure, typically 0.4 bara. The
gas mixture and pressure in the explosion vessel shall be adjusted so that at ignition the
atmosphere is nominally atmospheric.
7. Pressurise the dust injection vessel with compressed air. The pressure and volume of the dust
injection vessel is matched to the level and vacuum in the explosion vessel such that the
pressure at ignition is at nominal atmospheric pressure.
8. Close the fume cupboard sash and initiate the test.
9. Display the pressure traces and save the data to hard disk for analysis.
10. Open the outlet valve and release the pressure.
9
4 MINIMUM IGNITION ENERGY
4.1 TEST APPARATUS
The MIKE3 test apparatus, manufactured by Kuhner is normally used for the determination of
minimum ignition energy (MIE) of micron-sized dusts (Figure 5). However, the risk
assessment for handling nanopowders required improvements to the operating procedure for this
apparatus. This equipment has therefore been modified to enable its use with nanopowders.
Figure 5: MIE test apparatus
Dust dispersion chamber
The dust dispersion chamber was modified to minimised operator contact with the powder and
enable the safe handling, weighing of nanopowder and loading of the powder into the apparatus.
This was achieved using a glovebox to weigh and transfer nanopowder to the MIKE3 apparatus.
A two stage process was used to develop the modified dispersion chamber.
Development Stage 1: An enclosed barrel shaped chamber was constructed to completely

replace the Kuhner dispersion cup assembly (Figure 6).
10
Figure 6: Dispersion chamber
This comprised a stainless steel chamber split into three components, lower middle and top
barrels that were screwed together to form the assembled chamber. Two aluminium bursting
discs were sandwiched between the barrels to form a central chamber that held the nanopowder.
The assembly method was as follows:
Powder, dispersion chamber and balance were placed in a glovebox. The lower and middle
barrels were assembled with the lower bursting disc inserted between the barrels. Powder was
then deposited into the middle chamber, the second bursting disc is fitted and the top barrel is
screwed into position. The assembly was removed from the glovebox and transferred and
fitted to the MIKE3 apparatus. The compressed air line was connected to the lower chamber.
This provided the blast of compressed air that, during the test, ruptured the bursting discs and
dispersed the nanopowder into the glass tube. Following commissioning tests on the first design
it became clear that the efficacy of the system was not acceptable in that it was not possible to
obtain the established MIE of a standard test powder. This was likely to be the result of erratic
rupturing of the bursting discs and the increased distance of the initial formation of the dust
cloud to the ignition source.
Development Stage 2: A second dispersion system was therefore developed utilising the
existing Kuhner dispersion cup. A thin slide valve was manufactured and fitted to the top face
of the dispersion cup (Figure 7). The close fitting of the slider and fixed slides provided sealed
chamber once the valve was closed. This design had the benefit of utilizing the original
dispersion cup, nozzle and geometric locations of the dispersion and ignition points.
11
Figure 7: Slide valve dispersion chamber
Test procedure: In common with the first design, the powder, dispersion chamber assembly and
balance were placed in the glovebox. The air isolation valve was closed and the slide valve
opened. Nanopowder was then weighed, placed in the dispersion cup and the powder was
safely sealed within the cup by closing the slide valve. The assembly was removed from the
glovebox, transferred and fitted to the MIKE3 apparatus and the compressed air line was
connected to the lower chamber. The test procedure normally used in the operation of the
Kuhner test apparatus was then followed in accordance with the general principles of BS EN
13821 (2002). Ignition energy levels used were 1 mJ, 3 mJ, 10 mJ, 30 mJ, 100 mJ, 300 mJ and
1000 mJ.
12
5 FIRE HAZARDS
5.1 FIRE HAZARDS IN NANOPOWDERS
With highly mobile, easily dispersed materials it is difficult to draw boundaries between ‘fire’
hazards, ‘deflagration’ hazards and ‘explosion’ hazards.
‘Fire’ is normally understood to be the ignition and combustion of material in bulk - as layers
or within packages. Limiting consideration to these circumstances, there are three major fire
issues that might arise with nanopowders:
• New ignition routes due to the presence of bulk nanopowders – e.g. the possibility of
spontaneous combustion.
• The possibility of enhanced burning rates in bulk nanopowders compared with solids or
larger particulates and powders.
• The potential for fire-induced air flows to lift bulk materials into a dispersed cloud while
providing an ignition source for a nanopowder explosion.
The particular hazards in any fire situation are also threefold:
• heat, toxic gases and smoke that might lead to injury or death,
• heat and smoke that might cause property damage
• growth from a small, initial fire to a larger, more hazardous one.
At present, nanopowders other than carbon black are high value materials produced in relatively
small quantities. Even ‘large’ amounts in storage are limited to kilogram quantities or less. In
terms of fire load, the amount of combustible material present, the packaging of many
nanopowders (plastics, cardboard, shrinkwrap etc) and the storage infrastructure (pallets,
equipment, cables, even paints) are likely to represent the main hazard to life and to property.
If nanomaterial combustion is relatively slow, such small quantities will not release energy
quickly enough to be an unusual fire threat.
On the other hand, nanopowders may exhibit a much more rapid combustion than equivalent
materials at larger particle sizes. This means that they might release energy at a much faster rate,
tending asymptotically to the explosion scenario. It seems unlikely that the ‘fire’ end of this
range would cause significant harm with small amounts of nanopowders.
5.2 PRACTICAL ASSESSMENT OF FIRE PROPERTIES
Current nanopowders are typically highly mobile, requiring continuous containment to prevent
losses even through diffusion into the air. This high mobility means that nanomaterials within
process equipment would not be able to build up on surfaces without undergoing agglomeration
or other physico-chemical changes, which would – in turn - affect their fire properties.
Unchanged nanopowders are unlikely to “settle out” even in still conditions, and much less so in
13
any moving gas or air flows. Should nanomaterials escape from a process into a more open
building (factory/storage) environment, the normal air movement present in large structures will
help to reinforce their natural mobility and it is even less likely that any significant amount of
nanomaterials could form a combustible mass or layer.
The current high intrinsic value of almost all nanopowders means that there are also financial
drivers on producers and handlers to minimise any losses – again reducing the likelihood of a
significant amount of material being allowed to build up in process or storage.
To pose a fire risk, nanopowders need to be present in layers containing many grams, even
kilograms, of material. For the present, at least, such quantities will only be present for materials
in storage. Any fire assessment therefore needs to address nanomaterials in storage conditions.
The important fire properties (rate of heat production, rate of fire spread and formation of toxic
combustion products) all include processes that are scale dependent. For example, the effect of
thermal radiation is important in the spread of fire but is dependent on the square of the
distance. This means that small-scale testing does not reproduce the effects of a full-scale fire.
Consequently, fire tests need to be carried out at a representative scale.
Nanopowders that are likely to be important in the future are currently only available in limited
quantities at relatively high cost. The health and environmental risks associated with
nanomaterials have not been completely defined; despite considerable ongoing effort in the area
of nanotoxicology, many unknowns remain and it is recommended that a precautionary
approach is taken when handling nanomaterials in research and manufacturing workplaces
(Seaton and Tran, 2009).
The combination of high cost and potential risk means that only the minimum necessary amount
of work should be undertaken with nanopowders. Given the current low risk of harm from
nanomaterial fires simply as a result of the low quantities present in any given location, it is
difficult to justify fire testing to further quantify that risk.
14
6 ELECTROSTATIC PROPERTIES
6.1 ELECTROSTATIC CHARGE
The likelihood of accumulation of electrostatic charges and voltages in an industrial process

increases with increasing electrical resistivity of powder (Eckhoff, 1997). With regard to dusts
that penetrate into electrical and electronic equipment, the chance that they give rise to short
circuits and equipment failure, increases as the dust resistivity decreases. In both situations the
generation of ignition sources may result.
BS EN 61241-2-2:1996 states that a dust with electrical resistivity equal to or less than 103 Ωm
is considered to be a conductive dust and a dust with electrical resistivity greater than 103 Ωm is
considered to be a non-conductive dust.
The CCPS (2005) publication reports that resistivities above 108 Ωm are indicative of the
potential for significant electrostatic charging. Increased humidity in the atmosphere where
powders are handled may help to reduce the potential for electrostatic discharges by increasing
the conductivity of the powders. PD CLC/TR 50404 (2003) divides powders into 3 groups
depending on their volume resistivity: a) low resistivity powders with volume resistivities up to
about 106 Ωm, b) medium resistivity powders in the range 106 Ωm to 1010 Ωm and c) high
resistivity powders with resistivities of 1010 Ωm and above. NFPA77 (2007) has a similar 3-
category classification for powder resistivity. They indicate that low resistivity powders with
resisitivities of up to 108 Ωm can become charged during flow. Powder resistivities within the
range 108 to 1010 Ωm are considered to be medium resistivity powders. High resistivity powders
have resistivities greater than 1010 Ω m.
The significance of resistivity is discussed in NFPA 77, the main points being:
a) Low resistivity powder can become charged during flow but rapidly dissipates when
conveyed into a conductive grounded vessel. Powder conveyed into non-conductive
container can result in an incendive spark.
b) For medium resistivity powders coming to rest in bulk, the retained charge depends on
the resistance between the powder and ground. Medium resistivity powders do not
produce bulking brush discharges or sparks.
c) High resistivity powders do not produce spark discharges themselves. However, they
can produce corona, brush, bulking brush, and propagating brush discharges.
Electrostatic charges are generated on solids during transport due to the rubbing of particles
against particles or particles against equipment and the internal surfaces of pipes (CCPS, 2005).
The charging of particles may take place independently of whether they are conductive or
insulating materials. Equipment may also become charged by virtue of coming into contact
with charged particles. However, an electrostatic charge by itself does not necessarily represent
an ignition hazard – this only exits when the charge is sufficiently high that a discharge occurs
due to the breakdown of a high voltage electric field.
The criteria described above relates to micron-scale powders. These parameters have been used
to assess the potential for charge build-up in a range of nanopwders, and the chance of
nanopowders becoming potential ignition sources. Two test rigs have been constructed to
assess charging and resistivity.
15
6.2 CHARGING TEST
In order to measure and demonstrate electrostatic charging of nanopowders in a realistic

situation, a test apparatus has been constructed capable of generating particle movement and
measuring the charge build-up. This incorporates a 25 mm diameter closed-loop continuous
circulating PVC pipe incorporating a fan system to pneumatically circulate the powder. The
overall length of the loop is approximately 1000 mm. A charge measuring station is
incorporated into the pipe. Nanopowder was introduced through a valved port to be circulated
within the loop. The aim is for any charge developed to be measured at an intermediate point in
the loop. The system was developed, initially using four in-line fans, but later replaced with a
centrifugal fan driven by a 55W single-phase motor (Figure 8).
Figure 8: charge test apparatus
A number of techniques were employed to measure the deposition of charge at the measurement
station during the commissioning of the test equipment. The technique that gave the most
reliable results was to connect a capacitor between the measurement station and Earth; the
voltage developed across the measurement capacitor is related to the charge accumulated on it.
A number of measurement capacitor values were used during the tests, in each case the
capacitance was accurately measured using a Wayne Kerr model 6425 LCR multi bridge.
The voltage across the measurement capacitor was measured using a Keithley model 6514
electrometer. The data was captured using a PC running National Instruments’ Labview
SignalExpress software and a National Instruments model USB-6221 data logger. The data
logger captured data from the analogue output of the electrometer at a rate of 500Hz. Due to the
high impedance of the electrometer it was necessary to provide an Earthed screen of wire mesh
around the charging loop to control electrical noise.
The relationship between the voltage across the capacitor and the supplied current is given by:
dV
I =C
dt
Where I is the charging current (amps), dV/dt is the rate of change of voltage on the capacitor
and C is the capacitance of the measurement capacitor (farad).
16
The rate of change of voltage in the powder charging cycle can be determined from the graph of
voltage against time. Hence the streaming current can be calculated.
Each powder was tested using a new, clean set of pipes. Before the test equipment was loaded
with powder the resistance between the measurement station and earthed casing of the fan was
measured using a Robin model 3075DL insulation and continuity meter. In each case the
resistance was found to be greater than 2000 MΩ at 500V.
The fan speed was set at the start of the tests and was not changed. The velocity of the air within
the pipe was 28 m/s (measured using a TSI Airflow instrument) and is assumed to have been
constant throughout the testing.
The ambient environmental conditions were measured using a Rotronics Hygropalm 0 portable
temperature and humidity meter.
6.3 RESISTIVITY TEST
Equipment has been designed and manufactured specifically to measure the bulk resisitivity of
nanopowders in layers. The design is based on the equipment described in BS EN 61241-2-
2:1996, but has been designed on a smaller scale to allow testing of small quantities of
nanopowder.
The equipment consists of a PTFE test cell (Figure 9) with two stainless steel bars. The bars
have nominal cross sectional dimensions of 25 mm x 25 mm and are 25 mm long. These
formed the electrodes and are located with a 10 mm long gap in a PTFE block. The block is 75
mm long x 90 mm wide with a channel in which the electrodes were located. Each electrode
has a terminal connection for a calibrated resistance meter.
Operator contact with nanopowders during these tests was minimised by locating the test cell in
sealed test chamber within a glove box. Where necessary the atmosphere within the glovebox
was inerted using argon. The humidity of the test chamber was adjusted using a desiccant and
was monitored using a humidity meter.
A weighed quantity of nanopowder was deposited within the cell and excess powder was
removed by running a straight edge along the top of stainless steel electrodes. The bulk density
of the sample was calculated by weighing the deposited sample. The electrodes were connected
to the resistance meter and the resistance was measured over a range of humidity levels.
The resistivity was calculated in accordance with the principles described in BS EN 61241-2-
2:1996 and is described as follows.
Where Ro is greater than 10 Rs, and the resistivity of the dust is calculated from the equation:
ρ= 0.001 Rs [H × W/L]
where
ρ is the resistivity in Ω m
Ro is the resistance of empty test cell in Ω
Rs is the resistance of the filled test cell in Ω;
H is the height of the electrode in mm;
W is the length of the electrode in mm;
L is the space between electrodes in mm.
17
A range of nanopowders was assessed and, for comparison purposes, a number of micron-scale
powders were also tested.
Figure 9: Resistivity test cell
18
7 TEST MATERIALS
Nanopowder test materials were obtained including metal powders and organic materials and
are listed in Table 2. Micron-scale powders used in the tests are listed in Table 3.
Initial tests were done using Lycopodium, a well-established micron-sized standard test powder.
This was supplied by Fluka Reference 19108 (Explosion Safety Unit Sample No EC/026/08.
Intrinsiq Materials Ltd (previously Qinetiq Nanomaterials Ltd) supplied a quantity of

aluminium nanopowder from their early production runs in which some carbon contamination
had occurred. The material was useful for early commissioning of the test apparatus.
Intrinsiq Materials Ltd also supplied commercial grade aluminium nanopowder. This material
is passivated during production by exposing the material to an oxygen/argon atmosphere. This
produces a partially oxidised material such that it can be exposed to the atmosphere during
experiments without the risk of rapid oxidation and burning.
19
Table 2: Nanopowders
Material HSL Supplier Supplier’s Information from supplier
Sample Reference
Number
Iron (passivated) EC/147/07 Nanostructured & Amorphous 0266JY Iron powder (Fe) (partially passivated)
Materials, Inc. 16840 Clay Road, Suite Purity: 99.5% (metal basis),
113, Houston, TX 77084, USA APS: 25 nm
SSA: 40 - 60 m2/g
Color: black
Morphology: spherical
Bulk density: 0.10 - 0.25 g/cm3

True density: 7.87 g/cm3
Copper (passivated) EC/148/07 Nanostructured & Amorphous 0296JY Copper (Cu) (partially passivated)
Materials, Inc. 16840 Clay Road, Suite Purity: 99.8% (metal basis)
113, Houston, TX 77084, USA APS: 25 nm
SSA: 30 - 50 m2/g
Color: black brown
Morphology: spherical
Bulk density: 0.15 - 0.35 g/cm3

Silicon Carbide EC/159/07 Nanostructured & Amorphous 4620KE Purity 97.5%; APS 45-55 nm; colour Grey white: Bulk
Materials, Inc. 16840 Clay Road, Suite Density 0.068 g/cm3; True density 3.22g/cm3;
113, Houston, TX 77084, USA Morphology spherical; Specific surface area 70-90 m2/g
Zinc Oxide EC/149/07 Nanostructured & Amorphous 5810MR Zinc Oxide (ZnO) Purity 99.5%; APS: 20nm; SSA: 50
Materials, Inc. 16840 Clay Road, Suite m2/g; Colour: milky white; Morphology: nearly spherical;
113, Houston, TX 77084, USA Bulk density: 0.3-0.45 g/cm3; True density: 5.606 g/cm3
Iron Oxide EC/150/07 Nanostructured & Amorphous 2652FY Iron Oxide (Fe3O4); Purity 99.5%; APS 15-20 nm; Colour:
Materials, Inc. 16840 Clay Road, Suite black; Morphology : spherical; Bulk density : 0.8-0.9
113, Houston, TX 77084, USA g/cm3; True density 4.8-5.1 g/cm3
Zirconium Oxide EC/151/07 Nanostructured & Amorphous 5930LQ Zirconium Oxide (ZrO2); Purity 99.95%; APS: 29-68 nm;
Materials, Inc. 16840 Clay Road, Suite SSA: 15-35 m2/g; Colour: white; Morphology: spherical;
113, Houston, TX 77084, USA Bulk density: 0.74 g/cm3; True density 5.89 g/cm3
Zinc EC/152/07 Nanostructured & Amorphous 0303WF Zinc (Zn); Purity 99.5%; APS : 130nm; SSA: 6.4 m2/g;
Materials, Inc. 16840 Clay Road, Suite Colour : Grey; Morphology: spherical; Bulk density: 0.7-
113, Houston, TX 77084, USA 0.85 g/cm3; True density 7.14 g/cm3
Multi-walled carbon EC/153/07 Nanostructured & Amorphous 1229YJ MWNT Purity: > 95% ; Outside diameter: 20-30 nm;
nanotubes Materials, Inc. 16840 Clay Road, Suite Inside diameter: 5-10 nm; Length: 10-30 um; SSA: ~ 110
113, Houston, TX 77084, USA m2/g; Colour: black; True density: ~2.1 g/cm3
Carbon nanofibres EC/158/07 Nanostructured & Amorphous 1190JN Carbon nanofibers; Purity 95%; Outside diameter: 80-200
Materials, Inc. 16840 Clay Road, Suite nm; Core diameter: 1-10 nm; Length 0.5-20 um;BET: 25-
113, Houston, TX 77084, USA 35 m2/g; Electronic Resistivity: 0.75-0.1 Omu.cm (20-
180MPa); Colour: black; Bulk density: 0.06-0.08 g/cm3;
20
Table 2 (continued): Nanopowders
Carbon Nanofibre EC/042/08 Pyrograph Products Inc PR-19-SLD Diameter 100-200 nm
Length 30-100 micron
Carbon Nanofibre EC/116/08 Applied Sciences Inc. PR-25-XT- Diameter 70-200 nm
PS-AM Length 2-5 micron
Carbon Nanofibre EC/117/08 Applied Sciences Inc PR-24-XT- Diameter 70-200 nm
HHT-AM Length 2-10 micron
Aluminium EC/104/08 Intrinsiq Materials Ltd QNA 2607 Nominally 100nm
Minimum 60% Al, remainder Aluminium oxide
surface layer.
Trace elements Fe, Na, Zn, Ga.
Aluminium EC/011/09 Intrinsiq Materials Ltd QNA 2798 Nominally 210 nm
Minimum 60% Al, remainder Aluminium oxide
surface layer.
Trace elements Fe, Na, Zn, Ga.
Aluminium EC/060/07 Intrinsiq Materials Ltd Ref: Estimated 73-109 nm
PQN301
Batch 17
Carbon Black EC/076/07 Cabot Monarch 800 17 nm immediately after formation in the combustion
process.
Table 3: Micron-scale powders

Material HSL Supplier Supplier’s Information from supplier
Sample Reference
Number
Aluminium EC/085/08 Alpoco Batch 8- Specification 99.7% Al, 8 micron.
137X
Iron EC/003/09 Ronald Britton Ltd Code 100% <150 micron
B10121
Batch
831302
Zinc oxide EC/112/08 Fluka 96479
Copper EC/004/09 Ronald Briton Ltd 100% < 75 micron
Zirconium oxide EC/111/08 Acros Organics 190522500
Carbon EC/113/08 Sigma Aldrich 484164-50G 2-12 micron
Lycopodium EC/026/08 Fluka Sieved to 63 micron
Zinc Stearate EC/118/08
Coal EC/120/08
Toner EC/122/08
21
8 RESULTS
8.1 COMMISSIONING TESTS – EXPLOSION SEVERITY
Initial commissioning was done using micron-scale aluminium nanopowder and lycopodium
powder shown in Table 4. The injection pressure was maintained at 20 barg. The variables were
dispersion, dust concentration, ignition source, the valve timing, ignition delay, and vacuum
pressure. The maximum rate of pressure rise and maximum explosion pressure were measured
in the 2 litre sphere for each test powder over a range of conditions. The objective was to
achieve a set of conditions that would produce a rate of pressure rise and maximum explosion
pressure that could be related to the results obtained from the 20 litre sphere. Appendix 1 shows
the range of test conditions used to explore this objective.
Table 4: Commissioning powders
Powder Sample Number Supplier Comment
Lycopodium EC/026/08 Fluka Sieved to 63 micron
Aluminium EC/060/07 Intrinsiq Ref: PQN301 Batch 17

nanopowder
Estimated 73-109 nm
8.1.1 Ignition delay
The ignition delay is the difference between the time the control system initiates the opening of
the dust injection valve and the time of the initiation of the ignition source. The ignition delay
was varied within the range 60 – 110 ms. For the maximum rate of pressure rise, the optimum
ignition delay was found to be 75 ms.
8.1.2 Ignition source
Chemical igniters manufactured by Sobbe GmbH were used in the test programme.
Additionally, a low energy ignition source - an electric fuse head - was assessed.
The strength of the ignition source will influence the rate of propagation of flame through the
dust cloud and will therefore produce different explosion characteristics of the dust in an
explosion severity test. For example a very strong ignition source may result in a much greater
rate of pressure rise than that produced from a relatively weak ignition source. This is
demonstrated by tests done using lycopodium powder in the 20 litre sphere (Table 5) where the
rate of pressure rise was higher with correspondingly greater strength of ignition. The Pmax was
relatively flat. The 10 kJ igniter is used in the standard 20 litre sphere test (BS EN 14034-1:
2004 and BS EN 14034-2: 2006) and clearly produces the highest rate of pressure rise. Without
the presence of dust, the rate of pressure rise is approximately 100 bar/s with a peak pressure of
approximately 1 bar.
22
Table 5: Ignition strength in the 20 litre sphere (lycopodium powder)
Ignition source Pmax (barg) dP/dt (bar/s) K (bar m/s)
42 J electric fuse head 6.5 243 66
1 kJ Sobbe 7.3 551 150
5 kJ Sobbe 7.0 620 168
10 kJ Sobbe 7.3 673 183
A range of tests were performed in the 2 litre sphere with lycopodium (sample EC/026/08) and
aluminium (sample EC/060/07) to assess the effect of the ignition strength on the rate of
pressure rise and maximum explosion pressure (Table 6). As expected, the rate of pressure rise
increased with increasing ignition strength. Table 4 shows that with an electric fuse head, the
highest rate of pressure rise achieved with lycopodium was only 200 bar/s, and in some cases
ignition was not achieved. At the other extreme, with a 5 kJ igniter both dusts tended to result in
high rates of pressure rise and in the case of lycopodium the results were close to the 20 litre
sphere but the aluminium far exceeded the 20 litre sphere results. The Pmax of the lycopodium
was generally much greater than the 20 litre sphere tests whilst the aluminium was comparable.
The 5 kJ igniter therefore had a tendency to over-drive the explosion such that an ignition
source smaller than this value would be required for the nanopowder tests.
Table 6: Effect of ignition strength on test dusts – 2 litre vessel
Ignition source Pmax (barg) dP/dt (bar/s)
Lycopodium ( sample EC/026/08)
Electric fuse head 8.2 200
1 kJ Sobbe 8.7 882
5 kJ Sobbe 9.1 1500
Aluminium (sample EC/060/07)
1 kJ Sobbe 10.8 1450
2 kJ Sobbe 9.8 1950
5 kJ Sobbe 9.5 5000
Additionally, to assess the pressure development of the igniter itself, a number of igniters were
tested in the 2 litre vessel, without the presence of a dust cloud. Table 7 shows the results of the
tests with each of these igniters. The use of a 10 kJ Sobbe igniter in the 2 litre vessel would
result in an over-driven ignition of the dust and was therefore not considered. The highest
strength considered was a 5 kJ Sobbe igniter.
23
The 1 kJ Sobbe igniter selected for the nanopowder test programme since this would not over-
drive the explosion. The igniter alone produced a pressure rise of 1.05 barg and a maximum
rate of pressure rise of 75 bar/s. During nanopowder explosion tests, it is assumed that the effect
of the 1 kJ igniter can be ignored after 1.05 barg, thus the rate of pressure rise of the
nanopowder explosion is evaluated at pressures above this value.
Table 7: Ignition strength in the 2 litre vessel (without dust).
Ignition source Pmax (barg) dP/dt (bar/s)
Electric fuse head 0.2 3.3
1 kJ Sobbe 1.05 75
2 x 1 kJ Sobbe 1.5 154
5 kJ Sobbe 4.2 250
8.1.3 Dp/dt, KSt value and Pmax
From the commissioning tests, a set of test conditions were established for main series of
nanopowder tests. These are shown below are shown below:
Dust injection pressure: 20 bar g
Initial test chamber pressure: 0.1 - 0.4 bar a
Dust injection valve open: 10 ms
Dust injection valve closed: 65 ms
Ignition ON: 83 ms
Ignition OFF: 180 ms
Igniter: 1 kJ Sobbe chemical igniter
Logging rate 50 kHz
The rates of pressure rise used to assess the explosion severity were lower than were needed for
direct calculation using the “cubic law”equation (2.1). This is mainly the result of flame
impingement and quenching at the vessel wall. Consequently, it is not possible to calculate the
KSt value directly using the 2 litre sphere. Table 8 and Figure 10 show the comparison of data
from the 20 litre sphere and 2 litre sphere for a range of dusts using these test conditions. A
scaling factor was established to relate the rate of pressure rise measured in the 2 litre sphere
with the KSt value measured in the 20 litre sphere. To obtain the equivalent KSt value from the
rate of pressure rise in the 2 litre sphere a scaling factor of 0.268 is applied.
The expectation was for the maximum explosion pressure in the 2 litre sphere to be generally
less than in the 20 litre sphere due to the greater cooling effects at the walls of the smaller test
24
chamber compared with the 20 litre chamber. However, Figure 10 shows that the maximum
explosion pressures in the two vessels were broadly comparable.
Table 8: Commissioning results
Material 20 litre sphere results 2 litre sphere results
Rate of KSt (bar Pmax (bar) Rate of Pmax (bar)

pressure m/s) pressure
rise (bar/s) rise (bar/s)
Lycopodium 673 183 7.3 700 7.7

EC/026/08
Aluminium 1200 326 9.8 1450 10.8

EC/060/07
Zinc 1080 293 7.6 1400 8.2

stearate
EC/118/08
Coal 558 151 7.4 600 7.3

EC/120/08
Toner 714 194 7.5 725 6.3

EC/122/08
Carbon 382 104 7.8 320 6.2

black
EC/076/07
Aluminium 2368 643 12.0 2000 11.2

EC/104/08
700 13
Kmax 20 litre sphere vessel (bar m/s)
600 12
20 litre vessel - Pmax (bar)
500 11
400 10
300 9
200 8
100 7
0 6
500 1000 1500 2000 5 6 7 8 9 10 11 12
Rate of pressure rise in 2 litre vessel (bar/s) 2 litre test vessel - Pmax (bar)
Figure 10: Comparison of 20 litre and 2 litre test apparatus
25
8.1.4 Pressure-time history
A typical pressure-time trace, taken from a 2 litre sphere test, is shown in Figures 11 and 12;
the test material is aluminium nanopowder EC/104/08. Two pressure curves (red and blue) are
logged from the two pressure transducers and an average value is displayed (green). The initial
pressure rise is the injection of the dust into the test chamber which raises the pressure to
atmospheric after which the ignition is initiated. This is followed by a rapid pressure rise due to
the ignition of the dust.
The pressures traces illustrate that the peak values of explosion pressure and rate of pressure rise
can occur at different dust concentrations. In Figure 11, with a dust concentration of 3000 g/m3,
the pressure rises to 11.2 bar g with a maximum rate of pressure rise of approximately 1625
bar/s. In Figure 12, with a dust concentration of 2500 g/m3, the maximum explosion pressure is
9.1 bar g with a maximum rate of pressure rise of 2000 bar/s.
Figure 11: Pressure – time trace Figure 12: Pressure-time trace
26
The tests on lycopodium (sample EC/026/08) are a good example of the scaling tests in the 2
litre and 20 litre spheres. Tests on lycopodium using the 20 litre sphere apparatus resulted in a
rate of pressure rise of 673 bar/s, KSt of 183 bar m/s and a Pmax of 7.3 bar (Figure 13).
Similar tests in the 2 litre sphere (Figure 14) resulted in a pressure rise of 700 bar/s and a Pmax
of 7.7 bar. Application of the scaling factor to the rate of pressure rise results in a KSt value of
187 bar m/s.
Figure 13: Lycopodium - 20 litre sphere test
Sample number EC/026/08 Sample Number EC/026/08

Lycopodium powder Lycopodium powder
Fluka 19108 Fluka 19108
2 litre sphere test

2 litre sphere test
8.0 800
700
7.5
Rate of pressure rise (bar/s)
Explosion pressure (barg)
600
7.0
500
6.5
400
6.0
300
5.5
200
5.0 100
200 400 600 800 1000 1200 1400 200 400 600 800 1000 1200 1400
Dust concentration (g/m3) Dust concentration (g/m3)
Figure 10: Lycopodium - 2 litre sphere test
27
8.2 EXPLOSION SEVERITY OF NANOPOWDERS
A range of nanopowders have been tested in the 2 litre sphere. The results are presented in
Table 9 and plots of the individual test are shown in Figures 15 - 24.
Table 9: Nanopowder test results in the 2 litre sphere results
Sample Material Pmax (bar g) dP/dt (bar/s) Equivalent

number KSt (bar m/s)
EC/011/09 Aluminium 12.5 1677 449

nanopowder
(210 nm)
EC/104/07 Aluminium 11.2 2000 536

nanopowder
(100 nm)
EC/147/07 Iron 2.9 68 18

nanopowder
EC/152/07 Zinc 5.6 377 101

nanopowder
EC/148/07 Copper 1.2 10 3

nanopowder
EC/042/08 Carbon 5.2 62.5 17

nanofibre
EC/158/07 Carbon 6.0 112 30

nanofibre
EC/116/08 Carbon 6.9 591 158

nanofibre
EC/117/08 Carbon 5.6 137 37

nanofibre
EC/153/07 Multi-walled 6.4 339 91

carbon
nanotubes
28
Sample number EC/011/09 Sample number EC/011/09
Aluminium nanopowder Aluminium nanopowder
Average particle size 210 nm Average particle size 210 nm
Supplier: Intrinsiq Supplier: Intrinsiq
13 1800
1600
12

Explosion pressure (bar g)
1400
11
1200
10 1000
800
9
600
8
400
7 200
0 1000 2000 3000 4000 5000 6000 7000 0 1000 2000 3000 4000 5000 6000 7000
Figure 15: Aluminium nanopowder sample EC/011/09

Aluminium nanopowder Aluminium nanopowder
Average particle size 100 nm Average particle size 100 nm
Supplier: Intrinsiq Supplier: Intrinsiq
Batch: QNA 2607 Batch: QNA 2607
12
2200
11
2000
10 1800
1600
9
1400
8
1200
7
1000
6
800
500 1000 1500 2000 2500 3000 3500 4000
500 1000 1500 2000 2500 3000 3500 4000
Figure 12: Aluminium nanopowder sample EC/104/08
29
Sample number EC/147/07 Sample number EC147/07
Iron nanopowder Iron nanopowder
25 nm Average particle size 25 nm
Supplier: Nanostructured & Amorphous Matl Supplier: Nanostructured & Amorphous Matls.
3.0
100
2.5
80

2.0
60
1.5
40
1.0
20
0.5
0 500 1000 1500 2000 2500 3000
0
Dust concentration (g/m3) 0 500 1000 1500 2000 2500 3000
Dust concentration (g/m3)
Figure 13: Iron nanopowder sample EC/147/07

Zinc nanopowder Zinc nanopowder
APS 130 nm APS 130 nm
6 400
350
5
300
250
4
200
3 150
100
2
50
0
1 0 1000 2000 3000 4000 5000 6000 7000 8000
0 1000 2000 3000 4000 5000 6000 7000 8000
Concentration g/m3 vs dp/dt bar/s
Concentration g/m3 vs Pmax bar
Figure 14: Zinc nanopowder sample EC/152/07
30
Sample number EC/148/07 Sample number EC148/07
Copper nanopowder Copper nanopowder
APS 25 nm APS 25 nm
1.25
100
1.20
80

1.15
60
1.10
40
1.05
1.00 20
0.95
0
0 500 1000 1500 2000 2500 3000 3500
0 500 1000 1500 2000 2500 3000
Figure 15: Copper nanopowder sample EC/148/07

Carbon Nanofibre Carbon nanofibre
Supplier: Pyrograph Products Inc. Pupplier: Pyrograph Products Inc
PR-19-LD-PS PR-19-LD-PS
70
5.5
60
5.0
50
4.5
4.0 40
3.5 30
3.0 20
2.5 10
2.0
0
1.5
0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200 0 200 400 600 800 1000 1200 1400 1600 1800 2000 2200
Figure 16: Carbon nanofibre sample EC/042/08
31
Carbon nanofibre Carbon nanofibre
Supplier: Nanostructured & Amorphous Materials Inc Supplier: Nanostructured & Amorphous Materials Inc
Stock No 1190JN Stock No 1190JN
Lot No 1190-111907 Lot No 1190-111907
6.5 120
6.0
100

5.5
80
5.0
60
4.5
40
4.0
20
3.5
0 200 400 600 800 1000 1200 1400 1600
0 200 400 600 800 1000 1200 1400 1600
Figure 17: Carbon nanofibre sample EC/158/07

Carbon Nanofibre Carbon nanofibre
Supplier: Applied Sciences Inc. Supplier: Applied Sciences Inc
PR-25-XT-PS-AM PR-25-XT-PS-AM
Batch 1323 Box5 Batch 1323 Box5
700
7.0
600
6.5
500
6.0
400
300
5.5
200
5.0
100
4.5
0
0 500 1000 1500 2000 2500
0 500 1000 1500 2000 2500
Figure 18: Carbon nanotube sample EC/116/08
32
Carbon nanofibre Carbon nanofibre
Supplier: Applied Sciences Inc. Supplier: Applied Sciences Inc
PR-24-XT-HHT-AM Batch 155 PS 1131
Batch 155 PS 1131 Box1
160
5.8
140
5.6

5.4 120
5.2
100
5.0
80
4.8
4.6
60
4.4
40
4.2
4.0 20
0 200 400 600 800 1000 1200 1400 1600
3.8
0 200 400 600 800 1000 1200 1400 1600 Dust concentration (g/m3)
Figure 19: Carbon nanotube sample EC/117/08

Multi walled carbon nanotubes Multi walled carbon nanotubes
Supplier: Nanostructured & Amorphous Materials Inc. Supplier: Nanostructured & Amorphous Materials Inc.
Stock No 1229YJ Stock No 1229YJ
Lot No 1229-101603 Lot No 1229-101603
400
7
350
6
300
5
250
4
200
3 150
2 100
1 50
0
0
0 200 400 600 800 1000 1200 1400 1600
0 200 400 600 800 1000 1200 1400 1600
Figure 20: Carbon nanotubes sample EC/153/07
33
8.3 COMMISSIONING MIE TEST RESULTS
To prove the efficacy of the modified MIE apparatus, initial tests were done using lycopodium
powder, a well-established micron-sized test powder. The ignition circuit inductance was set at
zero, L=0 mH. The purpose was to compare the performance with results using the conventional
Kuhner test apparatus for the standard micron-sized test powder. Results for the conventional
dispersion system are shown in Figure 25; the MIE for this powder is within the range 10 – 30
mJ.
Barrel dispersion chamber
The MIE obtained using the enclosed barrel dispersion chamber were 30 – 100 mJ. This result
was greater than the conventional system and consequently the system was developed further
and a slide valve dispersion chamber was manufactured.
Slide valve dispersion chamber
The MIE obtained using the slide valve dispersion chamber were equivalent to the conventional
apparatus i.e. MIE = 10 – 30 mJ. This modified system was therefore used in subsequent tests.
.
Figure 25: MIE test on lycopodium (File MO20813A.XYZ)
34
8.4 MIE TEST RESULTS – NANOPOWDERS
Minimum ignition energies of nanopowders are presented in Table 10. Examples of tests shown
in Figures 26 and 27 illustrate the different burning characteristics of two metal nanopowders.
Iron nanopowder tended to produce many incandescent particles and less flame than zinc
nanopowder which produced a very bright flame.
Table 10: Nanopowder minimum ignition energies
Sample number Material Minimum

Ignition Energy
(mJ)
EC/104/08 Aluminium < 1 mJ

nanopowder
EC/011/09 Aluminium < 1 mJ

nanopowder
EC/153/07 Carbon >1000 mJ

nanotubes
EC/116/08 Carbon > 1000 mJ

nanotubes
EC/147/07 Iron nanopowder < 1 mJ
EC/152/07 Zinc 3 – 10 mJ
nanopowder
EC/148/07 Copper > 1000 mJ

nanopowder
35
Figure 26: Iron nanopowder
Figure 27: Zinc nanopowder
36
8.5 ELECTROSTATICS
8.5.1 Resistivity test results
8.5.1.1 Aluminium
Test data for aluminium nanopowder and micron-scale aluminium powder is presented in Table
11. and in Figure 28.
Table 11: Aluminium resistivity

Aluminium nanopowder EC/104/08 Aluminium standard powder EC/085/08
Relative Relative
humidity (%) Resistance (Ω) Resistivity (Ω.m) humidity (%) Resistance (Ω) Resistivity (Ω.m)
70.0 5.50E+07 3.44E+06 75.0 3.03E+03 1.89E+02
60.0 9.48E+07 5.93E+06 65.0 2.89E+03 1.81E+02
50.0 2.21E+08 1.38E+07 55.0 3.36E+03 2.10E+02
40.0 7.01E+08 4.38E+07 45.0 3.97E+03 2.48E+02

30.0 5.65E+10 3.53E+09 35.0 3.93E+03 2.46E+02
20.0 3.86E+11 2.41E+10 27.0 3.71E+03 2.32E+02
8.1 5.82E+12 3.64E+11 23.0 2.90E+03 1.81E+02
Bulk density 0.221 g/cm3 9.1 2.94E+03 1.84E+02
Bulk density 0.968 g/cm3
R e si st i v i t y v R e l a t i v e H u m i di t y f o r A l u m i ni um P owd e r s
Aluminium nanopow der EC/104/08 Aluminium standard pow der EC/085/08
1.00E+12
1.00E+11
1.00E+10
1.00E+09
1.00E+08
Ω
1.00E+07
1.00E+06
1.00E+05
1.00E+04
1.00E+03
1.00E+02
1.00E+01
1.00E+00
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
R e l a t i v e H umi di t y ( %)
Figure 28: Aluminium resistivity
37
8.5.1.2 Iron
Test data for iron oxide nanopowder, iron nanopowder and micron-scale iron powder is
presented in Table 12. and in Figure 29.
Table 12: Iron resistivity

Iron oxide nanopowder EC/150/07 Iron standard powder EC/003/09
Relative Relative
humidity (%) Resistance (Ω) Resistivity (Ω.m) humidity (%) Resistance (Ω) Resistivity (Ω.m)
75.0 3.85E+09 2.41E+08 65.0 51.5 3.22E+00
65.0 3.87E+09 2.42E+08 60.0 51.7 3.23E+00
55.0 3.96E+09 2.48E+08 55.0 51.8 3.24E+00
45.0 4.17E+09 2.61E+08 45.0 51.7 3.23E+00
35.0 4.46E+09 2.79E+08 35.0 51.8 3.24E+00
25.0 5.44E+09 3.40E+08 25.0 51.6 3.23E+00
16.6 7.63E+09 4.77E+08 5.0 51.2 3.20E+00
7.7 5.51E+10 3.44E+09 Bulk density 3.497 g/cm3
Iron nanopowder EC/147/07D (inert atmosphere)

Relative
humidity (%) Resistance (Ω) Resistivity (Ω.m)
7.8 1.45E+04 9.06E+02
Resistivity v Relative Hum idity for Iron Pow ders
Iron nanopow der EC/147/07D Iron standard pow der EC/003/09

Iron oxide nanopow der EC/150/07
1.00E+10
1.00E+09
1.00E+08
1.00E+07
1.00E+06
Ω
1.00E+05
1.00E+04
1.00E+03
1.00E+02
1.00E+01
1.00E+00
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
R elat ive Humid i t y ( %)
Figure 29: Iron resistivity
38
8.5.1.3 Zinc
Test data for zinc oxide nanopowder, zinc nanopowder and micron-scale zinc powder is
presented in Table 13 and in Figure 30.
Table 13: Zinc resistivity

Zinc oxide nanopowder EC/149/07 Zinc oxide standard powder EC/112/08
Relative Relative humidity
humidity (%) Resistance (Ω) Resistivity (Ω.m) (%) Resistance (Ω) Resistivity (Ω.m)
76.7 2.23E+06 1.39E+05 90.0 1.93E+09 1.21E+08
66.0 3.00E+06 1.88E+05 85.0 1.77E+09 1.11E+08
60.0 3.18E+06 1.99E+05 78.1 1.93E+09 1.21E+08
54.0 3.71E+06 2.32E+05 70.0 2.13E+09 1.33E+08
50.0 3.93E+06 2.46E+05 63.2 2.44E+09 1.53E+08
43.5 4.30E+06 2.69E+05 52.3 3.52E+09 2.20E+08
40.0 4.87E+06 3.04E+05 44.9 5.42E+09 3.39E+08
35.7 5.24E+06 3.28E+05 40.2 7.61E+09 4.76E+08
25.7 7.68E+06 4.80E+05 28.8 2.48E+10 1.55E+09
22.2 1.11E+07 6.95E+05 23.5 6.10E+10 3.81E+09
18.9 1.72E+07 1.07E+06 19.9 1.14E+11 7.13E+09
17.4 2.52E+07 1.58E+06 16.6 2.29E+11 1.43E+10
15.6 3.50E+07 2.19E+06 14.6 4.68E+11 2.93E+10
14.9 5.01E+07 3.13E+06 8.5 1.82E+13 1.14E+12
8.7 9.83E+10 6.14E+09 Bulk density 0.690 g/cm3
Zinc nanopowder EC/152/07 (inert atmosphere)
Relative
humidity (%) Resistance (Ω) Resistivity (Ω.m)
5.7 6.00E+04 3.75E+03
Resistivity v Relative Humidity for Zinc Oxide powders
Zinc oxide nanopowder EC/149/07 Zinc oxide standard powder EC/112/08
1.00E+13
1.00E+12
1.00E+11
1.00E+10
1.00E+09
1.00E+08
Resistivity (Ω.m)
1.00E+07
1.00E+06
1.00E+05
1.00E+04
1.00E+03
1.00E+02
1.00E+01
1.00E+00
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
Relative Humidity (%)
Figure 30: Zinc resistivity
39
8.5.1.4 Copper
Test data for copper nanopowder and micron-scale copper powder is presented in Table 14 and
in Figure 31.
Table 14: copper resistivity

Copper nanopowder EC/148/07A (inert atmosphere)
Relative humidity (%) Resistance (Ω) Resistivity (Ω.m)
6.6 3.75E+03 2.34E+02
Copper standard powder EC/004/09
70.0 121.9 7.62E+00
60.0 120.7 7.54E+00
50.0 119.8 7.49E+00
45.0 116.3 7.27E+00
40.0 115.9 7.24E+00
30.0 115.4 7.21E+00
25.0 114.7 7.17E+00
10.0 113.6 7.10E+00
Resistivity of Copper Pow ders versus Hum idity
Copper nanopow der EC/148/07A Copper standard pow der EC/004/09
1.00E+05
1.00E+04
1.00E+03
Ω
1.00E+02
1.00E+01
1.00E+00
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
R e l a t i v e H u m i di t y ( %)
Figure 31: Copper resistivity
40
8.5.1.5 Zirconium oxide
Test data for zirconium oxide nanopowder and micron-scale zirconium powder is presented in
Table 15 and in Figure 32.
Table 15: zirconium oxide

Zirconium oxide nanopowder EC/151/07D

70.0 2.35E+07 1.47E+06
60.0 2.58E+07 1.61E+06
50.0 2.80E+07 1.75E+06
40.0 3.34E+07 2.09E+06
30.0 5.29E+07 3.31E+06
23.0 1.64E+08 1.02E+07
8.8 1.20E+09 7.51E+07
Zirconium oxide standard powder EC/111/08

75.0 1.85E+07 1.16E+06
70.0 1.88E+07 1.17E+06
65.0 1.93E+07 1.20E+06
60.0 1.96E+07 1.23E+06
50.0 2.05E+07 1.28E+06
40.0 2.21E+07 1.38E+06
28.3 2.64E+07 1.65E+06
22.5 3.98E+07 2.49E+06
18.3 5.08E+07 3.18E+06
16.9 5.92E+07 3.70E+06
7.5 3.95E+10 2.47E+09
Bulk density 1.716 g/m3
Resistivity v Relative Humidity For Zirconium Oxide Powders
Zirconium oxide nanopowder EC/151/07D Zirconium oxide standard powder EC/111/08
1.00E+10
1.00E+09
1.00E+08
1.00E+07
Resistivity (Ω.m)
1.00E+06
1.00E+05
1.00E+04
1.00E+03
1.00E+02
1.00E+01
1.00E+00
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
Relative Humidity (%)
Figure 32: Zirconium oxide resistivity
41
8.5.1.6 Carbon
Test data for carbon nanomaterials and micron-scale carbon powder is presented in Table 16
and in Figure 33.
Table 16: Carbon resistivity

Multi-walled carbon nanotubes EC/153/07C

75.0 1.71E+02 1.07E+01
65.0 1.64E+02 1.03E+01
55.0 1.70E+02 1.06E+01
45.0 1.69E+02 1.06E+01
35.0 1.78E+02 1.11E+01
25.0 4.64E+01 2.90E+00
15.0 4.46E+01 2.79E+00
7.6 4.44E+01 2.78E+00
Carbon nanofibres EC/158/07A

75.0 8.44E+02 5.28E+01
64.0 7.98E+02 4.99E+01
55.0 7.66E+02 4.79E+01
45.0 7.52E+02 4.70E+01
35.0 7.41E+02 4.63E+01
25.0 7.40E+02 4.63E+01
17.1 7.20E+02 4.50E+01
7.8 3.05E+01 1.91E+00
Pyrograf III nanofibres EC/042/08

75.0 8.36E+04 5.23E+03
65.0 8.39E+04 5.24E+03
55.0 8.58E+04 5.36E+03
45.0 8.88E+04 5.55E+03
33.0 8.73E+04 5.46E+03
25.0 9.20E+04 5.75E+03
19.6 9.23E+04 5.77E+03
9.8 2.48E+04 1.55E+03
Carbon micron powder EC/113/08

70.0 3.25 2.03E-01
60.0 3.19 1.99E-01
50.0 3.15 1.97E-01
40.0 3.11 1.94E-01
28.0 2.93 1.83E-01
21.5 2.8 1.75E-01
8.5 3.2 2.00E-01
42
Resistivity v Relative Hum idity for Carbon Pow ders
Multi-w alled carbon nanotubes EC/153/07C Carbon nanofibres EC/158/07A

Pyrograf III nanofibres EC/042/08 Carbon micron pow der EC/113/08
1.00E+04
1.00E+03
1.00E+02
Ω
1.00E+01
1.00E+00
1.00E-01
0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0
R el at i ve Humi d it y ( %)
Figure 33: Carbon resistivity
43
8.5.2 Charging test results
8.5.2.1 Commissioning
Base-line test
As part of the commissioning tests on this equipment, a base-line test with no powder was
initially carried out; this was done regularly during the tests to ensure the instrumentation was
functioning correctly. Although this test was performed with new, clean pipes, surface and air
contaminants seem to have contributed to the occurrence of some charging. An example of the
plot of voltage against time is shown in Figure 34 and the test data is shown in Table 17. The
airflow was maintained at a velocity of 28 m/s for the test programme.
100m
50m
-50m
- 100m
- 150m
- 200m
- 250m
- 300m
- 350m
- 400m
- 450m
- 500m
- 550m
- 600m
- 650m
- 700m
- 750m
- 800m
- 850m
- 900m
- 950m
-1
-1.05
-1.1
-1.15
-1.2
-1.25
0 50 100 150 2 00 250 300 350 4 00 45 0 500 550 600 65 0 700 750 800 8 50 88 0.2
Time (s)
Figure 34 : Base-line test without powder
Table 17 : Base-line test without powder

Rate of change of voltage (V/s) Capacitance (F) Current (A)
-6.29E-03 1.02E-08 -6.40E-11
Ambient environmental conditions:
Humidity: 67.0%RH
Temperature: 19.9ºC
44
Lycopodium (EC/026/08)
To test the system, micron-scale lycopodium powder was tested. This was free flowing and did
not aggregate or accumulate in any significant quantities. During testing a very fine layer of
powder coated the inside wall of the pipes.
Figure 35 shows the plot of voltage against time and Table 18 presents the test results. An initial
period of high rate of change of voltage was noted in the first few seconds following the fan
being started, the cause of this was not investigated. This was followed by a period of steady
gradient.
11.5
11.25
11
10.75
10.5
10.25
10
9.75
9.5
9.25
9
8.75
8.5
8.25
8
7.75
7.5
7.25
7
6.75
6.5
6.25
6
5.75
5.5
5.25
5
4.75
4.5
4.25
4
3.75
3.5
3.25
3
2.75
2.5
2.25
2
1.75
1.5
1.25
1
750m
500m
250m
0
-250m
0 10 20 30 40 50 60 70 80 90 100 110 120 130 14 0 150 160 170 18 0 190 200 210 220 230 240 250 255.7
Time (s )
Figure 35: Lycopodium (EC/026/08)
Table 18: Lycopodium (EC/026/08)

1.76E-01 3.84E-09 6.77E-10
45
8.5.2.2 Charging tests
Micron-scale zinc oxide (EC/112/08)
The micron scale zinc oxide powder was observed to be highly cohesive; small quantities could
be compressed together to form a layer on the surface of a spatula.
A mass of 0.1g of powder was tested. Charge was generated for less than 1.5 seconds. The
powder quickly accumulated around any obstructions and rapidly removed the powder out of
circulation; this was particularly apparent where the plastic pipe was connected onto a metal
tube at the powder injection port. Some powder was also deposited on the inner wall of the
pipes. The test results are shown in Table 19.
Table 19: Micron-scale Zinc Oxide (EC/112/08)

8.12E-01 1.02E-08 8.26E-09

Humidity: 63.7%RH
46
Zinc oxide nanopowder (EC/149/07)
Zinc oxide nanopowder had a very fine consistency with very little cohesion under
compression. A mass of 6g of powder was tested. Charging was observed for the whole duration
of the test (Figure 36). A number of phases of charging each with different gradients are
apparent from the data. There was little powder accumulation observed but a significant layer of
powder formed at the inner walls of the pipe. A good proportion of the powder was observed
resting in the bottom of the pipes following the test. The test results are shown in Table 20.
8 60m
8 40m
8 20m
8 00m
7 80m
7 60m
7 40m
7 20m
7 00m
6 80m
6 60m
6 40m
6 20m
6 00m
5 80m
5 60m
5 40m
5 20m
5 00m
4 80m
4 60m
4 40m
4 20m
4 00m
3 80m
3 60m
3 40m
3 20m
3 00m
2 80m
2 60m
2 40m
2 20m
2 00m
1 80m
1 60m
1 40m
1 20m
1 00m
80m
60m
40m
20m
0
-20m
-40m
0 20 40 60 80 1 00 120 14 0 160 18 0 200 220 2 40 260 2 80 300 32 0 340 360 380 400 4 20 43 3.1
Time (s)
Figure 36: Zinc oxide nanopowder (EC/149/07)
Table 20: Zinc oxide nanopowder (EC/149/07)

8.92E-03 1.05E-06 9.37E-09
Humidity: 41.7%RH
47
8.5.2.3 Zirconium Oxide
Micron-scale zirconium oxide (EC/111/08)
The powder had a very fine consistency with very little cohesion under compression.
A mass of 1g of powder was tested. A rate of charging greater than anticipated was observed;
the voltage across the capacitor rapidly increased to a value greater than could be measured. The
powder was deposited in a thick layer on the inner wall of the pipe.
Following the test the resistance between measuring station and the Earthed fan casing was
measured as greater than 2000MΩ.
The test results are shown in Table 21.
Table 21: Micron-scale zirconium oxide (EC/111/08)

4.95E+00 1.02E-08 5.04E-08
Humidity: 63.7%RH
Zirconium oxide nanopowder (EC/151/07)
A quantity of 1g of powder was tested. The powder accumulated at the pipe joints forming a
significant build-up. Charging was recorded for approximately 2 seconds. There was little
powder was deposited on the inner wall of the pipe.
Further quantities of powder were added to try to overcome the problems of powder build and
hence reduced powder circulation. A maximum of 6g of powder were used, these tests resulted
in very similar behaviour and charging magnitudes.
The test results are shown in Table 22.
Table 22: Zirconium oxide nanopowder (EC/151/07)

-1.84E-01 1.05E-06 -1.93E-07
Humidity: 41.7%RH
48
8.5.2.4 Carbon
Micron-scale carbon (EC/113/08)
The powder was very cohesive; this was evident in the difficulty experienced in releasing the
powder from the powder loading port.
A quantity of 3g of powder was tested. The voltage across the capacitor rapidly increased to a
value greater than could be measured. The powder was deposited in a significant layer on the
inner wall of the pipe. A plot of the test is shown in Figure 37 and the test results are shown in
Table 23.
It was noted that the voltage on the capacitor returned to 0V within 1 second once the fan was
stopped. The resistance between measuring station and the Earthed fan casing was measured as
0.50MΩ. The carbon deposits on the inner walls of the pipe has significantly reduced the
resistance to Earth.
10 .74
1 0.5
10 .25
10
9 .75
9.5
9 .25
8 .75
8.5
8 .25
8
7 .75
7.5
7 .25
7
6 .75
6.5
6 .25
6
5 .75
5.5
5 .25
5
4 .75
4.5
4 .25
4
3 .75
3.5
3 .25
3
2 .75
2.5
2 .25
2
1 .75
1.5
1 .25
1
75 0m
50 0m
25 0m
0
-263.4m
1 1:36:55.739 11:36:57.739 11:36 :59.739 11:37:01.739 11:37:03. 739 11:37:05.739 11:37:07.739 11:37:09.739 11:37:11.544
Time (s)
Figure 37: Micron Scale Carbon (EC/113/08)
Table 23: Micron Scale Carbon (EC/113/08)

1.28E+01 1.02E-08 1.31E-07
Humidity: 65.3%RH
49
Carbon Pyrograf III nanofibres (EC/042/08)
The powder was observed as being made up of mixed sized particles from fine dust to
agglomerated particles of approximately 2mm in size. Powder could not be aggregated further
when compressed.
A quantity of 2g of powder was tested. A plot of the test is shown in Figure 38 and the test data
is shown in Table 24. A continuous flow of powder was observed, there was no apparent
powder accumulation. A significant layer of powder was deposited on the inner wall of the pipe.
A quantity of mainly larger sized particles where present on the bottom of the pipe once the fan
was stopped.
100m
50m
-50m
-100m
-150m
-200m
-250m
-300m
-350m
-400m
-450m
-500m
-550m
-600m
-650m
-700m
-750m
-800m
-850m
-900m
-950m
-1
-1.05
-1. 1
-1.15
-1. 2
-1.25
-1. 3
-1.35
-1. 4
-1.45
-1. 5
-1.55
14: 00:36.256 14:01:26 .256 14:02:16.256 14:03:06.256 14:03:56.2 56 14:04:46.256 14:05: 36.256 14:06:26.25 6 14 :07:16.256 14:08 :11.156
Time (s)
Figure 38: Carbon Pyrograf III nanofibres (EC/042/08)
Table 24: Carbon Pyrograf III nanofibres (EC/042/08)

-3.42E-02 1.05E-06 -3.59E-08
Humidity: 65.3%RH
50
Carbon nanofibres (EC/158/07)
In common with sample EC/042/08, the powder was observed as being made up of mixed sized
particles from fine dust to agglomerated particles of approximately 2mm in size. Powder could
not be aggregated further when compressed.
A quantity of 1g of powder was tested. During the test, a continuous flow of powder was
observed but a layer of powder was deposited on the inner wall of the pipe. A quantity of
mainly larger sized particles where present on the bottom of the pipe once the fan was stopped.
The plot of the test is shown in Figure 39 and the test data is presented in Table 25. It was
observed that the voltage across the capacitor dropped away rapid during the test. The resistance
between measuring station and the Earthed fan casing was measured as 0.12MΩ. The carbon
deposits on the inner walls of the pipe has significantly reduced the resistance to Earth.
20m
-20m
-40m
-60m
-80m
-100m
-120m
-140m
-160m
-180m
-200m
-220m
-240m
-260m
-280m
-300m
-320m
-340m
-360m
-380m
-400m
-420m
-440m
-460m
-480m
-500m
-520m
-540m
-560m
-580m
-600m
-620m
-640m
-660m
-680m
-700m
-720m
-740m
-760m
-780m
-800m
-820m
0 10 20 30 40 50 60 70 80 90 100 110 120 1 30 140 150 160 170 180 19 0 200 210 220 230 240 25 0 2 60 26 8.7
Time (s )
Figure 39: Carbon nanofibres (EC/158/07)
Table 25: Carbon nanofibres (EC/158/07)

-1.16E-02 1.05E-06 -1.22E-08
Humidity: 65.3%RH
51
Carbon multi-walled nanotubes (EC/153/07)
The powder was observed as being made up of mixed sized particles from fine material of
approximately 2mm in diameter. Powder could not be aggregated further when compressed.
A quantity of 1g of powder was tested. A continuous flow of powder was observed. A

significant layer of powder was deposited on the inner wall of the pipe and a quantity of mainly
larger sized particles where present on the bottom of the pipe once the fan was stopped.
During the test there was audible evidence of electrostatic discharges. However, there was no
apparent visible evidence of discharges.
measured as 0.15MΩ. A plot of the test is shown in Figure 40 and the test data is presented in
Table 26.
20m
-20m
-40m
-60m
-80m
-100m
-120m
-140m
-160m
-180m
-200m
-220m
-240m
-260m
-280m
-300m
-320m
-340m
-360m
-380m
-400m
-420m
-440m
-460m
-480m
-500m
-520m
-540m
-560m
-580m
-600m
-620m
-640m
-660m
16:23:15.287 16 :23:35.287 1 6:23:55.287 16:24:15.287 16:24:35.287 16:24:55.28 7 16:25:15.2 87 16:25:35. 287 16:25:55 .287 16:26:1 5.287 16:26 :35.287 16:2 6:55.287 16:27:15.287 16:27 :48.987
Time (s)
Figure 40: Carbon multi-walled nanotubes (EC/153/07)
Table 26: Carbon multi-walled nanotubes (EC/153/07)

-2.51E-02 1.05E-06 -2.63E-08
Humidity: 65.3%RH
52
8.5.2.5 Copper
Micron-scale copper (EC/004/09)
A quantity of 5g of powder was tested. A continuous flow of powder was observed; there was no
apparent powder accumulation. A very fine layer of powder was deposited on the inner wall of
the pipe. The test data is presented in Table 27.
Table 27: Micron-scale copper (EC/004/09)

5.65E-02 1.05E-06 5.93E-08
Humidity: 61.8%RH
Copper nanopowder (EC/148/07)
The powder was naturally aggregated into approximately 0.5mm material. The powder could
not be aggregated further when compressed.
A quantity of 2.5g of powder was tested. The powder was initially observed to flow reasonably
well. The inner wall of the pipe was coated with a fine layer of powder shortly after starting the
test. When the test was complete there was little evidence of any free powder remaining in the
pipes.
measured as greater than 2000MΩ. The test data is presented in Table 28.
Table 28: Copper nanopowder (EC/148/07)

-4.89E-02 1.05E-06 -5.14E-08
Humidity: 61.8%RH
53
9 DISCUSSION
9.1 EXPLOSION TEST EQUIPMENT
A 2 litre explosion test facility has been established. Equipment has been manufactured and
where necessary existing equipment has been modified and configured to allow the safe
handling and testing of nanopowders. Sealed systems have been designed for use in
conjunction with a glovebox that have allowed nanopowders to be weighed and handled under
an inert atmosphere. This has enabled non-oxidised materials to be safely handled and held
under inert conditions until the point of ignition.
The 2 litre explosion test apparatus has been used to measure the explosion pressure and rates of
pressure rise of nanopowders using a set of conditions established and calibrated against a range
of dusts tested in both the 2 litre and 20 litre vessels. The small volume of the 2 litre vessel has
reduced the quantity of test material required for the test procedure. Typically, 50 – 100 g of
material is required for a series of tests covering a wide range of dust concentrations. This is
approximately 10% of the material requirements for a standard 20 litre sphere test.
The small volume of the test chamber was a concern in that quenching of the flame at the vessel
wall was possible and could lead to lower combustion rates. This proved to be the case and the
rates of pressure rise were lower than might be expected from a small vessel. However, a
scaling factor has been obtained from the test data and equivalent KSt values have been
obtained. The maximum explosion pressures were generally comparable with those measured
in the 20 litre sphere.
The minimum ignition energy of nanopowders was measured using a modified Kuhner MIKE3
test apparatus. The equipment was modified to allow nanopowder to be safely handled within
the test apparatus. The modifications included enclosing the dispersion cup with a slide valve to
fully contain the nanopowder and allow handling within a glovebox. Argon was used for
inerting the material. The full test chamber assembly was configured such that it was removable
as a complete sealed assembly for cleaning purposes.
9.2 EXPLOSION SEVERITY AND MIE
A comparison of the nanopowder data generated in this test programme with published micron-
scale powder data (BIA-Report 1997) is summarised in Table 29. Generally, the KSt values of
nanopowders are similar to conventional micron-scale powders. The Pmax values are a little
lower due to flame quenching effects at the vessel wall. Surprisingly, the large surface to
volume ratio has not produced greater reactivity than the equivalent material at micron-scale.
However, the minimum ignition energies of some nanopowders have been lower than the
equivalent material at micron-scale. This may be explained by consideration of dispersion and
agglomeration (Preining, 1998). Although powders are initially well dispersed in the 2 litre
chamber, the rate of re-agglomeration may be such that nanopowders are dispersed only for a
very short duration and rapidly begin to re-agglomerate. Thus, combustion may well take place
initially with dispersed particles but some degree of re-agglomeration of the particles may take
place before combustion has been completed. This hypothesis would need to be checked by
sampling the dispersed dust cloud and measuring particle size using equipment that measures
the size distribution and number concentration of fast dynamic aerosols.
Equipment needed for measuring the particle size of the dispersed dust cloud was considered
during this project. The current bench-mark instrument for measurement of nanoparticles in
54
their airborne state is considered to be the Scanning Mobility Particle Sizer (SMPS). This
instrument measures a size and number distribution from around 5 nanometres up to 1 micron in
diameter.
The SMPS, comprises a Condensation Particle Counter (CPC) and a Differential Mobility
Analyser (DMA). The latter component invariably employs a radioactive source to produce a
known charge distribution on the sampled aerosol through ionisation of the sampled air.
However, the SMPS cannot be used to obtain a particle size distribution in anything like real-
time. Full particle size data is derived from a series of scans during which a varying voltage is
applied to the selection electrode in the Differential Mobility Analyser. In general, instruments
of the SMPS type require around three minutes to complete a full particle size scan. This means
that the information gathered by the SMPS will be spurious if the concentration and size range
of the aerosol being sampled are dynamic and not constant. This is particularly disadvantageous
when measuring nanoparticles dispersed into the air during explosion testing. Under these
circumstances, the distribution of sizes will be extremely transitory due to the tendency of the
primary particles to rapidly agglomerate. Therefore the SMPS equipment could not be used for
assessment of the particle size and agglomeration.
An instrument is required that will capture the transient particle size distribution. The
instrument identified for this work is a Fast Mobility Particle Sizer Spectrometer. A sample in
near real-time is taken and is therefore an ideal tool to measure an aerosol composed of
nanoparticles, either in the laboratory or in the field, under rapidly changing conditions. It is
recommended that further work is done to assess the particle size/agglomeration of the dust
cloud at the point of ignition using this instrument.
In common with micron-scale powders the maximum rate of pressure rise and maximum
explosion pressure of nanopowders sometimes occurred at different dust concentrations. The
metal dusts tend to have relatively high optimum dust concentrations compared with carbon
nanopowders, for example the optimum dust concentration for aluminium is within the range
2000 – 3000 g/m3 whereas carbon is typically within the range 400 –1500 g/m3.
9.2.1 Aluminium
The aluminium nanopowders have demonstrated explosion properties comparable with the
micron-scale powders. Published data (BIA, 1997) of micron-scale aluminium powders
typically have KSt values within the range 300 – 700 bar m/s, and have maximum explosion
pressures within the range 7 – 12 bar. KSt values of 449 bar m/s for 210 nm aluminium
nanopowder and 536 bar m/s for 100 nm nanopowder powder are well within this range of
values. As might be expected, the finer aluminium nanopowder had the higher KSt value. Other
research workers (Nanosafe, 2008) have found the KSt of some aluminium nanpowders to
decrease with decreasing particle size. This has been attributed to the thin oxide layer wrapping
the passivated nanoparticles; this may make them less explosible than micron-scale aluminium
nanopowders. The aluminium nanopowders are easily ignited, burning with a highly
incandescent flame and are firmly in the St3 Dust Group. Aluminium nanopowders, as with
micron-scale aluminium powders, have very severe explosion characteristics.
The minimum ignition energies of aluminium nanopowders were comparable with micron-scale
aluminium with MIE < 1 mJ. The aluminium nanopowders are therefore very sensitive to
ignition by electrostatic discharge and extreme caution is required when handling these
powders.
55
9.2.2 Zinc
Visually, zinc nanopowder burned in a similar manner to aluminium, producing a very bright
flame. The maximum explosion pressure was 5.6 bar, the maximum rate of pressure rise was
377 bar/s resulting in an equivalent KSt value of 101 bar m/s. The nanopowder is thus classified
as an St1 dust. The zinc nanopowder characteristics are typical of published micron-scale zinc
powders (BIA 1997). The minimum ignition energy was 3 – 10 mJ and was considerably lower
than published data for micron-scale zinc powder; 100-1000 mJ and 300-1000 mJ have been
quoted. The ignitability of zinc nanopowder is therefore greater than conventional micron-scale
zinc powder.
9.2.3 Iron
Iron nanopowder produced an explosion pressure of almost 3 bar and a rate of pressure rise of
68 bar m/s resulting in an equivalent KSt value of 18 bar m/s. This was lower than published
data for micron-scale iron where explosion pressures 5.1 - 5.2 bar and KSt values of 41 - 50 bar
m/s have been published. Iron nanopowder is clearly an St1 dust. The minimum ignition
energy of iron nanopowder was less than 1 mJ. Although the explosion data is not as severe as
micron-scale powder, the low MIE value is lower than micron-scale iron powder and
demonstrates it is potentially a very ignitable nanopowder.
9.2.4 Copper
Copper nanopowder proved to have very low explosion characteristics to the extent that it is
considered non-explosible. This is consistent with micron-scale copper powder; test data in the
BIA database indicates micron-scale copper powder is non-explosible.
9.2.5 Carbon
The carbon nanopowders generally had a very agglomerated appearance and had poor handling
characteristics in that they would not flow in the manner of free-flowing powders. KSt values
were within the range 17-158 bar m/s and are therefore classified as St1 dusts and are weak to
moderately explosible. The difference in the reactivity of the two extremes may be due to the
length of the fibres. Sample EC/042/08 was the lowest rated carbon nanofibre and had relatively
long fibres; 30-100 micron long with a diameter of 100-200 nm. The highest rated material was
carbon nanofibre sample EC/116/08; this had a diameter of 70 – 200 nm but a significantly
shorter length of 2-5 micron. Therefore, as might be expected, the shorter carbon nanofibres
were more explosible than the longer fibres. The minimum ignition energies of these materials
were greater than 1000 mJ and therefore the materials are not sensitive to ignition from
electrostatic ignition sources.
56
Table 29: Comparison of nanopowders with micron-scale powders
Nanopowder Micron-scale powder (typical range of data)
Material Particle size Pmax (barg) Equivalent Material Particle size Pmax KSt
KSt(bar.m/s) (barg) (bar m/s)
Aluminium 210 nm 12.5 449 Aluminium Median 7-12 300-700

nanopowder
<10-100μm
(210 nm)
Aluminium 100 nm 11.2 536

nanopowder
(100 nm)
Iron 25 nm 2.9 18 Iron Median 12μm 5.2 50

nanopowder
Median 32μm 5.1 41
Zinc 130 nm 5.6 101 Zinc Median 160μm 0.7 2

nanopowder
Median 10μm 7.3 176
Copper 25 nm 1.2 3 Copper Median 25 μm No No

nanopowder ignition ignition
Carbon Dia 100-200 nm 5.2 17 Carbon 100%<63μm 8 151

nanofibre
Length 30-100μm 100%<63 μm 7.1 43
EC/42/08
Carbon Dia 80-200 nm 6.0 30

nanofibre
EC/158/07 Length 0.5-20μm
Carbon Dia 70-200 nm 6.9 158

nanofibre
Length 2-5μm
EC/116/08
Carbon Dia 70-200 nm 5.6 37

nanofibre
Length 2-10 μm
EC/117/08
Multi-walled Dia 20-30 nm 6.4 91

carbon
nanotubes Length 10-30μm
EC/153/07
57
9.3 ELECTROSTATIC ISSUES
9.3.1 Resistivity
The resistivity data demonstrates that with increasing relative humidity the resistivity of
aluminium, zinc oxide and zirconium oxide nanopowders decreases. Carbon nanopowders
tended to be relatively insensitive to changes in humidity. The effect of high relative humidity
on reducing the resistivity of powders is a well-known characteristic in micron-scale powders.
PD CLC/TR 50404 (2003) indicates that high relative humidity decreases the surface resistivity
of many powders. But for this to be effective as a means of charge reduction, most materials
need a relative humidity well in excess of 70 % and it is often impracticable to operate powder
processing units at these high levels of humidity. NFPA 77 comments that at humidities of 65%
and higher, the surface of most materials absorb enough moisture to ensure a surface
conductivity that is sufficient to prevent accumulation of static electricity.
Generally, nanopowders have greater resistivity values than conventional micron-scale powders.
For example, the three carbon nanopowders all had greater resistivity values than the micron
scale carbon powder. The test results have shown that nanopowders have lower packing
densities than micron-scale powders, i.e. Table 16 shows densities of 0.047 g/cm3 to 0.105
g/cm3 for carbon nanopowders compared with 0.577 g/cm3 for micron-scale carbon. Since the
low packing densities of nanopowders are accompanied by greater voids (air gaps) between the
particles, when compared with micron-scale powders, this may be responsible for the higher
resistivity values.
Table 30 is a summary of the data over the range of humidity levels. Aluminium, iron oxide,
and zinc oxide were medium resistivity powders but touched on high resistivity at low humidity
levels. The individual materials are discussed in more detail below.
Table 30: Summary of nanopowder resistivity

Powder Sample number Range of experimental Comment
resistivity values (Ωm)
Aluminium EC/104/08 3.44E+06 to 3.64E+11 Generally a medium resistivity powder.

nanopowder
Iron nanopowder EC/147/07 9.06E+02 Conductive
Iron oxide nanopowder EC/150/07 2.41E+08 to 3.44E+09 Medium resistivity powder
Zinc nanopowder EC/152/07 3.75E+03 Conductive
Zinc oxide nanopowder EC/149/07 1.39E+05 to 6.14E+09 Medium resistivity powder
Copper nanopowder EC/148/07 2.34E+02 Conductive
Zirconium oxide EC/151/07 1.47E+06 to 7.51E+07 Medium resistivity powder

nanopowder
Multi-walled carbon EC/153/07 2.78E+00 to 1.11E+01 Conductive

nanotubes
Carbon nanofibres EC/158/07 1.91E+00 to 5.28E+01 Conductive
Carbon nanofibres EC042/08 1.55E+03 to 5.77E+03 Conductive
58
9.3.1.1 Aluminium
The resistivity of aluminium nanopowder was 3.64 x 1011 Ωm at 8% RH drecreasing to 3.44 x

106 Ωm at 70% RH. Micron-scale aluminium powder was insensitive to changes in humidity
and had a relatively low resistivity in the order of 2 x 102 Ωm. The aluminium nanopowder
appears to be sensitive to changes in humidity and can be classified as medium resistivity for
the medium RH levels. At low RH it can be classified as a low-resistivity powder and at high
RH it can be classified as a high-resistivity powder. It was generally more resistive than micron-
scale aluminium powder.
9.3.1.2 Iron
Iron nanopowder had a resistivity of 9 x 102 Ωm at 7.8% RH and was more resistive than the
micron-scale iron powder. Iron nanopowder is considered to be a conductive powder. Iron oxide
nanopowder had a resistivity of 3.44E+09 Ωm at 7.7 % RH making this powder a medium
resistivity powder. The micron-scale iron powder was insensitive to changes in humidity.
9.3.1.3 Zinc
Zinc nanopowder had relatively low resistivity with value of 3.75 x 103 Ωm at a relative
humidity of 5.7% and is classified as a conductive powder. The zinc oxide nanopowder had
higher resistivity, decreasing with increasing humidity and is classified as a medium resistivity
powder.
9.3.1.4 Copper
Copper nanopowder and micron-scale copper both had very low resistivity values (2.34 x 102
Ωm and 7.1 Ωm respectively) and are classified as conductive powders. The micron-scale
powder was insensitive to humidity levels.
9.3.1.5 Zirconium oxide
Zirconium oxide nanopowder and micron-scale powder closely mirrored each other. At low
humidity, both had similar resistivities, with the nanopowder having a resistivity of 7.51 x 107
Ωm and the micron-scale powder 2.47 x 109 Ωm. They are classified as medium resistivity
powders. Both had decreasing resistivities with increasing humidity.
9.3.1.6 Carbon
Carbon nanofibres, nanotubes and micron-scale carbon powder all had low resistivities. Their
resistivities appear to be relatively insensitive to changes in relative humidity. The resistivities
were within the range 0.2 Ω for micron-scale carbon to 5.7 x 103 Ωm for the nanofibre sample.
The samples were all considered to be a conductive powders.
Should the conductive nanopowders penetrate into electric and electronic equipment, they could
give rise to short-circuit problems and may lead to the generation of ignition sources. The
possibility of nanopowders penetrating into electrical and electronic equipment may be greater
due to their reduced particle size.
9.3.2 Charging
The results of the charging tests are presented in Table 31. The resistivity results for each
powder are also shown in this table.
59
Table 31: Charging tests
Powder Sample Number Resistivity (Ωm) Current (A)
Copper
nanopowder EC/148/07 2.34E+03 -5.14E-08
Copper micron
scale EC/004/09 7.49E+01 5.93E-08
Carbon nanotubes EC/153/07 1.06E+01 -2.63E-08
Carbon nanofibres EC/158/07 4.79E+01 -1.22E-08
Carbon Pyrograf
III nanofibres EC/042/08 5.36E+03 -3.59E-08
Carbon micron
scale EC/113/08 1.97E-01 1.31E-07
Zirconium Oxide
nanopowder EC/151/07 1.75E+06 -1.93E-07
Zirconium Oxide
micron scale EC/111/08 1.23E+06 5.04E-08
Zinc Oxide
nanopowder EC/149/07 2.46E+05 9.37E-09
Zinc Oxide micron
scale EC/112/08 1.53E+08 8.26E-09
All the powders produced charge, indicated by the current in Table 31: some materials
developed negative and some developed positive charge. Generally, the charge developed by
nanopowders was comparable with the micron-scale powders.
Various graphs of resistivity against current have been plotted but no relationship between these
two properties can be identified from this data. The chargeability of powders is a notoriously
complex subject with numerous factors (humidity, surface contaminants, particle size etc.)
influencing a powder’s tendency to charge. The British standard BS5958-1:1991 refers to the
charging characteristics of powders and states “The charging characteristics are often
determined at least as much by surface contamination of the particles as by the chemical
composition of the powder itself and charge generation is usually difficult to predict”. It is
therefore likely that the charging characteristics of the nanopowders will be significantly
influenced by their surface characteristics.
BS5958-1:1991 indicates that the most important parameter in electrostatic phenomena in

powders is the mass charge density of the powder (i.e. the charge to mass ratio). The tendency
for many of the powders to accumulate around small obstructions, therefore reducing the net
powder flowing in the system, meant that it wasn’t possible to ensure that each test was
performed with the same quantity of powder. This tendency for particle deposition also limited
the duration of many tests, which meant it was difficult to identify whether a steady state of
charge accumulation had been reached.
BS5958-1:1991 provides details of suggested test apparatus for determining the mass charge
density of a powder. This apparatus is not suitable for use with nanopowders given the potential
risks from exposure and the small size of the particles. However, with further work and the
experience gained in this test programme, the principles of this apparatus could be employed in
new test apparatus designed specifically for work with nanopowders.
60
10 CONCLUSIONS
Equipment has been manufactured and where necessary existing equipment has been modified
and configured to allow the safe handling and testing of nanopowders. The equipment
comprises:
• 2 litre explosion test apparatus for measuring the rate of pressure rise and the maximum
explosion pressure.
• Minimum ignition energy test apparatus. The minimum ignition energy of nanopowders
was measured using a modified Kuhner MIKE3 test apparatus that allowed nanopowder
to be safely handled within the test apparatus.
• Sealed systems that allow the safe handling of oxidized and non-oxidized nanopowders
under inert atmospheres.
KSt , Pmax and MIE values have been obtained for a range of metal and carbon nanopowders. The
KSt is an equivalent value scaled from the results obtained in the 2 litre vessel. Generally, the
explosibility (maximum explosion pressure, rates of pressure rise and KSt) of nanopowders are
broadly similar to conventional micron-scale powders. Surprisingly, their large surface to
volume ratio has not produced greater explosion violence than the equivalent material at
micron-scale. However, the minimum ignition energies of some nanopowders have been lower
than the equivalent material at micron-scale. This indicates that the nanopowders may be more
susceptible to ignition but once ignited the explosion violence is no more severe than micron-
scale powders.
Resistivity measurements on nanopowders have demonstrated that with increasing relative

humidity (RH) the resistivity of aluminium, zinc oxide and zirconium oxide decreases. Carbon
nanopowders tended to be insensitive to changes in humidity and were classified as conductive
powders. Aluminium, iron oxide, and zinc oxide were medium resistivity powders but touched
on high resistivity at low humidity levels. There is a tendency for nanopowders to have higher
resistivity values than conventional micron-scale powders. For example the three carbon
nanopowders all had greater resistivity than the micron scale carbon powder. The lower packing
densities of nanopowders and the accompanying voids between the particles may be responsible
for the higher resistivity values.
The conductive nanopowders, such as the carbon nanopowders, are not likely to be an
electrostatic hazard but should these powders penetrate into electric and electronic equipment,
they could give rise to short-circuit problems and may lead to the generation of ignition sources.
The possibility of nanopowders penetrating into electrical and electronic equipment may be
greater due to their reduced particle size.
The following is recommended:

a) Further nanopowders should be tested to broaden the body of knowledge relating to the
explosion characteristics of nanopowders, and to develop further expertise and
knowledge, particularly as new powders are developed.
b) Although powders are initially well dispersed in the 2 litre explosion test chamber, the
rate of re-agglomeration may be such that nanopowders are dispersed only for a very
short duration and rapidly begin to re-agglomerate. Thus, combustion may well take
place initially with dispersed particles but some degree of re-agglomeration of the
particles may take place before combustion has been completed. This would need to be
checked by sampling the dispersed dust cloud and measuring particle size using
61
equipment that measures the size distribution and number concentration of fast dynamic
aerosols. It was not possible to assess the particle size/agglomeration of the dispersed
dust cloud at the point of ignition using the SMPS equipment due to the transient nature
of the dispersed dust cloud. An instrument is therefore required that will capture the
transient particle size distribution. The instrument identified for this work is a Fast
Mobility Particle Sizer Spectrometer. It is recommended that further work is done to
assess the particle size/agglomeration of the dust cloud at the point of ignition using this
instrument.
c) In terms of fire characteristics, as nanopowders grow in use, balance of risk to benefit
may change and further assessment of fire behaviour may be justified. The most
important areas for study are then likely to be a) the potential for a fire to lead to a
release of nanopowders that would subsequently form an explosion hazard of an
environmental risk and b) the potential for nanopowders to form a source of ignition for
a fire in more conventional materials. Perhaps the most important early tests should
therefore be an examination of the effects of transient fires on nanopowder packaging –
is it possible to perforate storage materials without igniting the contents? Also, an
examination of the potential for bulk nanopowders to react with air and form an ignition
source through mechanisms analogous to the ‘spontaneous combustion’ processes that
can occur through oxidation in some other bulk materials.
62
11 APPENDICES
APPENDIX 1 : Commissioning data - Lycopodium sample number EC/026/08

Test Dust Igniter t_open t_close t_ignition Vacuum dP/dt Pmax Comments
run concn. (barg) (bar/s) (barg)
(g/m3) (ms) (ms) (ms)
1 500 Electric 10 40 120 -0.65 100 7.9

fuse head
2 500 Electric 10 40 120 -0.65 170 8.2 Dispersion

nozzle fitted with
fuse head deflector
3 500 Electric 10 55 120 -0.65 - 7.7
4 500 Electric 10 55 105 -0.65 40 4.5

fuse head
4a 500 Electric 10 55 95 -0.65 20 3.6

fuse head
4b 500 Electric 10 55 85 -0.65 - - Dust not

fuse head ignited
4c 500 Electric 10 55 80 -0.65 - - Dust not

fuse head ignited
5 500 Electric 10 55 100 -0.65 166 7.5

fuse head
6 No 1kJ Sobbe 10 55 100 -0.65 - 0.9

dust
6a No Electric 10 55 100 -0.65 - 0.05

dust fuse head
7 1500 Electric 10 55 100 -0.65 - - Rebound

fuse head nozzle. Dust
not ignited
8 1500 Electric 10 55 100 -0.65 200 7.5

fuse head
8a 1250 Electric 10 55 100 -0.65 200 7.7

fuse head
8b 500 1kJ Sobbe 10 55 100 -0.65 Misfire
8c 500 1kJ Sobbe 10 55 100 -0.65 250 7
63
(g/m3) (ms) (ms) (ms)
9 1000 1kJ Sobbe 10 55 100 -0.65 333 8.75
9a 1500 1kJ Sobbe 10 55 100 -0.65 375 8.75
9b 1500 1kJ Sobbe 10 65 100 -0.65 200 7.75
9c 1500 Electric 10 65 90 -0.65 183 6.5

fuse head
9d 1500 1kJ Sobbe 10 65 80 -0.65 666 7.7
9e 1500 1kJ Sobbe 10 65 75 -0.65 - - Misfired

(3 attempts)
9f 1500 1kJ Sobbe 10 65 75 -0.65 750 8.2
10 500 1kJ Sobbe 10 65 75 -0.65 576 8.4
10a 1000 1kJ Sobbe 10 65 75 -0.65 750 8.5
10b 1000 1kJ Sobbe 10 60 75 -0.65 882 8.5
10c 1000 1kJ Sobbe 10 60 70 -0.65 714 8.4
11 Misfires
12 Misfires
13 1000 5kJ Sobbe 10 60 75 -0.65 1300 9
13a 1000 5kJ Sobbe 10 60 75 -0.65 888 9.1
13b 1500 5kJ Sobbe 10 60 75 -0.65 - - Misfire
13c 800 5kJ Sobbe 10 60 75 -0.65 1500
13d 400 5kJ Sobbe 10 60 75 -0.65 833 8.4
13e 555 5kJ Sobbe 10 60 75 -0.65 1500 9.5
13f 300 5kJ Sobbe 10 60 75 -0.65 944 8.6
64
(g/m3) (ms) (ms) (ms)
14 360 5kJ Sobbe 10 60 75 -0.9 666 7.4
14a 400 5kJ Sobbe 10 60 75 -0.9 1100 8.3
14b 625 5kJ Sobbe 10 60 75 -0.9 650 7.3
14c 700 5kJ Sobbe 10 60 78 -0.9 625 7.8
14d 550 5kJ Sobbe 10 60 78 -0.9 875 7.8
14e 550 5kJ Sobbe 10 60 78 -0.9 850 7.4
65
APPENDIX 1 : Commissioning data – Aluminium sample number EC/060/07
(g/m3) (ms) (ms) (ms)
16a 250 5 kJ Sobbe 10 60 78 -0.9 972 6.7
16b 500 5 kJ Sobbe 10 60 78 -0.9 1800 8.6
16c 750 5 kJ Sobbe 10 60 78 -0.9 2100 9.4
16d 1000 5 kJ Sobbe 10 60 78 -0.9 3100 8.9
16e 1250 5 kJ Sobbe 10 60 78 -0.9 2250 8.9
16f 1500 5 kJ Sobbe 10 60 78 -0.9 3700 9.0
16g 1750 5 kJ Sobbe 10 65 83 -0.9 5000 9.5
17a 250 1 kJ Sobbe 10 65 83 -0.9 20 1.7
17b 500 1 kJ Sobbe 10 65 83 -0.9 400 4.7
17c 750 1 kJ Sobbe 10 65 83 -0.9 588 6.7
17d 1000 1 kJ Sobbe 10 65 83 -0.9 375 7.7
17e 1250 1 kJ Sobbe 10 65 83 -0.9 1100 7.7
17f 1400 1 kJ Sobbe 10 65 83 -0.9 1250 9.6
17g 1500 1 kJ Sobbe 10 65 83 -0.9 1400 9.6
17h 1600 1 kJ Sobbe 10 65 83 -0.9 1200 10.5
17i 1750 1 kJ Sobbe 10 65 83 -0.9 1450 10.8
18a 250 2 kJ Sobbe 10 65 83 -0.9 212 4.2
18b 500 2 kJ Sobbe 10 65 83 -0.9 750 6.4
18c 750 2 kJ Sobbe 10 65 83 -0.9 833 6.5
18d 1000 2 kJ Sobbe 10 65 83 -0.9 1666 7.9
18e 1250 2 kJ Sobbe 10 65 83 -0.9 1250 8.0
18f 1500 2 kJ Sobbe 10 65 83 -0.9 1250 8.25
18g 1750 2 kJ Sobbe 10 65 83 -0.9 1950 8.6
66
APPENDIX 1 : Commissioning data – Aluminium sample number EC/060/07
(g/m3) (ms) (ms) (ms)
18h 2000 2 kJ Sobbe 10 65 83 -0.9 1700 8.4
18i 1625 2 kJ Sobbe 10 65 83 -0.9 1300 9.4
18j 1875 2 kJ Sobbe 10 65 83 -0.9 1500 8.6
18k 1812 2 kJ Sobbe 10 65 83 -0.9 1025 8.9
18l 1750 2 kJ Sobbe 10 65 83 -0.9 1000 8.6
19a 1750 2 kJ Sobbe 10 65 100 -0.9 1200 8.9
19b 1750 2 kJ Sobbe 10 65 90 -0.9 1425 9.8
19c 1750 2 kJ Sobbe 10 65 120 -0.9 misfire
19d 1750 2 kJ Sobbe 10 65 120 -0.9 750 9.6 Comparing

tests18g and
19a-d , the
optimum
ignition point
= 83 ms
1750 3 kJ Sobbe 10 65 83 -0.9 1700 9.5

20a
20b 1625 3 kJ Sobbe 10 65 83 -0.9 1800 9.1
21a 1500 1 kJ Sobbe 10 65 83 -0.9 1400 9.1

+ 0.1 Mg
21b 1500 1 kJ Sobbe 10 65 83 -0.9 850 8.4

+ 0.3 Mg
67
12 REFERENCES
BS 5958-1:1991 Code of practice for Control of undesirable static electricity – Part 1: General
considerations.
BS EN 61241-2-2 (1996). Electrical apparatus for use in the presence of combustible dust – Part
2: Test methods – Section 2.2 Method for determining the electrical resistivity of dust in layers.
BS EN 14034-1 (2004). Determination of explosion characteristics of dust clouds – Part 1:

Determination of the maximum explosion pressure ρmax of dust clouds.
BS EN 14034-2 (2006). Determination of explosion characteristics of dust clouds – Part 2:

Determination of the maximum rate of pressure rise (dp/dt)max of dust clouds.
BS EN 13821 (2002). Potentially explosive atmospheres – Explosion prevention and protection

– Determination of minimum ignition energy of dust/air mixtures.
Barton J. (2002). Dust explosion prevention and protection a practical guide. Published by
Institution of Chemical Engineers ISBN 0 85295 410 7.
BIA-Report (1997). Combustion and explosion characteristics of dusts. ISBN 3-88383-469-6
CCPS (2005). Guidelines for safe handling of powders and bulk materials. Centre for Chemical
Process Safety of the American Institute of Chemical Engineers. ISBN 0-8169-0951-2.
Eckhoff R K (1997). Dust explosions in the process industries. ISBN 0 7506 3270 4.
Nanosafe Report (2008). European Strategy for Nanosafety. Dissemination Report February
2008 DR-152-200802-2. Project ID: NMP2-CT-2005-515843
NFPA 77 (2007). Recommended Practice on Static Electricity 2007 Edition. National Fire
Protection Association.
Preining O (1998). The physical nature of very, very small particles and its impact on their
behaviour. J. Aerosol Science., 29, 481-495.
PD CLC/TR 50404 (2003). BSI published document. Electrostatics – Code of practice for the
avoidance of hazards due to static electricity.
Seaton, A., L. Tran, et al. (2009). "Nanoparticles, human health hazard and regulation."
J.R. Soc. Interface doi:10.1098/rsif.2009.0252.focus.
Published by the Health and Safety Executive 02/10

Health and Safety
Executive

of nanopowders
Nanotechnology is a rapidly expanding technology

in which existing and novel materials are engineered
at the nanoscale, typically in the range of 1 to 100
nanometres. Engineered nanomaterials include uniquely
manufactured products with unique shapes and
enhanced physical and chemical properties, compared
with conventional materials of the same composition.
There is currently little available information on the
explosion risks of these materials. The Health and
Safety Executive therefore commissioned this project
to investigate the potential fire and explosion hazards
associated with nanopowders. Test equipment and
procedures were developed to assess the key properties
of a selected number of nanopowders. A specialised
2 litre test vessel was developed to determine the
explosion characteristics and modified standard test
apparatus was used to measure the minimum ignition
energy of nanopowders. Resistivity and electrostatic
charging characteristics were assessed using specially
designed test apparatus. Key information including
KSt, Pmax and MIE values were obtained for a range
of metal and carbon nanopowders. Generally, the
explosibility (maximum explosion pressure, rates of
pressure rise and equivalent KSt) of nanopowders were
found to be broadly similar to conventional micron-scale
powders. However, the minimum ignition energies of
some nanopowders were found to be lower than the
equivalent material at micron-scale. It was demonstrated
that with increasing relative humidity the resistivity
of most nanopowders decreases. There was also a
tendency for nanopowders to have higher resistivity
values than conventional micron-scale powders. All the
powders produced electrostatic charge. Generally, the
charge developed by nanopowders was comparable
with the micron-scale powders.
This report and the work it describes were funded by

the Health and Safety Executive (HSE). Its contents,
including any opinions and/or conclusions expressed,
are those of the authors alone and do not necessarily
reflect HSE policy.
RR782
www.hse.gov.uk

Fire and Explosion Properties PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fire and Explosion Properties PDF

Uploaded by

Copyright:

Available Formats

Health and Safety

Fire and explosion properties

Prepared by the Health and Safety Laboratory

Fire and explosion properties

P Holbrow, M Wall, E Sanderson, D Bennett, W Rattigan,

First published 2010

All rights reserved. No part of this publication may be

Applications for reproduction should be made in writing to:

4 MINIMUM IGNITION ENERGY ................................................................. 10

5 FIRE HAZARDS ....................................................................................... 13

6 ELECTROSTATIC PROPERTIES ............................................................ 15

7 TEST MATERIALS ................................................................................... 19

2. Explosion properties of a range of nanopowders have been characterised including

3. Electrostatic properties of nanopowders have been considered. Dust layer resistivity

REACH is a new European Union regulation concerning the Registration, Evaluation,

3.1 STANDARD TEST

KSt = (dP/dt)maxV1/3 (2.1)

No commercial equipment is available to satisfy these requirements, so a specially designed test

The test vessel incorporates a number of ports and flanged connections:

3. A port incorporating a glass viewing window with a diameter of 40 mm.

4. A dust injection port is located at the bottom of the vessel.

5. A pressure release valve and an evacuation port.

Figure 1: Vessel general arrangement

Figure 3: Test vessel

3.3 CONTROL AND INSTRUMENTATION

Table 1: Pressure transducers

3.4 TEST PROCEDURE

The following is a summary of the general operational procedure.

5. Non-oxidised nanopowder is handled under an inert atmosphere.

1. Prepare the ignition source.

8. Close the fume cupboard sash and initiate the test.

10. Open the outlet valve and release the pressure.

4.1 TEST APPARATUS

Figure 5: MIE test apparatus

Dust dispersion chamber

Development Stage 1: An enclosed barrel shaped chamber was constructed to completely

5.1 FIRE HAZARDS IN NANOPOWDERS

The particular hazards in any fire situation are also threefold:

• heat and smoke that might cause property damage

• growth from a small, initial fire to a larger, more hazardous one.

5.2 PRACTICAL ASSESSMENT OF FIRE PROPERTIES

6.1 ELECTROSTATIC CHARGE

The likelihood of accumulation of electrostatic charges and voltages in an industrial process

In order to measure and demonstrate electrostatic charging of nanopowders in a realistic

Figure 8: charge test apparatus

6.3 RESISTIVITY TEST

Figure 9: Resistivity test cell

Intrinsiq Materials Ltd (previously Qinetiq Nanomaterials Ltd) supplied a quantity of

Bulk density: 0.10 - 0.25 g/cm3

Bulk density: 0.15 - 0.35 g/cm3

Table 3: Micron-scale powders

8.1 COMMISSIONING TESTS – EXPLOSION SEVERITY

Table 4: Commissioning powders

Powder Sample Number Supplier Comment

Lycopodium EC/026/08 Fluka Sieved to 63 micron

Aluminium EC/060/07 Intrinsiq Ref: PQN301 Batch 17

8.1.1 Ignition delay

8.1.2 Ignition source

Ignition source Pmax (barg) dP/dt (bar/s) K (bar m/s)

42 J electric fuse head 6.5 243 66

1 kJ Sobbe 7.3 551 150

5 kJ Sobbe 7.0 620 168

10 kJ Sobbe 7.3 673 183