See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/265888322

Origins and Effects of Potroom Dust

Article in TMS Light Metals · January 2005

CITATIONS READS
5 453

2 authors:

Margaret Hyland Mark Taylor

Victoria University of Wellington University of Auckland
183 PUBLICATIONS 2,919 CITATIONS 120 PUBLICATIONS 957 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Response of Cryolite-based Electrolytes and Side-ledges to Flexible Potline Power Shifts at Smelters View project

Two-phase Flows in an Aluminum Reduction Cell View project

All content following this page was uploaded by Mark Taylor on 10 February 2015.

The user has requested enhancement of the downloaded file.

 Light Metals 2005 Edited by Halvor Kvande TMS (The Minerals, Metals & Materials Society), 2005

Origins and Effects of Potroom Dust

Margaret Hyland and Mark Taylor
Light Metals Research Centre
and
Department of Chemical & Materials Engineering
University of Auckland, Private Bag 92019, Auckland, New Zealand

Keywords: Dust, fluoride, particulate, potrooms, environment

Abstract cell – that is, alumina or anode cover. Such fines can cause
serious upsets to the process and probably contribute directly and
Dust is defined as the fine particulate material that escapes indirectly to dust, but to date there is no clear evidence of the
confinement and causes a nuisance or a hazard in the workplace. relationship. There is no doubt that all three of these issues -
This predominantly fine material, which is enriched in fluoride environmental dust, occupational health dust and process fines -
and sulfur, impacts on the potroom working environment, and as are important and interconnected.
it escapes from the roof as fugitive particulate, on the wider
smelter environment. This paper reviews the current state of Legislation and company imposed limits on dust release are
knowledge of the composition and origins of dust and the factors tightening. The Light Metals Research Centre has recently
that may control its release into the environment. These include completed a Scoping Study on Potroom Dust (P791) that was
the fines content of the materials fed, but are also related to the commissioned by AMIRA on behalf of the Alumina Roadmap
smelter transport and containment systems. The role of alumina as Committee. This paper reviews the current state of understanding
a component of dust and its role in dust generation are of on the composition of dust, and the reasons for concern about its
particular interest. Process operations and cell technology that release. The sources of dust and the factors that may affect its
can affect the nature and mechanisms of dust release are also generation are also considered.
considered.
Potroom Dust Characteristics
Introduction
Surprisingly few studies have been published on the composition
The evolution of understanding regarding potroom dust has been of potroom dust [2-6]. Despite differences in sampling locations,
relatively slow for a number of reasons. The motivations for times and process operations studied in the published literature
understanding dust origins and its effects are many: environmental several common features emerge.
pressure is probably still the most pressing, with process and
mechanical problems caused by dust quite a long way behind this. Dust is made up of >100 micron particles. A notable point is that
Then behind this again is the concern regarding the effect of dust 7% [6] or more of this is the very fine size fractions including the
on the occupational health of people working in the area – some respirable fraction. Potroom dust is composed primarily of
companies have been more proactive in this regard than others. alumina, electrolyte, carbon and a variety of non-process specific
But in general the wearing of personal protective equipment such components such rust, soil and marine aerosols. Analysis in our
as respirators on a voluntary or compulsory basis is still seen as a labs of settled dusts also showed the same basic components,
solution to this last problem, rather than a symptom of it. Figure 1. The relative amounts of these components may differ
depending on what operation was sampled and the cell,
There are various definitions of dust in an industrial context and transportation and containment technologies in use in the smelter.
these are not yet converged to a single methodology – at least as
far as the industry is concerned. By dust, we mean material, A characteristic feature of potroom dust is the presence of very
typically fines, which escapes confinement and causes a nuisance fine material that often decorates alumina and crushed bath
to the environment and/or people. The ASTM definition is helpful particles (see Figure 1) or appears as agglomerates. This material,
because it recognises that dust may be intermittently airborne, which may have individual particle sizes on the 10-100 nanometer
“...solid particles predominantly larger than colloidal and capable scale [6], is derived from condensed electrolyte vapours, carried
of temporary suspension in air or other gases” [1]. Fugitive with draft air - pot fume - and is a significant source of fluoride.
particulate material in the potroom may settle on surfaces, but has
the potential to become airborne through air, machine or people Pot fume is a critical component of potroom dust. In addition to
movement. There is no guarantee either that the dust which is condensed bath, it will carry with it some alumina, anode cover
important from a potroom roof emission point of view is the same and carbon. It will have at least 50% of the particles below 20
dust which appears to be associated with occupational health microns [4, 7] and is enriched in F, S and other impurities such as
concerns – and specifically respiratory health - for the people in Fe and P. The fluorides are present in four forms [4, 8,9]:
the work area. • Condensed NaAlF4 (a metastable phase)
• Chiolite, Na5Al3F14 from condensed vapour
It is useful here to make an arbitrary distinction between ‘dust’ • Na3AlF6, as entrained bath droplets or anode cover
and ‘fines’. In the smelting context, ‘fines’ are considered to be • Aluminium hydroxyfluorides - surface reacted fluoride
the -45 micron or -20 micron fraction of particulate inputs to the on secondary alumina, AlFx(OH)3-x.yH2O.

141
Crushed Bath Carbon Table 1 summarises the fluoride emissions for two plants
employing prebake, point feed technologies. The particulate roof
emissions – potroom dust – make up one-third of the total fluoride
emissions from the plants and about 70% of the total particulate
fluoride emissions.

Alumina The contribution of potroom dust to roof particulate fluorides is

becoming a more critical issue. At the same time that capacity
creep is resulting in increased total fluoride emissions, regulatory
authorities are tightening the allowable emission levels.
Recommended European limits for prebake technologies are
proposed to be 1.0 kg/T Al total F and 0.5kg/T Al HF in 2007 and
by 2010 reducing to 0.6 kg/T Al and 0.4 kg/T Al for total fluoride
and HF, respectively [10].
Bath Droplet
A similar picture is seen for the contribution of fugitive potroom
dust as a source of total particulates. Fifty percent or more of the
Alumina + Fume Cover Bath + Fume total particulate emission comes from fugitive roof emissions.
Typical emissions for prebake, point feed technologies are 0.5
Figure 1: Settled dust from roof louvers, showing some kg/T Al from the stack and 0.8 kg/T Al from the roof [11].
typical components. European limits on total particulate levels (dust) for prebake
technologies are proposed to halve from 2 kg/T Al in 2007 to 1
kg/T Al in 2010 [10].
Alumina is consistently listed as a component of potroom dust.
There are a number of possible sources of the alumina component. From an occupational health point of view, the dust is of concern
• Cell feed that escapes prior to reaching the cell, during once it becomes airborne. Particle size is important, as it
transportation or storage in bins; determines the extent of penetration into the respiratory tract. The
• Alumina that is carried out with rising hot fume as it is inhalable fraction, that can be inhaled by nose or mouth typically
being fed; has an aerodynamic diameter <100 micron. The respirable
fraction is particularly problematic having an aerodynamic
• Alumina as a component of cell cover, lost during
diameter of <10 micron, small enough to lodge in the gas-
transport, delivery to the cell or during cell tending
exchange region of the lung. Smelters will typically monitor the
operations.
total particulate loading and the sub 10 micron loading as part of
Aluminas from these sources can be distinguished from each other
their occupational hygiene monitoring. The respirable fraction is
by careful analysis of the particle morphology and chemistry. For
also the fraction that will carry high loadings of fluoride and
example, alumina in cell cover will have a higher α-alumina
sulfur. The question of the cause-effect relationship between
content, being partially calcined, or may have adherent angular
dust and respiratory illness is still a matter of debate, but at least
crushed bath particles on the surface.
two of the suspected agents are present in the respirable fraction:
fluorides [12,13] and ultrafine or nano-particles [5,6,14,15].
Bath components in potroom dust, in addition to the condensed
and entrained bath vapours associated with pot fume, will be
sourced from cell cover – as crushed bath recovered from anode
Dust Generation
butts. Its distinctive shape, angular particles, mean it is easily
distinguished from fume, which is very fine, usually aggregated
The origin of the dust is not always obvious. There are at least
spherical particles coating other particles.
three routes for dust release:
1. Spillage of alumina, usually after it has been through the dry
Environmental and Health Significance of Potroom Dust
scrubbing system and has absorbed fluoride from the pot
fume. This spillage occurs in the handling systems –
Potroom dust becomes an environmental concern when it is
conveyers, bins, ore hoppers, etc. – which bring the alumina
transported to the potroom roof and escapes as fugitive roof
to the cells.
emissions. In this way it contributes significantly to the smelter
2. Spillage of fine bath cover material, which is ground to a size
total particulate emission and the total F emission.
similar to fine alumina by some of the bath crushing systems
in use in modern smelters (apparently because it is then
Table 1: Fluoride emissions in kgF/Tonne Al for two prebake,
easier to handle in dense phase lines);
point feed technologies.
3. Leakage of pot fume from inside the cell cavity, due to
Total F ROOF EMISSIONS STACK disturbances in pot operation, or operations such as cell start-
emission, EMISSIONS up, anode replacement or bath tapping.
kgF/T Al F, F, F, F,
Particulate Gas Particulate Gas The release of pot fume into the operator’s environment generally
Plant 1 0.54 0.18 (34%) 0.19 0.08 0.09 coincides with substantial heat release as well - and occurs when a
Plant 2 0.71 0.21 (30%) 0.23 0.08 0.19 cell’s air suction is overwhelmed by the thermal convection of hot
fume upward and out of the cell. Typically this occurs when bath
or hot anode surfaces become exposed for some reason. An

142
extreme instance of this is the start-up of a pre-bake cell. When
operators remove resistor coke used during the bakeout of the cell,
they will be exposed to large volumes of pot fume. These
practices are still common although mechanisation sometimes
removes the operator from the most highly concentrated area of
fume release.

More commonly, pot fume will be released during more routine

operations, such as anode changing, metal tapping or cell
dressing, or at any time when hoods do not fully seal.
Measurements of particulate emissions peak during anode change
[5], as fume is released from the unhooded cell and from the
removed anode butt.

The second major source of dust - bath cover spillage - is far too
common, given the containment systems which now exist within
point fed smelters. Bath spillage occurs primarily at anode setting
when the anodes are covered with crushed bath. However, the cell Figure 3: Fine alumina escape during cell hopper filling from
dressing operation and anode/butt movements also spread bath overhead bins, themselves filled from a Day Bin with segregation
dust into the potroom environment. In modern smelters, the dust of fines possible.
which collects on horizontal surfaces everywhere is often sourced
predominantly from the crushed bath stream into the potline, and obvious dust source. Figure 3 shows fine alumina escaping from
for the existing handling systems the cause of this problem is an overhead ore bin. Spilled alumina is found both onto the top of
the bin and onto the superstructure of a cell. The alumina fines
shown in Figure 2.
are often implicated both as the cause of the spillage and as the
As the data in Figure 2 show, there is no coarse fraction in this problem component of the dust, but despite many operational
crushed bath. When transferred from a bin to any other container, investigations of these events, at this point the evidence is still
circumstantial. Traditionally the alumina ‘fines’ have been
this material very often fluidizes although the onset is
defined as the -45 micron fraction, as measured by a screen or a
unpredictable. Loss of control over the flow properties then
occurs, along with spillage onto deckplates, hoods, pits, roadways, laser sizer – making up roughly 6-8 wt% of the alumina.
anode beams and into the heated air flume above the cells. The
However, this fairly coarse material is not generally what arrives
consequences of fine crushed bath are also evident in the
at the cell – because of the impact of dense phase conveying and
behaviour inside the cells – causing over-covering of anodes and
collapse of cover in certain circumstances. This situation causes also the dry scrubbing system on the alumina particles, and
exposure of bath and immediate fume release into the cell. It is because of the other condensed fume particulates which are
also likely that the heat of the fume released will overwhelm the caught in the bags and recycled along with the alumina. The
amount of -45 micron fraction can increase by two or more [16],
fume containment system of the cell, releasing fume into the
environment. often where significant dense phase conveying installations exist.

Finally, spillage of alumina during transport or feeding is another There are now quite a number of “events” recorded where an
increase in the -45 micron content of secondary alumina has
caused a process problem – particularly with alumina feeding. A

Figure 2: Particle Size Distribution for Crushed Bath produced Figure 4: The tell tale signs of a Non-Breaker within a 170 kA
from an Autogenous Mill, before blending with a proportion of cell. The alumina heaped in the centre channel due to unsuccessful
feeds has been raked onto the floor for removal.
alumina.

143
common occurrence in some potlines is shown in Figure 4. This is the mechanisms discussed here, the generalisation of dust sources
a ‘nonbreaker’ – where the breaking mechanism preceding from different potrooms and technologies, and the ways in which
alumina addition has failed to break the crust. The frequency of potroom dust can be minimised.
non-breakers increases markedly if the alumina feeding from the
cell hoppers becomes variable or uncontrollable. The subsequent
impact inside the cell is severe because the alumina fed does not References
reach the bath, and anode effects followed by other unsuccessful
feeding events soon follow. The net result of 20-30 tonnes of fine 1. ASTM American Society for the Testing of Materials D
secondary alumina is that a whole section of cells can experience 1356-91a, “Standard Terminology Relating to
a temperature increase of 10-20˚C within hours of cell hopper Atmospheric Sampling and Analysis.’
filling.
2. Kurthy-Komlosi, J., et. al, ‘Investigation of settling and
It is not clear that the loss of control of the flow of alumina from flying dusts collected implants of aluminium industry
the cell hoppers in Figure 4 is due to the – 45 micron fraction aiming at environment and labour protection.’ Magyar
itself, however. The proportion of even finer material in this Aluminium 1990, 27, 10, 360-372.
fraction could be implicated.
3. Kaufmann, G. and Mannweiler, U. ‘The characterisation
of particulate samples collected in the working
A Framework for Understanding Potroom Dust Generation environment of pot workers.’ Light Metals 1981, 1011-
1019.
The picture which emerges from this discussion is a complex
interplay involving a number of dust sources and a variety of 4. Roach, G. and J. Cornell, ‘Dust and dustiness testing of
mechanisms for dust release, both being affected by smelter smelter-grade alumina.’ Proceedings of the 2nd
technology and practices and the materials properties of the dust International Alumina Quality Workshop, Oct 14-19,
sources. This relationship is illustrated in the schematic in Figure 1990, Perth, Australia, 247-262.
5.
5. Thomassen, Y. et al, ‘Are we really monitoring the
The sources of potroom dust, shown on the bottom tier in Figure 5 health relevant contaminants in workroom atmospheres?
and the relevant materials’ properties contributing to dust Experiences from Al-smelters and suggestions for
monitoring strategies.’ Australian Institute of
Occupational Hygienists (AIOH) 2001 Annual
AIRBORNE POTROOM DUST Conference, Wollongong, 2001, 212-218.
Alumina Bath Rust, etc
Carbon 6. Heinrich, U. et al, ‘Characterisation of the exposure to
ultrafine particles at workplaces of a primary aluminium
MECHANISMS SOURCE MATERIALS’ smelter : Final Report.’ Fraunhofer ITEM Project No
Spillage in transport, Re-suspension of PROPERTIES 112 062, May 2004.
spilt material by Fines
handling
transport vehicles Aeratability content
Entrainment of loose Entrainment of spilt Flowability
7. Less, L.N. and J. Waddington, ‘The characterisation of
Moisture
material and fines in material by potroom air Attrition content
aluminium reduction cell fume.’ Light Metals 1971,
cell gases
GY SMELTE 499-508.
T ECHNOLO R PRACTIC
SMELTER ES
8. Heiberg, A., G. Wedde, O.K. Bøckman and S-O
SOURCES Strømmen, ‘Pot gas fume as a source of HF emission
Condensed
Electrolyte Alumina Cell form aluminium smelters – laboratory and field
Anode Cover
Vapours and Material Feed investigations.’ Light Metals 1999, 255-261.
droplets Spent Miscellaneous
Anodes
9. Gillespie, A., M. Hyland and J. Metson, “Irreversible
HF adsorption in the dry scrubbing process.” Journal of
Figure 5: Schematic showing the relationship between the Metals 51 (5), 30-32, 1999.
sources of potroom dusts, the mechanism of generation and the
source materials properties. 10. Nordheim, Eirik ‘Environmental regulations and
performance for European smelters.’ Light Metals 2005,
generation will be common to most smelting operations. The this volume.
relative contributions will depend on the cell technology and
practices in a given potroom and on the smelter infrastructure in 11. Wedde, G., 18th International Course on Process
terms of attrition of material in handling systems and fine grinding Metallurgy of Aluminium Ed Harald Oye, NTNU,
in crushing equipment second tier in Figure 5). How the potline Trondheim, Norway, June 1999.
technologies interact with the properties, such as particle size
distribution, aeratability, attritablility, flow and moisture content, 12. V. Søyseth, J. Kongerud, ‘Prevalence of respiratory
of the material fed to the cell is also important. These principles disorders among aluminium potroom workers in relation
will be employed in the follow on to the Potroom Dust Scoping to exposure to fluoride.’ Br J Ind Med; 49: 1992, 125-
Study. This will be a full plant based study to examine some of 130.

144
 13. K. Lund, J. Boe, M. Refsnes, P. Søstrand, P. Schwarze,
T. Sandström, ’Experimental hydrogen fluoride
exposure induces airway inflammation in normal
subjects.’ Am J Respir Crit Care Med; 153(2): 1996,
A484.

14. Brouwer, D., J. Gijsbergs and M. Lurvink, ‘Personal

exposure to ultrafine particles in the workplace:
Exploring sampling techniques and strategies.’ Ann.
Occup. Hyg., 48(5), 439-453, 2004.

15. Vincent, J., ‘Characterisation of coarse, fine, very fine

and ultrafine workplace aerosols.’ Australian Institute
of Occupational Hygienists (AIOH) 2001 Annual
Conference, Wollongong, 168-177.

16. Lindsay, S., ‘Bridging the gap between refiners and

smelters.’ 8th Australasian Aluminium Smelting
Technology Conference and Workshops, Yeppoon,
Australia, Oct 2004, 152-162.

145

View publication stats