Int. J. Miner. Process.

93 (2009) 1–5

Contents lists available at ScienceDirect

Int. J. Miner. Process.

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j m i n p r o

Recovery of valuable materials from aluminium salt cakes

W.J. Bruckard ⁎, J.T. Woodcock
CSIRO Minerals, Box 312, Clayton South, Vic. 3169, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Salt cakes, which are nominally waste products derived from aluminium dross melting furnaces, are complex
Received 17 February 2009 mixtures of some 20 different compounds made up of many different elements. Normally they are regarded
Received in revised form 29 April 2009 as waste products and they are disposed of in toxic waste dumps. However, it is shown here that some
Accepted 13 May 2009
components are readily recoverable as high-grade products for recycling or sale and that the residues thus
Available online 19 May 2009
generated can be non-toxic. Recoverable components include metallic aluminium, salt (halite (NaCl) and
sylvite (KCl)), alumina-containing compounds, and possibly hydrogen gas.
Keywords:
Salt cake
Metallic aluminium is soft and malleable and is not reduced in size by crushing and grinding, whereas the
Aluminium other components in salt cake are soft and brittle or are readily dissolved in water. Hence the coarse metallic
Waste processing aluminium can be readily recovered by crushing and screening and the finer metallic aluminium can be
recovered by fine grinding and screening, froth flotation, or possibly electrostatic separation. Aqueous acid or
alkaline leaching has also been proposed to recover aluminium metal from salt cake.
The halite and sylvite are easily extracted by leaching ground salt cake with cold water and filtering off the
saline solution. This solution can be sent to solar evaporation ponds where the water is evaporated and the
dry salt harvested for recycling to dross treatment furnaces or other markets.
Some of the water-insoluble or oxide aluminium compounds present are soluble in Bayer-type leach
solutions and could possibly be sent to a Bayer-type leach plant for production of high-grade alumina for
aluminium production. Alternatively, because the oxide aluminium compounds are inert they could be sent
to a non-toxic dump. The possibility of integrated flowsheets to recover more than one product in a single
plant is also discussed.
Crown Copyright © 2009 Published by Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. General characterisation of salt cakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. Recovery of sodium and potassium chlorides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4. Recovery of metallic aluminium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.1. Aluminium metal recovery by screening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
4.2. Aluminium metal recovery by flotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.3. Aluminium metal recovery by electrostatic separation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.4. Aluminium metal extraction by aqueous acid or alkaline leaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
5. Recovery of oxide aluminium compounds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
6. Hydrogen gas recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
7. Decomposition of reactive compounds in salt cakes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
8. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1. Introduction

Aluminium salt cakes are solid waste products derived from the
⁎ Corresponding author. Tel.: +61 3 9545 8500; fax: + 61 3 9562 8919. smelting of aluminium drosses under a salt cover to recover some of
E-mail address: [email protected] (W.J. Bruckard). the metallic aluminium entrained in the dross. However, the solidified

0301-7516/$ – see front matter. Crown Copyright © 2009 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.minpro.2009.05.002
2 W.J. Bruckard, J.T. Woodcock / Int. J. Miner. Process. 93 (2009) 1–5

salt cake itself contains metallic aluminium as well as many other Table 2
substances, some of which have an intrinsic value, some of which are Alphabetical list of phases identified in typical Australian salt cakes by qualitative XRD
and other methods.a
toxic, and some of which react with water to give off toxic gases.
Whilst a number of workers have considered the recovery of valu- Phase Formula Relative abundanceb
ables, mostly metallic aluminium, from aluminium dross, e.g., Zhang Aluminium calcium AlCa bDL
(2006), Peterson and Newton (2002), Schlesinger (2000), and Bruckard Aluminium metal Al Trace
Aluminium nitride (25–1133)c AlN Medium
et al. (2004), few studies have been conducted on the salt cakes derived
Aluminium oxide Al2O3 bDL – trace
from them. Those that have, e.g., Bahr and Kues (1978) and Gil (2005, Aluminium oxide nitride (48–1581)c,d Al5O6N bDL – trace
2007), have noted the complexity of salt cakes and the relatively low Aluminium oxide nitride (48–686)c,d Al5O6N bDL – trace
tonnages produced in most areas. These two factors have contributed Aluminium nitride (00-46-1200)c AlN Medium
to the slow progress in developing economic treatment routes for the Bayerite Al(OH)3 bDL – minor
Corundum Al2O3 Major
re-processing of aluminium salt cakes. There are, however, several Cryolite Na3AlF6 Trace – minor
commercial operations treating salt cake or high-salt aluminium dross Diaoyudaoite NaAl11O17 Minor
(for example JBM International Ltd, UK; Engite Technologies, Italy; Elpasolite K2NaAlF6 Trace – medium
Alustockach, Germany; KVS-Ekodivize, Czech Republic; Alumitech Inc, Fluorite CaF2 bDL
Gibbsite Al(OH)3 bDL
USA), with each operation using different technologies to cope with the
Halite NaCl Major
different feed stocks. Iron metal Fe bDL
It is the main purpose of this paper to review and discuss the various Silicon metal Si bDL
ways in which some of the more valuable materials can be recovered Sodalite Na6[Al6Si6O24].2NaCl bDL
from salt cakes. It may be possible to recover more than one material in a Sylvite KCl Minor – medium
Villiaumite NaF bDL
single processing plant. These materials include metallic aluminium,
a
sodium and potassium chlorides, and an alumina-containing solid or Other methods included SEM examination, electron probe micro analysis (EPMA),
chemical analysis by various methods, and hand sorting.
solution which could be suitable as Bayer process plant feed. In addition, b
The decreasing order of relative abundance is as follows: major, medium, minor,
leach conditions which decrease the toxicity caused by compounds such trace, and bDL, where bDL was below the detection limit by the XRD conditions used
as aluminium nitrides are proposed so that final residues can be safely (say 3–5%). This phase was detected by one or more of the other analytical methods.
c
sent to a non-toxic waste dump. Bracketed numbers are the International Center for Diffraction Data (ICDD) file
numbers.
d
These phases have the same composition but different crystal forms.
2. General characterisation of salt cakes

Salt cakes contain many different components, and most of the scanning electron microscope (SEM) examination, electron probe micro
elements present are listed in Table 1 for two separate Australian analysis (EPMA), various chemical techniques, and hand sorting of the
samples. These components have been determined by many different coarser size fractions of crushed salt cake to remove relatively coarse
analytical techniques as discussed by various investigators such as pieces of metallic aluminium. Of particular note is the method of
Bruckard and Woodcock (2004, 2007), Hagni (2002), and Manfredi et al. determination of the fine-sized metallic aluminium by cold bromine
(1997). Although the two samples are broadly similar in composition methanol extraction followed by analysis of the extract solution.
there are many subtle differences in detail. These differences presumably It is known that particles of metallic aluminium up to about 2.5 mm
result from differences in feedstocks and operating practices. in size can occur in salt cakes (Bruckard and Woodcock, 2007) and that
Table 2 lists the various phases identified in the two Australian particles down to 38 μm can occur in aluminium drosses from which
salt cakes studied in this work using qualitative X-ray diffraction (XRD), salt cakes are produced (Bruckard et al., 2004). Table 3 gives an
estimate of the contribution of the various aluminium-bearing phases
present in the two salt cake samples to the total aluminium assay.
Table 1 These results together with those in Table 1 indicate that between
Chemical analyses of two Australian salt cakes.
1% and 3% of the feed is possibly recoverable as metallic aluminium.
Element Sample A Sample B
Major elements (N 1%)
Al (total) 37.2 36.8 Table 3
Al (metallic) 1.22 2.79 Estimated contribution of the various aluminium-bearing phases present in the salt
Cl 9.39 6.79 cakes to the total aluminium assay.
Na 8.52 5.20
N 7.53 1.96 Aluminium-bearing phase Sample A Sample B
F 5.50 5.50 (% Al in feed) (% Al in feed)
K 3.18 3.74 + 20 mm metallic Ala 0.50 0.88
Mg 2.59 0.70 − 20 + 0.85 mm metallic Ala 0.36 0.09
Si 2.07 1.03 − 0.85 mm metallic Ala 0.36 1.82
Fe 0.82 5.85 Total metallic Alb 1.22 2.79
Carbide aluminiumc 1.59 1.21
Intermediate elements (0.1–1.0%) Nitride aluminiumd 4.8 1.26
Ca 0.72 0.50 Water-soluble aluminiume 0.28 0.30
C 0.55 0.42 Other aluminiumf 29.3 31.24
Ti 0.13 0.50
Total aluminium in feed 37.2 36.8
Minor elements (b 0.1%) a
Determined by direct analysis of the screen size fraction.
Zn 0.089 0.012 b
Sum of aluminium in the three coarsest size fractions.
P 0.065 0.020 c
Calculated by assuming that all of the carbon present is there as aluminium carbide
Mn 0.057 0.047
(Al4C3).
Ba 0.017 0.018 d
Calculated by assuming that all the nitrogen present is there as aluminium nitride
S 0.020 0.020
(AlN), but some aluminium oxide nitride (Al5O6N) is also present.
Pb 0.019 0.024 e
Present in an unknown form, but calculated as the Al2O3 equivalent.
V 0.013 b DLa f
Calculated by the difference between the aluminium accounted for here and total
a
Less than the detection limit by the method used. aluminium content of the feed.
 W.J. Bruckard, J.T. Woodcock / Int. J. Miner. Process. 93 (2009) 1–5 3

Table 3 also shows that reasonably high levels of aluminium carbide In this circuit, cold salt cake from the melting furnace is first
(Al4C3), aluminium nitride (AlN), and aluminium oxide nitride (Al5O6N) crushed in closed circuit with a double deck screen fitted with 20 mm
are present. These and other compounds present can react with water and 10 mm aperture screens. The screen sizes here are nominal and
to give off noxious gases such as hydrogen, ammonia, methane, and may be varied to suit the particular salt cake feed (it is proposed that
gaseous sulphides and phosphides, and this must be taken into account material less than 10 mm may be a suitable size to wet mill for salt
in any proposed wet treatment process as discussed later. cake). Any + 20 mm material is mostly aluminium metal with some
Very little water-soluble aluminium is present, and this can probably entrained salt cake and can be recycled to the remelting furnace. The
be ignored in any treatment process. However, as shown in Table 2, −20 + 10 mm material is returned to the crusher for crushing to
appreciable levels of water-soluble halite (NaCl) and sylvite (KCl) are −10 mm. The −10 mm material is then wet ground in an open circuit
present and these could be relatively easily extracted by water washing tube mill with the aim of flattening any metallic aluminium particles
and recovered by solar evaporation for recycling to the dross smelting present, so that they are more readily screened out, and grinding the
furnace. non-metallic particles, which are relatively brittle, to a fine size.
Reasonably high levels of aluminium oxide, bayerite, corundum, Simultaneously the water-soluble particles are dissolved. The ground
and gibbsite were present. Whilst corundum is not soluble in strongly
alkaline solution, some of the other phases are, and these could
potentially be recovered for feeding to a Bayer process plant (Davies
et al., 2008).
With respect to magnesium, data from Table 1 indicate one of the
salt cake samples contained 2.59% Mg. No magnesium-bearing phases
were identified by XRD, but in general magnesium in salt cakes has
been known to be present as oxide phases or alloyed with aluminium.
On the basis of the potential recoverable products outlined above,
it seemed that it was worth reviewing the various methods suggested
for recovering these and other compounds from salt cakes and this is
done below.

3. Recovery of sodium and potassium chlorides

Recovery of sodium and potassium chlorides from salt cakes for

recycling back to the smelting process could be economically advan-
tageous and would reduce the amount to be disposed of in toxic waste
dumps. Bruckard and Woodcock (2007) have shown that simple water
washing of −2 mm crushed salt cake lowers the level of these
chlorides to below the level of detection by XRD. Whilst the chlorides
are efficiently removed by water washing, the process does not re-
move all of the sodium and potassium because these elements are
also present in other, non-water-soluble compounds. No other water-
soluble compounds are known to be present.
It should then be possible to recover the pure dissolved salts
by solar evaporation as practised commercially in many operations
throughout the world (e.g., Gilbert (1993) and Garrett (2001)).
The solids residue from the water leach stage may be suitable
for disposal in a non-toxic dump, depending on the absolute levels of
elements such as fluorine in the residue. Appropriate TCLP tests would
need to be conducted to validate this.

4. Recovery of metallic aluminium

Methods that can be used for the recovery or extraction of metallic

aluminium from a salt cake include screening, flotation, electrostatic
separation, and dissolution in strong alkali. These methods are dis-
cussed briefly in turn below.

4.1. Aluminium metal recovery by screening

As shown in Table 3, aluminium salt cakes contain metallic alu-

minium pellets ranging in size up to at least 2 cm in diameter and
down to an unknown, but very fine size. Because aluminium is a soft,
malleable metal it cannot be crushed or ground to a finer size. For-
tunately, the bulk of the salt cake itself is brittle and is readily crushed
and ground to liberate the metal so that most of it can be recovered by
screening. The sodium and potassium chlorides present are readily
soluble in water and their dissolution can also help liberate metallic
aluminium. A circuit to accomplish this is proposed by the authors and
is shown in Fig. 1. The flowsheet is a good example of recovering two
products in one unit operation. Fig. 1. Proposed flow sheet for recovering valuable materials from aluminium salt cakes.
4 W.J. Bruckard, J.T. Woodcock / Int. J. Miner. Process. 93 (2009) 1–5

product is fed to a DSM (Dutch State Mines) screen having 150 μm as discussed later, and could be worth further investigation. It would
wide slots. According to Pryor (1965) this is about the finest size that be necessary, of course, to first dissolve and remove the chlorides and
can be used commercially for wet screening. The +150 μm material some other elements before conducting an alkaline leach because
is mainly aluminium metal and can be dried and sent back to the these elements are not wanted in Bayer-type solutions.
melting furnace. The − 150 μm pulp goes to a salt recovery circuit
where solar evaporation is used to recover the salt as discussed later 5. Recovery of oxide aluminium compounds
in this paper.
It should also be noted that the grinding behaviour of aluminium Davies et al. (2008) studied the treatment of salt cakes by Bayer-
metal will depend on its content of alloying elements. It is likely, for type leaching and found that about 43% of the total aluminium present
example, that an aluminium-silicon alloy would be more brittle in was dissolved in a typical two-stage process with a leach at 100 °C in
comparison to pure aluminium. The grinding behaviour of the metallics the first stage and 140 °C in the second stage. Evidently, much of the
would also be influenced by the time of grinding and the type of grinding aluminium is present as phases which are not soluble in Bayer-type
device used. solutions. This was examined in detail by Davies et al. (2008), and are
not discussed here.
4.2. Aluminium metal recovery by flotation Although the amount extracted may be worth recovering in a
combined process as shown in Fig. 1, it is also possible that the tonnage
Only a few metals occur in nature in metallic form. These include may be too small to be of interest to bauxite mining and treatment
gold, silver, platinum, and more rarely mercury and copper. Metallic companies, which commonly treat several million tonnes per year.
aluminium does not occur in nature and so no work has been published
on its flotation from ores. Furthermore, no work has been published on 6. Hydrogen gas recovery
its flotation from salt cakes. However, Soto and Toguri (1986) and
Bruckard and Woodcock (2004) conducted work on floating aluminium It may be possible to produce hydrogen gas from salt cake using
from aluminium dross. Soto and Toguri used a Hallimond tube initially, a complex process analogous to that developed by Owada et al. (2008)
and subsequently a Wemco laboratory cell. Bruckard and Woodcock for producing “green hydrogen” from aluminium dross. They proposed
used a 3 L Denver laboratory flotation cell and obtained similar results to that hydrogen could be generated from the reaction of metallic
those of Soto and Toguri. aluminium and an alkaline aqueous solution according to Eq. (3).
It was found that metallic aluminium did not float with xanthate-
type collectors alone, but when copper sulphate was added the alu- 2Al þ 6H2 O→2AlðOHÞ3 þ 3H2 ð3Þ
minium was activated and floated readily. The activation process in
this instance was in fact copper cementation on the aluminium metal
The hydrogen could be used as a fuel and/or feed to fuel cells and the
surfaces. Best results were obtained on particles in the size range of
aluminium hydroxide sold in the market. Furthermore, the reaction is
110–275 μm, and good recoveries in good grade concentrates were
exothermic and the heat of reaction could be utilized for drying the
achieved.
aluminium hydroxide product. Whether or not such a process could be
These results need to be confirmed on salt cakes, but it seems
economic is unknown. Certainly there are numerous methods currently
reasonable to expect that they would be. Coarse aluminium that does
available for commercial hydrogen production (Anon, 1995), e.g. steam
not float could be recovered by screening, and, as noted previously,
reforming of natural gas. Moreover, as shown elsewhere in this paper,
dissolved halite and sylvite could be recovered by solar evaporation.
when salt cakes are mixed with water other gases are given off and these
may contaminate any hydrogen produced.
4.3. Aluminium metal recovery by electrostatic separation
7. Decomposition of reactive compounds in salt cakes
Electrostatic recovery of metallic aluminium from dross has been
studied by Mah et al. (1986) who developed a circuit which involved
Various compounds in salt cakes react with water in the atmo-
first crushing the dross to −840 μm. They achieved a recovery of 70% of
sphere, and especially in aqueous pulps, to give off ammonia and other
the metallic aluminium at a concentrate grade of 70% Al. However, dross
noxious gases. These gases need to be properly vented from any
usually contains much more aluminium than salt cake and at a much
treatment plant. Only one venting point is shown for example in Fig. 1,
greater size, partly because it also often contains scrap aluminium, so it
but a number may be required in practice.
was relatively easy to achieve a good result. It remains to be shown if a
Aluminium nitrides react with water according to Eqs. (4) and (5)
similar result could be obtained on salt cake, but it could be worthwhile
to give either ammonium hydroxide or ammonia. Eq. (4) is more likely
investigating this possibility.
to occur in more alkaline solutions (say above pH 8) and Eq. (5) is
more likely to occur below pH 8.
4.4. Aluminium metal extraction by aqueous acid or alkaline leaching
AlN þ 4H2 O→AlðOHÞ3 þ NH4 OH ð4Þ
As Habashi (1997) has pointed out, aluminium metal is amphoteric
and can dissolve in acid solutions (Eq. (1)) or alkaline solutions
AlN þ 3H2 O→AlðOHÞ3 þ NH3 ð5Þ
(Eq. (2)) with the generation of complex aluminium-containing ions
and the generation of hydrogen gas.
Aluminium oxide nitride (Al5O6N) reacts with water according to
þ 3þ
2Al þ 6H3 O þ 6H2 O→2½AlðOHÞ6  þ 9H2 ð1Þ Eq. (6) to give aluminium oxide and ammonia.
− −
2Al þ 2OH þ 6H2 O→2½AlðOHÞ4  þ 3H2 ð2Þ 2Al5 O6 N þ 3H2 O→5Al2 O3 þ 2NH3 ð6Þ

Either acid or alkaline leaching would also dissolve many of the

Aluminium carbide reacts with water according to Eq. (7) to give
other oxide aluminium compounds present in salt cakes. Recovery of
aluminium oxide and methane gas.
aluminium in metallic form from acid solutions is not easy, where-
as with alkaline leaching it may be that the solution is suitable for
feeding directly to a Bayer-type process plant. This might be viable, Al4 C3 þ 6H2 O→2Al2 O3 þ 3CH4 ð7Þ
 W.J. Bruckard, J.T. Woodcock / Int. J. Miner. Process. 93 (2009) 1–5 5

Beckman (1991) has reported that gaseous sulphides and phos- References
phides can be generated from some German salt cakes, and these Anon, 1995. Hydrogen production, 4th Edition. Kirk–Othmer Encyclopedia of Chemical
would need special care in their removal. Technology, vol. 13. John Wiley & Sons, New York, p. 852.
Based on the ratio of aluminium nitride to metallic aluminium in Bahr, A., Kues, J.,1978. Processing of salt slags from aluminium remelting plants. In: Jones, M.
J. (Ed.), Complex Metallurgy '78. In The Institution of Mining and Metallurgy, London,
the salt cakes considered in this study, one would expect about 5 times pp. 134–143.
more ammonia to be generated via Eqs. (4) to (7), than hydrogen via Beckman, M., 1991. Slag reclamation in the state of Nordheim-Westfalen, Aluminium-
Eq. (3). 67. Jahrgang 1981-6, pp. 586, 589–593.
Bruckard, W.J., Woodcock, J.T., 2004. Characterisation of metal-containing waste products
in relation to retreatment methods for metal recovery and recycling. Green Processing
8. Conclusion 2004. In The Australasian Institute of Mining and Metallurgy, Melbourne, pp. 217–224.
Bruckard, W.J., Woodcock, J.T., 2007. Characterisation and treatment of Australian salt
Methods proposed and tested for recovering valuable materials from cakes by aqueous leaching. Minerals Engineering 20, 1376–1390.
Bruckard, W.J., Walta, P., Woodcock, J.T., 2004. The recovery of aluminium metal from
salt cakes, waste products derived from the treatment of aluminium primary and secondary aluminium drosses by wet grinding and screening. In:
dross, are presented and discussed. Whilst the chemical and physical Tabereaux, A.T. (Ed.), Light Metals 2004. In The Minerals, Metals, & Materials Society,
complexity of salt cakes make re-processing or recycling challenging, Warrendale, pp. 1203–1208.
Davies, M., Smith, P., Bruckard, W.J., Woodcock, J.T., 2008. Treatment of salt cakes by
many workers have shown that valuable components can be extracted aqueous leaching and Bayer-type digestion. Minerals Engineering 21, 605–612.
in simple flowsheets which have both technical and possibly economic Garrett, D.E., 2001. By-products recovery from solar operations. In: Coogan, A.H.,
merit. Hauber, L. (Eds.), Fifth Symposium on Salt, vol. 2. The Northern Ohio Geological
Society, Inc, Cleveland, pp. 281–293.
Coarse metallic aluminium can be recovered by crushing and Gil, A., 2005. Management of the salt cake from secondary aluminium fusion processes.
screening, intermediate size aluminium can be recovered by grinding Industrial Engineering Chemistry Research 44 (23), 8852–8857.
and screening or electrostatic separation, and fine aluminium can be Gil, A., 2007. Management of salt cake generated at secondary aluminium melting
plants by disposal in a controlled landfill: characteristics of the controlled landfill
recovered by flotation. Recovery of aluminium metal by aqueous acid
and a study of environmental impacts. Environmental Engineering Science 24 (9),
or alkaline leaching has also been proposed. 1234–1244.
Sodium and potassium chlorides present can be extracted by cold Gilbert, K.J., 1993. Salt production by Penrice Soda Products Pty Ltd at Dry Creek, SA. In:
Woodcock, J.T., Hamilton, J.K. (Eds.), Australasian Mining and Metallurgy, vol. 2. The
water leaching and recovered by solar evaporation after removing the
Australasian Institute of Mining and Metallurgy, Melbourne, pp. 1414–1416.
insoluble components by thickening and filtration. Habashi, F., 1997. (Aluminium) reactions with aqueous solutions. In: Habashi, F. (Ed.),
Some of the oxide aluminium components are soluble in Bayer- Handbook of Extractive Metallurgy, vol. 2. Wiley-VCH, Weinheim, p. 1043.
type leach solutions and could be sent to a Bayer-type leach plant to Hagni, A.M., 2002. Fine-article characterization by Rietveld QXRD, CLM, and SEM-EDS
phase mapping. Journal of Metals 54 (12), 24–26.
extract alumina. Alternatively, since the water-insoluble phases are Mah, K., Toguri, J.M., Smith, H.W., 1986. Electrostatic separation of aluminium from
inert, they could possibly be disposed of in a non-toxic dump, given dross. Conservation & Recycling 9 (4), 325-324.
they pass appropriate TCLP testing criteria. Manfredi, G., Wuth, W., Bohlinger, I., 1997. Characterizing the physical and chemical
properties of aluminium dross. Journal of Metals 49 (11), 48–51.
Hydrogen gas has also been generated from salt cakes, but, to date, Owada, S., Tokoro, C., Otsuki, A., Genka, Y., Maeda, H., Yamada, S., 2008. Application of
only in laboratory studies. selective grinding and electrostatic separation of aluminium dross as a pre-
It seems possible that integrated flowsheets could be developed treatment for generating “green hydrogen”. In: Wang, Dian Duo, Sun Chuan Yao,
Wang Fu Liang, Zhan Li, Cheng, Han, Long (Eds.), Proceedings of XXIV International
whereby one or more of the above valuable components are recovered, Mineral Processing Congress. InScience Press, Beijing, pp. 3469–3474.
assisting positively with the process economics, and helping to close Peterson, R.D., Newton, L., 2002. Review of aluminium dross processing. In: Schneider,
the loop in terms of overall waste disposal. W. (Ed.), Light Metals 2002. InThe Minerals, Metals & Materials Society, Warrendale,
pp. 1029–1037.
Pryor, E.J., 1965. Mineral Processing, 3rd Ed. Applied Science Publishers, London, pp. 193–195.
Acknowledgements Schlesinger, M.E., 2000. Aluminium Recycling. CRC Press, Boca Raton.
Soto, H., Toguri, J.M., 1986. Aluminium recovery from dross by flotation. Conservation &
Recycling 9 (1), 45–54.
The Analytical Services and XRD Services Groups of CSIRO Minerals
Zhang, L., 2006. State of the art in aluminium recycling from aluminium dross. In:
in Melbourne and Perth are thanked for conducting the many chemical Galloway, T.J. (Ed.), Light Metals 2006. InThe Minerals, Metals & Materials Society,
analyses and XRD scans required in this work. Warrendale, pp. 931–936.