Note: Descriptions are shown in the official language in which they were submitted.
~ ~13Z54
At present the aluminum industry consists of two
quite distinct processes which can be, and indeed often are,
conducted in diff~rent locations. ~hey are
~i) The production of alumina.
(ii~ Its electrolytic reduction to metal.
5 Bauxite is the basic raw material for ~he production of alumina
o and the latter i9 obtained using the ~ayer process which in-
volves selective solution with caustic soda.
Many countries have little or no bauxite and, because
prices of both it and the alumina made from it have risen
10 sharply in recent years, alternative raw materials and proce6ses
are being widely 30ught~ Clays fairly high in alumina content
are the subject of most of the research and development work
done in the last 20 years because they are indigenous to most
countries and self-sufficiency i8 regarded a~ being important.
15 But the economics o producing from clay (by extraction with
~'''?. '
hydrochloxic, nitric or sulphuric acid) are calculated as beiny
marginally less good than those from a new Bayer-type plant:
~For a good survey of the present position relating to
extraction from clay see "Environmen~al Considerations of
Selected Energy Consexving Process Options: Vol VIII,
Alumina/Aluminum Indu~ry" A.D. Little Inc. Cambridge,
Mass. Published by U.S. Dept. of Commerce, National
~echnical Information Service, P8-264 274, Dec. 1976.
--2--
.. . . . . . , .
. , . . ..... ~, ~ - - .
... . . ~ , ~
3ZS4
- 3 -
moreover the lattcr process is well-established whilst no
full-scalc ~lant from clay has been operated. An improvement
in the economics of one of thc alternative processes is
therefore required if the competition of the Bayer process
using b~uxite is to be met. This is the objective of the
present invention.
This invention relates to the production from the com-
bustion residue of a solid carbon-containing fuel (such ascoal
ash) having a carbon content of 2% or less by weight and an
iron content calculated as Fe203 of 4% or less by weight, of
a nitric acid extract which is high in aluminum content and
low in other constituents of the residue. The solution of
aluminum nitrate obtained can be worked up by known methods
to give alumina of high purity.
Any freely available raw material containing a high
proportion of A1203 equivalent must be regarded as a potenti-
ally attractive source of alumina because of the increasingly
higher price of bauxite. The main raw material in this
category, whicll has up till now received a great deal of
attention, is clay. Coal ashes, which sometimes contain
30-40% A1203 equivalent - come into this group.
Both high-alumina clays and coal ashes contain about
30-40~ A1203 and 40-60% SiO2. To avoid solution of silica,
extraction with nitric acid has been thoroughly studied in
the case of clay. In this work the clay has been calcined
at a controlled temperature, normally in the range 500 -
800C depending on the type of clay being used. In the
calcination, free water and water of hydration etc. are re-
moved and the nitric acid attacks the clay more easily after
calcination than prior to it~
Many coals are contaminated by clay-like impurities and
coal ash might be a cheap alternative to the use of calcined
clay if it were possible to discover an economical way to
extract its aluminum. When using coal ash there are no
B
. . .. . . - ~
.
~ . . . . .
.
. .
3L3L~L.,.r 2~ ~
cos~s of clay mining no costly calcination step is required,
and we subtract from an existing environmental and disposal
problem instead of creating one, as is the case if we use clay
and discard the siliceous and other clay constituants. W~
would not expect coal ash to be as convenient a source of
A12O3 as is calcined clay ~ecause in the latter case the heat-
treatment is carefully tailored to the clay, and to the
process, for the sole purpose of facilitating the acid extrac-
tion which follows. In contrast, the production of coal ash
is not all tailored to provide a satisfactorily-extractable
O ash, as the objective in this case is rather to optimize 'the
generation of heat from the combustion of the coal. Ohviously,
a calcination process cannot be equated to a combustion
process in which the carbonaceous constituent of the coal is
burning and heating the clay-like portion of the coal. The
closest study of the inability to consider the ash from comb-
usted materials as an equivalent to calcined clay prior to
nitric acid attached is reported on pages 18 and 19 of the
review "Aluminum from Indigenous UK Resources", Report No.
LR 219 ~ME), 1976, by P. Christie and R. Derry, Warren Spring
Laboratory, wherein they state: .
"In this country Riggl9 has recently studied the extract-
'ion of aluminum from unburnt spoil materials by acid
leaching. He has shown that it is necessary to heat
unburnt spoil to between-600 and 800 prior to leach-
ing in order to enable high extractions to be achieved
but that if heated above 800C the'extraction of
2S aluminum falls. During the uncontrolled burning of
spoil in a tip, much of the spoil is likely to have
reached temperatures much above 800UC; thus it is to be
anticipated that aluminum extraction from burnt spoil
by acid leaching will not result in high extractions.
However, since preheating to 6-800C is beneficial the
possibility of autogenously burning the spoil under
controlled conditions prior to acid extraction of the
aluminum is worth considering. Unburnt spoil has a
calorific value of betwee~ 2.8-4 x 106 Btu per tonne
(2.g6-4.23 x 109 J tonne~ ) and the NCB have shown that
19. Rigg, T. Private Comm.
-4-
:: . . . - . ~ . . , - . .
,, . . . . . ~ ..
; . . . ~ .
: . ., .. ~ , . : . . . -
5 ~
it will burn without an~ extra fuel input in a fluidized
bed combustioll apparatus at a bed temperatuxe of around
800~. ~owever, even under such controlled condit~ons
indi~idual paxticles reach much higher temperatures and
since the-autogenous combustion will not proceed ~elow
an average temperature of 800C it is unlikely that this
particular process could be used to enhance the reactivity
of unburnt spoil prior to acid leaching."
The importance of the coal burning temperature can
be observed from the relative ease of extraction of alumina
from (a) fly ash and ~b) fluidized bed ash. The latter is
much easier to extract than the former and hence fluidized
bed ash in our process is extracted in hot or boiling nitric
acid at atmospheric pressure whilst fly ash requires to be
~, extracted at higher temperatures under pressure~ Ihi~ is the case
even though the chemical analysis of the two ashes can be
virtually identical. A study by X-Ray diffraction and micro-
scopy of the chemical phases actually present in the two ashes
provides the key to understanding the difference between the
two in regard to their ease of extractability. In the case
of a typical fluidized bed ash the presence of an amorphous
phase and of metakaolinite along with some quartæ can be
identified. In the case of fly ash the presence of a glassy
phase and of mullite in considerable quantity can be observed.
Fluidized bed ash 1s normally formed at 800 - 900C
and fly ash at 1300 - 1600C. W~ have heated the fluidized
bed ash at 1400C and find that after this treatment metakaolin
disappears and mullite becomes a major phase present: indeed,
fluidized bed ash calcined at 1400 is virtually ind~stinguish-
able from fly ash. We conclude that fly ash is produced at
temperatures which favor the formation of mulli~e, and when
~he latter is present the alumina in the ash is less readily
soluble than in the case when the temperature of formation
of the ash is too low to cause its formation - and this is the
--S
.
... . : .
. . .
. . . ~ .
case ~-hen fluidi~ed bed ash is formed. Th~ condi~ions required
for the forma~ion of metakaolinite and of mullite have been
reported on by'J~D.C. Mc~onnell and S.G. Fleet, Clay Minerals
(1970) 8 279-290` these authors point out that metakaolinite
formed at 800 and ~he amorphous phase o~tained by heating
it at about 900 are both finely porous, we conclude that this
will facilitate solution in nitric acid. Mullite begins to
be ~ormed at 950 and is present in large quantity at temper-
atures in excess of 1150: the surface area of this mullite-
containing product is much lower than that of the 800-900
product and the surface area becomes much lower still at
j temperatures of 1250 and above.
¦ As stated, coal ash is a by-product of a power-
I generatin~ operation, whilst calcined clay is a product
tailor-made for alumina-extraction; the latter will normally
be more suitable for the ùsë therefore. But in addition it
must be observed that in the United Sta~es Bureau of Mines
and other literature publications the economlc and other
calculations are done on the assumption of using "the
hypothetical clay".
For chemical analysis of this material see Margolin,
S.V. and Hyde, R.W., the A.D.L. Nitric Acid Process for
Recovery of Alumina from Aluminum. Bearing materials, ~MS-
AIME, Paper No. A74-49, 1974; Proc. of Light metals, 103 AIME
Am. Meeting, 1974 Vol. 2, pp.469-487. U.S.P. 3,586,481;
~ohnson, P.W., Peters, F.A. and Kirby, R.C., Methods for
Producing Alumina from clay. An evaluation of a nitric acid
process-~SBM, 1964, R.I. 6431. This "hypothetical clay"
material is assumed to cont~in about 2% of impurities and 98% of
alumina and silica equivalents: this composition is equivalent
-6-
...... . . .. .. . . . .
: . '. . '-: ~ .
- . ~ :.... . - : . :
. ~ . . . .
Z~4
to a calcined high-grade kaolinite which typically would
contain 44-45~O A1203 and.53-54~b Si02. Our own coal ashes
are much less pure than this and contain (A1203+Si02) equal
to only 80-95~ after making allowance for moisture and carbon
S contents and can be as low as 70~ after allowing for moisture
only. In coal ash the carbon content can be in excess of
10%, and the cationic purities rising to 15%. Accordingly,
in our invention there will be decarbonization and/or magnetic
separation when required and a nitric acid pretreatment
before the extraction. Needless to say, our preferred.coal
ash for this process would be ~i) low in carbon and cationic
impurities and ~ii) readily attacked by nitric acid in
regard to.rate of aluminum solubility.
In accordance with this invention there is provided a
lS process for preparing a solution of high aluminum content
from the combustion residue of a solid carbon-containing
fuel having a carbon content of 2~ or less by weight and
an iron content.calculated as Fe203 of 4% or less by weight,
which comprises:
ta) pretreating said residue with nitric acid at a
temperature lower than the next extraction:stage
which follows, in order to obtain optimum solution
of impurities combined with minimum solution of
aluminum, and removing the nitric acid solution
from the residue and then ~ -
(b) extracting said treated residue with nitric acid
o~ strength about 30% to 65% by weight strength
at about 100 to 225~C.
The amount of nitric acid employed in said pret-reatment step
and in said extraction step being in excess of the cations
actually dissolved in each said step.
According to the invention, the method of obtaining
BJ
:-;, ~ ~ :
a nitric acid solution containing a con~iderable proportion
of aluminum and few impùri~ies involves some or all of the
following ~hrea preliminary sequential purification steps
followed l~y the aluminum extraction stage. Of the three
purification steps, oni~- the "wash staye" or"pretreatment
stage" prior to extraction i5 essential to the invention.
~ he first Purification Step involves the removal
(if necessary) of carbon from the coal ash by subjecting
it to heat in the presence of air. ~his is done at a
temperature high enough, and for a period long enough, to
C~ I reduce the content of carbon and other carbonaceous matter
I to a low level. If ~his is not done there is wastage of
¦ ni~ric acid in the later processing stages. The carbon in
~ question is that which remains in coal ash after combustion
and we prefer it to be less than 2~ by weight, otherwise
decarboniæation ls desirable. The exact proportion of carbon
i which can be tolerated depends, amongst other things, on the
local cost of nitric acid. Carbon causes it to decompose, for
example in accordance with the equation
3C ~ 4HNO3-4No + 3CO2 + 2H2O
The Second Purification Step involves the removal
if necessary of a su~stant~al part of the iron-containing
compounds in the ash by magnetic separation. ~he final
alumina product must have a low Fe~Al ratio and this i9
achieved partly by careful control during the stages of
processing the aluminum nitrate solution. This processing
is easier ~o operate if the Fe~Al ra~io in the aluminium
nitrate solution is itself as low as possible. The magnetic
separation operation contributes to achieving a low ratio
by removing as much iron as is economically possible in this
-8-
~ . . ,
~`'`'' ' : . - ' `'
,
z~
l~urif lcation Step . A typical r~duc~ion i.s to one-quarter o F
the iron ori~inally present, or example from 6~ Fe2O3 to
1-1/2% Fe2O3 a~ the expense of discardlng about 10-30% of a
high-iron fraction of the coal ash. We prefer to do the
magnetic separation in the aqueous slurry state but a dry
separation can be used if ~hat happens to be more convenient.
It is possible t if carbon removal and iron-contain
ing compound rem_val are both required before the "wash stage"
or "pretreatment stage" to effect these removals in any
de~ired order. However, the preferred method is to effect
carbon removal from and then proceed to the removal of the
iron-containing compound.
The Third Purification Step (the "Wash Stage"or
"Pretreatment Stage")involves treating the coal ash, (after
decarbonization and magnetic separation as necessary), with
a nitric acid which can be weaker than the nitric aci.d used
in the ~luminum Extraction Stage. The temperature of the
acid is also lower than that used in the later Aluminum
Extraction Stage. ~he purpose of the Third Purification
Stage is to dissolve away from the coal ash as much as possi~le
of such impurities as calcium, magnesium, sodium, potassium,
titanium etc. and, particularly as much as possible of the
iron remaining after the magnetic separation. If these im-
purities are not separated in this step they appear in the
aluminum nitrate solution which is the end-product of this
invention. ~he strength of acid and the temperature and
time used in the Third Purification Step are chosen to give
optimum solution o~ impurities combined with minimum solution
of a~wnin~. The actual optimum values o acid strength,
temperature and time used in the wash stage vary somewhat
_g_
. .
. . . -
.
~ Z5~
.....
from ash to ash, and particularly in respect of the temperature
achieved b~ the ash d~tring the coal--combustion process.
In this pre~reatment process the nitric acid is used
i.n excess over the amount required to dissolve the desired
proportion of cations in this pretreatment stage. The amount
of nitric acid u~ilized is equivalent to about 100 to 300%
required to dissolve all the A1~03 present in the combustion
residue. As a practical matter, the pretreatment only
dissolves about 10% by wei.ght of the cations present in the residue and
in turn the cations in the residue are present only to the extent
of about 40~ by weight of the total residue. Accordingly,
¦ the true excess of nitric acid actually employed is in the
¦ range of 4 to 8 times the amount o~ nitric acid solution
necessary to dissolve the cations actual~y removed from the
residue durin~ this pretreatment step.
The nitric acid solution of impurities is removed
! from the process, for example by settling and filtration,
prior to use, as a by-product; or example as an ingredient
of nitrogen-containing.fertilizers o.r for recovery of nitric
acid by the action of heat.
The Aluminum Extraction Stage comes now. It in-
vol~es treating the coal ash, purified as above, with nitric
acid of 30-65~ HNO3 strength, and preferably of 55-60~
strength to minimiæe evaporation costs at a later stage.
2S Once again the choice of acid strength, temperature and time
o extraction depends on the nature of the ash and part~cularly
on its temperature of formation. To take an example, fly
ash is produced at about 1100C and is found to be relatively
difficult to extract and therefore requires a h~ler extraction temperature
e.tc; on the other hand ash produced ~y fluidi2ed bed combustion
--10--
- . ~ .
- : .-'' ' -
:'
32~i~
is forme~ at ~bout 9~0C and is much easier to extract.
The extraction of fly ash i5 best done al_ 150-225C
and prefer~bly 160-200C; the ~rea~ment ~akes place under
pressure. In the case of batch extraction ~he time is normally
S chosen to permi~ at least half the aluminum present in the
purified fly ash to be ta'~en into solution and is typically
3 hours.
The extraction of fluidized bed ash i~ best do~e
at 100-13~C and preferably at 105-122C; the latter range
relates to the boiling points of nitric acid of 30-65~
, concentration under atmospheric pressure. Alternatively a
¦ staged counter-current extraction process may be used to
¦ minimize the consumpkion of nitric acid per ton of aluminum
~ extracted.
In the extraction stage an excess of nitric acid
over that required to dissolve the aluminum cation is
employed. The excess preferably employed is 300% to 600% of
the amount of nitric acid solution necessary to dissolve
the aluminum cations taken up from the residue during
extraction. ~he amount of aluminum extracted depends on
pxactical considerations~ These are the economically accept-
able amounts of aluminum dissolved in the course of the time
and temperature employed.
~he solution obtained as a resùlt of the ~xtraction
Stage i9 separated, for example ~y settling and filtration.
It is then worked up by known processes to give a high purity
alumina, for example by recrystallization of Al(NO3)3 9H2O
followed by heat decomposition of the latter to give alumina
and most of the nitra~e content as nitric acid. The final
propor~ion of nitrogen need~ to be driven off as oxides of
.: , .-
Z~ ~L 1
nitroge~; the lattex also ~eing recycled af~er r~conversion
to nitric acid. ~he recrystallization of aluminum nitrate
is to sep~ra~e the small amounts of iron, calcium and other
contaminant~ and to permit the production of an alumina product
wh~ich meets the desired specifica~ion in regard t~ chemical
impurities. As usual, any contaminated nitric acld liquors
can be used in ertilizer manufacturing processes to take
advantage of their nitrogen values. The solid remaining
after the Alumlnum Extraction Stage is relatively low in
aluminum and impurities and high in silica. It can be
discarded or used in the production of refra~tories or zeolites
etc.
According to the procass of this invention the ratio
of aluminum content to combined cations is improved by a
lS factor of 9 to 15 in the case of fly ash and by a factor of
2 to 5 in the case of the fluidized bed ash,
The fly ashes chosen for the process of this inven-
tion preferably contain 30 to 35~ by weight A1203 equivalent,
2 to 8% Fe203 equivalent, 45 to 55% SiO2, 8 to 12% other
impurities, with the remainder being carbon and water.
Thesé "other impurities" are generally calcium, magnesium,
potassium, sodium and titanium values.
In the final solution from the fly ash the rates of
Fe203 to A1203 has been reduced by a factor of about 10
whilst the remaining (or other) impurities has been reduced
by a factor of 10 to 15.
The fluidized bed combustion system is favored for
coals and colliery reæidues which have a relatively high ash
content. It has been our experience that ashes high in alumina
and relatively low in iron content are avai~able from fluidized
-~.2-
~,i., . ,.. - - , . ~ .
~'~ . ',
: . :
-
. , . ~ . : .
~,: ' , :. . ' '
bed combuc;tion thar. in the ca~e from pul~reriæed r-u~l burning.
~ he improvement in the propoîtion of alumina to iron
and other impurities achieved in the c~se of fly ash has not
been achieved in the case of fluidized bed ash for 2 main
reasons (a) the aluminum in the fluidized bed ash dissolves
relatively ~uickly and therefore the efective separation from
iron and the impurities realized in the case of fly ash cannot
be achieved in the case of the fluidized be~ ash and (b) the
impurities in fluidized bed ash tend to be at a lower level to
begin with and therefore the opportunities for improvement
are less available.
Our experience is that in the case of the fluidized
bed ash the improvement in the ratio-or A12O3 to Fe2O3 is about
2 and in the case of "other" impurities the improvement is
approximately 5.
Coal ash is a product which is produced in coal-
burning power stations and similar processes. The commonest
form of coal ash is fly-ash, which is the solid product remain-
ing from the combustion of pulverized coal. Another form
of coal-ash is the solid product obtained from the combustion
of a coarser size of coal in a fluidized béd unit: the ash
obtained from this kind of plant has frequently been exposed
to rather lower temperatures than is the case with fly ash
and the use of this kind of ash in our process proceeds
rather differently as has already been indicated. The fluid-
i~ed bed process of combustion i9 only in its infancy at
present but the prospects for it are excellen~. This is on
account of its ability to accept very low grade coals and
even colliery wastes, in which the ash content i5 considerahly
greater than the content of com~ustible ma~ter~
-13-
.
z~ :
llhe major ingredients of fly ash are alumina and
silica in the form of mi~ed compounds such as mullite
(3A12O~ 2SiO2) and vitreous materials. The impurities are
pxesent in smaller amount than are the two major ingredien~s
and iron oxides and carbon are two of the mai.n ones. By way
of example, the analysis of a certain South African fly ash ~ .
very high in carbon (from an old power station) is as in Table
1. It has heen ~ried at 130~C.
Table 1
. A12O3 ....... 27.8
SiO2 ;........... 38.7
C) Fe O -- 5~0
Carbon ....... 16.0 ;
Other ignition loss 6.2
Minor ingredients 6.3
CaO ~1.6)
MgO (1.8)
K20 (1. ~)
~ . Na2O (0.6)
; ~ : 20 . TiO2 (0~5)
~:~0 ' 100
~`
~ (The minor ingredients are estimates)
: , . .
;~ ' ' .
,
. -14-
.. ,.,........................................ .. : :~ ~ . . ,. . . : .
... .. . .
In the light of what has been said he~ore it is clear
that this particu~ar ash i5 much too high in carbon to be process-
ed without ~he use of the decarbonization step. The heating
for decarbonization can, for example, take place in a fluidized
bed or other calcinat~on equipment. Some coal ashes will
contain sufficient carbonaceous material to prov~de the heat
for this stage. If the carbon content is too low to provide
the n~cessary heat then the decarboniæation step will frequent-
ly be unnecessary. By way of example the South African
fly ash described above has the analysis given in Table 2 after
calcination at 950C:
Table 2
A12O3 ~ - 35 7
SiO2 .......49.8
Fe2O3 .......6.4
Minor inyredients 8.1
lon . o
A high-alumina fluidized bed ash has the composition
given in column 1 o~ Table 3.
Table 3
1. 2. 3.
23 38.2 31.6 36.0
SiO2 48.6 58.0 S3.g
Fe2O3 2.4 2.9 1.7
CaO ~0.1 1.3 1.3
MgO ~0.1 1.4 1.4
TiO2 0.5 1.8
Na2O 0~2 0.3 0.2
K O 1.9 2.2
2 -15~
. .
Carbon 6.6 ) )
) 0.4 ) 0.4
Moisture 1.0
This is from the interceptor cyclones following the
combustion and is unduly high in carbon. For most of our work
we have therefore used an ash such as is given in column 2,
which is from ~he ~ed itself rather than from the cyclone
separators. After having been sub~ected to SOMe screening and
to a magnetic txeatment to reduce the level of ~e2O3 the com-
position of the ash was as given in column 3 and it has been
used for much of our practical work.
Two ways of extracting alumina from coal ash are -
(i) by batch extraction and (ii) counter-current flow extrac-
tion. Our process can use both procedures and examples of
both are given later.
lS Thé use o 1uidiæed bed ash in our process brings
with .it one enormous advantage as compared with the use of
fly ash: it is that the alumina is relatively easily extract-
ed at atmospheric pressure by nitric acid at the boiling
point or even below. But an important disadvantage accompanies
the advantage. It is that the alumina is so readily dls~olved
; ~ that a purification depending on the rela~i~e solubilities
~; of Fe2O3 and A12O3 in the nitric acid wash stage, such as
applies in the case of fly ash, i8 difficult to achieve~
The wash stage in the case of fluidized bed ash i9 therefore
largely to reduce impurities such as calcium, magnesium,
~odium and calcium and gives very little improvement in the
Fe2O3/A12O3 ratio in the washed solid. In contrast the
Fe2O3~A12O3 ratio in pulverized fuel ash can be ~onsiderably
lowered in the wash stage.
.
-16-
`i :
,
.~. ' ,'. ,' : '
.
.
Example 1.
100 Parts of dry fly ash (~ in Table 4 below) were
subjected to a very simple magnetic separation,process. 11
Parts of the ash were removed on account of their high iron
content and 89 parts of material B remained. The carbon
content of this ash was low and therefore no decarbonization , '
step was necessary.
In the nitric acid purification step material B
was subjected to extraction in a batch treatment process
using 35% HNO3 at the boiling point of 110C. Each batch
treatment used 9 1'/2 liters of acid liquid to 1 kilogram
of material B. In this treatment the weight loss of the
¦ solid was 12.6% and 78 parts of residue C'(dry basis) were
obtained. The latter was now subjected to the aluminum extract-
tion stage using 61% HNO3. A batch treatment was again used,
230~C being the extraction temperature and 2 hours using
20 liters of acid per kilogram of C being the conditions for
each batch operation. The residual solid was separated
~rom the acid by filtration and weighing showed a weight loss
of 28.2%, leaving 56'parts of D. The solution of aluminum
- nitrate in nitric acid was not subjected to any further
purification in this instance but was simply evaporated,to
dryness and calcined at 1000C prior to analysis to determine
the amounts of aluminum and of impurities dissolved by the
treatment. The calcined material represented 18 parts by
weight of the original ash and its X-Ray diffraction pattern
showed it to be alpha-alumina. The analysis of the calcined
material is given at E. The remaining 4 parts of C not
accounted for correspond to the oxidation of carbon content
in the aluminum extraction stage.
,, -17-
: . , .
~ '' I '
.
z~
- T~ble 4
A B C D E
Al O 3~.7 30.9 27.6 10.6 95.6
';i2 50.9 53.3 61.7 83.3 0.6
CaO 5.9 5.9 1.5 2.4 1.3
MgO 2.0 2.2 0.6 0.6 0.6
Fe23 6.1 3.2 2.4 1.3 '1.8
TiO2 1.0 0.9 1.4 0.7 0.2
Na2O/K2O0. 6
L.O.I. 2.2 3.1 4.6 Nil Nil
(~) 100.4 99.7 99.8 98.9 100.1
Example 2.
100 Parts of dry fly ash (A in the tab~e below) were
subjected to a commercial-scale magnetic separation process
whexeby ~9 parts of B were removed and 71 parts of C remained.
The latter was then treated in the nitric acid purification
step with 35~ 'IN03 in a batch treatment process. Two hours
at 110C (boiling point) was given in each stage. Approx-
imately 3 liters of nitric acid-containing liquor was used
G 20 in each stage per kilogram of ash. In this purification
step the weight loss of the solid was 9.5% and 64 parts of
dry residue D were obtained. The~ filtrate, a~ter drying and
calcining had the composition E: the effectiveness of im-
purity removal can be observed.
The solid B was now subjected to the aluminum
extraction stage using 61~ HN03 by batch extraction using
' a ratio of 3 liters of acid li~uor per kilogram of solid
in each batch. The extraction time of each batch was 4
hours and the temperature 165C. The solid ~ ~53 parts)
-18-
~ .
, .
. ~ ,
was separ~l~ed from the acid li~uor ~nd had the composition
shown. Th~ iltrate so]utlon, without any attempt at further
purification, was evaporated to dryness and calcined at 1000C
to give B parts of solid G. X-~ay diffraction study of G
showed it to be alpha-alumina. The resnaining 3 parts of D
not accounted ~or correspond to the oxiclation of ~arbon content
in the aluminum extraction stage.
A B C D E F G
A123 24.6 21.6 25.8 26.0 38.9 17.6 94.5
SiO2 55.7 47.8 58.9 64.7 0.4 76.3 0.2
¦ CaO 4~4 4.9 4.2 0.5 41.4 0.9 1~3
¦ MgO 1.4 1.7 1.3 0.4 11.5 0.5 0.6
Fe23 21.0 1.6 1.3 6.3 1~1 2.6
TiO2 N.D N.D N.D 1.7 1.2 1.7 0.1
lS L.O.I. 3.6 1.3 ~.6 5.3 Nil 1.6 Nil
97.6 g8.3 96.4 g9.9 99.7 9~.7 99.3
Example 3.
The fluidized-bed ash used in this example was the
400-800 micron fraction from the bed itself. This fraction
comprised 40% of the to~al bed ash. Its composition i9 given
in column 2 of ~able 3. N.B. 98% of the total bed ash had
a composition A12O3 = 30.6 + 1%, Fe2O3 ~ 3.1 + 0.2~ and
L.O.I. 0.4-0.6%. The 2% of -200 micron material had L.O~I. =
2.2%. After magnetic separation at 7000 gauss in a disc
separator the low-iron residue had the composition given
in column 3 of Table 3 and this raw materia~ was used now
for acid treatment. ~he low content means that no decarbon-
i~ation step is required in this case.
The ash was first sub~ected to a wash stage as
shown in Fig~ 1. 61% HNO3 was used at 70'C for 3 hours. ~t
_~ 3_
.
.: .
will be i~.een ~hat althc,ugh consider~ble solu~ion of calcium
took place there was also considera~le solution of alunlina
and no lmpro~ement in the ~e20/~1203 of the residual so-id book place. A seriesof extraction stages at ~le boiling point was ~hen done using 61~ HN~3 at 120C,with the first ex~-action in any stage heing d~ne usin~ a co~bination of the
liqulds from the seoond extraction and t~e wash from the previous stage
and the solution rom this extraction containing the
product aluminum, The solution usually contained %8-95%
A1203 with some iron and other impurities. This solution
is now suitable for further conversion to alumina by known
techniques such as crystallization and ion exchange to reduce
the iron content, hydrolysis by heat to give nitric acid for
recycling and calcination to provide pure alumina. It will
be noted that in the second stage o extraction the A1203
15recovery in the product solution is 1-83x1~0 ~27~ . Higher
tO.61+6.12)
recoveries can be achieved if required by variatlons of
temperature, ti.me and acid strength but some o~ these bring
obvious disadvantages. Since coal ash is a waste~product
we have so far preferred to minimize the operating problems
at the expense of discarding half to three-quarters of
aluminum originally present in the ash.
In this trial therefore a typical product solution,
when calcined for purposes of analytical examination, contain-
ed 90% A1203, 4.6~ Fe203 and 5~ other constituents, mainly
K20. The 5~ can be reduced somewhat by better liquid-solid
separation operations than were used in this example. It
will be noted that failure to opèrate the nitric acid pre-
treatment step would have given a product solution o~ con~
-siderably lower purity than that actually ohta~ned.
-20-
~,
.,. . ~
:
