Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
9i~L
The present invention relates to the disposal of
fluoride-containin~ wastss and in particular to the waste
rssulting from the reaction of fluorspar with strong sulphuric
acid to generate hydro~en fluorida. This reaction is carried
out on a large scale in connection with the production of
aluminium fluoride for use in the electrolyte of electrol~tiG
reduction cells for the production of aluminium.
In the treatment of fluorspar with strong sulphuric
acid a substantial exces~ of acid i~ employed. Consequentl~
the solid residue, resulting from the process, is highly
acidic. Despite the use of e~cess sulphuric acid, the waste,
which is primarily composed of anhydrou~ calcium sulphate,
contains substantial quantities of unreacted calcium fluoride
and free hydrogen fluoride. ~his waste therefore presents a
substantial problem for disposal in an ecologically acceptable
manner. In order to satisfy that requirement the pE of the
acidic waste should be brought to a value above p~ ~ and more
preferably above pH 5, for example pE 6 and the reæidual
fluoride should be "locked" in the neutralised waste so that it
does not leach out to ~uGh a~ extent as to raise the ~luoride
content of neighbouring water courses to an unacceptable level.
We have found that a useful test for audging whether the
fluoride is adequately locked into the waste i8 to subaect a
batch of neutralised waste material to water leaching for 48
hours. If at the e~d of the third or fourth such leachi~g
treatment the extraction of fluoride is found to have reduced
to less than 1% of the fluoride content the treatment may be
considexed accaptable. However the proce~s of the present
in~ention is capable of achieving far superior results to that
above indicated.
~ he pre~ent invention ~eeks to lock in the residual
fluoride content by neutralisation of the calcium sulphata
Y.~-
~. . .
: . , , ., . .:
.: . : , . . . .
10~39~L
anhydrite waste to a compaat, rather impervious rock-like
state. ~o achieve this result it is necessary to convert the
acidic anhydrite waste to gypsum (CaS04.2E20). Conventionally
the acid anhydrite waste has been treated with lime or with
calcium carbonate, either dry or under aqueous conditions.
Using dry neutralisation it has been necessary to employ a
briquetting technique in order to bring the waste into a form
in which it could be conveniently transported. ~owever, the
~riquettes are u~stable in the waste disposal area and therefore
hazardous from an ecological viewpoint. When using neutrali-
sation under aqueous conditions the resultin~ residue has to
be stored in unsightly "white mud lake~", leading to pos~ible
contamination of water courses not only by dissolved fluoride
but al80 by unde~irable finely divided solid particles~
We have now apprecia-ted that the neutrali~ation of the
acidic anhydrite waste can be very conveniently effected by the
use of a waste material, the so-called "rea mud" re~ulting from
-khe Bayer pxocess for the production of alumina from bauxite.
As discharged from a Bayer process plant the red mud is in the
form of a slurry containing 20-25% solids a~d 1-2/o soda in
solution in the aqueous phase. ~h~ red mud solids co~tain
sodalite ~a sodium aluminium silicate complex) which readil~
reacts with mineral a¢ids such as sulphuric acid and hydro-
fluoric acid.
~he present in~ention is based on neutralising the free
acid content of the acid anhydrite waste by reaction with the
sodalite conte~t of the red mud solids. ~he product~ of this
reaction have a catalytic effect on the hydration of anhydrite
to gyp8um. ~his reaction sequence imparts unique rheological
properties to the mix~
According to the present invention, there is provided
a method for dispo~ing of acid anhydrite waste, resulting from
.
: .
.
the treatment of fluor~par with stro~ sulphuric acid, which
co~prises mixing said waste with red mud solid~ from the pro-
duction of alumina from bauxite to form a slurry haYi~g 65-8~h
~olids, said waste being comminuted either before or i~ the
course of mixi~g with said red mud solid~, the proportions of
acid a~hydrite w~ste and red mud ~olid~ bei~g selected 80 that
at the end of the mixing procedure the slurry has a pH v~lue of
at least 3, the resultant red mud/acid anhydrite waste slurry
then being pumped to a receiving area and allowed to ~et into
a de~se, rock-like ma~s.
It is a surprisi~g feature of the present inventio~ that
the high-solids slurry formed by mixing the red mud ~olids and
acidic anhydrits waste ~hould be pumpable.
It is pre~exred that the red mud/acid anhydrite waste
slurry should be a~ nearly neutral as possible and for that
purpose it is preferred that the relative proportions of the
two components should be cho~en to provide a ~lurry which~ after
reaction, ha~ a pH in the range of 5-9. ~owever it i~ in no
way critical that the pH of the slurry should be withi~ that
range.
It is k~own that alumi~ium sulphate and ~odium sulphate
act as activator~ for the tran5formatio~ of &nhydrite to gyp~um ~:
and these substance~ are produced in the neutralisation of the
exce~s sulphuric acid content of the anhydrite waste~ I~ this
trans~ormation a substantial proportion of the water conte~t of
the ælurry iæ absorbed as water of crystallisation, so that the
slurry on standing at the receivi~g area sets into a rock-like
mass which is relati~ely impervious to the passage of waterO
~he imperviousne~s is attributed to the release of silica in the
30 form of an amorphous gel when sodalite is broken down by reac
tion with æulphuric acid. On subjecting the æolidified material ~:
to the leaching test sug~ested above, it is found that the rate
--3-- :
.
..
. . ., ~ ~. .
9~;1
of extractio~ of both fluoride and sulphate is well within
acceptable limits, so that any percolatio~ of the wa5te disposal
area by rain or stream water will not lead to unacceptable
¢ontamination of water courses.
While the present procedure for the neutralisation of
acidic anhydrite waste may be carried out with a red mud slurry
taken directly from the Bayer process plant, it is preferred to
filter the red mud slurry, or use other separation methods
such as settli~g, to recover part of the soda-containi~g a~u~ous
pha3e a~d the~ to re-slurry the moist separated solids i~ a
little water, maintaining a high-~olids slurry, before mlxing
with the acid anhydrite waste. In this way, the ratio of
anhydrite waste to red mud solid~ can be readily adjusted to a
desirable level. ~he percentage of solids i~ the final mix can
then be controlled by a suitable addition of water. hnother
ad~antage of this approach is that more sodalite, and less soda
from liquor, is used for the neutralisation and thus more
aluminium sulphate and silica are formed. It i8 believed that
these two compounds are beneficial in the setting of the mix.
In fact, in test results it has bee~ found that the rate of
leaching from the ~eutralisation product, after setti~g, shows
some further improveme~t when filtered red mud solids are
employed as compared with the use of the raw red mud ~lurry
taken direct from the Bayer process plant.
In one series of experiments de~igned to simulate
commercial scale eperation acid a~hydrite waste at a tempera-
ture in the range of 150-205C ~typical of the discharge
temperature from a reactor for generation o~ ~F) was miæed with
a red mud slurry withdrawn from a ~ayer process plant at a
temperature in the range of 55-70C. ~he resulting slurry had
a solids conte~t in the range of 60-75% and after reaction the
pH of the slurry ranged from 102 to 8.6. I~ all cases where
--4--
896~
the pH value of the slurry was above 3 the mix harde~ed within
5 days. Satisfaotory results were obtained with acid anhydrite
wastes having residual free sulphuric acid content~ a~ hîgh as
1~/o and free re idual calcium fluoride contents as high a8 1~/o
and as low as 0-5%O In all cases the ~lurries had an ~pparent
viscosity ~ufficiently low ~or them to be pumpable7 When a
slurry of less than 65% solids conte~t was allowed to set i~
la~ers of a thick~ess of 1-3 inches a thin layer of ~uper~atant
water appeared on the top of the set material. It was there-
fore concluded that for practical operation it WAS preferableto operate with a slurry having a solids content i~ the region
Of 65~ /o.
~ he operation of the process is further described with
reference to the accompanying diagrammatic drawing of a plant
designed to put the invention into e~fect~
Red mud slurry from a Bayer process plant enters a
storage tank 1 via conduit 2. ~imited settlement of the ~lurry
may take place in tank 1 and soda-containing liquor ma~ be
withdraw~ via overflow pipe 3. Red mud slurry is withdrawn from
the bottom Of tank 1 by a slurr~ pump * and forwarded to a
paddle mixer 5 at a predetermined rate through a variable
opening valve 6 under control of a flow metor 7, excess slurr~
being returned to the tank 1 through the conduit 8~
Acidic anhydrite waste is supplied through conduit 10
and is supplied via a screw conveyor 11 and magnetic hump 12
(a device to trap an~ tramp iron in the anhydrite waste) to
paddle mixer 5, in which the two compone~ts are thoroughly
mixed for several mi~ute~ to allow i~itial reaction to take
place with relea~e of generated C02. ~he slurry is then for-
warded to a disintegrator 14 in which the mixi~g is co~tinuedand any remaining coarse lumps of the anhydrite waste or com-
pacts ~ormed by red mud/a~hydrite waste reactio~ in mixer 5 are
--5--
broke~ down to ensure complete conversion of the anhydrite to
gypsumO In some instances, particularly where the acid content
of the anhydrite waste is 1QW ~and consequentl~ the propor-
ti~n of red mud slurry is also low),, the paddle mixer and
disintegrator may be replaced by a ball mill. The ælurry from
the disintegrator or ball mill is then forwarded by a slurry
pump 15 to the disposal area. On reaching the disposal ~ite
the slurry is deposited in a thin layer to allow setting to
take place quickly, so that it is usually deposited on a mass
from earlier operations.
A plant te~t was carried out using the equipment shown
in ~igure 1. Several hundred to~s of mix were deposited in a
trench-like depression ~nd the physical and economic parametsrs
were monitored. ~he mix behaved as predicted in the above small
scale tssts with no measurable contamination o~ atmosphere, of
sub-surface waters, or of run-offO The compressive strength of
mix was over 2~000 pounds per square inch after 20 days of
hardening.
--6--
' ' ~
