Note: Descriptions are shown in the official language in which they were submitted.
1~5~5~
BACI'~G~OU~D OF T~ NV~NTION
As is well known, since the qovernment has lifted the
price on gold from $35.00 an ounce, the price of yold has
multiplied. As a result, many gold mines ~hich were forced
out of operation by the $35.00 an ounce-ceiling have now
resumed operations, and gold exploration and mining has greatly
increased.
Because most gold ores contain less than a few ounces of
gold per ton of ore, large amounts of gangue must be processed
in order to recover the gold. In addition to the low grade of
gold ores, the gold is usually present as very fine particles.
Thus, gravity processes for the separation of gold from gangue
are inefficient. This is due to the high viscous drag forces
acting on small particles in water relative to the force of
gravity.
Typically, large amounts of water are needed for bene-
ficiating gol~ ores, particularly placer gold ores. This is
a significant problem in recovering gold from low
,
~, , '' , ' .
: '
.
dm: b~
'-'~.1
, 1~ ~ . . .
.
,- . ~ ;. ,.
--
~1~a$9S~
grade ores particularly plac-~rs existing in arid areas su-~h
as deserts. There is, therefore, considerable time and
expense involved in recovering gold from its ores.
The above conditions have created a need for
improved and more efficient beneEiciating procedures for
the recovery of gold from low grade ores, i.e., gold
- associated Wit}l ~oreign materials with which the gold
exists in small percentages. Also, a process which operates
dry would be especially useful, because it would provide
a method for recovering gold which is located in deserts.
Accordingly, it is a principal object of this
invention to provide an economically feasible method for
separating gold rom foreign material by selectively enhancing
the magnetic susceptibility of the gold particles so that
they may be successfully separated from the foreign material
by magnetic separation.
SUMI~ARY OF Tll~ INVENTION
The magnetic susceptibility of gold associate~
with oreign materials is increased to the point where
magnetic separation of gold particles from the foreign
material is feasil~le. The magnetic susceptibility of the
gold particles is increased by contacting a mixture of
particulate gold ~nd foreign materials, such as occurs
with placer deposits, with an iron carbonyl like iron
pentacarbonyl under conditions at which general decompo-
sition of the iron carbonyl into metallic iron and carbon
.
monoxide is not appreciable. ~he carbonyl-treated mixture
is then passed through a ma~netic separator for removal
of the gold particles.
~: , . .
, . . .
' .
~:
l~S~5~ ~
Placer ~old ores usually do not require grinding
to achieve liberation; however, if required~ they may be
ground. The liberated ore is then contacted with carbonyl
vapors in a gas treating chamber, either alone or by means
of a gas that is inert to the process, which is used to
carry the iron carbonyl vaporsO Physical separation between
gold and forei~n material follows in a magnetic separator.
DESCI~IPTION O~ THE; PREF13RRED E:~5BODIMENTS
The invention is particularly useful for re-
covering gold from placer type gold deposits wherein gold
particles which are either free or have an e~posed surface
exist in small percentages with large amounts of sand and
other particulate material including dolomite, albite,
muscovite, gypsu~, and calciteO In the case o placer gold,
grinding can ordinarily be dispensed with. The invention
i5 apylicable to recoverin~ gold from quartz, granite, other
type rocks, and other mat~rial to which it is attached;
however, in the case of these materials it is ordinarily
first necessary to grind the material to a sufficiently
fine particle si~e to liberate particulate gold. This
process also includes the recovery of more than one metal
value at a time from an ore or mixture. The terrn "mixture"
- as used herein includes ore.
It is not known why the process of the invention
énhances l:he magnetic susceptibility of the gold particles.
It is well known that neither gold nor iron carbonyl are
magnetic. It is probable that the ~old is coated with a
thin shell oE metallic iron, which, of course, is magnetic.
What is not known is why there should be a selective deposi-
tion of a film of magnetic material on the gold while under
. ~ . . .
' ' . . ':
' , . . .
- , :
~,,~ .. . ... _._ ._ .. . . , . .. ,_ __ _ _ . _ . __,
.
' ~ , ' '" ~ ~, . ' .
,
, , . .:
, ~
S95~
.
essentially the same conditions there i5 not decomposition of
iron carbonyl producing a magnetic ilm on all ~he ore
- particles. Of course, a rapid and complete decompositicn of
iron carbonyl would ~esult in coating partic}es of both gold
and the fo~eign material with iron so tha~ an effective
magnetic separation would be obviated. Other metal car-
~onyls may be used such as those o~ the Group ~III metals
nickel and cobalt.
Iron carbonyl decomposes under the proper tempera-
ture conditions in accordance with the following reaction:Fe(CO)5 ~_ Fe ~ 5C~
The process is applied by contacting the mi~ture
of gold and foreign material with iron carbonyl under
conditions wherein the iron carbonyl decomposes to form
a magnetic skin on the gold particles but not on the foreign
material. These conditions are determined ~y the temperature,
the type of carbonyl used, pressure, gas composition, etc.
ordinarily~ th~ reaction occurs at a teraperature just
below the substantial decomposition temperature of the
~ carbonyl in the presence of an ore. Various types of
available equipment can be used for contacting the gold
and foreign material with iron carbonyl vapors, such as,
a rotating kiln used as B reaction vessel with the material
being contacted directly with iron carbonyl vapors or the
~5 vapors carried into contact with the tumbling contents of
the kiln by a gas such as nitroc3en which is inert to the
j reaction process. It has been found that the material
: which enhances the magnetic susceptibility of the gold
particles exercises a preferential selectivity for the
~¦ gold particles over the particles of the foreign material.
.. .`
:'
, ,'', " ".'
.~
. .. . ..
i I ' i . '; ; ;, .! I
~ ~ - ~
' .
.. .
1~4~95~ ~:
The process must be carried out at a tempexature
below the temperature o major decomposition of the carbonyl
under the reaction conditions so that there is no opportunity
or decomposition o~ the carbonyl on a nonselective basis
or, ~erhaps, for its reaction with some material to produce
the magnetic material with which the gold particles are
coated. Obviously, if ~he temperature is allowed ~o rise
above thc decomposition temperature of the carbonyl for
sufficient time, complete decomposition of the carbonyl
will occur with the result that the particles of the
foreign material as well as the gold will be coated with
metallic iron to give both types of particles an enhanced
¦ magnetic suscep~ibility, thus preventing their effective
separation magnetically.
The amount of carbonyl used and the time of treat-
ment can be varied to efect substantially complete magneti-
zation of the gold presentO l`he time, temperature and
injection rate o the treatment is a balance between the
reaction rate and the economics of the magnEtic separation
' process. Carbonyl will be added in an amount of from
about 0.1 to about 128 kilograms per metric ton of feed
with from about 0.25 to about 8 ]cilograms per metric
_ ~ ton o feed being preferred and from about 0.5 to about
4.0 kilograms per metric ton of feed being more preferred.
~dditionally, it is preferred to in~ect the carbonyl into
I the reactor durin~ the first ha].f of the roast period
¦ and it is more preferred if it is injected during the
j~ first quarter of the roast and most preferred if injected
during the first tenth of the roast period~
~' .
~, I
~iiq~ I ' , ', , .5
i'`i~ I
~ .-
,, . .. , . ~ .
..". ,.
; , ' ~. ' ' .
', .
.: ' ., . ' . ' ' ~ '
.: : . : , ' . ,
. . . '. ' : ' '- . .: ' ' ,:
S95~
Generally, a reaction time not in excess of about
two hours is adequate, with a reaction time not in excess
of one hour being preferxed and a reaction time not in
excess of a hal~ hour being most preferredO The temperature
at which the reaction is formed at atmospheric pressure can
vary between about 100-250C, a preferred temperature rang?e
is from about 100 to about 150C and a more preferred
temperature ran~e is from about 110 to about 130C.
Generally, the higller the temperature, the more complete
the gold recovery with lower gold concentration in both
the tails and the magnetic concentrate and the larger the
amount of magnetic concentrate. Therefore, for any
feed material, the economics of the situation will have
to be consi~ered and conditions set to produce the most
favorable balance between the grade and recovery.
If desired, prior to treating the gold and foreign
material with iron carbonyl, the mixture of gold and foreign
material can be magnetically cleaned to remove any magnetlc
impurities. Thereafter, the non magnetic fraction o~ the
~0 mixture is treated wlth the iron carbonyl. After the feed
mixture containing the gold has been treated with a metal
carbonyl, it is then subjected to a magnetic separation
process to efEect the separation of gold. Any of many
comrnercially available magnetic separators can be used to
remove the gold from the gangue. Por example, low or
medium intensity separations can be made with a permanent
magnetic drum separator (field strengths up to about
2,500 gauss), electromagnetic drum separators (field
strengths up to about 7,000 gauss), induced roll separators
(field strengths of about 11,000 gauss) or other confi~-
urations known to those skilled in the art. Additionally,
newer high-~Jradierlt magnetic separators are espacially good
.
?
.. . .
_ _, ,, _ _ _ . . _ . . _ . . ..
, ~ ,__ ... . _._
' ' ' ~ .' .
:, ' . ' :
.
' ' : ~ ' ' ' ~ . :
~ ' , '
,,
, ~'
5:L
for separating ~ines, although they are generally operated
wet. A dry matJnetic separation process for gold iB
generally preferrcd. This avoids the exp~nse o~ de-
waterin~ and also allo~s for the reco~ery of gold from
deserts.
The invention is illustrated by the examples
I presented below in which samples of placer gold and
associated foreign material were treated by the process
. of the invention. The examples are illustrative of the
invention bu~ not limiting thereofO
e~ .
~X~MPL~ 1
In this example, a sample of placer gold concen-
trate was diluted with gangue of essentially silicon
dioxide and aluminum dioxide. The resultant sample con-
tained 4050 grams gold per metric ton of placer ore. ~or
- ~ the purpose of a blank, a comparative magnetic separation
~ was made on an untreated portion of the sample. Rnother
~~ ! portion of the sample was treated with the process of the
? invention at 135C and a third portion of the sample was
;~t , z~ treated by the process of the invention at temperatures up
to 145 C. Both of the treated samples were subjected to
magnetic separation as in the first test, and the magnetit,
and nonmagnetic ractions of each test were analy~ed as
, to ~old content with the gold distributlon for the magnetic
and nonmagnetic fractions of each test cGmputed. By "Gold
Distribution" is meant the percentage of gold in the entire
,; ~ beginning sample which is partitioned to the specified final
~ fraction. The following table sets forth the results obtaint
.. 1 . - ~
~"~ 7
.-9
. ,..... ;., . . . - :
. . , !. ' ~
595~L
'I'ABI,E 1
Weight Gold Gold
Treatment % of AssayDistri-
of_Samplt~ FractionsSample oz/tonbution
No Treatment Magnetic 11.54 21~26 1.92
Non-Mat7netic 88.46142 04 98.08
Low TempeOa-
turc 135 C,
30 min Magnetic11.95 132.8713.29
32 kgfm. ton
Fe(CO)5 Non-Magnetic 88.05117.68 86.71
High Tempera-
turQ up to
145~C in
23 min Magnetic14.18 200.8326.79
~otal 23 kg/
m.ton FetCO)5 Non-Magnetic85.82 90.70 73.21
EX~MPLE 2
To provide a test sample for this example, 4.7
grams of the non-magnetic fraction of a placer gold con-
centrate was blended with 195 grams of sandO Analysis of
this material showed a gold content of 84 grams per metric
ton. A one-~tn~ split of the above material (52 grams)
was placed in a rotating glass reactor and heated to
150C under nitrogen. At this temperature, the mixture
was exposed to vapors of iron carbonyl ~or one half hour
at an amount equal to about 32 kilograms of carbonyl per
metric ton oE material. Cool down was under nitrogen.
After treatment, magnetic separation was efEected by using
a Dings crossbelt separator with a 4.5 amp setting. Two
recleanings of the magnetic material were made.
The results of the above tests are set forth
in the following table:
.
' ' ,' ' , ' . , ' ' ..
,. . .
.. .. ... .... _ . .. __ _~_....... ... ....... _ ... . . _ .. . ......... .. ........ . _
~,.,.. ` ' , - ' : ' ' :
.
-
s~
T~BL~ 2
yio~d ~ (Oz/Ton~ D~
col.cent~:a~e a8 3
5(~agnetic) ~o96 225O
[Non-Magnetic) 99.04 0.29 11.7 ?
EX~MP~E 3
The sample of Clear Creek placer gold from
Colorado was diluted with silica to yield a gold content
of 1.0 kilogram per metric ton. This placer yold ore
1 was treated with 1 kilogram of iron pentacarbonyl per
¦ metric ton feed at a temperature of 122C for 15 minutes.
The iron carbonyl was injected in l.S minutes coincident
1~ with the start of the lS minute roast and the reaction
chamber was purged with nitrogen during heating and cool
down. The reactor product was magnetically separated
1 yielding a magnetic con~centrate of 57O4 kilograms per
$1 metric ton of gold (1676 ounce per ton~ and 1.6~ of the
. 20 feed. The non-magnetic tails contained 660 9 grams per
/ metric ton gold (1.95 ounce per ton). The overall gold
~, I recovery was 93.3%.
EXJ~/IPLE 4
To several non-magnetic fractions of 28- x lS0-
mesh Vulture placer from Arizona spiked with non-magnetic
28- x 150-mesh Clear Cxeek gold concentrate to a total
of 891 grams per metric ton (26 ounce per ton) wexe added
various co~mon mineralsO Several samples were made from
this mixture by adding an eYcess of loe of each of the
~ '.'.,>~
L
, ~ .. . . ~ . . . ..
.. ..
~ . . . ~ . . ~ . .
. . . ; . ..... ... , ~ ~
gangue minerals so that the gold content oE the
composited placer ore was approximately 823 grams
per metric ton (24 ounce per ton). Each of the
mixtures was then treated with 1 kilogram of iron
pentacarbonyl per metric ton of feed for 15 minutes
at 122C in a small glass rotary reactor. The treated
ore was then separated usin~ an induced magnetic roll
(IMR) separator and the products assayed for gold.
The results are glven below in Table 3.
'
-
~:~
!~
,
. ~ . .
. .
. . ~
, , .
n j;
" ~
~ : . ' . ' , '
. .:~J
- . . . ~ . .
. .
:
.. :
.
TA~L13 3
. Gold Gold
Added .M~q etic ~t % o~ ~ of Total
Muscovite Nonma3netic 278 6 ,82 698 ag.3
Calc head 100.026.5
Gypsum No~magnedic 100 0 26.6
llen~atite ~agnet;lCtic 48 7 310 8 76.5
Calc head 100.022.3
Albite l~onmat3netiC lloO.. 82o 125.0 87.4
J Dolomite ~laynetic 19.0 91.2 85.9
Nonmagnetlc 81.
Calc head 100.020.2
",~ 20 Calcite Nogmay letlc 810 o 92330 9367 86.7
'~ Silioal Nonmagnctic 98 4 1 95
Calc head 100.028.7
¦ Vulture Nonmaylletic 79 6 ~ 78.0
Vulture~ gnetict 729 82 106 0 ~ 82.0
Nonrna~i1nead100 0 26.0
Clear Creek yold collcentrate added to silica sand, :
no Vulture placer.
Assay data not available ~or tails.
~ . . . .
. ~.. ~, . . .
.. . .
9S~L
EX~MPLE S
~ synth~tic placer containing 891 grams of
gold per metric ton (26 ounce per ton) was diluked with
magnetically sc,alped Vulture placer to 27.1 grams of gold
per metric ton of feed ~0.79 once per ton)O The gold
particles con~ained in this feed material were 28- x
150-mesh. A second sample of a placer containing a low
gold content was prepared by adding 49 flakes of
65- x 100-mesh gold ~hand picked from Clear Creek
concentrate) to one kiloyram of magnetically scalped
Vulture placer. This resulted in a placer ore contain-
ing 3.4 grams of gold per metric ton of feed (0.098
ounce per ton). One kilogram samples of each of these
mixtures was then separately treated at 122C for 15
minutes with an iron pentacarbonyl dosage of one kilogram
pex metric ton of feed. The carbonyl was injected into
the reactor by a syringe pump calibrated to deliver the
required amount oE iron carbonyl in the first 1.5 minutes
of the roast. The test~results are presented in Table 4.
.
2 O , T2~BLE
Gold
Yield, Gold ~ssay, Recovery
PractionWt ~i _ oz/ton ~i of ~'otal
Magnt-3tic11.6 5.82 85.7
25 . Nonmaglletic 88.4 0.127
Calc head100.0 0.787
Mac3netic14.1 0.49 69.9
Nonmagnetic 85.1 0.034
Calc head100.0 0.098
,
. 1~ ' ' .
_ , .... . ... . . . _ ._,.. ~__._.. ..
. . ~ , ,
5~s~
.
EXAMPLE 6
Eight 90 gram samples of a simulated gold placer
ore with a si~e range of 28- x 150-mesh were subjected to
two different ~oast d~rations, iOe~ 15 minutes and 60
minutes. For each of these times two injection rates
were used, additionally ~he effec~ of varying roast time
and injection rates were analy~ed with respect to different
size fractions. All of the samples were treated with
4 kilograms of iron carbonyl per metric ton at a tempera-
ture of 120-122C. ~or the "slow" injection rate, the
iron carbonyl was injected during the entire run, while
for the "fast" injection rate, all the iron carbonyl
was injected in the first 11% of the roast time, i.e.
1.65 minutes for the 15 minute run and 6.6 minutes for
the 60 minute run. The results are given below in Tables
5 and 6. ~
: .
.
: ` .
!
;
~.
'
j.-
.~.~ ....... . . .
- , . .
~ - . -
~l~S~5~L
, .
~ ~1
~o ~ ~ ~ o ~9 o
o~ o ~ o ~r o .
~,
o
.
o o ~- ~ o co ~ o 1- ~ o
a ~ , O ~ O ,~ -r O o~ ~ o
~ I ~ OD O W ~ O ~D ~r o ~1 o~ o
a ~c ~ ~ a ''~ a ~ a
, a nl ' ~D ~" a '~'
o ~ ~ 3 ~ JO
H
~''" a a a ,a
` :.1 . . :
~:l ' ' '' ' :
14
.
~ ~ .
~ L5Sa5~L
V V
N ¦
~1
~:
I~ ~ 'r '~ ` 'I ~
1 0 1
'I
31 j
''.~ i`~ O ~
9~ o 3
~ ,, ' '; . 15
~ ~ r
:
.
~4~5~
~X~MPLE 7
~ synthetic gold placer ore was prepared from
the nonmagnetic fraction of Vulture placer spiked to ap-
proximately 920 grams gold per metric ~on f~ed with the
nonmagnetic portion of Clear Creek gold concentrate. The
size range o the feed was 28- x 150-meshO Four dif-
ferent samples were treated with 4 kilograms iron penta-
ca~bonyl per metric ton of feed for a period of 15 minutes
- at various temperatures. The results are given below in
Table 7.
~ - T~LE 7
¦ Treatment Gold Gold Distri-
Tempgratu~e Yield ~ssay, bution
C Fraction W _ oz/ton
110 Magnet.ic 9.42245. 86.2
No~lmaanetic 90.58 4.09 13.8
Calc head 100.0 26.8
115 Ma~netic 9.37 276.93.7
Nonmaanetic 90.63 1.93 6.3
Calc head 100.0 27.6
125 Maqnetic, 25.42 93.0 99.2
Nonma(Jnetic 74.58 0.26 0.8
- Calc head 100.0 23.8
135 Ma~ne-tic 73.42 39.299.98
N nmahnadic 100 0 28 82 ~0,02
:~ , ' ' :
1'1 '
!.
" ' ' ,'' '
: .'
':~' . ..
' 16
.~ , '.
. ....
r. ~
. . . . . .
.
. .
i
.~ ~
P~
A synthetiC 9 pi~od to ap-
OnmagnetiC ~aCtion tric ~on o~ feed (2
pro .th a nonmagnetic P
ullce p ~he Size range of ctor for
lS ml 5 ~ rlbed irOn ~sr
130oc at a pr6 tacarbonyl per
'i~1 ox 4 kilograms 5 minute5 of the roa
t c 5~paratin ~
d r With the ma9 n i e raction5
being Carried i fractin three PasSes
d ~ver an indUced ma9n h e testS are
an the cOmposited~size
marlze
_'3 ~.." '
.''','',^" ~ ' .
'~ ~ j; ; .;
~ ' ' ''
~"~.".'.''" ' ';
P~ ' ' ,
:, '
~ =i
1~
~ ~` C. ~
.
:. . ' ; ' ' ~: ' ~ . : : . ' '
' ~ . '~ ' ' . '
5~
o O a~ 1- ~ o o o
o ~
.
~n ~ n ~ ~ o ~ ri ~ o
~I cr~ o ~ ~ oo o o n ~r1 0 ~I ~ O O~ ~ O O O O
. ~ 3: Ir) ~ o N ~ o~ n o ~D ~ o ~`I o ~9 r7 o o:~ ~ o
~J 8 '~ ' F ~ c~o ~ U . s-,
c ~ ~ r I N o o o ~ o
. H ~,) ~ ,y
O 0 o O 0, o
F~ -
~ l ~
~ ~o
j ` . ..
~ I ~ 18 .:
~.. ~
~ .~,, i ~ = -- ~''
.: . . , ~ , . ~: . ,. . . :
. . : . : . . :
~ .: . . .:
- .' . ~.: : :
~:-. . ,
~5~
, '' " ' '
~a u r~ I o ,~
dP co o ~ n ~
I 0~ ~ ~ er Ir~ O ~ ~ ~ ~D ~ Ul 1- CO 'n
~ ~C t j o~ O
"~ ~ a~ o ~ r- o ~ ~ o c~ o ~r ~D o ~ ~- o r~ ~ o
~I ~ O .-1 OD ~ O ~ O _~
~ J . ~ D U U
,~ r~l
~- . . Ht)a~
~ ~ ' ~t
~ ~oU o o o o o o o
Ir .
._ j . , .
~' :~ ~ ~
,
: r 1 ~ ~ l
~ 'f~ . '
' ' '
'~ `,t~ j ' ' ' ' :
, . . ' ' . 19, . . .
~' . ~
., _ .. ~ _ .. __. _. _._.. _.. _ ~ .. _ .__. ,, . .. ., .. , ,_ , . . .
.. . .
~ L59~
EXAMPLE 9
R p~acer ore containing 446 grams gold per
~etric ton of fee,d (13 ounces per ton) which had been
treated with 4 ki.lograms of iron pentacarbonyl per
metric ton of ~eed for one hour at a temperature o~
125 to 130C in a large reacto~ was subje,ctecd to abr~sive
and weathering conclitions prior to magnetic separation o~
the gold. The results are given below in Table 9.
T ~
Gold Gold
Sam~]e Treatment Yield ~ssay, Recovery
Be~ore t on ~ oz/Ton
Magnetic Separation Frac 1
_ ' ' in Dry ~ir , Nonmagnetic 60 3 0.78
~ged lwo~iMro NonmacJnctic 64 1 0 87
~cJ d T~o Month'din24 yonmaqnetic 65 1 11 9
D1y ~ir, ~Y~l~osed 48 Nonmaclnet~jic 62 8 30 5 91,4
'.^~ llumidity, Stored
' ~ 24 llours in Vial
~:~ . , . .:
~<~ i ~ ~
~,, . :~
,;-3~ I . .
,'.~
.. . ~ . .
~ ' ' . ~' .
' . " ' '' , '
~I r
. ~ . . .
. .. ~
~.
9~
~X~ML'LE 10
Sam~les oE a non-magnetic fraction o~ 28-
x 65-mesh Vulture placer were spiked ~ith non-magnetic
28- x 65-mesh Clear Creek gold concentrate ~o ob~ain
1.99 kilograms of gold per metric ton of synthetic
placer (58 oz/ton~. The synthetic placer was then wet-
screened at 55-mesh to remove fines. Thereafter, each
sample was treated with iron carbonyl at varying levels
at a 122C in a small glass rotary reactor for 15 minutes.
In each case, the carbonyl was injected during the first
1.5 minute of the roast. Results are given below.
I ~ TABLe 10
¦ Carbonyl Yield, Wt. % ~ Recovery,
k/a9ton Maqnetic magnetic Magnetic mag ~
1 9.3 90.7 5608.14 59.5 87.6
2 9.6 90.4 ~915.06 51.7 91.2
a lo.l as.o 36:2.59 60 3 94.2
I
~'~ ' . ,.
U ; ' '
~' -} ' ' ,
~ ~ 21
'I :
,
:: '
