Numerical modeling and microwave roasting of refractory gold ore

INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may
be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in
reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6" x 9" black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly
to order.
U niv ersity M icrofilm s In te rn a tio n a l
A Bell & H owell In fo rm a tio n C o m p a n y
3 0 0 N o rth Z e e b R o a d . A n n A rbor. Ml 4 8 1 0 6 -1 3 4 6 U SA
3 i3 /7 fii.4 7 n n
flnn.'59i.nfinn
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
O rder N u m b er 1359880
N um erical m odeling an d m icrow ave ro astin g of refra cto ry gold
ore
Honaganahalli, Puttanna S., M.S.
University of Nevada, Reno, 1994
UMI
300 N. Zeeb Rd.
Ann Arbor, MI 48106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
University of Nevada
Reno
N um erical M odeling & M icrow ave R oasting
of
R efractory Gold Ore
A thesis subm itted in partial fulfillm ent of the requirem ents for the
degree o f Master o f Science in Metallurgical Engineering
3y
Puttanna S. Honaganahalli
Manoranjan Misra - Thesis Advisor
August 1994
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The thesis of Puttanna S. Honaganahalli is approved:
Hi
Thesis Advisor
D epartm ent Chairm an
^
s z ji
< ?.
/X
e e ^ Q
Dean G raduate School
U niversity o f Nevada
Reno
August 1994
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ii
A cknow ledgm ents
The author wishes to express his profound gratitude to Dr. M anoranjan M isra
and Dr. Indira C hatteijee for th eir continuos encouragem ent and suggestions
during the course of this study.
The auth o r also wishes to acknowledge the partial financial support provided
for this research by the Mining and Minerals Research Institute.
I would like to express my sincere th a n k s to m y fath e r, S h iv ap rasad
Honaganahalli and my late mother, Savitri Shivaprasad, for inspiring m e to be
an achiever, my grandparents for their love and blessings, w ithout w hich I
would not be able to achieve this distinction and my m ost co-operative sister,
P u shpa R am u an d brother-in-law , Dr. Ram u, who took up my dom estic
responsibilities and freed me to attain my life ambition.
And last, b u t not least, this acknowledgment would not be complete w ithout
m ention of my wife, Vany, who stoically stood behind m e irrespective of the
travails and trium phs. Her patience, fortitude and understanding catalyzed the
completion of this thesis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A bstract
Microwave roasting of refractory ores is an attra c tiv e altern ate to
conventional th e rm a l roasting. T em p eratu re m easu rem en ts an d hence
tem perature control is a major problem which is obstructing the introduction of
th is new technology to m in e ra l processing. T h is th e sis p re se n ts a
m athem atical model of determ ining tem perature continuously and suggests
control of tem perature as a function of time of roasting.
The m athem atical model was applied to two types of refractory gold
ores viz. carbonaceous and sulfidic gold ores. The model was tested for three
different pow er levels of 133, 298 and 531 kW /sq.m. The tem p eratu res
predicted by th e model and th e tran sien t tem p eratu re plots w hich enable
control of tem perature are presented here. The tem peratures for sulfidic ore
was always found to far higher th an th a t for carbonaceous ore for a given input
power and tim e of irradiation. Power absorption efficiency for sulfidic ores was
a t 55% while for carbonaceous ore it was a dism al 20%. S teady state was
achieved very rapidly in all cases and irradiation beyond this tim e would only
lead to dissipation losses of energy.
P relim in ary investigations qualitatively confirm ed th e theoretical
results. The recovery of gold with sulfidic ore was higher because it got roasted
a t th e lim ited power levels th a t was applied. Scanning electron microscope
studies showed size reduction of the ore particles which suggested capabilities
of improved mass transfer of th e lixiviant and promise of enhanced recoveries.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table of Contents
A cknow ledgm ents
A b strac t
.............................................................................................. ii
............................................................................................................... iii
T able o f C o n ten ts
L ist o f T ables
..............................................................................................iv
...................................................................................................... vii
L ist o f F ig u re s
.................................................................................................. viii
1. In tro d u c tio n
....................................................................................................1
1.1
Significance
1.2
Gold Ores
1.3
Processing of Gold ores
.........................................................................4
1.4
Pretreatm ent Processes
........................................................................ 6
1.5
.............................................................................................1
................................................................................................3
1.4.1
Roasting
........................................................................................ 8
1.4.2
Bio-Oxidation
1.4.3
Pressure Oxidation
Research Objectives
2. L ite ra tu re S u rv ey
.................................................................................8
....................................................................... 9
.............................................................................10
.........................................................................................11
2.1
Background
.......................................................................................... 11
2.2
Microwave Heating
2.3
Roasting of Sulfidic Gold Ores
2.4
Temperature Measurement
3. T h eo retical C o n sid eratio n s
.............................................................................. 14
............................................................... 19
...................................................................... 22
3.1
Theory of Dielectric Heating
3.2
Numerical Modeling
3.2.1
.............................................................16
...............................................................22
............................................................................ 24
Electromagnetic Model
............................................................... 24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
V
3.2.2
H eat Conduction Model
3.2.3
Results
3.2.4
Inferences
.......................................................................................... 36
.......................................................................................59
4. Preliminary Investigations
................................................................ 62
4.1
Objectives
4.2
Experimental Design
4.3
Materials and Equipment
4.4
..............................................................................................62
4.3.1
Ore Samples
4.3.2
Microwave Oven
Experiments
4.4.1
.......................................................................... 62
.................................................................. 64
............................................................................... 64
.........................................................................64
.............................
66
Phase-I: Pre Cyanidation Treatm ent Process
4.4.1.1
........................66
Stage 1: Construction of Tem perature C harts
.............66
........................................................................ 66
4.4.1.1.1
Carbon
4.4.1.1.2
Pyrite
.......................................................................... 66
4.4.1.1.3
Sand
............................................................................70
4.4.1.1.4
W ater
.......................................................................... 70
4.4.1.1.5
Inferences
4.4.1.2
4.4.2
............................................................. 34
...................................................................70
Stage 2: Roasting of Ores
................................................. 70
4.4.1.2.1
Carbonaceous Ores
4.4.1.2.2
Sulfidic Ores
.............................................................. 71
4.4.1.2.3
SEM Studies
..............................................................72
4.4.1.2.4
Inferences
...................................................................72
Phase-II: Cyanidation Process
4.4.2.1
Procedure
...................................................70
.......... ^.................................... 74
............................................................................74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.2.2.................Results
....................................................................75
4.4.2.3 Discussion and Inferences
5. C onclusions
................................................... 75
.............................................................................. 78
6. A p p e n d ix : Comparison of Process Economics of Microwave and
Conventional Roasting of Refractory Ores.................................. 79
7. R eferen ces
............................................................................. 82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
L ist o f Tables
Table 1.
Dielectric constant, dielectric loss and electrical conductivity of the
constituents of the mineral ore..............................................................31
Table 2.
Thermal conductivity, specific h eat and density of the constituents
of the mineral ore. .............................................................................. 35
Table 3.
Incident electric fields, incident power and absorbed power and
absorption efficiency for sulfidic ore......................................................60
Table 4.
Incident electric fields, incident power and absorbed power and
absorption efficiency for carbonaceous ore..........................................60
Table 5.
Microwave power rating determination - recommended and observed
conditions - results.................................................................................. 65
Table 6. Tem perature time correlation data for the constituents of the ore.
.................................................................................................................. 67
Table 7.
Roasting tem peratures of carbonaceous ore attained a t different
intervals of time...................................................................................... 71
Table 8.
Roasting tem peratures of sulfidic ore attained a t different intervals
of time...................................................................................................... 72
Table 9.
Gold recovery from microwave roasted and cyanided sulfidic ore.
.................................................................................................................. 76
Table 10. Gold recovery from microwave roasted and cyanided carbonaceous
ore............................................................................................................ 77
Table 11. Gold recovery from conventionally roasted and cyanided sulfidic ore.
............................................................................................................... 77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Vlll
List of Figures
Pig. 1.
World gold production.................................................................................. 2
Pig. 2.
Gold industry process flow sheet................................................................. 7
Fig. 3.
Electromagnetic spectrum......................................................................... 12
Fig. 4.
Schematic of the microwave heating arrangem ent of the m ineral ore
sample......................................................................................................... 25
Fig. 5.
Cross-section of the ore sample showing the coordinate system
Fig. 6.
Cross-section of the mineral ore sample divided into cells....................27
Fig. 7.
Sulfidic ore model........................................................................................32
Fig. 8.
Carbonaceous ore model............................................................................ 33
Fig. 9.
Power deposited in sulfidic ore a t incident electric field intensity
of lOkV/m.....................................................................................................37
Fig. 10.
Power deposited in carbonaceous ore at incident electric field
intensity of lOkV/m................................................................................... 39
Fig. 11.
Steady state tem perature distribution contours in sulfidic ore a t
incident electric field intensity of lOkV/m.............................................. 40
Fig. 12.
Steady state tem perature distribution contours in carbonaceous ore
a t incident electric field intensity of lOkV/m......................................... 41
Fig. 13.
T ransient tem perature curves for sulfidic ore a t incident electric field
intensity of lOkV/m...................................................................................43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
Fig. 14.
T ransient tem perature curves for carbonaceous ore a t incident
electric field intensity of lOkV/m..............................................................44
Fig. 15.
Power deposited in sulfidic ore a t incident electric field intensity of
15kV/m....................................................................................................... 45
Fig. 16.
Power deposited in carbonaceous ore a t incident electric field
intensity of 15kV/m...................................................................................46
Fig. 17.
Steady state tem perature distribution contours in sulfidic ore a t
incident electric field intensity of 15kV/m............................................... 47
Fig. 18.
Steady state tem perature distribution contours in carbonaceous ore
a t incident electric field intensity of 15kV/m.......................................... 48
Fig. 19.
Transient tem perature curves for sulfidic ore a t incident electric field
intensity of 15kV/m....................................................................................49
Fig. 20.
T ransient tem perature curves for carbonaceous ore a t incident
electric field intensity of 15kV/m.............................................................. 50
Fig. 21.
Temperature difference between the low, average and high
temperature cells for sulfidic ore..............................................................52
Fig. 22.
Temperature difference between the low, average and high
tem perature cells for carbonaceous ore................................................... 52
Fig. 23.
Power deposited in sulfidic ore a t incident electric field intensity
of20kV/m....................................................................................................53
Fig. 24.
Power deposited in carbonaceous ore at incident electric field
intensity of 20kV/m.................................................................................... 54
Fig. 25.
Steady state tem perature distribution contours in sulfidic ore a t
incident electric field intensity of 20kV/m............................................... 55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fig. 26.
Steady state temperature distribution contours in carbonaceous ore
a t incident electric field intensity of 20kV/m......................................... 56
Fig. 27.
T ransient tem perature curves for sulfidic ore a t incident electric field
intensity of 20kV/m................................................................................... 57
Fig. 28.
T ransient tem perature curves for carbonaceous ore a t incident
electric field intensity of 20kV/m............................................................. 58
Fig. 29.
Tem perature / time correlation chart for pyrite.................................... 68
Fig. 30. Tem perature / time correlation chart for sand....................................... 69
Fig. 31. Scanning electron microscope picture of unroasted sulfidic ore..........73
Fig. 32. Scanning electron microscope picture of roasted sulfidic ore............... 73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
Chapter 1
INTRODUCTION
1.1 Significance
The United States of America has become th e second larg est producer
of gold in the world producing 14% of world’s gold. F ig . 1. depicts the world gold
production. D uring 1991 the U.S. gold production increased a t a ra te of 2%
while the rest of the world registered a growth rate of little more th a n 1% 1. The
U.S. gold production recorded an all time high of 10.28 million ounces during
th is period th a t was ten times the production a decade ago1. This phenom enal
increase was made possible to a considerable extent by th e new developments
and positive advancem ents in th e processing technology of precious m etals.
H eap leaching, activated carbon adsorption and th e Z ad ra process of gold
recovery th a t were developed in the early 80's, have played a m ajor role in
enhancing the production of gold in U.S.A. As a re su lt of th is increase in
production th e U.S. is transform ed from a n e t im porter of gold to a n et
exporter. However, in future with the decrease in surface ore bodies, increase in
d eep er deposits and m ore findings of refracto ry ores in ad d itio n to
environm ental restrictions will all place a continuing u pw ard p ressu re on
extractio n and processing costs. If this new found p ro sp erity should be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
United States
Other
14%
22 %
Australia
South Africa
11%
27%
Canada
8%
Brazil
USSR/CIS
4%
11%
Papua New Guinea
3%
Fig. 1
World gold m ine production
to
3
consolidated and achieve higher goals the industry will have to look for more
economic ways of extraction and processing.
1.2 Gold Ores
There are various ways of classifying gold ores and one of th em is to
classify them in to three m ain categories viz. oxide ores, sulfidic ores and
carbonaceous ores. In sulfide ores very fine (sub-microscopic size) gold
particles are locked in a sulfide matrix. In carbonaceous ores th e gold particles
are adsorbed onto organic carbon. In some cases, th e ore m ay contain both of
these constituents. The oxide ores are easy to process by th e conventional
cyanidation m ethod while the la tte r types are tough to process by any of the
conventional techniques. Hence they are called refractory ores.
Sulfide ores are probably the largest group of refractory ores2. The
refractoriness of these ores is due to the complex nature of the locking of gold in
the sulfide matrix. In many cases native gold is very finely dissem inated in the
sulfide m inerals such as, pyrite, pyrrhotite, and arsenopyrite. In o th er cases
gold is associated with tellurides or contained in base m etal sulfides of lead,
copper and zinc. The finely disseminated gold particles m ay be locked a t the
grain boundaries of the m inerals. These m inerals affect th e cyanidation
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4
characteristics of th e ore3. I t h as been shown th a t th e efficiency of gold
cyanidation is directly related to the removal of sulfur from the ore2-4>5.
N ative gold is also found in association w ith carbonaceous m atter.
G enerally ores containing 0.25 - 0.8 percent of organic carbon are called
carbonaceous ores. This carbon is active and either adsorbs gold on to it or if
the gold is elsewhere in the ore it takes away (robs) th e cyanide complexes of
gold from p reg n an t solutions (preg robbing)6 causing a decrease in gold
extraction and increase in cyanide consumption. Thus even a sm all am ount of
organic carbon can poison the entire leach circuit.
In order to process refractory ores by conventional cyanidation some
form of pre tre a tm e n t such as oxidative roasting, p ressu re oxidation, bio­
oxidation or fine grinding is essential. Such treatm en t will probably still be
necessary after flotation or prior to chloride or thiourea leaching.
1.3. Processing of Gold Ores:
Cyanidation which replaced chlorination in 1887 is still th e m ost widely
used method of gold extraction for various reasons one of which being economy
of process and operational simplicity. The conventional cyanide processing of
am enable gold ores involves agitation leaching of the crushed and ground ore
with alkaline cyanide solution in air sparged tanks, usually for no more th an 24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
hours. D uring th is process cyanide extracts (complexes with) th e gold present
in the ore an d brings it into the aqueous medium. Recovery of gold from th e
aqueous m edium is accomplished either by carbon adsorption or cem entation
w ith zinc metal.
G ranular activated carbon is contacted counter-currently w ith th e leach
pulp w hen gold cyanide complexes load onto the carbon. The loaded carbon is
th en stripped by high tem perature caustic cyanide solution. The gold is th en
electrowon from the strip solution onto steel wool cathodes. The cathodes are
dried, m elted in a furnace w ith fluxing agents to remove im purities, and th e
molten gold poured into bars.
T he la te r process is also known as th e M erril-Crowe process. This
process requires filtration of the leach pulp to remove all suspended solids,
followed by vacuum deaeration to remove dissolved oxygen. M etallic zinc is
then added to the solution, and the resulting precipitate recovered by filtration,
usually in p late and fram e filter presses. The precipitate is acid leached to
remove impurities, and gold bullion poured.
Several alternate methods of extraction have been p u t forw ard b u t none
have found favor with the gold industry. Chlorination has problems of excessive
corrosion and thiourea requires an acidic circuit and is sensitive to tem perature
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
and iron concentrations in the leach slurry. F u rth er the recovery of gold from
thiourea leach liquors requires considerable investigations. However, the
sulfidic and carbonaceous ores are not amenable to direct cyanidation and need
to be pretreated . The pre treatm en t introduces an additional stage in the
process flow sheet and escalates the cash operating cost.
The general process flow sheet for extraction of gold is shown in F ig. 2 .
The final product is obtained after passing through five stages with th e crushed
and ground ore directly tak en to cyanidation tanks. W ith refractory ores a pre
tre a tm e n t stage needs to be introduced prior to cyanidation in order to
maximize recovery of m etal values.
1.4. P re tre atm en t Processes
All p re tre a tm e n t processes are based on th e principle of oxidative
destruction of th e refractory m inerals. The p re tre a tm e n t causes a porous
m atrix which greatly enhances the mass transport of the cyanide ions which in
tu rn increases th e probability of complexation w ith the gold particle by many
folds. This type of oxidation can be carried out therm ally (i.e., roasting),
chemically or biologically.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Cyanide Amenable Ore
Refractory Ore
Crushing
Crushing
Grinding
Grinding
Pretreatment
Operation
Cyanide
Leaching
Purification/
Concentration
Recovery
Refining
Fig.2 Gold industry process flow sheet.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
1.4.1 ROASTING
This is th e oldest of the pretreatm en t processes and has been used to
process sulfidic ores and concentrates since as early as 19252. Over th e years
the single and m ultiple-hearth roasters have been replaced w ith fluidized-bed
roasters. Generally the roasting tem peratures are m aintained in th e range of
450-750°C w ith ample supply of oxygen. To ro a s t arsenic ores two stage
ro astin g is commonly employed. The firs t sta g e is o p erated a t a low
tem perature and under oxygen deficient conditions. This helps in removing
arsenic as volatile AS2O3. The second stage is operated a t a high tem perature
and with ample supply of oxygen. These conditions ensure complete oxidization
of all the sulfur and carbon present in the ore. The calcine is th en subjected to
conventional cyanidation. This method of tre a tm e n t yields 90% gold recovery
a t Je rrit Canyon 7
1.4.2 BIO-OXIDATION
Thiobacillus Ferrooxidans and thiobacillus thiooxidans are two types of
bacteria th a t can oxidize sulfide m inerals. The bacteria thrive u n d er acidic
conditions and a t am bient tem perature. They require CO2 , N and P for their
growth. The rate of oxidation is generally very slow. As sulfuric acid is produced
during the process of oxidation, a n interstage n eu tralizatio n is introduced to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
keep the acidity of the solution under control. All th e dissolved sulfates are
precipitated w ith lime or limestone prior to waste disposal. Gold extraction as
high as 94% h as been reported for pyritic concentrates2. C u rren tly this
technology is being employed a t Tonkin Springs, Nevada, in Brazil, G hana
A ustralia an d South Africa. Biohydrometallurgy is gaining acceptance as it
shows promise of an economic way of processing sulfide ores.
1.4.3 PRESSURE OXIDATION:
Sulfide m inerals can be made to decompose rapidly in acidic media a t
elevated tem p eratu re and pressure, using oxygen as the principal oxidant.
Refractory sulfide ores containing greater th an 4% sulfide sulfur can be treated
autogenously to liberate gold and render the ore amenable to cyanide leaching.
The process is ru n a t high tem peratures of 180-225°C. The high tem perature
ensures th a t no elem ental sulfur is formed which could tu rn out to be
detrim ental. The base m etals are oxidized to a higher state. Since a gas phase
is involved m ass tran sfer of oxygen to th e m ineral surface could be a rate
controlling step. Pyrite and arsenopyrite are oxidized, solubilized and th en
precipitated as hem atite and iron and arsenate. This liberates most of the gold
for subsequent recovery. The autoclaved ore is neutralized and its pH raised for
subsequent cyanide leaching. Gold extraction a t S herrit Gordon is reported to
be between 87.0% to 98.5% for the ores and 92.4 to 99.4% for concentrates8.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
1.5 Research O bjectives
Among all the pre treatm ent options discussed above roasting rem ains
th e preferred process by th e gold in d u stry p rim arily because of th e v ast
experience acquired in operating roasting processes since over a century.
However conventional roasting is an energy intensive operation. This research
is an attem pt a t introducing m odem heating technology to m ineral processing
industry.
Microwave heating is an old concept which h as been tak en advantage of
by the food processing, textile, wood, paper and o th er in d u stries. Efforts a t
introducing this new technology to mineral processing are ham pered by lack of
precise tem p eratu re m easuring devices. It is th e aim of th is research to
develop a technique th a t determines and predicts th e tem perature distribution
in a mineral sample when irradiated by microwaves.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
11
Chapter 2
LITERA TU RE SURVEY
2.1 Background
The p a rt of the electromagnetic spectrum in the frequency range of 300
MHz - 300 GHz a re called as microwaves. F ig . 3 shows th e position of
microwaves on th e electromagnetic spectrum.
M icrow aves h a v e found m ajor ap p lic a tio n s in th e field of
com munications. T herm al applications of microwaves are steadily growing9.
Microwaves are popular for their therm al applications and the microwave oven
has found its place in alm ost every kitchen of the country.
The discovery th a t microwaves can be used for heating came about in
1921 w hen during a communication experim ent generation of excessive heat
was observed. The first p aten t for manufacturing microwave oven w as filed in
A m erica in 1951. I t was th e n m ainly employed for vulcanization of rubber.
W ith a b etter understanding of the power aspects of microwaves it came to be
w idely applied by th e food industry for cooking, thaw ing an d tem pering,
vaporization an d preservation. It was in the field of cooking th a t the therm al
applications of microwaves found their widest use9.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12
E le c tr o m a g n e tic R a d ia tio n
F r e m ip n o v (v ) H?.
Cosmic Rays
20
3 x 10
------y -ra y s
19
3 x 10
"
X - rays
16
3x10
"
U ltraviolet Light
14
7.89 x 10
**
' '
Visible L ight
14
3.84 x 10
Infrared Rays
a
j*
3x10
11
^
M crowaves
3 x 10
8
"*■
‘
Radio W aves
3
o
x 10
Fig. 3
..........
Electrom agnetic Spectrum
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13
C u rren tly m icrow aves a re used for d rying p urposes in paper, wood,
p harm aceu tical, tobacco, an d tex tile in d u strie s, vulcanizing of rubber,
polymerizing of plastics and sintering of ceramics. Microwaves have im portant
applications in tissue heating, the m ost promising being hypertherm ia for the
treatm en t of m alignant tissues (cancer). Agriculture is y et another field where
microwaves have found application in the trea tm en t of soils, enhancing the
rate of germination, crop protection and disinfestation10.
Effort is now on to introduce microwaves to th e m ineral processing
industry. Because microwaves h e a t specific m aterials only it is possible to
h eat or cause chemical or phase transform ation of th e desired constituent of
an ore w ithout directly affecting other m inerals11’12. T his ability to h e a t the
desired component of a m aterial spells savings in operating cost and the ability
to process those m in erals th a t were n o t am enable for processing by
conventional methods. For example, a t present, sulfidic ores are treated by
pressure oxidation or roasted in fluidized bed ro asters, or concentrated by
flotation techniques an d tre a te d pyrom etallurgically. Upon microwave
irradiation of a sulfidic ore the pyrite particle rapidly transform s into pyrrhotite
w hile th e o th er constituents of th e ore rem ain unchanged. P yrrhotite is
param agnetic and can easily be separated by a m agnetic separator12. As it is
chemically very reactive i t is more suitable for leaching th a n pyrite. This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
ability to h e a t th e desired constituent of a m a te ria l throw s open v ast
opportunities for application of microwaves in mineral processing.
2.2 Microwave Heating
The heating mechanism in the case of microwave h eatin g is different
from th a t of conventional heating. The process of h eatin g in conventional
heating is by conduction while in microwave heating it is due to dielectric
heating which is caused by dipole rotation and molecular vibration.
D ielectric h eatin g depends directly on the dielectric loss factor, an
intrinsic property of the m aterial. The pioneering study by Chen et. al.12, on
forty different m inerals shows th a t most silicates, carbonates an d sulfates,
some oxides and some sulfides are transparent to microwave energy (does not
heat) b u t other oxides are readily heated. Most sulfides, arsenides, sulfosalts
and sulfarsenides are readily heated and this causes th e ore to oxidize. They
also observed th a t the behavior of m inerals in a microwave environm ent
depends on th e ore composition. T em perature w as m easu red u sin g a
therm ovision cam era which only gave th e surface te m p eratu re. Because
heating is selective and depends on the dielectric loss of the m aterial different
p arts of the ore body will have different tem peratures and th u s th e surface
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15
tem perature is not an accurate m easure of the tem perature attain ed by the
pyrite particle.
The energy lost and the h eat generated in a m aterial depends also on the
microwave power. Published research13 has shown th a t h eatin g rate s vary
directly w ith the applied power except for some very high loss and very low-loss
m aterials. The sulfides showed a rapid increase in heating rates as power was
increased. Silica an d carbonates, th e common gangue constituents, were
tran sp a re n t to microwaves. This m aterial selectivity featu re of microwave
heating of m ineral ores enables heating of only the desired p art, th e sulfides,
without heating the associated gangue which suggests the possibility of energy
savings and economy of process.
A possible reason for th e rap id increase in h eatin g ra te s could, as
reported, be the lowering of the activation energy14*15. The precise reason for
th is is not known, b u t postulates include higher local tem p eratu res on the
m olecular level15 an d the faster heating ra te s affecting th e chemical rate
constants16.
The microwave heating of m inerals can benefit ore processing by either
bringing about a chemical reaction and oxidizing the sulfides into sulfur-dioxide
or causing cracks in the ore along and across the grain boundaries due to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16
therm al stresses which alter the liberation characteristics of the m ineral ore17
or both of the above.
2.3 Roasting of Sulfidic Gold Ores.
Long ago it was discovered th a t heating sulfidic ores in air converted
them to a form whereby they could be easily reduced by charcoal to a metallic
fo rm 18. Pyritic and arsenopyritic gold ores an d concentrates have been
processed by roasting since 19252. Over the decades single and m ultiple h earth
roasters have been replaced w ith fluidized-bed roasters.
The chem istry of p y rite roasting has been in v estig ated by m any
w orkers19. The mechanism of roasting depends on oxygen availability. U nder
oxidizing conditions (low sulfur dioxide content in th e gas phase) pyrite,
m arcasite, pyrrhotite and arsenopyrite are directly oxidized to m agnetite and
then furth er to hem atite19*20
3FeS2 + 8O2 = Fe304 +6SO2
3FeS + 5 0 2 = Fe304 + 3 S 0 2
12FeAsS + 2902 = 4Fe304 + 6AS2O3 + I 2SO2
The magnetite formed is further oxidized to form hematite:
4Fe304(s) + 02(g) = 6Fe203(s)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17
U nder reducing conditions, i.e., in a su lfu r dioxide rich atm osphere,
pyrite/arsenopyrite decomposes to pyrrohtite and sulfur in a process termed
desulfurization/dearsenification process. T his leaves a porous pyrrhotite
behind21.
FeS2 = FeS(g) + S(g)
The sulfur m igrates to the surface of the m ineral grain w here it volatilizes,
leaving a porous pyrrhotite structure. The volatilized sulfui: is rapidly oxidized
to sulfur dioxide in the presence of oxygen:
S(g) + 0 2(g) = so2
Arsenopyrite decomposes to pyrrhotite.
FeAsS(s) = FeS(s) + As(g)
The arsenic diffuses through the therm ally extended lattice and is volatilized a t
the surface, leaving porous pyrrhotite:
4As(g) + 3 S 0 2 = 2As2C>3(s)
Depending on conditions in the roaster, the arsenic trioxide m ay be oxidized to
arsenic pentoxide:
A s20 3 ( s ) + 0 2(g) = A s 2C>5(s)
This reaction is significant for it m ay lead to a fu rth er undesirable reaction
betw een hem atite and arsenic pentoxide to form ferric arsenate, typically a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18
non-porous and stable solid which tends to occlude gold and reduce subsequent
gold extraction:
Fe203(s) + As2(>5(s) = 2Fe AsO^s)
O ther sulfides - Cu, Zn and Pb - oxidize to form their respective m etal oxides.
2MS + 3 0 2 — > 2MO + 2S 02
Gold is less commonly associated with the above m inerals th a n w ith iron and
arsenic sulfides and consequently the above reaction rarely has a significant
effect on gold extraction.
The microwave roasting of a pyritic and arsenopyritic gold ore was
studied by Chen et.al.,12 in an air deficient silica tube. Pyrite transform ed to
pyrrhotite while arsenopyrite changed to pyrrhotite and iron-arsenate Similar
experim ents performed in a well aerated environm ent resulted in m agnetite
and hem atite in the calcine while arsenic was oxidized to arsenic trioxide which
w as found deposited on the walls of the oven22. Thus it is clear th a t microwave
roasting follows the same mechanisms as conventional therm al roasting giving
th e same gaseous and solid end products.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19
2.4 Temperature M easurements
Pyrom etallurgical processes are v ery sen sitiv e to v a ria tio n s in
tem perature and therefore demand accurate monitoring and precise control of
tem perature. Excessive heating leads to sintering which causes closure of the
pores and subsequent reduction in gold extraction, w hile a low tem perature
roast m ay be unsuccessful in liberating the gold particle as it does not oxidize
the sulfur to the desired extent.
Available tem p eratu re m easuring devices a re well su ited for the
conventional therm al heating. This process of h eatin g is brought about by a
transfer of h eat across a tem perature gradient from th e surface to th e interior
of the ore particle. Atomically, h eat energy is carried from one atom to another,
tow ards th e interior of the particle, due to vibrations of atom s about th eir
m ean position in the crystal lattice. Hence th e stead y sta te tem p eratu re
m easured a t the surface is representative of th e tem p eratu re of th e whole
particle.
G eneration of h e a t in a m aterial due to m icrowave irrad iatio n , as
m entioned earlier, is prim arily due to rotation of th e dipoles - a phenomenon
th a t occurs sim ultaneously in all the molecules th a t compose th e m aterial w hereas the extent of heat generated in the different molecules depends on the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
dielectric loss of the molecules. Thus if a m aterial is composed of different
regions w ith each region having molecules or atoms of a kind, these different
regions will a tta in different tem peratures despite being present in the same
m aterial. The tem perature gradient established between regions leads to h eat
tran sfer in th e conventional sense b u t only as a secondary process. Thus a
tem peratu re m easuring device should be able to m easure tem peratures in the
desired region, for e.g., the sulfide region in a sufidic ore particle, which is located
farther inside the ore particle. Given the nature of the problem it is clear th a t a
physical device cannot be employed to m easure tem perature. M easurem ent
done
12,13,22
u sin g any such physical devices gives only th e average
tem perature, if not the surface tem perature. The surface tem perature in the
case of a sulfidic ore particle is different from the actual tem perature a t the
sulfidic particle.
T hus in a situation where physical m easurem ent is not a possibility,
te m p e ra tu re s can be m easured indirectly to n ea r accuracy by employing
numerical modeling techniques.
C alculation of electromagnetic energy deposited in m a tte r h as been
perform ed by num erous w orkers for different m aterials, such as hum an
tissu es23, for th in lossy m aterials24 and in general for lossy m aterials25. The
approach to obtain the power deposited involves solving Maxwell's equations
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21
which give the distribution of electrom agnetic fields in th e region under
consideration from which the power deposited is computed. C hatteijee, I. and
M isra, M.26, have used this technique to calculate the electrom agnetic field
distribution and hence the tem perature distribution in a sam ple of coal. These
a u th o rs were th u s able to predict th e te m p eratu re d istrib u tio n during
microwave (hying of coal.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
Chapter 3
THEORETICAL CONSIDERATIONS
3.1 Theory of Dielectric Heating
The extent of absorbency of microwaves by a dielectric and the degree of
heating brought about in it is related to the m aterial's complex perm ittivity, e*
(F/m). £* is composed of a real p a rt e' (dielectric constant) and an im aginary
p a rt e"(dielectric loss factor).
e* = E '- je " = e 0(e,r -je"eff)
e’
eo
where j = V
e"
= Er
and
------ = E"e ff
e0
-l,
Eo = perm ittivity of free space (= 8 .86x l 0'12 F/m),
e'r = relative dielectric constant and
e"eff = effective dielectric loss factor14.
As microwaves p en etrate and propagate through a dielectric m aterial
th ey induce local electric fields under the influence of w hich the dipoles
(perm anent an d induced) are set into rotational motion. The resistance to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
these induced motions, due to inertial, elastic and frictional forces which are
frequently interdependent, results in a hysteresis betw een applied field and
polarization. This causes loss of electromagnetic energy and a tten u a tes the
electric field. The lost energy is transformed into h e a t resulting in volumetric
heating of the dielectric m aterial. The h eat generated in th e m edium by this
prim ary process is transm itted in the m aterial by secondary processes such
as, conduction, convection, and radiation. Convection occurs only in liquids and
ra d ia tio n occurs a t th e surface of the m aterial. The en erg y losses are
commonly described by the loss tangent (tanS)
n
e'r
27ifeoe’r
w here a is the total effective conductivity (S/m) an d f is th e frequency in
(Hz)14. The extent of heating is directly dependent on th e dielectric loss factor
of the m aterial. Since power is a practical u n it of m easure of energy, th e E.M.
power dissipated as h eat per u n it volume is proportional to th e incident
electromagnetic power P(W/m3) (i.e., to the square of the local electric field Eioc)
penetrating the same volume of the material,
P = | o l E ]oc2|
where Eioc (V/m) is the magnitude of the local field.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
O ptim um h eatin g is obtained when e" is maximum, an d not w hen tan (8) is
maximum; this is due to th e fact th a t e' decreases as resonance approaches.
3.2 Numerical Modeling
F ig . 4 shows the exposure configuration as conceived and on the basis of
which th e model is developed. The subsequent p art of this chapter is devoted to
developing a m ath em atical model the objective of w hich is to enable the
prediction of tem perature distribution in the ore. The modeling consists of two
distinct parts:
i) Electromagnetic Model:
determines the electric field inside the
material, i.e., Eioc.
ii) H eat Conduction Model
predicts the tem perature a t the point under
consideration when the electric field is known
at the point, i.e.,(Ei0C).
3.2.1 ELECTROMAGNETIC MODEL
M axwell's equations are the basis for the solution of electromagnetic
problems. These equations are solved for a 2-D situ atio n as th e num erical
solution of a 3-D model of the ore is almost prohibitive a t the higher frequencies
used in th e ex p erim en tal m easurem ents because of excessive storage
requirem ents and exceedingly long computation times.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
E M SOURCE
k (Propagation Vector)
Nj/ \ /
\ 1/ \ /
Air
"if
(Electric Field)
1.2 cm
Mineral Ore Sam ple
Fig. 4— S chem atic of th e microwave heating arran g em en t of th e mineral ore sam ple
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
26
F u rth er, a t these higher frequencies th e effect of diffraction of E.M.
waves in a direction perpendicular to the direction of propagation is rath e r
lim ited. Hence a 2-D model is a good approxim ation, p articularly w ith the
transm ission line applicator, which couples through a small contact area27.
Consider a harmonic wave incident in free space on a dielectric cylinder
of arb itrary cross section as shown in F ig . 5. For a sinusoidal wave time
harmonics are represented by (exp)1“*. The incident electric field is assumed to
be polarized along th e z direction (i.e., out of the plane of the paper). Therefore
a t any point inside the body, the total transverse electric field is represented by
the following equations.
E ‘ = zE1(x,y)
(1)
(Note: Bold letters indicate a vector quantity.)
z = unit vector in the z direction.
The ore sam ple is assum ed to have th e same perm eability as th a t of freespace (|i = Ho) and th e dielectric m aterial (ore sample) is lin e ar and isotropic
along th e z direction, b u t i t m ay be inhomogeneous w ith respect to the
transverse coordinates as follows:
e* = e * (x,y)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(2)
27
1.2 cm
Fig. 5
Cross section of the ore sample showing the coordinate system
24
1.2 cm
Fig. 6
The cross section of the ore sample is divided into 144 cells
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
In order to calculate the total incident electric field, E J, we should know the
scattered electric field, E 8, because the total field, E, is
E = E 1+ E s.
(3)
U nder the assum ed conditions the total and scattered electric field intensities
will have only the z components.
There are various ways of determ ining the scattered field28. However,
R ic h m o n d 's 28 po in t m atching tech n iq u e is chosen for reaso n s of its
applicability to dielectrics of an arb itrary shape and for accuracy of results.
The technique is based on the solution of the electric field integral equation for a
dielectric cylinder of arb itrary cross section shape. The dielectric cylinder is
divided into square cells which are sm all enough so th a t th e electric field
intensity is nearly uniform in each cell. The total electric field intensity within
each cell is initially considered to be an unknow n quantity. A system of linear
equations is obtained by enforcing a t the center of each cell th a t th e total field
m u st equal th e sum of the incident an d scattered fields. This system of
equations is solved to evaluate the electric field intensity in each cell.
The scattered field may be generated by an equivalent electric current
density J radiating in unbounded free space, where
J = jf'o)(e -
£o)E
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(4)
29
co = 2nf, angular frequency
J = polarization current density
The field of an electric current filam ent dl parallel to the z axis in free space is
given by
dE« = - z ^ H 0(2>(kp)dI
(5)
H0® (kp) = Hankel function of zero order,
p = the distance from the current filament to the observation point and
k = coVuoEo = ~ , X is the free space wave length.
A*
The increm ent of electric current which generates the scattered field is given
by
dl = JdS =j'co(e - eo)EdS
(6)
w here dS is th e increm ent of surface area on the cross section of the dielectric.
From Eqn. 5 and 6 the scattered field is given by
E»(x,y) =
4
Jter* - l)E(x',y')H0<®(kp)dx'dy'
w here (x,y) and (x',y') are the coordinates of the observation and source point
er* = complex relative dielectric constant.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(7)
30
p = V (x - x')2 + (y - y')2
Eqn. (7) is valid for the scattered field a t any p o in t inside or outside the
dielectric region. The integral field equation for th e to tal field E is obtained by
substituting Eqn. (7) in Eqn. (3)
E(x,y) + i f
J(er* - l)E (x 'y )H 0(2) (kp)dx'dy' = Ei(x,y)
(8)
We now m ake an assum ption th a t the cells are so sm all th a t th e dielectric
constant and the electric field intensity are essentially constant over each of
the cells. The division into cells is shown in Fig. 6. If Eqn.(8) is enforced a t the
center of the cell m, the following expression is obtained :
N
Em+
(if) 5
r
> . - 1)E„
n=l
J h 0<2>(kpjdx'dy' = Emi
J cell n
(9)
e*n = complex relative dielectric constant a t the center of cell n.
En = Electric field intensity a t the center of the cell n.
p = VCx’ -X m ^ + Cy'-ym)2
By taking m = 1,2,3,......., N, Eqn. (9) yields N lin e a r equations w here N
represents the total number of cells. These can be solved to determ ine the total
electric field intensity at the center of each cell (E i,E 2 ,E 3 ,E 4 ,
E n).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
31
The m atrix equation, Eqn. (9), is solved using the Method of Moments. It
h as been shown th a t the m om ent m ethod solutions for th e E.M. energy
absorbed in an object can be obtained with high degree of accuracy23*29. In this
case th e object is gold ore, shown in F ig. 7 and F ig . 8 . It is represented by an
array of 144 cells. Each cell represents the constituents of th e ore such as,
silica, pyrite, w ater or m oisture and carbon by th eir volume percentages. The
electromagnetic problem is to calculate the E.M. energy deposited in each one
of these cells due to an electric field transm itted from the E.M. source.
T a b le 1 shows the values of the dielectric properties of the various
constituents of the model ore sample obtained from various sources.
T a b le 1:
D ie le c tric c o n s ta n t, d ie le c tric lo ss a n d e le c tric a l
c o n d u c tiv ity o f th e c o n s titu e n ts o f th e m in e ra l o re.
E*
e"
a
R elativ e
R elativ e
E le c tric a l C o n d u c tiv ity
D ie le c tric
D ie le c tric
(S/m)
C o n sta n t
L oss
S a n d 30
3.8
2.00x l 0-4
2.72x10-6
W a te r30
80.0
1.6400
0.223
P y rite 26
7.0
7.6300
1.040
C a rb o n 26
3.0
0.4475
0.061
M aterial
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SAND
(S)=
88.9%
WATER
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sulfide ore model
Fig. 7
(W) = 5.5%
CARBON
(C) = 1.4%
PYRITE
(P) = 4.2%
32
SAND
(S)=
88.9%
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Carbonaceous ore m odel
Fig. 8
WATER
(W) = 5.5%
CARBON
(C) = 4.9%
PYRITE
(P) = 0.7%
33
34
3.2.2 HEAT CONDUCTION MODEL
The conductive heat transfer rate a t a point w ithin a m edium is related
to the local tem perature gradient by Fourier's law.
( 10 )
G enerally complex cases require the formation of an energy equation which
governs the tem perature distribution. The general equation of h e a t conduction
with internal heat generation and time dependence is:
kV2T + hem + hloss = pC —
where
( 11 )
C = specific h eat
p = m ass density
k = therm al conductivity
T = instantaneous tem perature
h<5m= electromagnetic energy absorbed
hioss = loss of h eat a t the surface of the ore particles through the
mechanisms of convection and radiation.
Solution of th is equation yields th e distrib u tio n of te m p e ra tu re or the
tem p eratu re history w ithin th e solid stru ctu re. A nalytical solutions are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
available for only certain well-defined geometries and even there they are too
complicated and are in the form of infinite series m aking computed values
difficult to obtain. In m ost other cases of practical in te re st such analytical
solutions do not exist. The other way of solving this equation is by num erical
methods.
In solving the above equation the h eat losses by convection and
radiation are neglected. It is solved in rectangular coordinates by the standard
implicit finite difference technique, which is stable for all size tim e steps. In
order to reduce com putational effort the m atrix equation is banded. The
th e rm a l conductivities, specific h e a ts, an d d en sities, for th e various
constituents of th e coal sam ple were obtained from th e literatu re 31 and are
shown in Table 2.
Table 2
Thermal conductivity, specific heat and density o f the
constituents o f the mineral ore.
Thermal
M aterials
Density
M.
m3
Specific Heat
W-hr
kgoc
Conductivity
W
m<>C
Sand
2250
0.620
1.20
Water
1000
1.161
0.59
Pyrite
5000
0.147
0.37
Carbon
2260
0.192
1.60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
3.2.3 RESULTS
Several incident electric field intensities were assum ed in th e numerical
calculations. They were 10,15 and 20 kV/m. The frequency of th e microwaves
chosen w as 2.45 GHz. The electric field intensity is related to the free space
power density by the following relationship:
Accordingly th e selected electric field intensities correspond to power densities
of 133 kW/m2, 298 kW/m2 and 531 kW/m2 respectively. Plane wave incidence
w as assum ed.
R efractory gold ores can prim arily be classified as sulfidic and
carbonaceous ores. Although the constituents are n early th e sam e they vary
significantly in th eir composition which makes the problem unique to each type
of ore. Hence the modeling study was done on both these types separately. The
compositions were obtained from a chemical analysis performed on th e two ore
types. On the basis of on these compositional num bers th e models w ere built
and the results follow below.
F ig . 9 shows the deposition of electrom agnetic pow er in th e model
sulfidic ore a t an incident electric field intensity of 10 kV/m. The cells w ith
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Contoured from 0 - 128,000,000 W/sqm.
F ig. 9.
Interval = 8,000,00OW/sqm
P o w e r d e p o site d in su lfid ic ore a t Ejnc o f lOkV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
highest deposition of power coincide with cells containing the high dielectric loss
constituen t viz., pyrite. It can be seen th a t the siliceous m a tte r is quite
tran sp a re n t to microwave radiation and very little power is deposited in this
region as it is a m aterial of very low dielectric loss. This resu lt is in agreem ent
w ith the theory discussed above.
Fig. 10 shows th e deposition of electromagnetic power in th e model of
carbonaceous ore a t incident electric field of 10 kV/m. H ere again th e electric
field is concentrated in and around cells w ith high dielectric loss. Carbon,
however, because of its low loss does not concentrate much of th e electric field
as is evident from these figures and hence does not influence th e h eatin g
p a tte rn to a m ajor extent. This fact is borne o u t in th e p relim in ary
experimental studies performed during the course of this project. The results of
th e se ex p erim en ts are available in th e section title d "P re lim in ary
Investigations". In com parison, m oisture (or w ater) because of its higher
dielectric loss th a n carbon, absorbs more microwave energy and contributes
appreciably to the heating effects especially in the case of carbonaceous ores.
Fig. 11 and Fig. 12 show the steady state tem perature contours in the
cross section of the sulfidic and carbonaceous ores w hen irra d iated w ith a n
incident microwave electric field of 10 kV/m. It can be seen th a t for an
irra d iatio n for about 3 m inutes the highest tem p eratu re a tta in e d by th e
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
i
Contoured from 0 - 96,000,000 W/sqm.
F ig. 10.
I
Interval = 6,000,000W/sqm
P o w e r d e p o s ite d in ca rb o n a ceo u s o r e a t Einc o f
lOkV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dftoto 30
Cotrt®ute'
Intelstva^
.300 »C.
= 10°C
tio u
co^
to u ts
t u r e a i s « '* u
te tevnPera‘ f l 0uVltn-
E S P * - *"
O'WO®^'
3Pv r# l<
01 \.pe 001
Reptod^e0
F^ e ( ^
^
UC'
^
^
0Ul')e,n''SS'0"'
41
•120
00.
Contoured from 20 -190 °C. Interval = 10°C
Fig. 12.
Steady state tem perature distribution contours
in carbonaceous ore at Einc o f lOkV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
sulfidic ore is 378°C while w ith carbon in th e carbonaceous ore it is only at
100°C. As these tem peratures are far below the oxidizing tem perature of the
sulfides and carbons which is in the range of 550-700 °C32 higher tem peratures
are desired. It is clear from the tran sien t tem perature plots shown in F ig . 13
and F ig . 14 th a t higher tem peratures cannot be attain ed by any fu rth er
increase in the length of time of irradiation as steady state is attained w ithin 4
m inutes.
To a tta in higher te m p eratu res th e power of th e m icrowaves was
increased according to th e findings of McGill, S. L., e t al.,13. F ig s. 15 and 16,
show th e electromagnetic power distribution in the sulfidic and carbonaceous
ores a t incident electric fields of 15 kV/m and F ig . 17 and F ig . 18 are the
corresponding steady state tem perature contours respectively. Thus when the
incident electric field was increased to 15 kV/m th e h ighest tem p erature
a ttain ed by th e sulfidic ore which corresponded w ith the pyrite particle was
824°C w hile th e carbon in carbonaceous ore attain ed a tem perature of only
180°C.
T he tra n sie n t tem perature plots F ig . 19 and F ig. 20 reveal a very
s trik in g fact th a t w hen th e p y rite p article h a d a tta in e d its h ig h est
tem p eratu re th e siliceous m atter was only heated to 47°C in the sulfidic ore
and only to 30°C in carbonaceous ore. The average tem perature of the sulfidic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
Highest Ore Tempearature (Pyrirt Particle)
Temperature
(°C)
300
200
Average Ore Terapeaturc
100
Lowest Ore Tempearature
j--r—
1
.02
.04
Time
Fig. 13.
.06
.08
(Hrs)
Transient tem perature curves for sulfidic ore at
Einc of lOkV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.1
44
“ J ___ i— -j-—
-i—
|—
i—
i—
i—
|—
r
Highest Ore Tempearature (Pyrirt Particle)
Temperature
(°C)
150
100
Average Ore Tempeature
Lowest Ore Tempearature
.02
.04
Time
Fig. 14.
.06
.08
(Hrs)
T ransient temperature curves for carbonaceous
ore at Einc o f lOkV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Contoured from 0 - 300,000,000 W/sqm.
F ig. 15.
Interval = 10,000,000W/sqm
P o w er d e p o site d in su lfid ic ore a t Einc o f 15kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
46
Contoured from 0 - 220,000,000 W/sqm.
Fig. 16.
Interval = 10,000,000w/sqm
P o w e r d e p o site d in ca r b o n a ce o u s ore at Einc o f
15kV/m .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
47
120
120
Contoured from 30 - 660 °C.
F ig. 17.
Interval = 30°C
S tea d y s ta te tem p e ra tu re d istr ib u tio n c o n to u r s
in su lfid ic o re a t Einc o f 15kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
48
160
Contoured from 20 - 400 °C.
F ig. 18.
Interval = 20°C
S te a d y s ta te te m p e ra tu re d istr ib u tio n con tou rs
in c a r b o n a c e o u s ore a t Einc o f 15kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
49
800
Highest Ore Tempearature (Pyrirt Particle)
o 600
4>
3
(0_
>
0)
CL
E 400
<u
H
Average Ore Tempeature
200
Lowest Ore Tempearature
*-—i
~~1
.02
.04
Time
Fig. 19.
.06
.08
(Hrs)
Transient temperature curves for sulfidic ore at
Ejnc of 15kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
-j
i
j
I
I
i
|
r
Highest Ore Tempearature (Pyrin Particle)
300
Temperature
(°C)
400
i"-n
Average Ore Tempeature
100
Lowest Ore Tempearature
.02
.04
Time
Fig. 20.
.06
.08
(Hrs)
Transient temperature curves for carbonaceous
ore at Einc of 15kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.1
51
ore w as 247°C while th a t of the carbonaceous ore w as only 110°C. F ig . 21 and
F ig . 22 depict this drastic tem perature difference.
T ests were conducted a t a higher incident electric field of 20 kV/m on
b o th th e sulfidic an d carbonaceous ore. F ig. 23 an d F ig . 24 show the
electromagnetic power distribution in the sulfidic and carbonaceous ores a t an
incident electric field of 20 kV/m and Fig. 25 and F ig . 26 is th e corresponding
steady state tem perature contours respectively.
The py rite particles in the sulfidic ore a tta in e d a tem p eratu re of
1449°C. The highest tem perature reached in the carbonaceous ore was 728°C
which w as associated w ith the pyrite particle present in th e ore. Carbon was
still only a t 300°C. At this tem perature the pyrite particle in th e sulfidic ore
would sin ter causing th e porous iron oxide structure th a t is developed during
oxidation to collapse encapsulating gold w ithin th e particle an d reducing
subsequent gold recovery. Carbon in carbonaceous ores is known to cease to be
refractory w hen roasted conventionally a t conditions sim ilar to roasting of
sulfide ores33, i.e., 550 - 750°C. As carbon does not reach th is tem perature the
ore continues to stay refractory. F ig. 27 and Fig. 28 is a plot of th e tran sien t
tem perature when the incident field is 20 kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
52
900 y
800 -■
Temperature (C)
700
600 500
400
300 200
-■
100
-
0-
Low Temp Cell 71
Fig. 21.
Avg Temp Cell 64
High Temp Cell 144
T em perature d ifferen ce b etw een the low ,
average an d h ig h tem p eratu re cells for su lfid ic ore.
900 T
800
Temperature (C)
700 -f
600
500 -400 300
200
100 +
0
Low Temp. Cell 12
Fig. 22.
Avg. Temp. Cell 118
High Temp. Cell 79
T em perature d ifferen ce b etw een th e low ,
a verage and h ig h tem peratu re ce lls for
carb on aceou s ore.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Contoured from 0 - 510,000,000 W/sqm.
F ig . 23.
Interval = 30,000,Q00W/sqm
P o w e r d e p o s ite d in su lfid ic ore a t E inc o f 20kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
54
1
1
1
!
*
I
Contoured from 0 - 400,000,000 W/sqm.
F ig. 24.
?
•
'
II
t«
Interval = 20,000,000W/sqm
P o w e r d e p o s ite d in c a r b o n a c e o u s ore a t Einc o f
20kV/m .
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
If
^ = 60° °
Co»»“ied
. O cont°^s
tttre ^
Av s t a t e
,
s te » d? ,• „ ore a t
F ig .2 5'
io s u l f t ^ °r
0, Wec°PV'i9W
rtuced w"*1Pevrn'SS'°n
peptotW*0
Contoured from 0 - 720 °C.
F ig . 26.
Interval = 40°C
S tead y s ta te te m p e r a tu r e d is tr ib u tio n c o n to u r s
in c a r b o n a c e o u s o re a t Ei„c o f 20kV /m .
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
57
Highest Ore Tempearature (Pyrirt Particle)
1250
£
1000
o
a
(0
I.
o
a
E
0)
H
750
Average Ore Tempeature
500
250
Lowest Ore Tempearature
.02
Fig. 27.
.04
.06
Time (Hrs)
.08
T ransient tem perature curves for sulfidic ore at
Einc o f 20kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.1
58
i
iHighest Ore Tempearature (Pyrirt Particle)
Temperature
(°C)
600
400
Average Ore Tempeature
200
Lowest Ore Tempearature
.02
Fig. 28.
.04
.06
Time (Hrs)
.08
Transient temperature curves for carbonaceous
ore at E inc of 20kV/m.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
.1
59
3.2.4 INFERENCES
Clearly, from the results presented here, th e sulfidic ores render very
well for microwave roasting while the carbonaceous ores are unsuitable for this
new technique.
The significance of tra n sie n t tem p eratu re plots lies in th e fact th a t
tem peratu re of h eating can be determ ined as a function of tim e a t a given
in p u t electrom agnetic power level. Thus, the key to tem p eratu re control is
tim e (duration) of irradiation for a given in p u t power level. On the contrary if
the duration of irradiation and power level is known th en th e tem perature a t
the end of th a t duration can be computed.
These results are used to estim ate the am ount of microwave power th a t
will be required to sufficiently h e a t the sample so th a t th e sulfidic ore gets
roasted a t its oxidation tem peratures. T a b le 3 shows the total incident power,
the absorbed power and absorption efficiency for the th ree incident E-fields of
10, 15 an d 20 kV/m. T a b le 4 shows the incident power, th e absorbed power
and absorption efficiency for the three incident E-fields of 10,15 and 20 kV/m.
It is observed th a t the power absorption efficiency in the case of sulfidic
ore is in agreem ent with most microwave heating sources. However in the case
of carbonaceous ores the efficiency is too low. One of th e reason for low
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Table-3
Incident electric fields, incident power, absorbed
power and absorption efficiency for sulfidic ore
E in(. (V /m )
P in , (W )
P«h„ (W )
10,000
2,653
1,476
15,000
5,968
3,321
20,000
10,610
5,836
Efficiency of power absorption = 55%
Table-4
Incident electric fields, incident power, absorbed power
and absorption efficiency for carbonaceous ore
E inc (V /m )
P in e ( W )
P a b s (W )
10,000
2,653
547
15,000
5,968
1,230
20,000
10,610
2,186
Efficiency of power absorption = 20.6%
absorption efficiency is the low dielectric loss of the m aterial. Another reason is
th e sim plicity of the irradiation system considered, i.e., plane waves which
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
61
produce standing waves in the sample create hot spots and nulls. However this
work is considered preliminary.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
62
Chapter 4
PRELIMINARY EXPERIMENTAL INVESTIGATIONS
4.1
Objectives
The objectives of this preliminary investigation is to
I)
roast the ore in a microwave oven and compare the tem perature
attained by the ore with th a t predicted by the numerical model
II)
compare the leaching characteristics of the microwave roasted ore with
th a t of conventionally roasted ore.
4.2 E xperim ental Design
The experimental procedure was divided into two phases.
Phase-I: Pre cyanidation T reatm ent Process (Roasting)
Phase-II: Cyanidation Process
Phase-I consisted of roasting the ore for different intervals of tim e a t the
b u ilt-in pow er level of a domestic microwave oven and m e asu rin g th e
te m p eratu re a t the end of the tim e interval by in sertin g a therm om eter.
P h ase-II involved conventional cyanide leaching of th e ro asted ore and
subsequent determination of gold recovery by assaying the tailings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
63
In Phase-I various stages were conceived. T em perature m easurem ent
performed by any physical devices would not provide inform ation on the high
and low tem perature regions of the ore. This would lim it the verification to only
the average tem perature of the ore. Hence S ta g e 1 was developed to overcome
this lim itation according to which the constituents of the ore in th eir pure form
would be individually roasted for different intervals of time. The am ount of pure
substance ta k en was in proportion to th e ir composition in th e ore. The
tem peratu res atta in e d would be plotted as a function of tim e. W ith the
tem perature-tim e plots handy the tem peratures attain ed by the constituents
in a real ore roasting, which was our S ta g e 2 , could be approxim ately known
(as in a real situation various interactions come into play causing the h eat flow
p attern s to differ). Thus th e scope of verification of th e firs t objective is
widened considerably though limited to approximate values of all constituents.
Given the prelim inary n atu re of the study it w as deemed satisfactory if a near
to close verification was achieved.
Phase-II is a w ell-established stra ig h t forw ard process. I t involves
leaching the ore in cyanide solution and assaying the tail for gold.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
4.3 M aterials And Equipm ent:
4.3.1 ORE SAMPLES:
Two ore types were tested during this project. One of th e ores was a
known preg robbing carbonaceous gold ore with total organic carbon content of
4.9% an d sulfides of <0.5%. The sulfides were not of concern as they were
inactive. The ore assayed to 0.362 oz/t of gold.
The second ore was a sulfidic ore. Sulfur analysis performed on it showed
th a t it contained 4% sulfur and a fire assay test yielded a gold content of 0.070
oz/t. A lthough it was a dum p grade ore it was decided to te s t th is newly
developed technique of microwave roasting on this ore for reasons of working
w ith an extreme case.
The ores were pulverized and crushed to 80% 150p size. The finely
ground ore was then packaged in zip-lock bags and kept aside for later use.
4.3.2 MICROWAVE OVEN
A lkW kitchen microwave w ith a turntable and a mode stirre r in th e
roof was bought from Sears Roebuck and Co. and installed in th e laboratory.
The tu rn table and fan ensured th a t the microwaves generated inside the
cavity were kept well stirred to avoid concentrate in any one region causing hot
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
65
spots. The m anufacturer specified the power ratin g as lkW . However it was
found according to conditions shown in T a b le 5 th a t u n d er load th e power
inside the cavity was only 820 W.
Table 5
M icrowave Pow er R ating D eterm ination
Recom m ended and O bserved C onditions
R esult
CONDITIONS
International Electro Technical Committee
Practical Conditions
Recommended Conditions
Qty. of water
1000ml
1000ml
Vessel
vol.= 1000ml
vol.= 1000ml
Specifications
O.D = 100mm
O.D = 100mm
Thickness = 3mm
Thickness = 3mm
10°C +/_ 2°C
22.5°C
Initial Temp, of
water
Final Temp.
5°C > the am bient
93°C
tem perature
Other
Stir w ater continuously
W as not possible.
Determine
t, the tim e of microwave
360s
Formula / Result
heating
„ 4187.A©
P = t(s)
820 W
A0 = change in temp.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
66
4.4 Experiments:
4.4.1 PHASE-I: PRE-CYANIDATION TREATMENT PROCESS
4.4.1.1 Stage 1: Construction of Temperature Chart:
4.4.1.1.1 Carbon
A ctivated carbon obtained from N orit Carbon Co. was ground to 80% <
150)1 size. 4.9%g of this carbon was taken in a silica crucible and attem pted to
roast for 1 minute. The carbon got charged and arced causing a fire in the oven
cavity. The experim ent w as called off. R epeated efforts a t tem p eratu re
m easurem ents failed as carbon charging was observed every time. This step of
the experiment was then terminated.
4.4.1.1.2 P yrite
P yrite obtained from W ards M inerals Supplies Co. was crushed and
ground to <150m. 4.2g of th is ore was taken in a silica crucible and heated for
different intervals of time. T a b le 6 shows the tem perature d ata obtained for
pure pyrite ore. F ig . 29 shows the variation of tem perature w ith duration of
roasting for pyrite ore. The results obtained here are not com parable w ith
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
67
sim ilar experiments conducted by Walkiewicz, J. W. et al.,17, for, th e conditions
under which these experiments were performed vary from th a t of
the other group. F urther the tem peratures obtained during the course of these
experiments are, unfortunately not reproducible.
Table 6. Temperature / Time correlation data
Time
(m)
Temp. Pyrite Temp. Sand
(°C)
(°C)
Temp.
Carbon
(°C)
Temp. Water
(°C)
1
34.5
96.9
No D ata
No D ata
2
41.9
134.7
No D ata
No D ata
3
55.2
165.2
No D ata
No D ata
4
65.2
182.4
No D ata
No D ata
5
71.9
193.0
No D ata
No D ata
6
74.5
186.7
No D ata
No D ata
7
82.9
179.7
No D ata
No D ata
8
84.5
181.3
No D ata
No D ata
9
90.3
160.1
No D ata
No D ata
10
93.5
181.2
No D ata
No D ata
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
68
Tem perature / time correlation
-- oo
chart for p y rite.
-- os
Fig. 29.
CO
©
o
o
00
e
©
o
<N
©
O ) a jn ^ B ja d x u a x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chart for san d .
69
Fig. 30.
Tem perature / time correlation
t*
o
o
o
o
CO
o
O ) aan^Bjaduiax
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
70
4.4.1.1.3 Sand
Here again very fine grained river sand (80% < 200) w as roasted for
different intervals which is shown in T ab le. 6 . The mass of sand roasted was
89g. Fig. 30 shows the variation of tem perature with duration of roasting.
4.4.1.1.4 Wafer
H eating experim ents w ith w ater could not be conducted as 5.5ml of
w a te r w as too sm all a q u a n tity to perform a c cu rate te m p e ra tu re
m easurem ents and hence this experiment was abandoned.
4.4.1.1.5 Inferences
In view of the fact th a t tem p eratu res in case of two of th e chief
com ponents could not be determ ined th e objectives of th is prelim inary
investigations were scaled down and the S ta g e 1 was no longer pursued.
4.4.1.2 Stage 2: Roasting of Ores
4.4.1.2.1 Carbonaceous Ores:
lOOg of carbonaceous ore was tak en from th e prepared sample in a
silica crucible and roasted for different intervals of time. The T a b le 7 shows
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
71
the poor response of the carbonaceous ore to microwave roasting. This roasted
ore w as th e n ta k e n to P hase-II leaching operatio n s of conventional
cyanidation.
4.4.1.2.2 Sulfidic Ores
lOOg of sulfidic ore was taken in a silica crucible and heated for different
intervals of time. T a b le . 8 shows the tem perature attain ed by th e ore a t
different intervals of time. After roasting the ore was subjected to conventional
cyanidation. During the process of roasting pungent odor of sulfur dioxide was
em anating from the microwave cavity which suggested th a t oxidation to a
lim ited extent was going on. Q ualitative confirmation of th is fact came from
SEM studies of the roasted ore.
T ab le 7.
R o a stin g te m p e ra tu re s o f c a rb o n a c e o u s o r e a tta in e d a t
d iffe re n t in te rv a ls o f tim e.
M ass
<*)
Time o f
Irradiation (min.)
Tem perature
(°C)
100
3
123.0
100
4
151.7
100
5
179.0
100
6
168.0
100
7
212.2
100
8
216.0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
T ab le 8.
R o a stin g te m p e ra tu re s o f su lfid ic o re a tta in e d a t
d iffe re n t in te rv a ls o f tim e.
M ass
<g)
Time of
Irradiation (min.)
Tem perature
(°C)
100
5
155.9
100
6
190.7
100
10
2206
4.4.1.2.3 Scanning Electron Microscope (SEM) Studies
T h at th e microwaves were oxidizing th e sulfides and increasing the
surface area of the ore was confirmed from the SEM pictures of th e roasted
ore. F ig . 31 shows the unroasted ore under a scanning electron microscope.
F ig . 32 shows the roasted ore u n d er a scanning electron microscope. It is
evident th a t the particle size in the roasted ore is drastically reduced compared
to the unroasted ore. This reduction in size due to ro astin g increases the
surface area of th e ore particles and enables b etter m ass tra n sp o rt of the
lixiviant to the m etal sites which in tu rn enhances the gold recovery.
4.4.1.2.4 Inferences
The average ro astin g te m p eratu res atta in e d in both th e cases carbonaceous ore and sulfidic ores - were far below th eir respective oxidation
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
73
F ig . 31. S c a n n in g e le c tr o n m icro sco p e p ic tu r e o f u n ro a sted
su lfid ic o re.
F ig . 32. S c a n n in g e le c tr o n m icro sco p e p ic tu r e o f ro a sted
su lfid ic o re.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
74
tem peratures. The results of the carbonaceous ore roasting experim ents were
only reinforcing the conclusion of the num erical model th a t it w as not well
suited for microwave roasting. Carbon in the carbonaceous ore needs higher
energy input to oxidize and deactivate it.
W ith respect to sulfidic ores the tem peratures attained though were not
any w here close to the modeling resu lts, m ay have been ju s t sufficient to
partially oxidize the ore. The pungent odor of sulfur dioxide em anating from the
oven cavity was an indication of oxidation of the sulfides. This partial oxidation
could have been the result of an autogenous reaction triggered inside the pyrite
particle. The leaching resu lts of Phase-II, presented in th e n ex t section,
substantiates th is conclusion.
4.4.2 PHASE-II: CYANIDATION PROCESS
4.4.2.1 Procedure
The solid to liquid ratio was m aintained a t 25:75. High cyanide level of
lOOOppm was m aintained during all th e experim ents. A lthough the usual
leaching tim e is 16 hrs. all the leaching w as conducted for 24 hrs. by the
standard bottle roll tests. After leaching the pregnant solution was filtered. The
pulp w as w ashed twice to ensure th a t th e re w as no cyanide complexes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
rem aining on the pulp. The pulp was dried and the cake crushed and thoroughly
mixed by hand. This was then assayed for gold and recovery determined.
4.4.2.2 Results
T a b le -9 a n d T a b le -10 show the recovery achieved in th e case of
roasted sulfidic and carbonaceous ores respectively. T a b le - 11 shows the
recovery achieved in the case of conventionally roasted sulfidic ore. As poor
recovery was obtained in the case of microwave roasted carbonaceous ores no
a tte m p t w as m ade to com pare its resu lts w ith conventionally ro asted
carbonaceous ores.
4.4.2.S Discussion and Inferences:
Leaching results of carbonaceous ore show th a t w hatever roasting was
achieved h ad no effect w hatsoever on m itigating the refractoriness of the ore.
In other words, the oxidation tem perature of carbons was not attained during
th e course of roasting and carbon continued to be "preg robbing". F u rth er,
ro a stin g for longer d u ratio n was not elevating th e te m p eratu re in any
appreciable m anner and on the contrary caused higher dissipation losses.
These prelim inary results corroborate the conclusions of the num erical model
and confirm th a t carbonaceous ores are not amenable to microwave roasting.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Leaching results of sulfidic ores are very encouraging. On comparison
w ith conventional roasting and leaching the recovery obtained by microwave
roasting and leaching is significantly higher. The resu lts confirm th a t some
oxidation was occurring and th a t given sufficient energy in p u t in term s of
higher power all of the ore can be oxidized and higher recovery achieved. Due to
unavailability of a higher power microwave this could not be verified during this
research project and is a fertile area for future research. To conclude this new
technique of roasting is successful in roasting sulfidic ores and th e calcine is
amenable for conventional cyanidation.
Table 9.
Gold recovery from microwave roasted and
cyanided sulfidic ore.
Head grade = 0.070 oz/t
Pretreatm ent Roast
Fire Assay Results
Extraction
t(min.)
(oz/t)
(%)
0
0.020
20.0%
4
0.042
40 0%
5
0.038
45.7%
10
0.022
68.6%
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
77
Table 10.
Gold recovery from microwave roasted and
cyanided carbonaceous ore
Head grade = 0.362 oz/t
Pretreatm ent Roast
Fire Assay Results
Extraction
t(m in.)
(oz/t)
(%)
0
0.360
Poor
3
0.360
Poor
4
0.356
Poor
5
0.356
Poor
6
0.358
Poor
7
0.356
Poor
8
0.360
Poor
Table 11.
Gold recovery from conventionally roasted and
cyanided sulfidic ore.
Head grade = 0.070 oz/t
Pretreatm ent Roast
Fire Assay Results
Extraction
Temperature(°C)
(oz/t)
(%)
700
0.042
40%
500
0.034
51%
260
0.034
51%
100
0.034
51%
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
78
C hapter 5
C o n clu sio n s
The following conclusions can be drawn from this research:
• Microwave roasting can successfully be applied to ro ast sulfidic ores
a n d th e calcine yields h ig h er gold reco v eries com pared to
conventional roasting process.
• T em p eratu re control and m easu rem en t can be achieved as a
function of tim e of roasting a t a given in cid en t electrom agnetic
power.
•
Carbonaceous ores are not suitable for this new technique of roasting
as carbon gets activated and becomes more pregrobbing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
79
A p p e n d ix
COMPARISON OF PROCESS ECONOMICS OF MICROWAVE AND
CONVENTIONAL ROASTING OF REFRACTORY ORES
Energy Estimation for the Conventional Process.
T2
Q = n ^[CpdT
where
^ = A + BT+ -7^
for sand (silica) a t Tmax = 847K
A = 4.871
Ti = 273K
B = 5.365xl0-3
T2 = 773K
D = -l.OOlxlO*5
mol.wt. = 60.0848 g/mole
n = 1 mole
T2
Q = R f(A + BT + ^ )d T
Q = R(AT + |T 2 + (Y ))773273
Q = r((4.871)(773 - 273) + ^ § ^ ( 7 7 3 2 . 2732) + -1.001xl0^_ l_
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
_1_^
80
Q = ^foii r ( 2435-5 + 1402 9475 - 237.17119) K
299415 3
2.778xlO~7kW.h
Q=(29941^ t ) ( w ) ( :
= 1.384312x10-4
)
VW h
kW.h
Q = 138 ton
The am ount of ore sample used for conventional roasting was lOOg.
lOOg = 2 moles
Energy consumed to roast 1 ton of ore by conventional method is: 276 kW .h
Energy Estimation for the Microwave Process.
Weight of ore (m) = lOOg
Roasting tim e (t) = 10 min.
Power Rating (P) = 820 W
Energy Consumed to roast lOOg of ore:
E = Pt
= 820x( 10/60)
= 136.66 W.h/lOOg
Energy consumed to roast 1 ton of ore is: 1,367 kW .h
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
81
COST ESTIMATION
Cost per kW.h = $0.03
Therefore cost of roasting 1 ton of ore by microwave m ethod = $41.00
Cost of roasting 1 ton of ore by conventional method = $8.28
Microwave ro astin g is five tim es costlier th a n conventional m ethod but
microwave roasting takes only 10 minutes, while conventional roasting takes
several hours. F u rth e r microwave method yielded a recovery of 68.6% while
conventional method gave only 51%.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
82
REFERENCES
1. Dobra, J. L. and Thomas, P. R. "The U.S.Gold In d u stry 1992" N evada
Bureau of Mines and Geology Special Publication 14, pp. 4-5
2. Jh a , M. C. "Refractoriness of Certain Gold Ores to Cyanidation: Probable
C auses an d Possible Solutions", M ineral P rocessing an d E xtractive
M etallurgy Review, Vol. 2, pp. 331-352.
3. G asparrini, C., CIM bulletin, March 1983, pp. 144-153.
4. Robinson, P. C., Trans. Inst. Min. Metall., Vol. 92, Ju n e 1983, pp. C83-C89.
5. G uinivere, R., Chap. XIV, in Proc. F irst Intl. Symp. on Precious M etals
Recovery, Reno, Nevada, June 10-14, 1984.
6. Guay, W. J., "Gold and Silver Leaching, Recovery an d Economics", Ed. by
Schlitt,W. J., et al., SME-AIME, Vol. 254,1973, pp. 102-104.
7. B arr, D. S., "Comparison of Whole Ore Roasting A lternatives" Randol 90,
pp. 165-167.
8. Berezow sky, R.M.G.S. and Weir, D. R., M inerals an d M etallurgical
Processing, May 1984, pp. 1-4
9. Thuery, J ., "Microwaves: Industrial, Scientific and M edical Applications".
Pub., by Artech House, Boston, London, pp. 19.
10. M etaxas, A. C. and M eredith, R. J. (1983) in "In d u strial Microwave
Heating". Peter Peregrinus Ltd., London.
11. Sm ith, R. D. (1984) In Microwave Power in Industries "Report ERPI-EM
3645", Electrical Power Research Institute, Palo Alto, Calif.,
12. Chen, T. J., Dutrizac, J. E., Haque, K. E., Wyslouzil, W. and Kashyap, S.,
(1984)"The Relative Transparency of M inerals to Microwave Radiation"
C anadian Metall. Quart., Vol. 23, No. 3, pp. 349-351.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
13. McGill, S. L., e t al., 1988, "The Effect of Power Level on th e Microwave
H eating of Selected Chemical Sand M inerals" Microwave Processing of
M aterials Vol. 124, pp. 247-252, M aterials Research Society, Pittsburgh,
PA.
14. Sutton, W. H., "Microwave Processing of Ceramic M aterial" in Ceramic
Bulletin, Vol. 68, No. 2, pp. 376-386,1989)
15. Lewis, e t al., 1989, "Pure K inetics an d M echanical B ehavior of
Electromagnetically Processed Polyimides". Proc. 1st A ustralian Symp. on
Microwave Power Appln., pp. 117-120, U niversity P ress W ollangang
Australia.
16. Hu, C. L. J. 1979, On-line M easurem ents of the F a st Changing Dielectric
C onstant In Oil Shale Due to High -Power Microwave H eating. IEEE
Trans. MTT Vol. 27, No. 1, pp. 38-43.
17. Walkiewicz, J. W. e t al., 1988, Microwave H eating C haracteristics of
Selected M inerals and Compounds. M inerals and M etallurgical Processing
Feb. 1988
18. Pehlke, R. D., "Unit Processes of Extractive Metallurgy" Chp. 2., pp. 7
19. Jha, M. C. and Kram er, M. J. 1985, Recovery of Gold from Arsenical Ores.
In Precious Metals: mining, Extraction and Processing. Ed. by Kudryak, V.
et al., TMS, W arrendale, PA, USA.
20. Coleman, R. B. 1990 Roasting of Refractory Gold Ores and Concentrates.
In Proc. of Gold 1990, AIME, pp. 397-410.
21. M arsden, J. and House, I., 1992, "The Chem istry of Gold Extraction" Pub.
by Ellis Horwood, NY, USA., Ch. 5.8, pp. 241.
22. Haque, K. E., "Microwave Irradiation P retreatm en t of a Refractory Gold
Concentrate" in Proc. of the Intl. Symp. on Gold M etallurgy, Winnipeg,
Canada, 1987.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
84
23. Iskander, M. F., Turner, P. F., DuBow, J. B. and Kao, J., "Two-Dimensional
technique to Calculate th e EM Power Deposition in th e H um an Body", J.
Mic. Power, Vol. 17, No. 3, pp. 175-185,1982.
24. Fu, W. and M etaxas, A., "A mathematical Derivation of Power P enetration
D epth for T h in Lossy M aterials", J. Mic. Power an d E lectrom agnetic
Energy, Vol. 27, No. 4, pp. 217-222,1992.
25. Jia, X. and Jolly, P. "Simulation of Microwave Field and Power Distribution
in a Cavity by a Three-Dimensional Finite Elem ent Method", J. Mic. Power
and Electromagnetic Energy, Vol. 27, No. 1, pp. 11-22, 1992.
26. C hatteijee, I. and M isra, M. "Electromagnetic and T herm al Modeling of
Microwave D rying of Fine Coal" Minerals and M etallurgical Processing ,
May 1991, pp. 110-114.
27. ------," Electrom agnetic energy coupler for medical applications," Proc.
IEEE. Vol. 67, pp. 1463-1465, 1979.
28. Richmond, J. H., S cattering by a Dielectric Cylinder of A rb itrary Cross
Section Shape, IEEE T rans, on Antennas and Propagation, Vol. AP-13, pp.
334-341, 1965.
29. Hagman, M. J. and Levin, R. L., (1986) IEEE Transactions On Microwave
Theory and Techniques. Vol. MTT-34, pp. 653.
30. CRC Handbook of Chem istry and Physics, 71st Edition, 1990-1991
31. Perry, R. H. and Chilton, C.H., 1984 Chemical Engineers Handbook, Sixth
Edition, McGraw Hill Book Co., New York.
32. Chen, B. and Reddy, R. G., (1990) Roasting C haracteristics of Refractory
Gold Ores, Advances in Gold and Silver Proc., Ed. F u ersten au , M. C. pp.
201-214.
33. M arsden, J. and House, I., 1992, "The Chemistry of Gold Oxidation", Pub.
by Ellis Horwood, NY, USA., Ch. 5, pp. 241.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.