close

Вход

Забыли?

вход по аккаунту

?

Microwave segregation process for nickel laterites

код для вставкиСкачать
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 th is reproduction is dependent upon the quality of th e
copy subm itted. 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.
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.
ProQuest Information and Learning
300 North Zeeb Road. Ann Arbor, Ml 48106-1346 USA
800-521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MICROWAVE SEGREGATION PROCESS FOR
NICKEL LATERITES
by
JIA MA
A thesis submitted to the Department of Chemical Engineering
in conformity with the requirements for
the degree of Master of Science (Engineering)
Queen’s University
Kingston, Ontario, Canada
February, 2002
Copyright © Jia Ma, 2002
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1*1
National Library
of Canada
Bibliothdque nationals
du Canada
Acquisitions and
Bibliographic Sen/ices
Acquisitions et
services bibliographiques
395 Waffngton Stract
Ottawa ON K1A0N4
Canada
395, rua WeVington
Ottawa ON K1A0N4
Canada
VaurH t M n i i M n c i
Our a * N a m M m n c a
The author has granted a non­
exclusive licence allowing the
National Library of Canada to
reproduce, loan, distribute or sell
copies of this thesis in microform,
paper or electronic formats.
L’auteur a accorde une licence non
exclusive permettant a la
Bibliotheque nationale du Canada de
reproduire, preter, distribuer ou
vendre des copies de cette these sous
la forme de microfiche/film, de
reproduction sur papier ou sur format
electronique.
The author retains ownership of die
copyright in this thesis. Neither the
thesis nor substantial extracts from it
may be printed or otherwise
reproduced without the author’s
permission.
L’auteur conserve la propriete du
droit d’auteur qui protege cette these.
Ni la these ni des extraits substantiels
de celle-ci ne doivent etre imprimes
ou autrement reproduits sans son
autorisation.
0-612-69314-7
Canada
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
For nearly forty years the segregation process has been investigated as a possible method
to extract nickel from nickel laterite ores. This is complex vapour phase process.
Calcium chloride is usually added as the chloridizing agent and carbon is added as a
reductant. In almost all of the nickel segregation research performed to-date, the primary
source of heat was provided by electrical energy in the form of resistance heating.
Microwave heating is a relatively new energy source and has been applied to the
segregation process in the present research.
In this work, the dielectric constants and the microwave heating behaviors of both
limonitic and silicate laterite ores, the associated minerals and the segregation reactants
have been studied. The dielectric constants of both the ores and the associated minerals
were very low. Thus, the ores and most of the minerals are difficult to heat using
microwaves. On the other hand, magnetite and charcoal were readily heated using
microwaves. Preheating improved the microwave heating characteristics of goethite,
limonite and hematite. But there was no significant effect for serpentine, kaolin or
olivine. With regards to the effects of particle size, the bulk sample temperature of the
limonitic laterite was low at both small and large particle sizes. For the silicate laterite.
the bulk sample temperature increased dramatically at particle sizes ranging from
-1 0 0
to
+150 mesh (Tyler).
With regards to the microwave segregation process, the effects of the reaction variables,
such as processing time, microwave power, silicate laterite size, the amount of charcoal,
the charcoal size, the type of chloridizing agent, the amount of calcium chloride, and the
i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
argon flow rate on the recovery of the nickel were examined. The experimental results
indicate that the optimum reaction conditions are as follows: microwave power of 700W,
reaction time of 30 minutes, silicate laterite size in the range -100 +150 mesh, calcium
chloride addition of 10%, charcoal addition of 6 % with a size of -150 mesh, argon flow
rate of 1000 cm 3/min. The effect of these operating variables on the nickel recovery were
similar to the results observed by other researchers for the conventional segregation
process.
The nickel recovery in the microwave process was relatively low (38%) and this is
attributed to the non-uniform temperature distribution with microwave heating. The
nickel grade of the ferronickel in the microwave segregation product was studied and it
was found to be higher than that of the ferronickel in the conventional segregation
process. The ferronickel particles had a diameter of about one micron and were not
uniformly distributed. Again this is due to the steep temperature gradient in the sample in
the microwave process.
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I would like to express my sincere gratitude to Dr. Christopher A. Pickles for his advice
and guidance during the course of my study. His support and encouragement were
invaluable and greatly appreciated.
I would also like to thank Dr. Edward W. Grandmaison for co-supervising my study, his
academic guidance in my courses and his support of my studies in the Department of
Chemical Engineering.
Thanks are also due to Mike Niedbala for his unending support during the performance of
my laboratory work, to Ron Hutcheon for his assistance with the dielectric constant
analysis, to Charlie Cooney and Maritza Bailey for their assistance in analyzing my
samples.
I would also like to thank all my friends, from both the Departments of Chemical
Engineering and Mining Engineering for their support, patience, and friendship.
Finally, I would like to thank my family for their love, support, and motivation.
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
ABSTRACT.......................................................................................................................
i
ACKNOWLEDGEMENTS.............................................................................................
iii
TABLE OF CONTENTS.................................................................................................
iv
LIST O F FIG U R E S......................................................................................................... viii
LIST O F TABLES............................................................................................................
xi
1. INTRODUCTION......................................................................................................
1
1.1 Nickel Laterites.......................................................................................................
1
1.1.1 Nickeliferrous limonites or nickeliferrous limonitic ores............................
1
1.1.2 Silicate nickel ores or serpentine ores..........................................................
1
1.2 Extractive Metallurgy of Laterite Ores.......................................................
2
1.2.1 Hydrometallurgical processing...................................................................
3
1.2.2 Pyrometallurgical processing........................................................................
3
1.3 The Segregation Process.......................................................................................
4
1.4 Microwave Heating in Extractive Metallurgy.....................................................
5
2. LITERATURE REVIEW .........................................................................................
8
2.1 Conventional Segregation Process.........................................................................
2.1.1 Nickel segregation process on the laboratory scale...................................
8
9
2.1.1.1 Segregation behavior o f different nickel laterite ores..................................................
9
2.1.1.2 T he effect o f operating variables on the segregation process....................................
jj
2.1.1.3 M orphology of the segregated m etal..............................................................................
j7
2.1.1.4 Beneficiation o f the segregated p ro d u c t........................................................................
jg
2.1.1.5 Analytical m ethods..............................................................................................................
20
2.1.2 Pilot-plant scale nickel segregation process...............................................
21
2.1.2.1 Industrial applications o f the segregation process in Jap an.......................................
21
2.1.2.2 T he M IN PRO -PA M CO nickel segregation process....................................................
22
2.2 Microwave Heating Characteristics of Minerals of Relevance to the
Segregation Process..............................................................................................
23
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3. THEORETICAL CONSIDERATIONS...................................................................
30
3.1 Mechanism of Nickel Segregation.........................................................................
30
3.1.1 Thermodynamic considerations....................................................................
31
3.1.1.1 G eneration o f chloridizing gases.......................................................................................
33
3 .1.1.2 Formation and volatilization o f metal chlorides...........................................................
36
3.1.1.3 Reduction o f nickel chloride..............................................................................................
-jg
3.1.2 Kinetic considerations..................................................................................
40
3.1.2.1 Rate determ ining reaction..................................................................................................
4q
3.1.2.2 Reducibility o f nickel chloride and ferrous chloride....................................................
4j
3.1.2.3 Reduction o f nickel ox id e..................................................................................................
42
3.2 Microwave Fundamentals......................................................................................
43
3.2.1 Interactions o f microwaves with materials..................................................
43
3.2.1.1 M icrowave fundam entals...................................................................................................
43
3.2.1.2 Interaction o f m icrow aves with different m aterials.....................................................
4g
3.2.1.3 Selective heating and thermal runaw ay...........................................................................
4g
3.2.1.4 Tem perature m easurem ent.................................................................................................
48
3.2.2 Microwave heating mechanisms..................................................................
49
3.2.2.1 M icrowave heating m echanism s......................................................................................
49
3.2.2.2 Com parison o f conventional heating and m icrow ave h eatin g ..................................
50
3.2.3 Microwave components................................................................................
51
3.2.3.1 M icrowave g enerators.........................................................................................................
51
3.2.3.2 W aveguides............................................................................................................................
53
3.2.3.3 A pplicators.............................................................................................................................
54
3.2.3.4 Safety of m icrow ave equipm ent........................................................................................
54
4. EXPERIM ENTAL.....................................................................................................
56
4.1 Raw Materials........................................................................................................
56
4.1.1 Laterite ores..................................................................................................
56
4.1.2 Minerals.........................................................................................................
56
4.2 Microwave Heating Behavior of Laterites and Associated Minerals.................
57
4.2.1 Equipment....................................................................................................
57
4.2.1.1 M icrowave sy stem ................................................................................................................
57
4.2.1.2 C rucibles.................................................................................................................................
59
4.2.1.3 T herm ocouple........................................................................................................................
59
4.2.1.4 Infrared pyrom eter................................................................................................................
61
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.2 Experimental procedure fo r the microwave behavior o f laterites...........
61
4.2.2.1 M axim um microwave tem perature - microwave tim e.................................................
61
4.2.2.2 M axim um m icrowave tem perature - preheat tem perature..........................................
62
4.2.2.3 M axim um m icrowave tem perature - particle size o f laterite......................................
63
4.3 Nickel Segregation Tests.......................................................................................
63
4.3.1 Microwave segregation test.........................................................................
63
4.3.2 Conventional segregation tests.....................................................................
54
4.4 Analytical Techniques...........................................................................................
57
4.4.1 Dielectric constant measurements...............................................................
57
4.4.2 % Ni in Fe-Ni................................................................................................
68
4.4.2.1 Sam ple preparation...............................................................................................................
68
4.4.2.2 Electron m icroprobe analysis............................................................................................
68
4.4.3 Recovery o f N i...............................................................................................
69
4.4.3.1 C oncentration m ethods........................................................................................................
69
4.4.3.2 Chem ical analyses of Ni in the concentrate...................................................................
70
5. RESULTS AND DISCUSSION................................................................................
72
5.1 Microwave Behavior of Nickel Laterite...............................................................
72
5.1.1 Dielectric constant measurements...............................................................
72
5.1.1.1 Lim onitic laterite and its major m inerals........................................................................
72
5.1.1.2 Silicate laterite and its m ajor m inerals............................................................................
75
5.1.1.3 Segregation sam ple under standard conditions.............................................................
77
5.1.2 Microwave heating characteristics o f laterite and associated minerals...
79
5.1.2.1 Effect o f tim e.........................................................................................................................
79
5.1.2.2 Effect o f conventional preheat tem p era tu re ...................................................................
gg
5.1.2.3 Effect o f particle size o f laterite........................................................................................
94
5.1.3 Microwave heating characteristics o f charcoal..........................................
96
5.1.3.1 Effect o f tim e.........................................................................................................................
96
5.1.3.2 Effect o f particle size o f charcoal......................................................................................
99
5.1.4 Microwave heating characteristics o f calcium chloride.............................
99
5.2 Recovery of Nickel in the Microwave and the Conventional
Segregation Process............................................................................................... 103
5.2.1 Nickel recovery in the microwave segregation process............................ 103
5.2.1.1
Effect o f reaction tim e ............................................
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
104
5.2.1.2 Effect o f m icrow ave pow er...............................................................................................
104
5.2.1.3 Effect o f p article size o f silicate laterite.........................................................................
107
5.2.1.4 Effect o f am ount o f charcoal as reducing ag en t...........................................................
107
5.2.1.5 Effect o f charcoal particle size.........................................................................................
110
5.2.1 .6 Effect o f chloridizing agent...............................................................................................
110
5.2.1.7 Effect o f am ount o f calcium chloride..............................................................................
113
5.2.1 .8 Effect o f inert gas flow rate...............................................................................................
1 13
5 .2 .1.9 Effect o f preheating o f the briquette...............................................................................
116
5.2.1.10 Effect of w ater additives....................................................................................................
5.2.2 Effect o f temperature on the nickel recovery and the grade fo r the
conventional segregation process................................................................
Iig
120
5.2.3 Nickel grade in the ferronickel..................................................................... 123
5.2.4 Microwave experiments with small briquettes............................................ 125
5.2.5 Microwave experiments with large briquettes............................................ 129
5.2.6 Morphology o f the segregation product...................................................... 132
6.
CONCLUSIONS AND RECOMMENDATIONS................................................. 139
7. R E FER EN C ES..........................................................................................................
1 42
APPENDIX A.....................................................................................................................
A
APPENDIX B....................................................................................................................
B
VITA...................................................................................................................................
V
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 2.1: Schematic diagram of the conventional segregation experimental
set-up*6 1 ...........................................................................................................
12
Figure 2.2: Effect of reaction temperature on the nickel recovery from silicate
laterite*161 .......................................................................................................
15
Figure 2.3: Effects of the addition of Ca.C\z and coke on the nickel recovery from
silicate laterite*161..........................................................................................
16
Figure 2.4: Conventional segregated product*2 1 ..............................................................
17
Figure 3.1: Equilibrium diagram for Ni-H2-HCl and Fe-H2-HCl at 950°C*u l ..............
36
Figure 3.2: Equilibrium diagrams of Ni-H2-HCl (a) Effect of temperature
Ph20 = 0.3atm (b) Effect of water vapor pressure at 950°C
(c) Vapor pressure of nickel chloride at 950°C and Pnzo^.Batm *111........
37
Figure 3.3: Microwave energy absorption as a function of effective
conductivity*9*...............................................................................................
47
Figure 3.4: Interaction of microwaves with materials*451 ...............................................
47
Figure 3.5: A typical microwave heating system *3' 1 .......................................................
Figure 3.6: Magnetron of a conventional microwave oven f48l .......................................
51
53
Figure 4.1: Schematic diagram of the microwave apparatus...........................................
58
Figure 4.2: Relationship between thermocouple wire size and sample temperature
59
Figure 4.3: Temperature of the different zones in the silicate laterite sample after
microwaving for 5 minutes............................................................................
62
Figure 4.4: Schematic diagram of the microwave segregation experimental set-up
65
Figure 4.5: Schematic diagram of the conventional segregation experimental set-up...
66
Figure 4.6: Davis Tube Tester..........................................................................................
69
Figure 5.1: The dielectric constant of limonitic laterite and its associated minerals
vs temperature as measured by MPN...........................................................
73
Figure 5.2: The dielectric constant of silicate laterite and its associated minerals
vs temperature as measured by MPN...........................................................
76
Figure 5.3: The dielectric constant of silicate segregation sample for the standard
condition vs temperature as measured by MPN...........................................
78
Figure 5.4: Half-power depth of the silicate laterite segregation sample for the
standard conditions vs temperature as measured by MPN..........................
80
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.5: Effect of time on the microwave heating characteristics of silicate
laterite and limonitic laterite..........................................................................
gi
Figure 5.6: Effect of time on the microwave heating characteristics of major
minerals in the laterites..................................................................................
g4
Figure 5.7: Effect of time on the microwave heating characteristics of the minor
minerals in the laterites..................................................................................
85
Figure 5.8: Effect of conventional preheat temperature on the microwave heating
characteristics of the silicate laterite and the limonitic laterite.....................
gg
Figure 5.9: X-Ray Diffraction (XRD) analysis of limonitic laterite and limonitic
laterite preheated to temperatures of 300°C and 450°C............................
89
Figure 5.10: Effect of conventional preheat temperature on the microwave heating
characteristics of major minerals in laterite................................................
91
Figure 5.11: Effect of conventional preheat temperature on the microwave heating
characteristics of minor minerals in laterite.................................................
93
Figure 5.12: Effect of particle size on the microwave heating characteristics of
limonitic laterite............................................................................................
95
Figure 5.13: Effect of particle size on the microwave heating characteristics of
silicate laterite...............................................................................................
97
Figure 5.14: Effect of time on the microwave heating characteristics of charcoal
9g
Figure 5.15: Effect of time on the microwave heating characteristics of different size
of charcoal....................................................................................................
100
Figure 5.16: Effect of time on the microwave heating characteristics of CaCl2-2 H 2 0
and CaCl2.......................................................................................................
102
Figure 5.17: Nickel recovery versus reaction time for the microwave segregation
process for the standard conditions............................................................. 105
Figure 5.18: Nickel recovery versus power for the microwave segregation
process for the standard conditions............................................................. 106
Figure 5.19: Nickel recovery versus silicate laterite particle sizes for the microwave
segregation process for the standard conditions......................................... ^Qg
Figure 5.20: Nickel recovery versus the amount of charcoal added for the microwave
segregation process for the standard conditions..........................................
109
Figure 5.21: Nickel recovery versus charcoal particle sizes for the microwave
segregation process for the standard conditions........................................ I l l
Figure 5.22: Nickel recovery versus type of chloridizing agent for the microwave
segregation process (The conditions were: microwave power of 700 W,
reaction time of 5 minutes, 10 g silicate laterite of —10 mesh, 5%
chloridizing agent addition, 6 % charcoal addition of -150 mesh, argon
flow rate of 1800 cm3/m in)........................................................................
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.23: Nickel recovery versus amount of calcium chloride for the microwave
segregation process for the standard conditions.......................................... j
Figure 5.24: Nickel recovery versus argon flow rate for the microwave segregation
process for the standard conditions.............................................................
115
Figure 5.25: Nickel recovery versus different conventional preheat temperatures for
the microwave segregation process for the standard conditions.................
117
Figure 5.26: Nickel recovery for various water additions for the microwave
segregation process for the standard conditions.......................................... 119
Figure 5.27: The effect of temperature on the nickel recovery for the conventional
segregation process.......................................................................................
1 21
Figure 5.28: Comparison of the nickel recoveries and grade for the microwave
process with the conventional process.........................................................
122
Figure 5.29: Nickel grade in the ferronickel at different positions on the cross-section
of the segregated product.............................................................................. 124
Figure 5.30: Conventional segregation heating process and temperature distribution
in the sam ple................................................................................................. 126
Figure 5.31: Microwave segregation heating process and temperature distribution
in the sam ple................................................................................................ 127
Figure 5.32: Standard free energy change for reactions (3.1.4) and (3.1.5)................... 128
Figure 5.33: Nickel recovery visas temperature for the conventional process (the
microwave results are superimposed on the conventional curve).............. 130
Figure 5.34: Effect of temperature on the moisture removal for silicate laterite
131
Figure 5.35: Optical micrograph of the melted segregation sample from
large briquette............................................................................................. 133
Figure 5.36: MgO-SiOa-FeO Phase Diagram156*.............................................................
134
Figure 5.37: Optical micrograph of the conventional segregation product
a. 750°C and b. 1050°C.............................................................................. l 3 5
Figure 5.38: Optical micrograph of the microwave segregation sample
a. 2 minutes, b. 5 minutes............................................................................
136
Figure 5.39 SEM micrograph of the microwave segregation sample under the
standard conditions........................................................................................
137
X
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1.1: Chemical analyses of various zones
in nickel laterites1 1...........................
2
Table 2.1: Chemical analyses of the nickel laterite ores121.............................................
9
Table 2.2: Chemical analyses of Indian ore (in
wt% ) [61 .............................................
12
Table 2.3: Segregation roasting of laterite ore: experimental conditions
investigated161..................................................................................................
13
Table 2.4: Heating rates of various materials at 2.45GHz in a resonant cavity1261
24
Table 2.5: A compilation of microwave heating rates at 2.45 GHz and maximum
............................................................
temperatures for various materials
25
Table 2.6: Results of microwave heating experiments on ore minerals (microwave
frequency: 2.45GHz; exposure: 3-5 min)'28' ..................................................
27
Table 2.7: Results of microwave heating experiments on oxides and uranium
minerals (microwave frequency: 2.45 GHz; exposure: 3-5min)1281.............
28
Table 2.8: Effect of microwave heating on the
temperature of natural minerals1251
29
Table 3.1: Standard free energy data for the reactions in the nickel segregation
process1301.........................................................................................................
31
Table 3.2: Equilibrium constants for the reactions in the nickel segregation process...
32
Table 3.3: Standard free energy for the nickel segregation process...............................
32
Table 3.4: Equilibrium partial pressures of chlorine and hydrogen chloride over
sodium chloride as functions of temperature and partial pressures of
water vapor and oxygen1111..............................................................................
34
Table 3.5: Equilibrium partial pressures of chlorine and hydrogen chloride over
calcium chloride as functions of temperature and partial pressures of
water vapor and oxygen111' .............................................................................
34
Table 4.1: Compound or element analysis by XRAL......................................................
56
Table 4.2: Composition of the crucibles (mass percent) used in the research
experiments......................................................................................................
59
Table 5.1: Critical preheat temperatures of the minerals in laterite................................
94
Table 5.2: Variables and conditions used in the microwave segregation process
103
Table 5.3: Free energy values for decomposition reactions of various halide salts
at 900°C161.......................................................................................................
110
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1. INTRODUCTION
1.1 Nickel Laterites
Laterite ores, in which nickel occurs in the form of an oxide, represent about 80 percent
of the total nickel reserves111. Laterite deposits are formed by the chemical weathering of
nickeliferrous peridotite rock. The weathering process concentrates the nickel content in
certain zones11,21. In very general terms, the lateritic nickel ores are mixtures of two
principal mineralogical zones131: limonitic laterite and silicate laterite, which will be
discussed in detail below.
1.1.1 Nickeliferrous Limonites or Nickeliferrous Limonitic Ores
They lie at the upper zone of the laterite just below the hematite zone, containing high
concentrations of iron oxide and 0.5% Ni or less121. The chemical formula of this type of
ore is: (Fe,Ni)0(0H)-nH20. The iron oxide occurs in crystalline form as the mineral
goethite (a—FeO-OH) and this mineral contains the majority of the nickel lul. The nickel
content is usually no more than 1.5%; the iron content is more than 40%; the silica
content is less than 5%; the magnesia content is less than 5%, and the cobalt content is
relatively high (Co/Ni>0.1)1-'1. The nickel occurs in a solid solution with the iron oxide.
1.1.2 Silicate Nickel Ores or Serpentine Ores
They lie at a greater depth in the ore, just above the unaltered ultrabasic rock, containing
submarginal Ni. The formula of this fraction is: (Ni,Mg)6 Si4Oi0(OH)8. The mixed
nickeliferrous silicates is a colloidal mixture of silica and nickel hydroxide with a wide
range of nickel contents111. The nickel content is usually more than 1.5%, the iron content
is lower than 20%, the magnesia content is more than 15%, the cobalt content is
l
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
relatively low (Co/Ni <0.05)141. The nickel content occurs in varying proportions in the
silicates and it is difficult to extract.
Elemental compositional analyses in the various zones in the laterite are given in Table
1.1. Only the important constituents are listed. It can be seen that the limonitic ore is
characterized by high Fe concentrations, relatively low MgO and Si0 2 , and low to
medium Ni; the silicate ore is characterized by high MgO and Si0 2 , relatively low Fe and
medium to high Ni.
Table 1.1 Chemical Analyses of Various Zones in Nickel Laterites14'31
Approximate Analysis wt%
Laterite Zones
Hematite Cap
Limonite
Silicate
Unaltered Peridotite
S i0 2
Cr2C>3
MgO
—
>1
<5
2-5
1 0 -2 0
>20
2
-5
—
0 .2 -1
<0.5
0.5-5
15-35
35-45
Ni
Co
Fe
<0 .8
0 .8 - 1 .5
1.8-3
>0.25
<0 .1
0 . 1 -0 .2
>50
40-50
0 .1
0 .0 1 -0 .0 2
1.2 Extractive Metallurgy of Laterite Ores
The recovery of nickel from laterite ores is difficult because of their complex mineralogy
and the technology which is economically viable is limited151. The mineralogical
characteristics of the ore affect the metallurgical behaviour. The choice between
hydrometailurgical
and pyrometallurgical
processing
depends on
the chemical
dissemination of the nickel in the ore. Ores that have a low magnesia-to-iron ratio
(limonitic laterite), are preferentially processed by hydrometallurgy; for ores with a high
magnesia-to-iron ratio (silicate laterite), pyrometallurgy is usually selected. This choice is
based on the fact that the silicate laterites contain suitable slag-forming minerals that are
2
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
not present in the limonitic laterites11^1. These processes will be discussed in more detail
below.
1.2.1 Hydrometallurgical Processing
The laterites vary widely in composition. At the present time, the limonitic laterites are
treated exclusively by hydrometallurgical processes112-51. In those processes, the contained
nickel oxide is dissolved under specific temperature and pressure conditions.
Alternatively, the nature of the nickel in the ore is changed so that it is preferentially
dissolved121. This process involves either selective prereduction followed by ammoniacal
leaching or for ores with a very low magnesia content, high pressure sulfuric acid
differential leaching121.
1.2.2 Pyrometallurgical Processing
The silicate laterites are treated by pyrometallurgical processes112-5’’. The technique
employs either sulphidising and smelting to obtain phase separation of an iron-nickel
matte from the gangue or reduction and smelting to produce an iron-nickel alloy121. In this
process, the ore, after being preheated and calcined, is either differentially reduced by
coal in the electric furnace or is melted electrically and poured into ladles in which it is
selectively reduced by ferrosilicon additions141.
The extraction of nickel from the laterites by the above conventional methods has proven
to be difficult. Alternative processes for nickel extraction, such as vapour phase processes
have been investigated. These processes offer the potential of significant energy savings.
Segregation roasting followed by magnetic separation or by flotation may be used for the
treatment of lateritic nickel ores. It has been observed that when the reactions are
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
complete, a ferronickel alloy is precipitated on the surface of the reductants. The nickel
can be recovered in a high grade concentrate by suitable beneficiation methods.
1.3 The Segregation Process
The segregation process, which is a vapour phase process, achieved some commercial
success for copper ores but not for nickel-bearing materials. During recent years,
extensive research has been conducted on the application of this technology to nickelcontaining laterite ores
[26
/|. In this process, the ore is heated with a chloridizing agent
and a solid reductant, and as a result, hydrogen chloride is produced. The metal oxide in
the ore reacts with the hydrogen chloride to form volatile metal chlorides. Then, the metal
chloride is converted to metal on the surface of a reducing agent. Thus, the metal is
segregated from its original sites in the ore to the surface of the solid reductant. from
which it can be more easily separated and a relatively high grade concentrate is obtained.
The chemical reactions involved include the following^46'1:
(1) Reaction of the alkali chlorides with water vapor and gangue materials to produce
hydrogen chloride. For example, with calcium chloride additions, the reaction would be
as follows:
CaCl2 (s) + H20 (g) + S i0 2 (s) = CaSi0 3 (s) + 2HC1 (g)
(1.1)
(2) Reaction of the nickel oxide in the ore with the hydrogen chloride to form volatile
nickel chloride:
NiO (s) + 2HC1 (g) = NiCl2 (g) + H20 (g)
(1.2)
(3) Reaction of the water vapour with hot carbon to produce hydrogen and carbon
monoxide:
C (s) + H20 (g) = CO (g) + H 2 (g)
(1.3)
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4) Reaction of the metal chlorides with hydrogen to produce the metal and hydrogen
chloride which is regenerated and migrates back to the ore grains:
NiCl2 (g) + H 2 (g) = Ni (s) + 2HC1 (g)
(1.4)
Lateritic ores contain iron oxides in substantial proportions. Under the reaction
conditions, FeCl2 can readily form. In the temperature range which is required for nickel
segregation, iron can also segregate. Consequently, nickel does not precipitate as a pure
metal but always as a ferronickel alloy141. The ferronickel particles on the coke can be
collected and concentrated by magnetic separation or flotation141.
As mentioned previously, considerable research has been performed on the segregation
process at the laboratory scale and also at the pilot-plant scale.
In almost all of the
research performed to-date, the primary source of heat was provided by electrical energy
in the form of resistance heating. It is the purpose of the present research to investigate
the use of microwaves as the primary energy source in the segregation process. The first
part of the study involves the microwave heating behaviour of the laterite ores and
associated minerals, while the second involves a study of the microwave segregation
process for the silicate laterite. Also the grade and recovery in the microwave process are
compared with the results for the conventional process.
1.4 Microwave Heating in Extractive Metallurgy
Microwaves are a relatively new energy source which are being applied industrially in a
number of fields. Microwave heating offers many benefits which could be applied to
extractive metallurgy. The major potential advantages of microwave heating are as
follows:
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(1) Microwave heating is different from the conventional heating methods which involve
ro t
mainly conduction and convection 1
In microwave processing, the heat is generated
within the material, and the absorption of the microwaves occurs at the molecular or
atomic level. The heating rate is fast, and the interior temperature of the sample is higher
than the surface, in contrast, to conventional heating.
(2) Selective microwave heating of an individual component in a mixture can be achieved
under some conditions, since different minerals have different microwave absorption
characteristics.
(3) Since the microwave energy is generated from electrical energy, the energy source is
relatively clean. Microwaves do not generate any combustion gases. The only gases
which need to be cleaned are those generated by the reactions in the materials being
processed191.
(4) The modem sensing devices of microwave systems make them amenable to a high
degree of automation f9l .
In the present work, the potential application of microwaves to the segregation process
for laterite ores was investigated. The major objectives of the research were as follows:
(1) To determine the dielectric constants of the silicate laterite ores and the associated
minerals.
(2) To study the microwave heating behaviors of the silicate ores and the associated
minerals.
(3) To evaluate the effects of the following variables of the nickel recovery: processing
time, microwave power, composition of the reacting mixture, size of silicate laterite, size
6
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
of the charcoal, sample size, type of chloridizing agent, inert gas flow rate, and moisture
additions.
(4) To compare the morphology of the microwave product with the conventional
segregation product.
/
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2. LITERATURE REVIEW
In this chapter, the literature pertaining to the conventional segregation process is
reviewed. Results of laboratory scale studies on the nickel segregation process are
discussed. This includes: ore type; operating variables; morphology; beneficiation of the
segregated product; analytical methods. Pilot-plant results are also reviewed. Then, the
literature available on the microwave heating behaviour of minerals of relevance to the
segregation process is discussed. Finally, some predictions are made regarding the
potential of microwaves to heat laterite ores in the segregation process.
2.1 Conventional Segregation Process
As mentioned previously, the conventional segregation process is one of the possible
methods for the treatment of low-grade refractory oxide ores. In 1923, Moulden and
Taplin discovered the segregation process for copper oxide ores[41. Over the years,
numerous attempts have been made to investigate the basic chemistry and to develop
practical systems for the copper segregation process. It was also found that some oxidized
minerals of other metals such as Sb, Bi, Co, Cu, Au, Pb, Ni, Pd, Ag, Sn, Ti could be
treated by the segregation process11011’. In particular, in 1958, Diaz observed that nickel
oxide ores would undergo the basic reactions of the segregation process’12’. From the
1960’s, extensive research programs on the application of the segregation technology to
refractory nickel oxide ores (nickel laterites) were initiated. The fundamental chemical
behaviour of these ores was determined in order to establish the most suitable conditions
for processing. In the following discussion the nickel segregation process will be
reviewed as follows:
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
•
The nickel segregation process on the laboratory scale
•
The nickel segregation process on the pilot-plant scale
2.1.1 N ickel Segregation Process on the Laboratory Scale
2.1.1.1 Segregation Behavior o f D ifferent Nickel Laterite Ores
The segregation responses of eight different types of nickel laterite ores were described
by J.K.Wright*21. Chemical analyses of the ore types, designated as A to H, are given in
Table 2.1.
Table 2.1 Chemical Analyses of the Nickel Laterite Ores[21
Ore
A
B
C
D
E
F
G
H
Ni%
1.16
1.85
1.35
1.70
0.98
Fe%
10.7
14.1
14.7
15.8
1 .2 0
33.3
41.6
46.7
1.03
0.67
2 0 .6
S i0 2%
32.0
40.5
45.3
30.1
31.1
18.2
20.3
12.3
MgO%
26.4
9.4
12.3
20.3
19.4
a i 2o 3%
5.0
4.7
5.1
4.5
4.6
CaO%
0.80
0.45
0.05
0.58
6 .2
8 .6
1.65
0.51
8 .1
0.35
0.85
0.24
1 0 .2
0 .2 2
It can be seen from Table 2.1, that ores A to E with high silica and magnesia contents are
silicate type laterites, and ores F to H with high iron oxide contents are limonitic type
laterites.
In these experiments, the reactor was a laboratory rotary kiln and the total amount of the
ore and the reagents was 1 kg. The ore was charged separately and heated to the desired
reaction temperature. The chloride and coke reagents were added directly to the hot ore
by means of a long handled spoon and mixed with the ore by the rotary motion of the
kiln. The charge was held at a temperature of 1000°C for an hour. An impressed
atmosphere of nitrogen was maintained over the charge. After the reaction, the reactor
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was allowed to cool under nitrogen gas. The typical operating conditions were: roasting
temperature: 1000°C; roasting time: 1 hour; 5% calcium chloride addition; 2% coke
addition (-0.1 mm); ore size, - 1.7 mm; kiln rotation speed, 2 rpm.
Both magnetic separation and flotation were employed to concentrate the segregated
products. Magnetic concentration was used for the silicate laterite products. However,
magnetic separation has been found to be less effective for the concentration of the
segregated limonitic laterite, because a large portion of the iron oxide in these ores is
reduced to magnetite during the segregation reactions, which subsequently reports to the
magnetic fraction. The selectivity is thus very poor. Based on these reasons, flotation is
used for the limonitic laterite products. The effectiveness of the segregation roasting of
the various ores was evaluated in terms of the grades and the recoveries of the
concentrates. The following formula was used to calculate the separation efficiency:
Efficiency = ( % extraction ) x ( enrichment ratio )112
= R (C/f )
where,
1/2
R = Ni recovery in the concentrate (%)
C = grade of the concentrate (%)
f = laterite Ni feed grade (%)
Under the same conditions, the silicate ores exhibited a higher recovery than the limonitic
types. For the silicate ores, the concentrate grades ranged from 7.4% to 18.6% with
recoveries ranging from 71.4 to 83.7%; for the limonitic ores, the concentrate grades
ranged from 7.4 to 15.4% and the recoveries ranged from 16.7 to 60.2%.
In another study, the optimum conditions and the effect of iron oxide on the segregation
behaviours of a number of different nickel oxide ores were investigated1131. The
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
experimental procedure and the segregation apparatus were similar to those described
previously121. It was found that each ore required a specific set of segregation process
conditions for optimum results. It was concluded that under the same experimental
conditions, that the silicate ores required larger amounts of CaCU and smaller amounts of
coke than the limonitic ores. They also found that the concentration of iron oxides in the
ore had an important effect on the segregation performance. At an iron content of 15 to
18%, the highest segregation efficiencies were obtained*131. The effect of iron oxide on
the segregation behaviour of a number of silicate laterite ores was also studied by
Davidson*14*. It was found that iron oxide additions up to 25% Fe2C>3 (17.4% Fe)
increased the nickel recoveries. This observation also confirms the investigation by Sole
and Taylor*13*. They indicated those ores containing approximately 25% Fe? 0 3 (17.4%
Fe) yielded the optimum segregation results. The addition of iron oxide (Fe^C^) to silicate
laterite ores increases the nickel recovery since the iron oxide reacts with the serpentine
(Mg6 [Si4 0 io](OH)io) present in the ore to form an iron-rich magnesium silicate phase.
This interaction apparently rendered the nickel contained in the serpentine more
amenable to subsequent segregation.
2.1.1.2 The Effect o f Operating Variables on the Seereeation Process
Also, the effects of operating variables on the segregation process have been examined.
The variables that were investigated were as follows: inert gas flow rate: reaction
temperature and time; type and amount of chloridizing agent and solid carbonaceous
reductant; moisture content and treatment of the ore by preheating. These effects will be
discussed in the following sections.
(1) Limonitic Laterites
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The segregation roasting of a limonitic laterite was investigated on the laboratory scale
by Mehrotra and Srinivasan[61. The laterite was from Sukinda, India. Chemical analysis of
the ore is given in Table 2.2.
Table 2.2 Chemical Analyses of Indian Ore (in wt%)^6*
Ni
1.16%
Fe
53.04%
SiOz
11.26%
a i 2o 3
4.8%
It can be seen from Table 2.2 that the ore is a limonitic type laterite in which the iron
content is high. A schematic diagram of the experimental set-up, which used a tubular
resistance furnace, is shown in Figure 2.1.
vonoc
6 0 s Outtot
Lead WIros
Moms
Furnaco Tub*
Thormocoupto
■
In o rt Gas
Rooction Mlitur*
PyroQoiioi C0CI2
Column Column
Wator
S oturator
Coppor-G ougo
Furnoco
Flow Motor
Argon Cylindor
Figure 2.1 Schematic Diagram of the Conventional Segregation Experimental Set-up [61
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
An ore size of -100 +140 mesh was used in all the experiments. The total amount of the
reactant mixture was 50 g. Argon was employed as the purging medium. All the samples
of the segregated products were chemically analyzed. The efficiency of the segregation
process under the particular operating conditions was expressed in terms of the degree of
metallization of the nickel. A number of experiments were performed to study the effects
of changing a single variable. The variables that were investigated are given in Table 2.3.
Table 2.3 Segregation Roasting of Laterite Ore: Experimental
Conditions Investigated161
Variable
Values, Conditions or Materials
Inert gas flow rate (cmJ/min)
Roasting temperature (°C)
Reaction time (min)
Calcium chloride (wt%)
Coke or charcoal (wt%)
Water content
Pretreatment of ore
50,100,150,200,250,350,500
800,850,900,950,1000
15,30,60,90,120
4,5,7,9,11
1,2,3,4,5,6.7
Dry gas; moisture-saturated gas;
Preheated to: 500°C for lh:
500°C for 2h:
700°C for lh
Solid reductant:
M aterial
Size, mesh
Chloridizing agent
Coke o r Charcoal
-60+72. -85+100, -100+140
CaCI2, MgCl2, NaCl
In Mehrotra’s research, the optimum metallization was obtained under the following
conditions: a gas flow rate of 100 cm 3/min, a temperature of 900°C, and a reaction time
of 60 min, and a charcoal addition of 3%. The most effective chloridizing agent was
found to be CaCh (The degree of metallization increased with increasing addition of the
metal chloride). A controlled amount of moisture is essential for the segregation process.
Pretreatment of the ore by heating at 500°C increased the degree of metallization. The
nickel extraction was 85%.
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Research on the nickel segregation process for limonitic laterite ores was also performed
by Iwasaki et altl4!. A 250 g sample of ore, coke and chloridizing agent were mixed
thoroughly and added into a tube furnace. The heating rate was 10°C per min, and the
sample was held at the specified temperature for
1
hr. Nitrogen was passed through the
tube at a flow rate of 600 ml/min. The optimum conditions of the test were: a CaCI2
addition of
8 .8 %,
a coke addition of 5.5%, and a temperature of 950°C. The nickel
extraction was 93%.
(2) Silicate Laterties
A number of studies on the nickel segregation process for silicate laterites have been
performed by research groups at Fuji Iron and Steel Co. Ltd1151. Iwasaki used an ore size
of minus 48 mesh[l61; while Ser and Stojsic employed an ore size of minus 10 mesh^1'1.
During the segregation reactions, the metal chloride is transported from the ore particles
to the solid reductant surfaces in the vapor-phase. Pelletization of an ore which has been
thoroughly mixed with the reagents should provide an ideal system1151. For this reason,
Takahashi made a pellet of the ground silicate laterite ore (minus 60 mesh) together with
CaCU and coke1181, but the metallurgical results were no better than loose powders.
To achieve the highest recovery, the roasting temperature must be optimized. Due to the
different compositions of the ore and the different equipment, the optimum temperature
varied. In the patent literature Sugawara and Kagaya*191 reported the optimum
temperature to be 900°C; Iwasaki reported an optimum roasting temperature of 950oC[l<s|;
and Stojsic and Ser11'* reported the temperature range of 900°C to 1000°C, whereas
Nakaf20’ claimed the best temperature range was 1100°C to 1150°C. The effect of
reaction temperature on the nickel recovery from Iwasaki’s report is show in Figure 2.2.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
80
%
Ni M
20 _d«an«r cone.
u
-6 0
su
Ni R«e. in cf«an«r
40
20
%. Ni in
800
to il
850
900
950 IOOO
Roasting Temperature,*C
1060
Figure 2.2 Effect of Reaction Temperature on the Nickel Recovery
from Silicate Laterite1161
Calcium chloride is the preferred reagent1161. The size of calcium chloride is not a critical
factor since hydrogen chloride is the major chloridizing agent and its generation is not a
rate-determining reaction1211 . The optimum level of the calcium chloride addition is in
the range of 3-10%1161. Coal and coke appeared to serve equally well as a solid reductant
and their surfaces provide nucleation sites for metallic nickel deposition. The size of the
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
coke used by the Fuji research group was minus 48 mesh[16i. The level of coke addition
ranged from 2 to 5%
The effects of the addition of CaCh and coke on the nickel
recovery are shown in Figure 2.3.
8
70
6
uto
•
a 4
60
70
80
• 3.
712
860
669
846
617
804
2
0.6
>
0
Calcium Chlorida, percent
(b) P ercent Kickei Recovery in C leaner Conce
Figure 2.3 Effects of the Addition of CaCh and Coke on the
Nickel Recovery from Silicate Laterite'[161
Iwasaki concluded that the nickel recovery was more strongly dependent on the amount
of CaCh, the higher the CaCh addition, the higher the recovery. The grade of
concentrate, however, is almost exclusively dependent on the coke content: the lower the
coke content, the higher the grade [161
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.1.1.3 Morphology o f the Segregated M etal
The morphology of the segregated products has been studied in polished sections and by
both the electron microprobe and the scanning microscope techniques by Rey et al1221.
The segregation condition was: 2.5% coke, 6 % CaCI2 at 980°C. In the polished samples,
some of the ferronickel alloy particles were rounded and were attached to the coke; some
of the metal particles did not seem to be attached to the coke. The diameters of the
particles were in the range of 5-10 microns. The scanning electron microscope (SEM)
was also used to study the metal deposit. It was found that cubic crystals of a ferronickel
alloy were apparent at the beginning of the segregation period. Later, the metal alloy
formed rounded shapes, and some cauliflower-shaped or mushroom-shaped metal
particles were observed and were interspersed with small spherical balls1221. A
photomicrograph of the segregated nickel morphologies which formed in Wright’s
research121 is shown in Figure 2.4.
Figure 2.4 Conventional Segregated Product121
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
It can be seen that the majority of the metallic particles were associated with the carbon
particles.
The metallic particles from the segregation process for the various ore types with the
various alkali chloride reagent additions were analyzed by electron microprobe analysis
(EMPA) by Wrightf2i. The segregated metailics were found to consist almost completely
of ferronickel of varying composition. It was found that there was no obvious consistent
relationship between the grades, recoveries and the metallic nickel content. The content
of nickel in the segregated metailics depended strongly on the amount of coke used in the
segregation roasting process, particularly for the limonitic ore.
Nucleation and agglomeration of metallic nickel grains on the coal panicles in the
segregation process have been studied by Hudyma^23*. The samples that were used in the
experiments contained a constant amount of NiClz.6 H 2 0 and coal and varying amounts
of CaCl2.6 H 2 0 and silica. The samples were heated at 1123K, 1223K and 1323K. It was
found that temperature was the main factor in the growth and the agglomeration of the
metallic grains. The diameter of the metal panicles increased with the reaction
temperature. It was also found that the growth of the metailics depended on the amount of
CaCl2.6 H2 0 . The metal grains were small when the content of CaCl2.6 H 2 0 was
2
'4% .
Grain growth was also inhibited when the content of CaCl2.6 H 2 0 was higher than 8%.
2.1.1.4 Beneficiation o f the Segregated Product
Both magnetic separation and flotation are the normal methods for the concentration of
the segregated products.
(1) Magnetic Separation
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Since the segregation product is ferromagnetic, it may be concentrated by magnetic
separation, particularly the product segregated from the silicate laterite ores with
relatively low iron oxide contents. For limonitic laterites, however, it has been found that
magnetic separation is less effective than for the silicate laterite, because a greater portion
of the iron oxide in these ores is reduced to magnetite which subsequently reports to the
magnetic fractions during the segregation process. The magnetite affects the magnetic
separation results. The selectivity is thus very poor.
In Wright’s research*21, a Davis Tube Tester was used to recover the ferronickel particles.
The product samples were ground in a planetary mill to 90% minus 200 mesh. In
Crawford’s research*31, a Davis Tube Tester was also used and the sample size used for
the magnetic separation was minus 100 mesh. In Iwasaki’s research*14*, some of the
roasted products were ground to minus 150 mesh and concentrated with a Sala laboratory
magnetic separator; another series of products were ground to minus 60 mesh and
concentrated with a Davis Tube Tester. The results showed that the nickel recovery was
several percent higher in the Davis Tube Tester as compared to the Sala magnetic
separator.
(2) Flotation
Flotation using a sulfhydryl collector together with a heavy metal salt, such as copper
sulfate, as an activator was found to be an effective concentration method for both the
limonitic and the silicate laterites. Batch flotation tests were used for the concentration of
the limonitic ores in Wright’s research*2*. One hundred grams of the segregated product
were used. The procedure involved the activation of the metailics with copper sulphate,
conditioning the pulp at 70°C and floating at 40-50°C. Froth flotation, which has been
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
used by Japanese workers1141, is not feasible in laboratory-scale studies161, because the
amount of the samples used in each experiment was only 50 g and the nickel content was
less than lg.
2.1.1.5 Analytical Methods
(1) Electron Microprobe Analysis
In Wright’s research121, the sample for the microprobe test was prepared as follows:
powders of the segregation product were mounted in cold setting liquid epoxy, cross­
sectioned and metallographically polished. The metallic particles were analyzed by the
electron microprobe. Ten metallic particles were examined in each sample.
(2) Chemical Analytical Methods
An analytical method for the degree of metallization of nickel in the segregation product
based on selective leaching with concentrated nitric acid was standardized and used in
Mehrotra’s lab161. This method was found to be quite accurate since one part Ni in
400,000 parts water can be detected. It is a very satisfactory method for the determination
of the degree of metallization. However, it can not be used as a method to separate Ni
from the segregation product on a commercial scale because of the high cost.
Conventional ammonia leaching can also be used to measure the degree of metallization
of nickel in the segregated product1141. A 25 g sample was placed into 300 ml of
equimolar NHs-fNR*) 2CO 3 leach solution. The temperature of the pulp was maintained
at 70°C and the pressure of oxygen was 10 atm. The solution was held under these
conditions for 3 hours, and then the solution was analyzed for nickel.
The above two methods were used for analyzing the metallization of the segregation
product. For most of the research work, the nickel recovery of the concentrate from the
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
segregation product was the major parameter utilized for judging the efficiency of the
segregation process. The nickel grade in the concentrate was chemically analyzed, and
the nickel recovery was calculated.
2.1.2 Pilot-plant Scale Nickel Segregation Process
2.1.2.1 Industrial Applications o f the Seereeation Process in Japan
A method for the extraction of nickel from limonitic laterite ores and silicate laterite ores
by the segregation process has been developed at the pilot plant scale by Nagano et altl8'.
New Caledonia gamierite (silicate laterite) and Homonhom laterite (limonitic laterite)
were used as the raw ores.
(1). New Caledonia Ore
The nickel content of the New Caledonia ore was 2.89%. Ore with a size of -60 mesh
was mixed with 3.5% -48 mesh coke and 7% CaCl2.2 H2 0 , after which they were formed
into 20-25 mm in diameter pellets. The pellet was heated at a temperature of 980°C for
one hour in a rotary kiln. The rotary kiln had an inner diameter of 1.2 m and a batch
capacity of 100 kg. The segregated product was recovered by flotation using CuS04. The
nickel content of the concentrate was 22.0%, and the nickel recovery was 90.0%. The
flotation concentrate was further refined by magnetic separation to produce an ultra-high
nickel concentrate, the nickel content was 47.0% and the nickel recovery was 87.4%. The
concentrates could be effectively used for making pure metallic nickel or nickeliferrous
alloys.
(2). Homonhom laterite
The nickel content of Homonhom laterite was 0.86%. The experimental procedure for the
Homonhom ore was the same as that used in the segregation process for the New
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Caledonia ore. However, the reaction conditions were changed. The size of the ore was
-3 6 mesh, and was combined with 4% CaCl2.2 H2 0 and 2% coke. The roasting
temperature was 1050°C, and the retention time was 1 hour. The segregated product was
recovered by flotation. The nickel content of the concentrate was 20.33% and the nickel
recovery was 52.6%.
2.1.2.2 The MINPRO-PAMCO N ickel Seereeation Process
MINPRO, a small R&D company in Sweden, and PAMCO, Pacific Metals Co. of Japan
have together developed a practical solution to the nickel segregation process. They
started their research program on the nickel segregation process in the 1970's, and they
developed a new reactor called the mechanical Kilnf241. The first pilot plant for the
treatment of 60 kg/h was made from an old ball mill which was provided with a
refractory lining. The mechanical Kiln performed satisfactory and the nickel segregation
process could be carried out.
The process was patented by MINPRO in 1973. Oxide ore from Rio Tuba in the
Philippines was used. It was crushed to minus 20 mm and dried to about 18% moisture.
The ore was calcined at 950°C in a rotary kiln, and then the calcined ore with calcium
chloride and coke were charged into the Mechanical Kiln for the segregation process. The
segregation process was operated at a temperature of 950°C as compared to over 1600°C
in the conventional smelting process, and thus there were significant energy savings. The
ferronickel product was recovered by magnetic separation. The metallic concentrate
contained 55-60% Ni and could be upgraded to over 65% Ni after regrinding. The total
Ni recovery was about 90%.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.2. Microwave Heating Characteristics of Minerals of Relevance to the Segregation
Process
In the conventional segregation process, the energy is produced by resistance heating, and
there are heat transfer limitations because of the low thermal conductivity of the laterites.
Microwave heating is an attractive alternative to conventional heating methods which
involve convection and conduction because with microwaves the heat is generated within
the material itself and high heating rates can be obtained125’. However, different minerals
can have significantly different dielectric properties and thus some minerals readily
absorb microwaves while others do not. Thus, in this section the information available in
the literature on the microwave heating behaviour of those minerals relevant to the
segregation process is described.
In 1967, Ford and Pei used a 1600 W microwave oven to study the microwave response
of various oxides and sulphides at 2.45 GHz’26’. They found that dark coloured
compounds heated rapidly to a temperature of about 1000°C. Lighter coloured materials
required a longer time but were capable of reaching higher temperatures. The results are
shown in Table 2.4. It can be seen that carbon, magnetite and nickel oxide are readily
heated by microwaves. On the other hand, alumina, hematite and magnesia can be heated
to high temperatures but the heating rate is low.
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.4 Heating Rates of Various Materials at 2.45GHz in A Resonant Cavity ^261
Compound
A120 3
C (charcoal)
CaO
C 02 O 3
CuO
CuS
Fe 20 3
Fe30 4
MgO
M0 O 3
MoS 2
Ni20 3
PbO
U 02
ZnO
FeS
Colour
white
black
white
black
black
dark blue
red
black
white
pale green
black
black
yellow
Dark Green
White
black
Heating Time (min)
24
Max.Temp (°C )
1900
0 .2
1000
40
3
4
5
200
6
1000
0.5
40
46
3
13
500
1300
750
900
1300
900
0 .1
1100
900
800
600
0 .1
4
1100
6
800
Atomic Energy of Canada Limited measured the microwave response of some materials,
and classified the heating behavior into four categories: hyperactive; active; difficult to
heat and inactive12'1. The results are shown in table 2.5. Again it can be seen that carbon,
magnetite and nickel oxide have high heating rates. Silica has both a very low heating
rate and a very low maximum temperature.
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.5 A Compilation of Microwave Heating Rates at 2.45 GHz and
Maximum Temperatures for Various Materials127’
Material
Classification
(a)
Hyperactive
Heating Rate Maximum
Reported
Temperature (°C )
°C /S
uo2
200
150
MoS2
C (charcoal)
100
2
00
Fe30 4
20
FeS2
20
CuCl
M n0 2
(b) Active
°C /min
400
Ni20 3
300
Co 20 3
200
CuO
170
Fe20 3
135
FeS
120
CuS
(c) Difficult to °C /min
Heat
a i 2o 3
80
70
PbO
MgO
33
ZnO
25
M o0 3
15
-
(d) Inactive
CaO
CaC 0 3
S i0 2
°C /min
5
5
2-5
1100
900
1000
Reference and
Notes
Note that these
materials tend to
be black, with
high thermal
conductivities
500/1000
500
450
-
1300
900
800
Violent
Violent
1000
800
600
1900
900
1300
1100
750
200
130
70
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In 1984 Chen et ai[281 reported the results of the microwave heating behaviour of different
minerals in air. A test sample of 0.5 g to 1.0 g was used. An AGA thermovision infrared
camera system was employed to measure the temperature. Because the temperature
measurement was not accurate, the temperatures were not reported. However, Chen et al
recorded the heating behaviour of the materials and also performed chemical analysis
both before and after heating. The microwave heating behavior of the minerals could be
divided into two categories. In the first category, little or no heating occurred due to the
transparency or surface reflection of the minerals. Thus, the minerals were not affected.
In the second category, considerable heat was generated in the minerals. These materials
were either stable or reacted rapidly and dissociated. The heating results are shown in
Table 2.6 and Table 2.7. It can be seen from Table 2.6 that most sulfides are readily
heated by microwaves. On the other hand as shown in Table 2.7 some oxide minerals can
be heated while others do not readily absorb microwaves.
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.6 Results of Microwave Heating Experiments on Ore Minerals (Microwave
Frequency: 2.45GHz; Exposure: 3-5 min)1281
Minerals
Power
Heating Response
Product Examination
(W)
Arsenopyite
80
Bomite
20
Chalcopyrite
15
Covellite/anilite
(60 vol%)
Galena
100
30
Nickeline/cobaltite
(3 vol%)
Pyrite
30
Pyrrhotite
50
Sphalerite
(high Fe%)
Sphalerite
(low Fe%)
Stibnite
Tennantite
(Cu 42.8%)
100
Tetrahedrite
(Cu24.9%)
100
>100
>100
100
35
S and As fumes; some
fusion. Pyrrhotite, As, Fearsenide and arsenopyrite
Some changed to bomiteHeat readily
chalcopyrite-digenite;
some unchanged
Heats readily
with Two Cu-Fe-sulphides or
emission of sulphur pyrite and Cu-Fe-sulphide
fumes
to
single
Difficult
to
heat; Sintered
Sulphur fumes emitted composition of (Cu, Fe^Ss
Heats readily
with Sintered mass of galena
much arcing
Some fused: most unaffected
Difficult to heat
Heats, some sparking
Heats readily: emission
of sulphur fumes
Heats readily
with
arcing
at
high
temperature
Difficult to heat when
cold
Does not heat
Pyrrhotite and S fumes
Some fused; most unaffected
Converted to wurtzite
No change, sphalerite
No change, stibnite
Does not heat
Difficult to heat when Fused mass of tennanitechalcopyrite; arsenic fumes
cold
emitted
Fused mass of Ag-Sb alloy.
Heats readily
PbS, tetrahedrite, Cu-Fe-Zn
sulphide
and
Cu-Fe-Pb
sulphide
27
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Table 2.7 Results of Microwave Heating Experiments on Oxides and Uranium Minerals
(Microwave Frequency: 2.45 GHz: Exposure: 3-5min)[281
Mineral
Power
(W)
> 150
Allanite
40
Cassiterite
Columbite (40 vol%)- 60
pyrochlore in silicates
(almandine 40%)
>150
Fergusonite
50
Hematite
Magnetite
Monazite
Pitchblende (90vol%)
contains chlorite,
galena, calcite
30
>150
50
Heating response
Product examination
Does not heat
Heats readily
Difficult to heat
when cold
No change, allanite
No change, cassiterite
Niobium minerals fused;
most silicates unchanged
Does not heat
Heats readily;
arcing at high
temperature
Heats readily
Does not heat
Heats readily
No change, fergusonite
No change, hematite
No change, magnetite
No change, monazite
Some fused to UC^.UsOg,
ThC>2 and Fe-Al-Ca-Si0 2
glass ;others unchanged
Walkiewicz reported data on the heating characteristics of selected minerals1251. A 1000
W, 2450 MHz commercial oven was used. A sample of 25 g or a volume of 18 ml for
low-density materials was employed. The experiments were performed under an inert
atmosphere. A Type K thermocouple with an ungrounded tip sheathed in Inconel 702 was
inserted through the roof of the oven to measure the temperature of the sample. All solid
samples were powders. The results are shown in Table 2.8 and demonstrate that clay
minerals such as albite and orthoclase are not good microwave absorbers.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.8 Effect of Microwave Heating on the Temperature of Natural Minerals1251
Mineral
Albite
Arizonite
Chalcocite
Chalcopyrite
Chromite
Cinnabar
Galena
Hematite
Magnetite
Marble
Molybdenite
Orpiment
Orthoclase
Pyrite
Pyrrhotite
Quartz
Sphalerite
Tetrahedrite
Zircon
Chemical
Composition
NaAlSi30 8
Fe20 3.3Ti0 2
Cu2S
CuFeS2
FeCr204
HgS
PbS
Fe20 3
Fe30 4
C aC0 3
M oS2
A s 2S 3
KAISi30 8
FeS2
Fei.xS
S i0 2
ZnS
Cui2Sb4Si3
ZrSi0 4
Temperature
(°C)
Time (min)
82
290
746
920
155
144
956
182
1258
74
192
92
67
1019
886
79
87
151
52
7
10
7
1
7
8
7
7
2.75
4.25
7
4.5
7
6.75
1.75
7
7
7
7
It is difficult to compare the heating rates and the temperatures of the minerals as
reported in the literature, because the surface heat loss and the power coupling factors
were different1291. However, of relevance for the microwave processing of nickel laterites,
it can be concluded that pure magnetite, pure carbon, and pure nickel oxide are very good
microwave absorbers. On the other hand, silica, magnesia, alumina, hematite and clay
minerals are not good microwave absorbers. Therefore, it would be predicted that laterite
ores would be poor microwave absorbers since the magnetite and nickel oxide contents
are very low. Thus, it would be necessary to add carbon to improve the microwave
absorption characteristics of the ore.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3. THEORETICAL CONSIDERATIONS
3.1 Mechanism of Nickel Segregation
At the present time, the segregation mechanism of nickel in the conventional segregation
process for laterite ores is not well understood. The process is complicated and includes
the generation of chloridizing gases, the chloridization of metal oxides and volatilization
of metallic chlorides, the precipitation of metal in the vicinity of the carbon, and
hydrogen chloride regeneration. As was mentioned in Section 1.3, in the temperature
range for nickel segregation, iron also segregates to some extent. All of the reactions
involved in the nickel segregation process are listed below:
Generation of Chloridizing Gas
S i0 2 (s) + CaCl2 (s) + H20 <g) = CaSiOs (s) + 2 HCI (g)
(3.1.1)
Formation and Volatilization of Metal Chlorides
NiO (s) + 2HC1 (g) = NiCl2 (g) + HzO (g)
(3.1.2)
FeO (s) + 2HC1 (g) = FeCl2 (g) + HzO (g)
(3.1.3)
Reduction of Metal Chlorides
NiCl2 (g) + H2 (g) = Ni (s) + 2 HCI (g)
(3.1.4)
FeCl2 (g) + H 2 (g) = Fe (s) + 2 HCI (g)
(3.1.5)
Side Reactions
NiO (s) + H 2 (g) = Ni (s) + H20 (g)
(3.1.6)
NiO (s) + CO (g) = Ni (s) + C 0 2 (g)
(3.1.7)
Exchange Reactions
Fe (s) + NiCl2 (g) = Ni (s) + FeCl2 (g)
(3.1.8)
NiO (s) + FeCl2 (g) = FeO (s) + NiCl2 (g)
(3.1.9)
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hydrogen Formation
C (s) + H20 (g) = CO (g) + H 2 (g)
(3.1.10)
A general description of the mechanism of nickel segregation using both thermodynamic
and kinetic considerations will be given in the following paragraphs.
3.1.1 Thermodynamic Considerations
Thermodynamic considerations are important in the segregation process as they can
predict the feasibility of a reaction. They assist in defining optimum operational
parameters and also provide a general picture concerning the relative feasibility of the
process. The thermodynamic calculations for the segregation reactions are shown in
Tables 3.1 and 3.2. The standard free energy data for the reactions in the nickel
segregation process at 900°C and 1000°C were taken from literature130'.
Table 3.1 Standard Free Energy Data for the Reactions in the
Nickel Segregation Process'30'
Standa rd Free
En<-rgy
(kca /mol)
AG° at AG° at
900°C
1000°C
Reaction
1
2
3
4
5
6
7
8
9
Si0 2 (s) + CaCl2 (s) + H20 (g) = CaSi0 3 (s)+ 2HCI (g)
NiO (s) + 2HC1 (g) = NiCI2 (g)+ H20(g)
FeO (s) + 2HC1 (g) = FeCl2 (g) + HzO (g)
NiCl2 (g) + H 2 (g) = Ni (s) + 2 HCI (g)
FeCl2 (g) + H 2 (g) = Fe (s) + 2 HCI (g)
NiO (s) + H2 (g) = Ni (s) + H20 (g)
NiO (s) + CO (g) = Ni (s) + C 0 2 (g)
Fe (s) + NiCl2 (g) = Ni (s) + FeCl2 (g)
NiO (s) + FeCl2 (g) = FeO (s) + NiCl2 (g)
2 .0
0 .0
8.6
4.2
-2 0 .2
-3.0
-11.6
-11.0
-17.2
4.4
7.0
3.5
-19.4
-2.6
-12.4
-11.0
-16.8
3.5
The equilibrium constants were calculated using the equation:
Ln K = -AG°/RT
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.2 Equilibrium Constants for the Reactions in the Nickel Segregation Process
Equilibrium Constants
K at
900°C
K at
1000°C
0.9998
1
0.9999
0.9993
0.9999
0.9997
1 .0 0 2 0
1.0018
K 5= ( P HCl)2/(PH 2)(PFeC.2)
1.0003
1 .0 0 0 2
K<s= (P h 2 o ) /( P h 2)
1 .0 0 1 1
1 .0 0 1 1
K7= (PC02 ) /(Pco)
1 .0 0 1 1
1 .0 0 1 0
K g = (PFeCI2 ) / ( P N1CI2)
1.0017
0.9990
1.0016
0.9997
Reaction
1
2
K
S i0 2 (s) + CaCl2 (s) + HzO (g) = CaSi0 3 K,=
(s) + 2HCI (g)
( P h C|)2/ P h 20
NiO (s) + 2HC1 (g) = NiCl2 (g)+ H20(g)
k 2=
(PN 1C12) (P h2o) / (P h c i) 2
3 FeO (s) + 2HC1 (g) = FeCl2 (g) + H20 (g)
k 3=
(PFeCI2) (PH2o)/(PHCl) 2
4 NiCl2 (g) + H2 (g) = Ni (s) + 2 HCI (g)
5 FeCl2 (g) + H2 (g) = Fe (s) + 2 HCI (g)
6 NiO (s) + H2 (g) = Ni (s) + H20 (g)
7 NiO (s) + CO (g) = Ni (s) + C 0 2 (g)
8 Fe (s) + NiCl2 (g) = Ni (s) + FeCl2 (g)
9 NiO (s) + FeCl2 (g) = FeO (s) + NiCl2 (g)
K 4= < P h o ) “/ ( P h 2) (PNiCI2)
K 9 = (PNiC!2)/(P FeC12)
Assuming, for the present, that hydrogen chloride is the primary chloridizer and that free
hydrogen is available from the carbonaceous reductant, then the segregation process
simply involves a chloridization/reduction cycle. In the other words, reactions 3.1.2 and
3.1.4 may be combined to give overall values of AG0segregaiion- The standard free energies
of the reactions from Table 3.1 are employed. The overall free energies of the segregation
process for both nickel and iron are shown in Table 3.3.
Table 3.3 Standard Free Energy for the Nickel Segregation Process
Standard Free Energy, kcal/mol
AG° at 900°C
AG° at 1000°C
Reaction
NiO + 2HCI = NiCl2 + HzO
NiCl2 + H2 = Ni + 2HC1
NiO + H2 = Ni + HzO
FeO + 2HC1 = FeCl2 + HzO
FeCl2 + H2 = Fe + 2 HCI
FeO + H2 = Fe + HzO
(3.2.2)
(3.2.4)
(3.2.3)
(3.2.5)
8 .6
-2 0 .2
• 1 1 .6
4.2
-3.0
1 .2
7.0
-19.4
-12.4
3.5
-2 .6
0.9
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
From Table 3.3, it can be seen that the segregation of nickel is more favourable than that
of iron. Nickel chloride is more readily reduced than ferrous chloride, and nickel oxide is
more easily reduced than iron oxide.
3.1.1.1 Generation o f Chloridizine Gases
The initial step of the segregation process is generally recognized as that involving the
generation of hydrogen chloride gas by the reaction of moisture with calcium chloride.
The resulting calcium oxide then reacts with some acidic component of the ore, such as
quartz or clay minerals1311. The reactivities of the quartz or clay minerals influence the
efficiency of hydrogen chloride generation.
CaCI2 + H20 = CaO + 2 HCI
(3.1.11)
CaO + S i0 2 = CaSiOj
(3.1.12)
The overall equation for the generation of hydrogen chloride is given by equation 3.1.1.
From the equations, it can be seen that hydrogen chloride gas is the chloridizing gas in
the segregation process.
The equilibrium partial pressures of chlorine and hydrogen chloride over sodium
chloride, calcium chloride and silica under selected conditions are shown in Tables 3.4
and 3.5lu |. In these tables, the partial pressures of the gases for the segregation of nickel
laterite are given at 750°C and 950°C. According to Rey1311, the water vapor
concentration in a segregation furnace is often found to be about 30% by volume. In
order to illustrate the effect of water vapor as a parameter, the water vapor concentration
of 3% by volume is also included for both temperatures.
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3.4 Equilibrium Partial Pressures of Chlorine and Hydrogen Chloride over Sodium
Chloride as Functions of Temperature and Partial Pressures of Water Vapor and Oxygen1111
Log Phci
Temperature °C
Log P02
Log Pcu
PH20
In the absence of water vapor:
2 NaCl + 0.5O 2 + S i0 2 = Na2S i0 3 + Cl2 (1)
-1
-6.00
750°C
-5
-8.00
-4.68
-1
950°C
-6.68
-5
In the presence of water vapor: 2 NaCl + H2O + S 1O 2 = Na2Si0 3 + 2HC1 (2)
2 HCI + 0.5 0 2 = H20 + Cl2 (3)
-6.00
0.3
-2.71
-1
750°C
-2.71
-10
-10.50
-6.00
0.03
-3.21
-1
-10.50
-3.21
-10
-1.79
-4.16
0.3
-1
950°C
-8 .6 6
-1.79
-10
-2.29
-4.16
0.03
-1
-2.29
-10
-8.66
Log Ki
Log K 3
Equilibrium Constants:
Log K2
-5.50
-4.90
-0.60
at 750°C
-1.12
-4.18
-3.06
at 950°C
Table 3.5 Equilibrium Partial Pressures of Chlorine and Hydrogen Chloride over Calcium
Chloride as Functions of Temperature and Partial Pressures of Water Vapor and Oxygen1111
Log Pcu
Log P02
Log P h c i
Temperature °C
PhzO
In the absence of water vapor:
2 CaCl2 + 0.5O2 + S i0 2 = CaSi0 3 + Cl2 (1)
-1.77
-1
750°C
-3.77
-5
-1.51
-1
950°C
-3.51
-5
In the presence of water vapor: CaCU + H 2O + S 1O 2 = CaSi0 3 + 2HC1 (2)
2 HCI + 0.5 0 2 = H20 + Cl2 (3)
-1.77
-0.59
-1
0.3
750°C
-6.27
-0.59
-1 0
-1.77
-1.09
0.03
-1
-6.27
-1.09
-1 0
-1.52
-1
0.3
0
.2
1
950°C
-6 .0 2
-1 0
-0 .2 1
-1.52
0.03
-0.71
-1
-6 .0 2
-1 0
-0.71
Log K3
Log
Kt
Equilibrium Constants:
Log K 2
-0.60
-0.67
-1.26
at 750°C
- 1 .1 2
+0 .1 0
- 1.01
at 950°C
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
From Table 3.4 and Table 3.5, it can be seen that, in the absence of water vapour, the
decomposition of sodium chloride and calcium chloride by oxygen is possible in the
presence of silica. However, the partial pressure of chlorine is low, therefore, chlorine is
not an efficient chloridizing gas. In the presence of water vapor, the reaction proceeds
further and the predominant gaseous species is hydrogen chloride rather than chlorine at
the temperature of the segregation process. In reality, the partial pressures of hydrogen
chloride at both temperatures may be higher than those values listed in Table 3.4 and
Table 3.5, since the activities of the reactants and the products in the condensed forms
would be less than unity because of mutual solubility1211.
Table 3.4 and Table 3.5 also show that the partial pressures of both hydrogen chloride
and chlorine with calcium chloride are a few orders of magnitude higher than those with
sodium chloride. Similarly, Takashashi noted the same phenomenon during studies of the
laboratory segregation process
r-»2|
. This is one of the reasons why CaCb is found to be a
more effective segregation reagent for the treatment of nickel ores. Also. Iwasaki found
that higher roasting temperatures may be required for sodium chloride than for calcium
chloride1141.
Although substantial partial pressures of hydrogen chloride are required in the
segregation process, the hydrogen chloride alone may not be entirely responsible for the
segregation of nickel ore1301. Calcium chloride has a dual role in the segregation process.
In addition to providing a source of the hydrogen chloride needed for the chloridization
stage of the segregation process, it also stimulates certain lattice changes which render
the refractory nickel more amenable to attack by hydrogen chloride1331.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.1.1.2 Formation and Volatilization o f Metal Chlorides
The standard free energy for the nickel oxide chloridizing reaction at 1000°C, i.e. reaction
3.1.2, was calculated to be 7.0 kcal/mol by H anft301. This positive value indicates that the
thermodynamics of the process are marginal. Iwasaki^111 also found that the stability
fields of the iron and the nickel chlorides at 950°C overlap closely as shown in Figure
3.1. Thus it is surprising that nickel segregation occurs. In laterite ores, the complex
mineralization and the bonding mechanism of nickel will influence the activity of NiO
and thus thermodynamic calculations based on data for simple compounds are of limited
value.
+2
- —20
—20
NO
—16
—IS I
-2
—10
—10
NiO
a.
a.
—10 -
+2
—10
F igu re 3.1 Equilibrium diagram for Ni-H2-HCl an d Fe-H2-HCl at 950°C (Solid lines are
for PH2o=0-3atm, and the broken lines fo r PH2o=0.03atm)[u |
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A convenient way to visualize the thermodynamic relationship is to construct equilibrium
diagrams which show the stability regions of the oxides and the chlorides of nickel under
several conditions and superimpose these diagrams for comparison.
As shown in the previous section, hydrogen chloride is the predominant chloridecontaining reagent in segregation roasting. Equilibrium diagrams for the Ni-H 2-HCl
system are shown in Figures 3.2 (a), (b) and (c).
(O f V u c ,) (950°C,PHj0
= 0-3)
(a) Temperature (PHj 0 = 0-3)
IS «
--10
- —5 in
r -6
-10
N iC lj(s)
NiO
550
750
950°C
NiCI3(s)
NiO
NiO
0 -0 3
0-3
atm
0 100
538
2
Figure 3.2 Equilibrium Diagrams of Ni-H2-HCl (a)Effect of temperature
Ph2o = 0.3 atm (b) Effect of water vapor pressure at 950°C
(c) Vapor pressure of nickel chloride at 950°C and PH2o=0 .3 atm
[shaded region indicates P h c j over (CaCl2 + H 2O +Si0 2 ) ] tnl
It can be seen that the effects of temperature and water vapor partial pressures on the
positions of the equilibrium lines are relatively small.
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 2 (b) shows that the position of the equilibrium lines between the oxides and the
chlorides change with the partial pressure of water vapor. It appears that the lower the
partial pressure of water vapor, the lower the partial pressure of hydrogen chloride
needed to chloridize nickel, and hence the amount of the salt reactant is reduced.
Figure 3.2 (c) shows the lines representing the partial pressure of nickel chloride vapor in
equilibrium with its condensed phase at that temperature. By comparing the partial
pressure of hydrogen chloride given in Table 3.5 with Figure 3.2 (c), the partial pressure
of nickel chloride vapor at a given temperature may be calculated. For example, at 950°C,
when Phzo = 0.3 atm, from Table 3.5, log PHci = -0.21. Under these conditions it can be
seen from Figure 3.2 (c). that the partial pressure of gaseous nickel chloride vapor in
equilibrium with calcium chloride is calculated to be 20 mmHg. At 750°C, when Phzo =
0.3 atm, from Table 3.5, log
P h c i=
-0.59. Under these conditions, from Figure 3.2 (c), it
can be seen that the partial pressure of gaseous nickel chloride vapor is 10 mmHg. It
readily follows that the partial pressure of nickel chloride vapor decreases with
temperature. This may be interpreted as a result of the decreased partial pressure of
hydrogen chloride.
3.1.1.3 Reduction o f Nickel Chloride
Hydrogen, carbon and carbon monoxide exist in the nickel segregation process. Mildly
reducing conditions usually prevail in this process. From the TGA curves134*of NiClz in
hydrogen and carbon monoxide atmospheres, it has been shown that NiCh can be
reduced by hydrogen starting at 340°C, but not in carbon monoxide at that temperature.
Therefore, carbon monoxide can not be the reducing agent. Also, carbon cannot act as a
reducing agent for metal chloride according to the reaction:
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 MCI2 +
C —2M + CCI4
(3.1.13)
since CCL« is unstable. It appears, therefore, that hydrogen is the reducing agent. The
nickel chloride is reduced by hydrogen which is produced by the reaction of water vapor
with carbon or carbon monoxide and the reduction of nickel chloride proceeds according
to reaction 3.1.4.
A moderately reducing atmosphere is favourable for the nickel segregation process. A
closer examination of the equilibrium diagrams in Figure 3.2 (c) shows that the isobars of
nickel chloride vapor over the respective metallic phase have a positive slope. Therefore,
from the shaded arrow in Figure 3.2 (c), it can be seen that the partial pressure of the
nickel chloride vapor decreases as the hydrogen content in the atmosphere is increased. It
is known that the segregation reactions proceed via the formation of metal chloride vapor.
If the reducing atmosphere is too strong, this will result in a low partial pressure of nickel
chloride. In this case, the chloridization reaction will be suppressed. Also in a strongly
reducing atmosphere the nickel oxide will be reduced directly to metallic nickel.
Though the presence of hydrogen is important, it cannot replace carbon in the segregation
process. The presence of the solid carbon is essential in the process. The solid carbon
provides nuclei for the precipitation and growth of metallic nickel from the nickel
chloride vapor1331. Furthermore, the carbon reacts with water and generates hydrogen as
shown in reaction 3.1.10.
Due to the high temperatures and high pressures of HC1 in the nickel segregation process,
FeCh can form readily. Therefore, iron can also segregate with the nickel. As discussed
previously nickel chloride is more readily reduced than ferrous chloride^30*, and also
ferrous chloride is more stable than nickel chloride. This is an important factor in
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
enabling successful nickel segregation and beneficiation of the nickel with respect to
iron.
Cementation of nickel on a piece of iron inserted in a nickel silicate ore during
segregation roasting has also been reported1351. Thus, the exchange reaction between iron
and nickel chlorides may play an important role in the reduction of nickel chloride as
shown in equation 3.1.8. The thermodynamics of the reaction are quite favorable at a
segregation temperature of 950°C.
3.1.2 Kinetic Considerations
In the nickel segregation process, some of the reactions will be fast and achieve
immediate equilibrium, while others will be slow and equilibrium will not be reached.
For example, under the proper segregation conditions, reactions 3.1.6 and 3.1.7 will be
slow compared with reaction 3.1.2, which in turn is slow compared with reactions 3.1.1
and
3.1.4
which
are
always
at
equilibrium1301.
Reaction
3.1.10.
which
is
thermodynamically favourable, is believed to be comparatively slow in the nickel
segregation process where Ph:o and P ^ a re low.
3.1.2.1 Rate Determining Reaction
In the nickel segregation process, it was observed that the hydrolysis of calcium chloride
is extremely rapid and the equilibrium partial pressure of the gas is established in a
fraction of a second. The reduction of nickel chloride by hydrogen is also an extremely
rapid reaction. It may be deduced that the chloridization of the nickel oxide in the ore is
the rate determining step of the process1301 . The chloridization reaction is not limited by
the hydrogen chloride supply because hydrogen chloride is generated by the rapid
reactions mentioned above.
40
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
The activity of the nickel oxide species in the ore and the hydrogen chloride pressure in
the system will affect the rate of the chloridization reaction. A high activity of nickel
oxide in the silicate ores and a high pressure of hydrogen chloride will accelerate the
kinetics.
Increasing the reactivity of the nickel oxide in the ore enhances the nickel segregation
process by permitting lower operating temperatures. Calcium chloride serves to increase
the reactivity of the nickel oxide in the ore. As was mentioned, calcium chloride plays a
dual role in the segregation process. It stimulates certain lattice changes in the silicate
ore, which render the nickel oxide less refractory and thus the nickel oxide is more
amenable to chloridization130* .
The reactivity of the nickel oxide in silicate laterite ores is also found to be favorably
influenced by the addition of iron oxide114’. The added iron oxide reacts with the
magnesium silicates in the ore to form an iron-rich olivine phase rendering the nickel
oxide contained in the phase more amenable to segregation. This enhances the reaction
kinetics, enabling a lower segregation temperature to be used.
3.1.2.2 Reducibilitv o f N ickel Chloride and Ferrous Chloride
The enhanced reducibilitv of nickel chloride as compared to ferrous chloride is a key
factor in the nickel segregation process. It is observed that during the reduction step
nickel chloride is readily reduced to form segregated nickel whereas ferrous chloride is
not easily reduced to iron by hydrogen. So the nickel chloride is shown to be reduced to a
greater extent by hydrogen than the ferrous chloride. Furthermore, the rapid reduction of
nickel chloride will promote further chloridization of the nickel oxide while ferrous
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
chloride vapour will suppress further chloridization of the iron oxides. Consequently,
beneficiation of nickel will occur while iron segregation is suppressed.
3.1.2.3 Reduction o f N ickel Oxide
Side reactions involving direct reduction of nickel oxide can also potentially cause nickel
oxide to be reduced in situ according to reactions 3.1.6 and 3.1.7. Under the reducing
conditions of the nickel segregation process, the direct reduction of nickel oxide in the
ore to metal (while thermodynamically favourable) appears to be much slower than the
chloridization process to form nickel chloride1301. Therefore, the segregation reactions
proceed via the formation of metal chloride vapors. The reason why the kinetics of direct
reduction are slower than chloridization is difficult to understand. The slow process of
direct reduction is indeed a fortunate occurrence since otherwise nickel segregation
would not be feasible. But as was mentioned previously, a strongly reducing atmosphere
is not feasible since it may suppress the chloridization reaction and reduce the nickel
oxides directly to metallic nickel, i.e. reduction in situ.
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2 Microwave Fundamentals
In 1945, it was found that microwaves can generate heat in certain materials136’. The
heating of industrial materials by microwaves has been under investigation for many
years. Microwaves are a form of electromagnetic radiation with frequencies in the range
of 0.3 to 300 GHz’9’. The internationally agreed and recognized frequency bands which
can be used for microwave heating: the ISM (Industrial, Scientific and Medical) bands
are: 2450 MHz and 915 MHz’37’. The corresponding wavelengths in free space are 12.25
cm and 32.8 cm, respectively. Electromagnetic compatibility (EMC) requirements
impose severe limits on any emissions outside of these bands. These limits are much
lower than the health and safety limits. In most countries, compliance with the relevant
EMC regulations is a legal requirement’37’. For some specific applications, it has been
shown that microwave heating is efficient, clean, and easy to control. Research work on
microwave heating has been performed on food’38’, paper’39’, cement’40’, minerals’41’ and
in the chemical industries.
A general description of the interactions of microwaves with materials, the microwave
heating mechanisms, and microwave components will be discussed in the following
paragraph.
3.2.1 Interactions o f Microwaves with Materials
3.2.1.1 Microwave Fundamentals
The complex permittivity, e, of a material is defined by the following equation:
e = e'-je" = Bo (er'-jeeff")
(3.2.1)
and determines the degree of energy absorbed by a material from an electric field. In the
above equation, Eq is the permittivity of free space (8.86E-12F/m) and j = (-l)l/2. e is
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
comprised of two components; er' and fieff". e / is the real part and is called the dielectric
constant. It is a measure of the ability of the material to store the electrical energy. e eff" is
the imaginary part and is called the loss factor. It represents the loss of the electric field
energy in the material. The dielectric constant governs the electric field distribution
within the material and the loss factor determines the resultant heating rate1421.
The loss tangent (tan 5) is defined as follows:
tan 6 = Eeff" / er'= 6 / (2ftfe/E o)
(3.2.2)
and is the ratio of the effective loss factor to the dielectric constant. It is commonly used
to describe dielectric losses. In the above equation, 6 is the conductivity of the material
and f is the frequency of the incident wave. Furthermore, by combining equation 3.2.1
equation 3 .2 .2 , the complex permittivity can be related to the loss tangent as follows:
e = e0 e / (
1-j tan 6)
(3.2.3)
The penetration depth or half-power depth D is defined by the following equation:
D = 3k o / 8.68 ft tan 6 (er7eo) 44
(3.2.4)
where kQis the wavelength of the incident radiation. The penetration depth is the depth at
which the power of the electromagnetic wave is reduced by one-half*431.
The power absorbed by a material per unit volume or power density (Pv) is as follows:
Pv = 55.63 f E2 Eeff" x lO 12
(3.2.5)
where E represents the root mean square (rms) electric field intensity. This equation is
crucial in determining how a dielectric material will absorb energy when it is placed in a
high frequency electric field. The power density is proportional to the frequency of the
applied electric field and the dielectric loss factor, and is proportional to the square of the
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
local electric field. For a given system, the frequency is fixed and the dielectric loss
factor can be measured. The only unknown in this equation is the electric field, E. The
power dissipation is also related to the dielectric constant er\ because the electric field
intensity is a function of e ^ 44*. From this equation it can be seen that materials with high
values of the dielectric loss factor will absorb energy more readily than materials with
low loss factors.
The rate of temperature increase in the dielectric material caused by the conversion of
energy from the electric field to heat in the material is:
dT/dt = 0.239P/[(Cp)(q)]
(3.2.6)
where Cp is the specific heat of the material and
q
is the specific gravity. From this
equation, it can be seen that the relative heating rate depends not only on the dielectric
properties, but also on the specific heat and the densities of the materials.
The dielectric properties of most materials are a function of several variables. The
dielectric properties depend on the moisture content of the material, the frequency of the
applied electric field, the temperature of the materials, and on the density and the
structure of the materials[441. In hygroscopic materials, the amount of water in the
materials is generally a dominant factor. In granular particulate materials, the bulk
density is the most important factor. The dielectric properties of materials are also highly
dependent on their chemical composition and particularly on the permanent dipole
moments associated with any molecules in the materials.
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.1.2 Interaction o f Microwaves with Different Materials
The amount of microwave energy absorbed by a material depends on its effective
conductivity. The effect of conductivity on the microwave heating behaviors of materials
is shown in Figure 3.3{9!. For materials with a high conductivity such as metals the loss
factor is high; penetration depth is low and most of the energy is reflected. Insulators are
relatively microwave transparent and absorb very little energy. Semiconductors have an
intermediate conductivity (typically from 1 to 10 S/m), and they can be heated through
the interaction of the materials with microwaves.
Figure 3.4 also shows the interaction of microwaves with materials1451. The microwave
energy absorbed by a material can also be evaluated as the product of the loss tangent and
the dielectric constant, i.e. loss factor. A low loss factor indicates that the material will
not be heated while a high loss factor value material will only be heated on the surface. A
material to be processed by microwaves should have a loss factor between one and a
hundred.
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Transparent
Reflecting
Insulators
10-’ 0.2 10®
101 14 10*
Effective conductivity (S m*') at 2450MHz
Figure 3.3 Microwave Energy Absorption as a Function of Effective Conductivity [91
Material type
Penetration
TRANSPARENT
(Low loss
insulator)
Total
OPAQUE
(Conductor)
None
(Reflected)
ABSORBER
(Lossy Insulator)
Partial
to Total
ABSORBER
(Mlsed)
(a) Matrix * low loss Insulator
(b) Fiber/particles/additives *
(absorbing materials)
Partial
to Total
A/WWW
/ V \ A A / vaa^
Figure 3.4 Interaction of Microwaves with Materials1451
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.2.1.3 Selective H eatine and Thermal Runaway
In a mixture or even in a non-homogeneous material, some portions of the sample heat
faster than others. This phenomenon is called selective heating and it is an important
aspect of microwave heating. Selective heating is possible if the constituent materials in a
mixture have different dielectric properties. It can be seen from equation 3.2.5, that the
constituent with a higher value of the dielectric loss factor would absorb more energy
than the constituent with the lower loss factor. Also from equation 3.2.6, it can be seen
that the specific heats and the densities of the constituents also affect the relative heating
rates.
For some samples, it can be observed that at temperatures above a critical temperature
(TCnt). the dielectric loss factor and the loss tangent of the microwave-heated material
increases rapidly with increasing temperature. This phenomenon is called thermal
runaway. Many solid materials have very low dielectric losses, but the loss increases
rapidly with temperature. At a temperature above Tcm. the loss tangent begins to increase
rapidly, the material begins to absorb microwave energy more efficiently, and the sample
temperature increases rapidly. This causes the loss tangent to rise even faster. The net
result is that the energy absorption can increase exponentially. The rate of temperature
rise and Tcrit vary widely for different materials. Thermal runaway causes undesirable hot
spots within a material; it can also be used to heat materials at rapid rates[461.
3.2.1.4 Temperature M easurement
The determination of sample temperature during microwave irradiation is one of the
major problems in the microwave processing of materials. In the past, pyrometers and
infrared techniques were employed to measure the surface temperature, but serious errors
48
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
occurred in the measured accuracy of temperature. In some cases errors as great as
several hundred degrees are possible
f28 l
. The major reason for the error is that the heat is
generated rapidly internally, so that the internal temperature of the sample is much higher
than the surface. Thermocouples also can not be used because arcing occurs between the
sample and the thermocouple during the microwave process, and the arcing results in
thermal runaway leading to inaccurate temperature measurement. A method for
continuously monitoring sample temperature was developed by Walkiewicz125*. A Type
K thermocouple with an ungrounded tip sheathed in Inconel 702 or Stainless 440 was
used to measure the internal temperature of the oxide or sulphide samples. The
thermocouple was inserted through the roof of the oven directly into the sample. The
accuracy of the thermocouple data was within ± 2% for the sample. However, for
samples considered to be transparent to microwaves, an error greater than ± 2% would be
expected because microwaves would penetrate the sample and heat the surface of the
thermocouple sheath1281. A popular method to measure the temperature of the sample is to
turn off the magnetron during the temperature measurement. But it is obvious that some
decrease in temperature would be expected during the measurement period1291. Another
method for estimating the temperatures of the heated sample is to identify and analyze the
phases that are present in the samples. This method is also not accurate since the phase
composition depends not only on the temperature, but also a number of other factors.
3.2.2 Microwave Heating M echanisms
3.2.2.1 Microwave H eatine M echanisms
Microwaves generate heat by a variety of mechanisms: by polarization losses; by ohmic
heating from induced currents; by heating from arcs generated between particles. The
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
dominant mechanism is dipolar rotation, i.e., the rotation of the entire polar molecule in
the electric field
f4 2 l
. Atoms and molecules in dielectrics are macroscopically neutral, i.e.,
positive and negative charges do not travel freely. However, when external fields are
applied, the bound charges shift slightly relative to each other, creating electric dipoles.
The short-range displacement of the charge results in a polarization phenomenon. Many
dielectric materials can generate heat by polarization losses. The electromagnetic fields
change direction rapidly (4900 million times a second in a domestic oven)191, and the
polar molecules oscillate at the same rate. As a result heat is generated by the friction of
the intermolecular bonds due to the molecular agitation.
This heating mechanism is responsible for a special feature of microwave heating.
Microwaves in some materials can be generated efficiently and instantaneously, so that
the thermal phenomena of conduction, and convection play only secondary roles in the
heat generation.
3.2.2.2 Comparison o f Conventional Heatin e and Microwave H eatine
(1) Conventional heating
In conventional heating, the heat is generated outside of the material (e.g. by a flame or a
resistance heater) and transferred by conduction or convection. The surface of the object
is heated first, and then the heat flows to the inside by conduction and thus the
temperature gradient is from the surface to the center14' 1.
(2) Microwave heating
Microwave radiation penetrates into the object which is to be heated. Within the material
itself, the electromagnetic energy is transformed into heat by means of several complex
conversion mechanisms14' 1. The heat generated inside the material is transferred through
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the material by conduction to the outside surface and to the atmosphere. The interaction
between certain materials and microwaves can lead to very rapid heating. Microwave
transparent insulation can be placed around the sample in order to reduce the heat loss
from the sample to the surrounding air.
3.2.3 Microwave Components
Microwave system has three main components: a microwave generator, a waveguide, and
an applicator. The typical microwave heating system is shown in Figure 3.5[3/|.
Magnetron ♦
power supply
Multimode cavtty
Circulator
Waveguide
Turn table
Fig 3.5 A Typical Microwave Heating System13'1
3.2.3.1 Microwave Generators
The magnetron is a vacuum tube oscillator which generates high-power electromagnetic
signals in the microwave frequency range and is the most widely used microwave
generation component for heating materials. All domestic microwave appliances and the
large majority of industrial microwave furnaces contain magnetrons
[481
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The typical magnetron in a domestic microwave oven is shown in Figure 3.6t481.
Common magnetrons are tubes containing a hollow cylindrical anode with a metallic
cathode at the axis center. The anode is made up of the metallic cylinder and inward
vanes (a series of empty trapezoid-shaped regions). The regions between the vanes
determine the output frequency of the tube. The antenna is a probe connected to the
anode and it transmits microwave energy from the magnetron to an attached waveguide.
Magnetrons are available for generating microwave energies ranging from a few
kilowatts to a few megawatts. Its operation is based on the combined action of a magnetic
field applied externally and the electric field between its electrodes. The tube is a diode
having a cathode and an anode and is surrounded by an external magnet. Without this
external magnetic field, the tube would work much like a simple diode, with the electrons
flowing directly from the cathode to the anode. The magnetic field forces the cathodeemitted electrons to assume a curved path and thus creates a rotating electron cloud about
the tube axis. Typically, magnetrons operate at about 60—65% efficiency, where the
efficiency is defined as the ability to convert electrical power input to microwave power
output.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
C avity
C ath od e
Interaction space
Coupling loop
T
F o utput
A node block
Cavity
Slot
O u tp u t coupling system
B J
Filam ent
Figure 3.6 Magnetron of a Conventional Microwave Oven [48]
3.2.3.2 Waveguides
Waveguides are the transmission devices used for microwave propagation so that the
wave is forced to follow a path defined by the physical structure of the guide.
Waveguides rely on high-reflectivity walls for microwave propagation. Waveguides are
hollow metal tubes with metallic walls which are nearly perfect electrical conductors, and
contain low-loss dielectric air. Waveguides generally have constant cross sections with
dimensions at least one-fourth the operating wavelength.
The reflection of substantial power from the applicator to the magnetron can cause
damage and, to prevent this, a device known as a circulator is inserted between the
magnetron and the waveguide. The circulator is basically a one way valve which allows
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
power from the magnetron to reach the applicator, but stops any reflected power reaching
the magnetron (it is dissipated in the water load attached to the circulator). Furthermore,
some form of tuning device is often inserted between the waveguide and the cavity. This
is used to tune to a minimum any reflected power, thus ensuring that the system operates
with a high efficiency.
3.2.3.3 Applicators
The applicator is a device designed to ensure the transmission of electromagnetic energy
from the waveguide to the processed material. The most common form of microwave
applicator is a metal box or cavity as used in the domestic microwave oven. The heating
process is extremely complex. The materials to be heated are placed directly in this
cavity. A turntable is often used to average out (in time) any variations in electric field
that exist within the product. Furthermore, a device called a mode stirrer is usually
employed in the cavity to periodically change the standing wave patterns which exist
within it. Both these techniques improve the uniformity of heating of the product.
3.2.3.4 Safety o f Microwave Equipm ent
The human body is made up of dielectric materials. So it will absorb power at microwave
frequencies. Because the dielectric properties vary for different parts, different parts of
the body will absorb different amounts of energy for a given applied electric field
strength.
Only thermal effects are considered to be a hazard— microwave radiation has too little
energy to cause direct ionisation. In order to induce a rise in temperature in different parts
of the body, the term specific absorption rate (SAR) which is the power density (in watts
per kilogram) required is usedt3/|. This turns out to be in the range of 4 to 8 W/kg. The
54
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
safety limits are then set at a certain fraction of this, typically 1/10, i.e., 0.4 W/kg. In
order to arrive at a practical limit which can be measured, this SAR has then to be
converted to an equivalent power density (electric and magnetic field strength): typically
10 W/m2 for microwave radiation. Microwave heating equipment has to be constructed
so that any leakage is below this value and the leakage is measured on a regular basis.
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4. EXPERIMENTAL
4.1 Raw Materials
4.1.1 Laterite Ores
The nickel laterites employed in the present experiments were from the Falconbridge
operations in New Caledonia. They can be grouped into two categories: limonitic
laterites, containing a high proportion of iron oxides; and silicate laterites, containing less
iron oxide but more magnesium silicate. The chemical analyses of the laterite ore samples
by XRF100 (X-Ray Fluorescence) and ICAY50 (sodium peroxide fusion of the sample
with analyses by ICP-ES) from XRAL Laboratories are shown in Table 4.1.
Table 4.1 Compound or Elemental Analysis by XRAL
Cr
Fe
Ni
Co
Element
CaO
MgO
Si02
A120 3
Analytical
XRF100 XRF100 XRF100 XRF100 ICAY50 ICAY50 ICAY50 ICAY50
Method
%
%
%
%
%
%
%
Unit
%
Limonitic
1.29
45.7
1.8740
0.0243
0.08
0.57
4.24
1.9
Laterite
Silicate
19.7
24400
1520
0.49
16.3
9910
2.25
31.6
Laterite
4.1.2 Minerals
The mineralogy of laterites has been described in section 1.1. The dominant minerals in
limonitic laterites are goethite and limonite. Serpentine is the major mineral in the silicate
laterite. Minor constituents also exist in the laterite such as: hematite, magnetite, kaolin,
and olivine. In the limonitic laterites, there are no discrete nickel (Ni) minerals, and
nickel occurs in the form of finely disseminated oxides in the iron oxides. In the silicate
laterite, the nickel is in solid solution in the serpentine. The minerals employed for the
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
microwave heating behavior experiments were obtained from Ward’s Natural Science
Establishment, INC. and were as follows:
Goethite: Hydrous iron oxide, a-FeO(OH): 62.9% Fe
Limonite: FeOOH«nH2C>. It is mainly goethite with absorbed water, clay minerals and
other impurities.
Serpentine: Hydrous magnesium silicate, Mg6[Si4Ol0](OH )8
Hematite: Iron oxide, Fe20 3; 70% Fe; clay and sandy impurities are sometimes
present.
Magnetite: Iron oxide, Fe30 4; 72.4% Fe; the iron is sometimes replaced by a
small amount of magnesium or titanium.
Kaolin: Hydrous aluminium silicate, Al4Si4Ol0(OH) 8
Olivine: Magnesium iron silicate (Mg, Fe) 2Si04, with Mg in excess of Fe.
4.2 Microwave Heating Behaviors of Laterites and Associated Minerals
4.2.1
Equipment
4.2.1.1 Microwave System
The microwave equipment used for testing samples was a Sylvania microwave oven
(Model No. SM80701). It operated at a frequency of 2.45 GHz and had a maximum
power of 700W. The normal operating voltage of the air cooled magnetron is 3.7 kv. The
power and processing time can be varied. The size of the microwave chamber was 0.28 m
in length, 0.28 m in width, and 0.22 m in height. A schematic diagram of the apparatus is
shown in Figure 4.1.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(S>
<fi)
d>
<D
LEGEND
©
Microwave cavity
Quartz crucible
©
©
©
Magnetron
Sample
Alumina insulation board
Figure 4.1 Schematic Diagram of the Microwave Apparatus
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.1.2 Crucibles
Quartz and fireclay crucibles were used in the experiments. The major components of the
crucibles are shown in Table 4.2. Both quartz and fireclay crucibles are poor microwave
absorbers.
Table 4.2 Composition of the Crucibles (Mass Percent) Used
in the Research Experiments
Crucible
Cylindrical Quartz
Fireclay Boat
S i0 2
100
61
A120 3
35
The cylindrical quartz crucible had the following dimensions: 37 mm in height, 28 mm in
diameter, and 1.5 mm in thickness. It was used for the microwave heating behavior tests.
For the segregation tests, a fireclay boat crucible was used. The fireclay boat had the
following dimensions: 0.1 m in length, 0.01 m in width, 0.01 m in height.
4.2.1.3 Thermocouple
A type K (Chromel/Alumel) thermocouple was used to measure the temperature of the
sample. The size (diameter) of the thermocouple wire affects the accuracy of the
temperature measurement. In order to study this effect, thermocouples with different wire
sizes, were utilized to measure the temperature of a standard sample.
The silicate laterite sample was first heated to a known temperature in a conventional
resistance furnace for one hour, and then the sample was removed from the furnace and
the temperature was measured. For every test, the tip of the thermocouple was placed in
the center of the sample.
In the microwave process, it is not possible to measure the temperature with the
magnetron on. Since the power is off during the temperature measurement then there will
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
be some temperature drop. Therefore, in order to simulate the microwave system, the
sample temperature was measured outside of the conventional fumace.
The relationship between the diameter of the wire and the measured temperatures are
shown in Figure 4.2.
500
400
300
g. 200
•“
100
0
0.01
0.03
0.02
0.04
Wire Diameter (cm)
♦ Sample Temperature in the Fumace
■ Sample Temperature out of the Fumace
Figure 4.2 Relationship between Thermocouple Wire Size and Sample Temperature
From Figure 4.2, it can be seen that there is a temperature drop between the known
temperature and the sample temperature in the fumace. As the diameter of the
thermocouple decreases, the measured temperature approaches the known sample
temperature. The effect of thermocouple size is most likely due to increased response
time as the size becomes finer. However, the use of thermocouples with diameters of
0.15mm and 0.05mm was not feasible due to the low mechanical strength of the wire.
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Therefore, in the present experiments, a thermocouple wire with a diameter of 0.020cm
was used.
4.2.1.4 Infrared Pyrometer
A OMEGASCOPE Infrared Pyrometer was used to measure the sample temperature
through the optical transparent microwave system door under microwave conditions. The
OMEGASCOPE Infrared Pyrometer has a temperature range of -30°C to 1400°C. It is a
computerized hand-held, non-contact thermometer. In order to measure the temperature
of an object, the OMEGASCOPE is aimed at the object and the trigger is pulled. The
OMEGASCOPE collects the infrared energy that the object emits and it computes the
object’s temperature. The pyrometer was calibrated by measuring known sample
temperatures in a conventional resistance furnace through a optical transparent
microwave system door.
4.2.2 Experimental Procedure fo r the Microwave Behavior o f Laterites
4.2.2.1 M aximum Microwave Temperature - Microwave Time
13 g of the sample with a size of -10 mesh (Tyler) was placed in a quartz crucible. The
crucible and its content were placed on an alumina insulation board (type SALI: AI2O 3:
80%, SiOa: 20%) which is a low loss material and absorbs only a small amount of
microwave energy. The sample was heated for different times (from 1 to 25 minutes), and
the temperature was measured immediately after removing the sample from the oven
(about 10 seconds). It was observed that the temperatures varied throughout the
microwaved sample. To demonstrate this effect, the sample was divided into three equal
zones: top, middle, and bottom. For the silicate laterite sample which was microwaved
for 5 minutes, the temperatures at different positions in the sample are shown in Figure
61
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
4.3. it can be seen that the temperature is highest in the bottom zone. The temperature
gradient from the bottom to top is due to the high losses at the surface of the sample.
135 j
130
g
§
3
125
120-
|
115
J
110 •
105 -
100
-
Top zone
Middle zone
Bottom zone
Figure 4.3 Temperature of the Different Zones in the Silicate Laterite Sample after
Microwaving for 5 Minutes
For an accurate comparison of the microwave behaviour of different samples, the tip of
the thermocouple was placed at the center of the sample for every test. The temperature at
this position was record.
As was mentioned in Section 4.2.1.3 there is a temperature difference between the real
sample temperature and the measured temperature due to the heat losses and the response
time of thermocouple. So the sample temperature was measured immediately after the
sample was removed from the microwave system using the thermocouple with the wire
diameter of 0.20mm.
4.2.2.2 M aximum Microwave Temperature — Preheat Temperature
To study the effect of conventional preheating, a 13 g sample with a size of -10 mesh
(Tyler) was placed in the quartz crucible. Prior to heating by microwaves, the sample was
62
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
pre-heated in air in a conventional resistance furnace at varying temperatures for 60
minutes. After preheating, the sample was quickly transferred into the 700W microwave
oven and heated for 25 minutes and the maximum temperature reading was recorded.
4.2.2.3 M aximum Microwave Temperature — Particle Size ofLaterite
A 13 g laterite sample with different particle sizes was placed in a quartz crucible. The
sample was preheated in air in a conventional resistance furnace at a fixed temperature
for 60 minutes. The sample was then microwaved for 25 minutes and the temperature
measured immediately after the test (about 10 seconds). The maximum temperature
shown on the meter was recorded.
4.3 Nickel Segregation Tests
4.3.1 Microwave Segregation Test
In the microwave segregation test, a special reactor was designed and utilized. A
schematic diagram of the experimental set-up is shown in Figure 4.4. The reactor was
made of a Pyrex glass tube with a diameter of 45 mm and a length of 300 mm. During the
test, the reactor with the sample was placed on an alumina insulation board. Teflon tubing
which is a low loss insulator, served as an argon gas inlet and another as an outlet for the
reaction off-gases. The flowrate of the argon was measured by direct reading flowmeters
(Cole-Parmer, FM092.04). The off-gases were passed through a gas bubbler.
First the calcium chloride, charcoal and the silicate laterite were mechanically mixed.
Then the mixture was compressed in a floating mold at 2.75x10 Pa for a small size
briquette and 6.89xl07 Pa for a large size briquette. Two briquette sizes were employed: a
small briquette with a nominal size of 12.5 mm in length, 12.5 mm in diameter and a
density of 1.96xl03 kg/m3; a large briquette with a nominal size of 15 mm in length, 25
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
mm in diameter and a density of 1.35 xlO3 kg/m3. The briquette sample was placed on a
fireclay boat and then placed in the reactor. Some washed silica sand was placed between
the fireclay boat and the reactor to prevent thermal shock and cracking of the reactor. A
constant flow of argon was passed through the reactor containing the sample during the
tests in order to prevent oxidation. The sample was heated by microwaves for a fixed
time and then cooled to room temperature under an argon atmosphere.
Three experiments under the standard microwave segregation conditions were repeated.
The average metallic nickel recovery was determined, and the standard deviation was
calculated.
4.3.2 Conventional Segregation Tests
A quartz crucible reactor was used in the conventional segregation tests. A schematic
diagram of the experimental set-up is shown in Figure 4.5.
The reactor had a diameter of 45 mm and a length of 500 mm and was placed in a
resistance furnace. This furnace provides uniform heating. During the test, three sample
briquettes with the same proportion of reactants, as employed in the microwave tests,
were placed in the quartz crucible. The quartz crucible with the samples was then
positioned in the quartz reactor. A K type thermocouple was inserted in the reactor
directly in contact with the sample. The samples were heated for one hour with argon gas
continuously circulating through the reactor until the reaction was completed. The sample
was then allowed to cool to room temperature under an argon atmosphere.
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LEGEND
Washed silica sand
Argon cylinder
®
©
©
©
Gas bubbler
©
Magnetron
Microwave cavity
©
Sample briquette
Teflon tubing
©
Fireclay boat
Pyrex reactor
Q
Alumina insulation board
Figure 4.4 Schematic Diagram of the Microwave Segregation Experimental Set-up
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To Bubbler
Stopper
Quartz Reactor
Thermocouple
Quartz T ube
tsVsVV*
Quartz Crucible
W
///5
Sample Briquette
Washer! Silica SanH
Resistance Furnace
M
HUB
Figure 4.5 Schematic Diagram of the Conventional Segregation Experimental Set-up
66
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
4.4 Analytical Techniques
4.4.1 Dielectric Constant M easurements
As was mentioned in the section 3.2, the fundamental microwave parameters, such as the
real and imaginary parts of the complex dielectric constant (s' and e"), half-power depth
(D-half P), loss tangent (tana) of the material, are a measure of the ability of the material
to absorb the microwave energy. The fundamental parameters of both the silicate laterite
and the limonitic laterite and their associated minerals were measured by Microwave
Properties North (MPN). The cavity perturbation technique is used and this method is
based on knowing the difference in the cavity response between an empty sample-holder
and a sample-holder with the sample at the same temperature. The system can measure
the dielectric properties at two frequencies, 915MHz and 2450MHz. which are the
internationally agreed and recognized frequency bands, and up to a temperature of
1400°C.
A hot sample and its holder, which is made of high purity amorphous silica, are rapidly
removed from a conventional furnace and inserted into a high electric field region of a
thick-walled, well-cooled cavity. The resonant frequency and loaded Q of the cavity are
measured by a Hewlett-Packard 8753 network analyzer and stored for off-line analysis
which includes subtraction of hot empty sample holder effects. The sample and holder are
either left in the cavity for further measurements at lower temperatures as the sample
cools, or are quickly returned to the furnace for further processing. In the latter case, the
sample can be out of the furnace for as little as 2 seconds for a measurement at a single
frequency. If required the half-power depth, and the loss tangent can be calculated from
the dielectric constants e' and e".
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.4.2 % Ni in Fe-Ni
4.4.2.1 Sample Preparation
Prior to analysis, the segregation product (briquette) was first mounted in epoxy, ground
and polished and then examined with a metallurgical microscope at magnifications up to
400 times or with a scanning electron microscope (SEM). The sample was then analyzed
in an electron microprobe to determine the grade of Ni in the Fe-Ni deposit. The samples
chosen for the microprobe analysis were as the follows:
1 hour conventional heating at 1050°C with a the recovery of 61.81%
1 hour conventional heating at 750°C with a the recovery of 10.17%
5 minutes microwave heating with a recovery of 21.48%
2 minutes microwave heating with a recovery of 3.87 %
4.4.2.2 Electron Microorobe Analysis
Electron microprobe analysis (EMPA) is a non-destructive method for determining the
chemical composition of small amounts of solid materials. The quantitative chemical
composition of a phase within a material, such as a mineral grain or metal, can be readily
determined. The electron microprobe can quantitatively analyze elements from fluorine
(Z=9) to uranium (Z=92) at routine levels as low as 100 ppm.
EMPA uses a high-energy focused beam of electrons to generate X-rays which are
characteristic of the elements within a sample from volumes as small as 3 micrometers
(lO^m) in diameter. The electron beam current varies between 10 to 200 nanoamps. The
resulting X-rays are diffracted by analyzing crystals and counted using gas-flow and
sealed proportional detectors. Chemical composition is determined by comparing the
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
intensity of X-rays from standards (known composition) with those from unknown
materials and is corrected for the effects of absorption and fluorescence in the sample.
In this research work, the iron and nickel contents of the sample were analyzed. The
precision attainable on the instrument is about 0.5% relative.
4.4.3 Recovery o f N i
4.4.3.1 Concentration M ethods
The segregation products (briquette) were crushed and ground to -200 mesh (Tyler)
powder. About 6 g of the segregation product was concentrated by passing it through a
Davis Tube Tester.
Davis Tube Tester
The Davis Tube Tester is a laboratory equipment which is used to separate the magnetic
fraction of a sample from the non-magnetic fraction. The magnetic materials are held in
an electromagnetic field while the tailing materials (non-magnetic) are washed through
the tube. Figure 4.6 is a schematic diagram of a Davis Tube Tester.
Water
Glass
C -M agnet
Coils
Tailings Outlet
Figure 4.6 Davis Tube Tester
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Davis Tube Tester used in this laboratory work was manufactured by Dings
Magnetic Separator Co. The flow rate of water was lxlO'6 m3/s. The stroke rate was 70
strokes/min, and the operation time was 15 minutes. The sample was passed through the
tester only once. A concentrate and final tails were produced in the process. The
concentrate was kept for chemical analysis.
4.4.3.2 Chemical Analyses o f N i in the Concentrate
(1) Sample Preparation
Less than 0.2 g of the segregation concentrate product was placed in a beaker and
dissolved with a mixture of 30 ml concentrated HCl and 10 ml concentrated HNO3. The
solution was set to boil on a hot plate and digested for 30 minutes. The digested product
was washed and the solution filtered into a 250 ml flask. The %Ni in the solution was
measured by AAS.
(2) Atomic Absorption Spectrophotometer (AAS)
The atomic absorption spectrophotometer used
in the chemical
analysis was
manufactured by Perkin-Elmer Corporation.
The atomic absorption spectrophotometer measures the elemental concentrations in the
solution by spraying the solution into a long narrow hot flame. Light from a hollow
cathode lamp, containing the element to be analyzed, passes above the long and narrow
burner to a monochromator. The monochromator isolates the ground state radiation from
the hollow cathode lamp. The solution is sprayed into the fuel and air premixing chamber
which feeds the flame. Free atoms are produced in the flame because of the heat and
reducing conditions. If the atoms are the same as the element in the lamp, then light is
absorbed. The degree of absorption depends on the amount of the particular element
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
present in the original solution. The results are compared with standard solutions, which
allows concentration to be determined. It is able to detect the presence of up to 70
elements at levels ranging from very low concentrations to one hundred percent of the
element. The technique can be highly precise and accurate. The precision attainable is in
the vicinity of 1% relative.
The grade of Ni in the concentrate and recovery were calculated based on the following
equations:
.
Ni Grade = %Ni in the concentrate =
(ppm Ni in the solution) x (Volumeof the flask)
-------------------------------------------------------Weight of sample analyzed
Recovery of Ni in the concentrate
_ (Weight of overall concentrate) x (%Ni in the concentrate)
(Weight of the ore) x (%Ni in the ore)
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5. RESULTS AND DISCUSSION
5.1 Microwave Behavior of Nickel Laterite
5.1.1 Dielectric Constant Measurements
As discussed in section 4.4.1, The dielectric constants of the ore and the mineral samples
related to laterite processing were measured by Microwave Properties North (MPN). In
this chapter, the dielectric constants, which were measured at a frequency of 2450 MHz
are reported at processing temperatures up to 1150°C.
5.1.1.1 Limonitic Laterite and its M aior M inerals
The dielectric constants of goethite, limonite, and limonitic laterite are shown in Figure
5.1 and discussed below.
(1) Goethite (a-FeO(OH))
The values of s' show little change up to 300°C, with only a slight suggestion of a small
amount absorbed moisture which disappeared above 100°C. The peak in s' that occurs
between 325°C and 375°C is likely caused by a phase change that expels water from the
mineral. At 425°C, the value of e' returns to a slightly lower value, suggesting a loss of
material (as would be expected if water was evolved from the mineral with no volume
change). Above 550°C, the value of e' begins to increase, and this increase persists up to
about 1000°C, where there is a dramatic increase in the dielectric constant with
temperature. The products from the thermal dehydration of goethite (a-FeO*OH) at
different temperatures are as follows1491:
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
"« 30
0
200
400
600
800
1000
1200
T e m p e ra tu re (°C)
Goethite
Limonite
Limonitic Laterite
Figure5.1 The Dielectric Constant of Limonitic Laterite and its Associated
Minerals vs Temperature as Measured by MPN
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
„
180 - 250°C
2 a F e O - O H ------------------>
6 _
—<xFei/3(0H)02
+
2
—H 2O
— aFei/zlOW ^Oz^ + A H2° 8°Q-:1Q5-QOC-^ aFe20 3 +
5.5
5.5
|
5
h 20
400 - 600°C
------------------>
(5.1.1)
A comparison of Figure 5.1 with equation 5.1.1 indicates that the dramatic increase in the
dielectric constant at 328°C is due to the format of aFei/3(0H )02. This increase may be
due to the fact that the OH' ion in aFei/3(0H )02 is more loosely bonded than in
aFeO O H . The decrease at about 377°C is likely due to the significant reduction of OH'
content when aFei/3(0H )02 is converted to aFei/^OH) i/20 2j . Also from equation 5.1.1,
it can be seen that even at temperatures over 800°C water can still be evolved from the
sample.
(2) Limonite (FeO(OH)-nH20 )
The slight decrease in the values of e' up to about 200°C, is likely a result of the
evolution of a small amount of adsorbed or lightly bonded moisture which is quickly
evolved as the temperature rises. However, e' increases modestly between 240°C and
290°C. The peak in e' that occurs between 275°C and 300°C is likely caused by a phase
change that expels water from the mineral. It appears that this evolution of water
continues, at a decreasing rate, up to 550°C. At about 575°C, the values of e' reach a
minimum. Above 575°C, e' increases rapidly. Above 900°C, the rate of increase with
temperature increases dramatically. It can be seen from Figure 5.1 that the dielectric
constant of limonite is lower than goethite at all temperatures.
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(3) Limonitic laterite
As shown in Figure 5.1, up to 800°C, the dielectric properties of limonitic laterite are
very similar to limonite. The values of e ' decrease slowly up to about 225°C, likely the
result of a small amount of adsorbed / lightly bonded moisture which is quickly evolved
as the temperature rises. The peak in e ' that occurs between 275°C and 400°C is likely
caused by a phase change that expels water from the mineral. It appears that this
evolution of water is complete by about 450°C. At about 425°C, the value of s' reaches a
minimum. Above 450°C, e' increases rapidly. Above 800°C. the dielectric constant
increases rapidly with temperature.
5.1.1.2 Silicate Laterite and its Maior M inerals
The dielectric constants of silicate laterite and serpentine are shown in Figure 5.2 and are
discussed below.
(1) Serpentine (MgjSuOioKOHJg)
The values of e' increase slowly up to about 400°C. At about 400°C, e ' reaches a
maximum, while above 400°C, e ' decreases slowly. At 750°C, the value of e ' reaches a
minimum. Above 750°C, the dielectric constant increases slowly again and remains
relatively constant between 750°C and 900°C. The reaction for the thermal dehydration
of serpentine (MgatSLOioKOfDg) is as the follows1501:
(Mg6[Si4Oio] (OH)8) -» 3Mg2Si04 + S i02 + 4H20
(5.1.2)
Using thermal analysis it was shown that the above reaction occurs in the temperature
range of 700°C-800°C.
Since there is not significant change in the dielectric constant in the temperature range of
700-800°C it can be concluded that the dielectric constant of Mg6[Si4Ol0] (OH)g and
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14
12
10
8
6
4
2
0
200
400
600
800
1000
1200
T e m p e r a tu r e (°C)
S e rp e n tin e
S ilicate laterite
Figure 5.2 The Dielectric Constant of Silicate Laterite and its Associated
Minerals vs Temperature as Measured by MPN
76
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Mg2Si04 are similar. Also, surprisingly, the dielectric constant is not dependent on
temperature.
(2) Silicate Laterite
The values of e ' increase slowly up to about 135°C. At about 135°C, e ' reaches a
maximum, while above 135°C, the values of s' decrease slowly up to about 225°C, likely
the result of a small amount of adsorbed or lightly bonded moisture which is quickly
evolved as the temperature rises. This evolution of water is complete by 600°C. At about
450°C, the value of e ' reaches a minimum. Above 600°C, e ' increases rapidly. The
dielectric properties of the silicate laterite are significantly different from those of
serpentine and this can be attributed to the presence of other minerals.
The dielectric constant of the minor minerals in both the silicate laterite and the limonitic
laterite were also measured and were found to be very low.
5.1.1.3 Seereeation Sample under Standard Conditions
The test sample was a mixture of the silicate laterite and 5% CaC^ and 6% charcoal. The
dielectric constant as a function of temperature for the silicate laterite segregation sample
is shown in Figure 5.3. Measurements are only reported up to 700°C. Above this
temperature, iron chloride vapours were evolved from the sample and condensed on the
wall of the quartz sample holder. Thus the results above 700°C were unreliable.
The room temperature values of e ' are modest, but drop off above 150°C, undoubtedly
due to adsorbed moisture which disappeared above 150°C. The value of e' then returns
to a slightly lower value, suggesting a loss of material (as would be expected if water was
evolved from the mineral with no volume change). The value of e ' starts to increase at
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
100
CD
0
100
200
300
400
500
600
700
800
T e m p e r a t u r e (°C )
Figure 5.3 The Dielectric Constant of Silicate Segregation Sample for the Standard
Condition vs Temperature as Measured by MPN
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
375°C, which is likely caused by a phase change that expels bound water from the
mineral. The s' values of the segregation sample are much higher than the laterite ore or
the associated minerals (Figure 5.1 and Figure 5.2). This can be mainly attributed to the
excellent microwave absorption characteristics of charcoal. The half-power depth of the
silicate laterite segregation sample is shown in Figure 5.4. It can be seen that for
temperatures below about 350°C the half-power depth is above about 6 cm. Thus, for the
sample sizes employed in the present work, the initial half-power penetration depth is
longer than any sample sizes. Above about 400°C, the penetration depth remains constant
at about 1 cm and this should provide adequate prenetration for the sample since
employed in the present work.
5.1.2 Microwave H eating Characteristics o f Laterite and Associated Minerals
5.1.2.1 Effect o f Time
(1) Effect of Time on Microwave Heating Characteristics of Nickel Laterite
A 13g sample of each of the nickel laterites (-10 mesh (Tyler)) was prepared for testing.
Each sample was microwaved for different times in the microwave system and the bulk
sample temperature was recorded. Figure 5.5 shows the results. During the first minute
the temperature increased rapidly for both the samples. For the limonitic laterite ore, after
the first minute, the temperature remained constant at 95°C for about three minutes due to
the removal of free water. Once the free water had been removed the sample temperature
increased very rapidly and then more slowly. On the other hand, the temperature of the
silicate laterite increased from one to five minutes and then remained constant at about
120°C for three minutes (between five to eight minutes). This is likely due to the removal
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Half-Power Depth (cm)
18
16
14
12
10
8
6
4
2
0
0
200
400
600
800
1000
T e m p e r a t u r e ( °C)
Figure 5.4 Half-Power Depth of the Silicate Laterite Segregation Sample for the
Standard Conditions vs Temperature as Measured by MPN
80
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
200
180
160
Bulk Sample
—. 140
S»
€4> 100
g«
*”
80
60
40
20
0
0
5
10
15
20
25
Time (minutes)
—•-L im o n itic Laterite
Silicate Laterite
Figure 5.5 Effect of Time on the Microwave Heating Characteristics of
Silicate Laterite and Limonitic Laterite
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
of water from the sample. Once this water was removed, then the sample temperature
increased very rapidly and then more slowly. It should be noted that for both samples,
once the moisture was removed, then the sample temperature increased rapidly. For both
laterites the maximum temperature achieved was about 177°C, and this demonstrates that
the ability of these ores to absorb microwaves is limited.
Figure 5.1 and 5.2 show that peaks in e' occur at temperatures of 100°C for the limonitic
laterite and 130°C for the silicate laterite. These temperatures roughly correspond to the
constant temperature plateaus in Figure 5.5. Also from Figure 5.1 and Figure 5.2, it can
be seen that the e' of the limonitic laterite is slightly higher than that of the silicate
laterite. Thus as shown in Figure 5.5 the heating rate of the limonitic laterite is higher
than that of the silicate laterite. However, since e' is relatively low for both the silicate
laterite and the limonitic laterite thus only low sample temperatures can be obtained.
(2) Effect of Time on Microwave Heating Characteristics of Minerals in Laterite
The mineralogy of a material is an important factor with regards to the microwave
heating behavior because some minerals absorb microwave energy, while other minerals
remain unaffected. There are several minerals commonly found in laterite ores. In
limonitic laterites, the major constituents are limonite (FeO-OH.nHzO) and goethite
(a-FeO(OH)). There are no discrete nickel minerals, and nickel occurs in the form of
finely disseminated oxides in the iron oxides. The minor constituents in the limonitic
laterite are hematite (FezOs), magnetite (FesO,*) and serpentine (MgetSUOioKOHjg). In
the silicate laterite, serpentine is the major mineral. The nickel is mainly in solid solution
in the serpentine. The minor constituents are olivine ((Mg,Fe)2Si0 4 ), hematite, magnetite,
goethite, limonite and kaolin (ALSUOiofOH) g).
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Major Minerals
The heating characteristics of the major minerals in the laterites are shown in Figure 5.6.
For serpentine, which is the major component of the silicate laterite, in the first five
minutes there was a dramatic increase in the bulk temperature of the sample up to 175°C.
Further microwave heating from 5 to 25 minutes resulted in a modest increase to 189°C.
Goethite and limonite, which are the major minerals in the limonitic laterite, exhibited
similar trends to serpentine. In the first five minutes there was a dramatic increase in
temperature but not as high as for serpentine. The temperatures attained were 72°C for
goethite and 45°C for limonite. Extending the microwave time from 5 to 25 minutes
resulted in a slight increase in the bulk sample temperature to 88°C for goethite and 70°C
for limonite. The results indicate that serpentine is a better microwave absorber than
either goethite or limonite.
M inor Minerals
The minor minerals in the two ores are similar. The heating characteristics of the minor
minerals in the laterites are shown in Figure 5.7. Olivine, hematite and kaolin exhibited
similar characteristics. In the initial 5 minutes there was a dramatic increase in the
temperature with olivine attaining the highest temperature followed by hematite and then
kaolin. Further microwave heating beyond 5 minutes resulted in slight decreases of
temperature for the three minerals, which indicates that the minerals can not absorb any
further microwave energy.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
200
180
160
Bulk Sample
140
oO .
120
s
100
«
a
E
«
►-
i
40
0
5
10
15
20
25
Time (minutes)
| L i m o n i t e - a - S erpentine
Goethite
Figure 5.6 Effect of Time on the Microwave Heating Characteristics of
Major Minerals in the Laterites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
30
600
500
Bulk Sample
q
400
m 300
200
100
0
5
10
15
20
25
30
Time (minutes)
—
Hematite
Kaolin -e-O livine —
Magnetite
Figure 5.7 Effect of Time on the Microwave Heating Characteristics of the
Minor Minerals in the Laterites
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The behavior of magnetite is totally different from olivine, hematite and kaolin. It is wellknown that magnetite is a very good microwave absorber126,28,291. The heating rate of
magnetite in the first minute is extremely fast as indicated by the curve in Figure 5.7, and
the sample attained a temperature of 456°C after one minute. From one to five minutes
the rate of temperature increase was reduced and after 25 minutes, the maximum
temperature was 555°C.
Clearly, the minerals in laterite have quite different microwave responses. Kaolin,
olivine, goethite, limonite and hematite are poor microwave absorbers and are not readily
heated. On the other hand, magnetite which is a high-loss hyperactive material is a very
good microwave absorber. However, magnetite is only present in the ores in low
concentrations. In the nickel segregation process, iron oxide can be reduced to magnetite,
which helps the sample to achieve a high temperature.
5.1.2.2 Effect o f Conventional Preheat Temperature
As mentioned in section 3.2.1.3. the dielectric constant of a material increases with
temperature, resulting in an increase in the energy absorption. Thus, since the as-received
laterite ore is difficult to heat from room temperature, the samples were preheated before
microwaving. Therefore, 13 g samples o f -10 mesh (Tyler) limonitic laterite and silicate
laterite ore were prepared and preheated in a conventional oven for one hour prior to
heating in the microwave system for 25 minutes.
(1) Effect of Conventional Preheat Temperature on Microwave Heating Characteristics of
Laterites
The chosen preheat temperatures for the laterite samples were higher than the maximum
bulk sample temperatures that the sample could attain with microwave heating only, i.e.
86
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
over 200°C. Figure 5.8 shows the effect of conventional preheating on the microwave
heating behavior of the laterite samples. It is interesting to note that the limonitic laterite
sample readily absorbs microwaves when preheated to temperatures higher than about
265°C.
The preheated samples consistently attained bulk sample temperatures of about 860°C.
On the other hand, the silicate laterite samples did not absorb significant microwave
energy when the preheat temperature was less than 450°C. However, when the samples
were preheated at temperatures higher than 492°C, the samples absorb more microwave
energy and the maximum microwave temperature increased with increasing preheat
temperature.
Limonitic laterite which was preheated at a temperature of 300°C was analyzed by X-Ray
Diffraction (XRD) and the result is shown in Figure 5.9. The X-ray showed that the
major component of the sample was goethite. As shown in Figure 5.1 at 300°C the
dielectric constant of goethite is higher than that of limonitic laterite. Thus, preheating of
the limonitic laterite to 300°C results in the conversion of limonitic laterite to goethite
and thus the microwave absorption is higher. Limonitic laterite which was preheated to
temperatures higher than 450°C was also analyzed by XRD and the results are shown in
Figure 5.9. It can be seen that the major component was hematite. As will be discussed
later in this section, hematite samples preheated to temperatures higher than 450°C
readily absorb microwaves, resulting in temperatures of over 900°C and eventually
sintering. Therefore, since hematite at temperature of over 450°C is a good absorber, then
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1000
900
800
Bulk Sample
700
£
0)
bl
600
3
<5
0)
500
Q.
400
«
h-
300
200
100
0
200
400
600
800
1000
1200
P reheat Tem perature (°C)
•Limonitic Laterite
Silicate Laterite!
Figure 5.8 Effect of Conventional Preheat Temperature on the Microwave Heating
Characteristics of the Silicate Laterite and the Limonitic Laterite
88
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18-
Intensity (Counts) X 100
15-
G o e t h it e
H e m a tite
12-
L im o n i t ic L a te r ite a t 3 0 0 ° C
9
-
L im o n i t ic L a te r ite a t R o o m
T em p era tu re
6L im o n i t ic L a te r ite a t 4 5 0 ° C
3
-
20.00
30 00
40 00
50.0
Figure 5.9 X-Ray Diffraction (XRD) Analysis of Limonitic Laterite and Limonitic
Laterite Preheated to Temperatures of 300°C and 450°C
89
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the limonitic laterite can obtain high bulk sample temperatures under microwave
conditions.
Figure 5.2 shows the effect of temperature on the dielectric constant of the silicate
laterite. The dielectric constant, e', has low values when the sample temperature is lower
than 450°C.
The experimental results in Figure 5.8 show that when the sample is
preheated at temperatures below 450°C, then the sample does not absorb significant
microwave energy. Figure 5.2 shows that above 600°C, e' increases rapidly with
increasing temperature. Thus, when the silicate laterite is preheated above 600°C it
readily absorbs microwaves.
(2)
Effect of Conventional
Preheat Temperature on the
Microwave
Heating
Characteristics of Laterite Minerals
M ajor Minerals
Figure 5.10 shows the effect of preheat temperature on the bulk sample temperatures of
the major minerals in laterite. Preheated limonite and goethite exhibited similar
behaviors. However, there is a difference in the critical preheat temperatures. The
limonite samples preheated at temperatures below 200°C demonstrated low microwave
absorption. However, a dramatic increase in the temperature is observed when the preheat
temperature is higher than 200°C. Above 200°C the temperature remained constant. On
the other hand, the goethite samples which were preheated to temperatures less than
550°C had low microwave absorption. However, the microwave temperature in the
goethite sample increased dramatically and then remained constant when the preheat
temperature was higher than 550°C. Hence, goethite is observed to have a much higher
critical preheat temperature.
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1200 n
1000
O
800
li
600
0
0
Q.
E
.2
400 •
200
0
100
200
300
400
500
600
700
P reh eat T em perature (°C)
{-♦—Goethite - 0 —Serpentine
Limonite
Figure 5.10 Effect of Conventional Preheat Temperature on the Microwave
Heating Characteristics of Major Minerals in Laterite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The effect of preheat temperature on the temperature of serpentine is also shown in
Figure 5.10. At a preheat temperature of around 200°C, the sample absorbs only a small
amount of microwave energy since the bulk sample temperature attained remained
constant at 480°C. As expected. Figure 5.10 shows that the temperature increases slowly
with increasing preheat temperature from 200°C to 500°C. It is observed that as the
preheat temperature of the sample is increased above 500°C, then the sample can absorb
more microwave energy.
Minor Minerals
Among the minor minerals in the laterites, it is well-known that magnetite is a very good
microwave absorber126'28,291. This was corroborated by the test results obtained in this
research. As shown in Figure 5.7 it was observed that magnetite samples, without
preheating, can attain a maximum temperature of 555°C when microwave heated for 25
minutes. Other minor minerals such as hematite, kaolin and olivine, which do not readily
absorb microwaves, were preheated. The results are shown in Figure 5.11. The behavior
of hematite was similar to that of goethite and limonite as shown in Figure 5.10. The
critical preheat temperature was about 450°C. Preheating the hematite samples above
450°C resulted in microwave temperatures over 1000°C and the sample sintered. It is
known that when hematite is heated in air, the following reactions can occur:
3Fe20 3----->2Fe30 4 + 0.5O2
(5.1.3)
Thus during the preheating process some hematite is converted into magnetite. Since
magnetite is a good microwave absorber, then the preheated sample can be readily heated
by microwaves.
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1800
1600
Bulk Sample
Temperature (°C)
1400
1200
1000
800
600
400
200
■m
0
0
100
200
300
400
500
600
700
800
Preheat Temperature (°C)
—♦—Hematite
Kaolin -♦-O livine
Figure 5.11 Effect of Conventional Preheat Temperature on the Microwave
Heating Characteristics of Minor Minerals in Laterite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
On the other hand, for kaolin and olivine, the bulk temperature achieved with preheating
was low and relatively independent of preheat temperature. For kaolin it was about 83°C
while for olivine it was about 170°C.
The critical preheat temperatures of the minerals in the two laterite ores are summarized
in Table 5.1.
Table 5.1 Critical Preheat Temperatures of the Minerals in Laterite
Minerals
Critical
Temperature
(°C)
Serpentine
none
Goethite
205
Limonite
545
Hematite
448
Kaolin
none
Olivine
none
As shown in Table 5.1, goethite, limonite and hematite exhibit critical preheat
temperatures while serpentine, kaolin, and olivine do not. This is due to the fact that at
the critical preheat temperature goethite. limonite and hematite undergo phase changes
and the products of the phase change have better microwave absorption characteristics.
5.1.2.3 Effect o f Particle Size o f Laterite
Limonitic laterite of different particle sizes was preheated at about 450°C for one hour,
prior to microwave heating for 25 minutes. The relationship between the bulk sample
temperature and the particle size is shown in Figure 5.12. The sample with a particle size
of -100 mesh (Tyler) (< 0.147mm) absorbs the least amount of microwave energy while
the sample with a particle size range of -10 to +20 mesh (Tyler) (0.833-1.65mm) absorbs
the most microwave energy. Samples with particle sizes in the range of -10 to +100 mesh
(Tyler) (0.147-1.651mm) absorb similar amounts of microwave energy. This behavior is
due to the differences in the permeabilities (related to the dielectric constant) of the
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
1000
1
800
tt
Q.
E
600
® ^
a
400
.2
•- —
o
E
(S
(0
200
3
CD
0
0
0.4
0.8
1.2
1.6
Particle S ize (mm)
Figure 5.12 Effect of Particle Size on the Microwave Heating Characteristics
of Limonitic Laterite
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2
different sizes of low loss material. When the particle size is small, the permeability is
particularly low1511.
The relationship between the bulk sample temperature and the particle size of the silicate
laterite is shown in Figure 5.13. The samples were preheated to about 500°C prior to
microwave heating. The curve shows that the silicate laterite with particle sizes ranging
from -100 to +150 mesh (Tyler) (0.104-0.147mm), absorbs the most microwave energy.
Other particle sizes have low microwave absorption and the maximum microwave
temperatures are lower than the preheat temperature. This phenomenon is likely due to a
higher dielectric constant at a particle size of -100 to +150 mesh.
5.1.3 Microwave H eating Characteristics o f Charcoal
5.1.3.1 Effect o f Time
A 3 g sample was employed with a particle size of -150 mesh (Tyler). The effect of
microwave time on charcoal is shown in Figure 5.14. The sample was heated in the
microwave for three minutes and the temperature increased rapidly. The sample attained
a temperature of 380°C in the first minute, 814°C after two minutes, and a maximum of
959°C after three minutes of microwave heating. The heating rate in the first minute was
380°C/min„ 434°C/min in the second minute and 145°C/min in the third minute. The
curve also indicates that the charcoal sample temperature can exceed 950°C when heated
in the microwave for more than 3 minutes. The behaviour of the charcoal is consistent
with the results in the literature128,291. Charcoal is a hyperactive microwave absorber and
its conductivity is intermediate between that of insulators and metals. Materials with
conductivities in this range are particularly suitable for microwave heating.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Bulk Sample Temperature (°C)
700
600
500
400
300
200
100
0
0
0.1
0.2
0.3
0.4
0.5
Particle Size (mm)
Figure 5.13 Effect of Particle Size on the Microwave Heating Characteristics
of Silicate Laterite
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0.6
1200
1000
£
«w
3
800
1
tt
a
1
600
a>
a
I 400
CO
3
®
200
0
20
40
60
100 120
Time (seconds)
80
140
160
180
200
Figure 5.14 Effect of Time on the Microwave Heating Characteristics of Charcoal
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.1.3.2 Effect o f Particle Size o f Charcoal
The effect of charcoal size on sample temperature is shown in Figure 5.15. A sample of
0.6 g and sizes of -150 mesh, -48 +65 mesh and +10 mesh (Tyler) were heated in the
microwave for three minutes. The charcoal temperature was measured every 10 seconds.
It can be seen from Figure 5.15 that in the first two minutes, the heating rate increased
with particle size. After about two minutes, the samples achieved approximately the same
temperature of 740°C. Above this temperature the heating rate increased with decreasing
particle size. This phenomenon can be explained as follows. When the reaction time is
shorter than two minutes, the coarser-sized charcoal particles heat up more rapidly than
the finer ones since for a hyperactive material, coarser particles have higher dielectric
constants than finer ones1511. When the sample is heated to about 740°C. the charcoal
begins to combust. Since the finer-sized particles have more surface area, they react more
easily with the oxygen in the air than the coarser particles. It was also observed that when
carbon monoxide was produced a plasma formed on the top surface of the charcoal.
These trends indicate that when charcoal is used as a reductant in a microwave heating
system, then the size of the charcoal particles can affect the reaction temperature. Finer
particles are preferred when longer reaction times and high temperatures are required,
while for short reaction times and higher temperatures, coarser particles are preferred.
5.1.4 Microwave Heating Characteristics o f Calcium Chloride
Since CaCl2 is employed as a reagent in the segregation process then the microwave
heating characteristic of CaCh and CaCl2*2H20 were determined. A 2 g sample was used
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Bulk Sample Temperature (°C)
1200
1000
800
600
400
200
0
20
40
60
80
100
120
140
160
180
200
Microwave Time (Second)
-150 m e sh
+10 m esh
-48 +65 m e sh
j
Figure 5.15 Effect of Time on the Microwave Heating Characteristics of Different
Size of Charcoal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in the microwave tests. The samples were microwaved for different times. Figure 5.16
shows the effects of time on the microwave heating characteristics of CaCl2 and
CaCl2-2H20 .
Figure 5.16 shows that CaCl2 and CaCl2-2H20 exhibit totally different microwave
behaviours. CaCl2 is not a good microwave absorber and the temperature increases very
slowly with time. On the other hand, CaCl2-2H20 is a good microwave absorber.
However, above 45.3°C1521, CaCl2*2H20 decomposes into CaCl2 and 2H20 . Therefore,
initially this material absorbs microwaves, dehydrates and is converted into CaCl?. Thus,
the curves for CaCl2*2H20 and CaCI2converge at about 8 minutes.
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0)
ka
3
C8
a>
a
E
-
o>
g(Q
(/)
£
3
160
120
80 40 -
0
6
3
9
12
Microwave Time (minutes)
CaCI,
CaCI2-2H20
Figure 5.16 Effect of Time on the Microwave Heating Characteristics of
CaCl2‘2H20 and CaCI2
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2 Recovery of Nickel in the Microwave and the Conventional Segregation Process
5.2.1 N ickel Recovery in the Microwave Segregation Process
The literature regarding nickel recovery provided meaningful information on the effects
of various combinations of parameters on the processing of a given ore. The nickel
recovery is a measure of the efficiency o f the process. A number of microwave
experiments were performed in order to study the effects of changing a single operating
variable. The variables and the values which were used are given in Table 5.2.
Table 5.2 Variables and conditions used in the microwave segregation process
Variable
Reaction time (min)
Microwave power (W)
Silicate laterite size (mesh) (Tyler)
Charcoal, (wt%)
Charcoal size (mesh) (Tyler)
Chloridizing agent (5%)
Calcium chloride (wt%)
Inert gas flow rate (cnvVmin)
Preheat temperature (°C)
Water additives
Values, Conditions or Materials
1, 2, 3, 5, 10, 15, 20, 25,30,35,40
420, 490, 560, 630, 700
-10+28, -28+35, -35+48, -48+65,
-65+100, -100+150, -150+200,
-200+270, -400
1, 2, 4, 6, 8, 10
-28 +35; -48 +65; -65 +100; -100 +150; -150
MgCl2, FeCl2, NiCl2, NaCI, CaCl2
2, 5, 8, 10
100, 200, 500, 1000, 1500,1800
200, 400, 600
2ml H20 in the sample; 2ml H20 in the
crucible; CaCl2-2H20 as a chloridizing agent
The study of the effects of changing a single operating variable on the segregation
process is based on the following standard conditions:
Microwave power of 700 W; reaction time of 5 minutes; 10 g silicate laterite of -10 mesh
(Tyler); 5% calcium chloride addition; 6% charcoal addition of -150 mesh (Tyler); small
sample briquette; argon flow rate of 30060 mm3/s.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.1.1 Effect o f Reaction Time
The recovery of nickel was determined as a function of reaction time and the results are
shown in Figure 5.17. It can be seen that a maximum is reached at a reaction time of
about 30 min. For reaction times of less than 30 minutes, low microwave processing
times result in low energy absorption, therefore, the sample temperature was low. As was
mentioned in section 2.1.1.2, the processing temperature is one of the major factors
which affects the nickel recovery, and the optimum temperatures as observed by many
other researchers1161719201, is usually above 900°C. At low temperatures the kinetics of the
chloridization reaction is slow which results in insufficient generation of NiCh vapor and
thus low nickel recoveries. This result is in agreement with the theory discussed in
section 3.1. At microwaving times above thirty minutes the nickel recovery decreased.
This result is in agreement with Mehotra et al’s161 research and may be due to the
reoxidation of nickel.
5.2.1.2 Effect o f Microwave Power
The effect of microwave power on the nickel recovery is shown in Figure 5.18. The trend
indicates that the nickel recovery increases linearly with the microwave power. At the
higher microwave powers, a larger amount of energy is absorbed by the sample and thus
the temperature increases which results in an increase in the recovery.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
45
40
35
30
25
20
15
10
5
0
0
10
20
30
40
P rocessing Time (minute)
Figure 5.17 Nickel Recovery versus Reaction Time for the Microwave
Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
Metallic Nickel Recovery (%)
20
-
15 -
10
-
50
60
70
80
90
Microwave Power (% of 700 W)
Figure 5.18 Nickel Recovery versus Power for the Microwave Segregation
Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
100
5.2.1.3 Effect o f Particle Size o f Silicate Laterite
The effect of the particle size of the silicate laterite is shown in Figure 5.19. The sample
with a laterite particle size of -100 +150 mesh (0.104-0.147 mm) had the highest
recovery. Particle sizes larger than 65 mesh (>0.208 mm) exhibited relatively constant
recovery. Particle sizes smaller than 150 mesh (<0.104 mm) showed a dramatic decrease
in nickel recovery. As shown previously in Figure 5.13, the silicate laterite with a particle
size of -100 +150 mesh exhibited the best heating behavior. At this particle size range, a
higher microwave temperature is achieved resulting in a more efficient segregation
process.
5.2.1.4 Effect o f Am ount o f Charcoal as Reducing A sent
The effect of varying the amount of charcoal on the nickel recovery was also studied. The
amount of charcoal added was varied from 1 to 10%. The relationship between the nickel
recovery and the amount of charcoal added is shown in Figure 5.20. The curve shows that
the maximum nickel recovery is achieved when 6% by weight of charcoal is added.
Charcoal additions below 6 wt% do not produce a sufficiently reducing atmosphere and
the solid reductant surface area is not enough to reduce all of the NiCh vapor to nickel.
On the other hand, the addition of more than 6% by weight of charcoal results in the
generation of excess carbon monoxide. This leads to the in-situ reduction of NiO and this
hinders the chloridizing reaction. Therefore, the amount of NiO available for the
chloridizing reaction is reduced and thus less NiCU is produced.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
30
25
20
-
15
0
0.2
0.4
0.6
0.8
1
1.2
Laterite Particle Size (mm)
Figure 5.19 Nickel Recovery versus Silicate Laterite Particle Sizes for the Microwave
Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
25
20
0
2
4
6
8
10
12
Charcoal Addition(%)
Figure 5.20 Nickel Recovery versus the Amount of Charcoal Added for the Microwave
Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.1.5 Effect o f Charcoal Particle Size
The effect of the particle size of the charcoal reducing agent on the nickel recovery is
shown in Figure 5.21. It can be seen that the nickel recovery is low for charcoal particle
sizes larger than 150 mesh (> 0.104 mm) but for charcoal particle sizes of less than 150
mesh (< 0.104) the nickel recovery increased dramatically. As shown previously in
Figure 5.15, charcoal with a size of -150 mesh exhibited the highest temperature under
microwave conditions. Also, finer charcoal particle sizes provide a larger surface area for
the reaction, and as a result, the nickel recovery is higher.
5.2.1.6 Effect o f Chloridizine Agent
In order to determine the most effective chloridizing agent for the nickel segregation
reaction, various chloridizing agents (5%) were examined as follows: magnesium
chloride (MgCl2), ferrous chloride (FeClz), nickel chloride (NiCla). sodium chloride
(NaCl) and calcium chloride (CaCl2). The nickel recovery as a function of the
chloridizing agent is shown in Figure 5.22. The maximum nickel recovery was obtained
with CaCI2 as a chloridizing agent. CaCl2 is the most effective chloridizing agent since it
provides a higher HC1 pressure for a given water vapor pressure1111. The free energy
values for the decomposition of the chlorides to form HC1 at 900°C are summarized in
Table 5.3.
Table 5.3 Free Energy Values for Decomposition Reactions of Various Halide Salts at
900°C[61
Reaction
CaCh (s)+ H20 (g) + S i02 (s)= 2HC1 (g)+ CaO.Si02 (s)
2NaCl (s)+ HzO (g) + S i02 (s)= 2HC1 (g)+ Na20 .S i0 2 (s)
MgCl2 (s)+ H20 (g) +SiOz (s)= 2HC1 (g)+ MgO.SiOz (s)
FeCl2 (s)+ H20 (g) + S i02 (s)= 2HC1 (g)+ FeO.Si02 (s)
2NiCl2 (s )+ 2 H 2Q (g) +Si02 (s)= 4HC1 (g)+ 2NiO*SiQ2 (s)
Free energy change at
900°C, kcal/mol
-385.5
-339.5
-107.3
-7.62
-14.9
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
25
20
15
10
5
0
T
0
0.1
0.2
0.3
0.4
0.5
0.6
C harcoal Particle S ize (mm)
Figure 5.21 Nickel Recovery versus Charcoal Particle Sizes for the Microwave
Segregation Process for the Standard Conditions
ill
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
MgCI2
FeCI2
NiCI2
NaCI
CaCI2
Chloride Additives
Figure 5.22 Nickel Recovery versus Type of Chioridizing Agent for the Microwave
Segregation Process (The conditions were: Microwave power of 700 W, reaction time of
5 minutes, 10 g silicate laterite of -10 mesh, 5% chioridizing agent addition, 6% charcoal
addition of -150 mesh, argon flow rate of 1800 cm3/min)
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
It is evident from this table that in the case of the nickel segregation process, which
requires a higher partial pressure of HC1 for chloridization, CaCl2 is the most favourable
chloride. According to Brittan and Liebenberg^21', it also appears that CaCl2 stimulates
mineralogical changes that render the nickel oxide more amenable to chloridization.
Many researches have reported that CaCl2 is the most effective chioridizing
agent1” '13'22,53’.
5.2.1.7 Effect o f A m ount o f Calcium Chloride
The effect of varying the amount of CaCl2 in the reactant mixture was also investigated
and the results are shown in Figure 5.23. The trend indicates that the nickel recovery
increases with increasing CaCl2 addition. Larger amounts of CaCl2 result in the
generation of more HC1 vapor for the chioridizing reaction so that higher partial pressures
of HC1 are produced and thus more nickel oxide is chloridized.
5.2.1.8 Effect o f in ert Gas Flow Rate
The effect of varying the inert gas flow rate on the nickel recovery was also investigated.
The argon flow rate was varied from 1670 to 30060 mm3/s with the other operating
parameters as fixed variables. Figure 5.24 shows the relationship between the nickel
recovery and the gas flow rate. The result indicates that a flow rate of about 16700 mm3/s
is the optimum. Higher or lower flow rates result in lower nickel recoveries. The lowest
recovery of 15% was attained with a flow rate of 1670 mm3/s.
As mentioned in section 3.1, a reducing environment is required for the segregation
reactions. Displacement of oxygen by the inert gas in the reactor is necessary. Flushing
argon gas continuously through the reactor before, during and after heating is a standard
procedure. At low argon flow rates, the oxygen may not be completely displaced from the
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
30
10
5
----------------------------
0
0
2
4
6
8
10
12
Calcium Chloride Addition (%)
Figure 5.23 Nickel Recovery versus Amount of Calcium Chloride for the Microwave
Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
35
30
25
20
15
10
5
0
0
5000
10000
15000
20000
25000
30000
35000
Argon Row Rate (mm3/s)
Figure 5.24 Nickel Recovery versus Argon Flow Rate for the Microwave Segregation
Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
reactor. Any residual oxygen can react with the segregated nickel and consume the solid
reductant. On the other hand, excessively higher argon flow rates cause more of the metal
chlorides to be carried away by the gas before reduction occurs on the surface of the solid
reductant. This is particularly critical when the surface area to volume ratio of the sample
is high. Therefore, the argon flow rate in the experiments was maintained at the optimum
value of 16700 mm3/s. Mehrotra et al’s161 research showed a similar trend for the effect of
varying the inert gas flow rate on the nickel recovery.
5.2.1.9 Effect o f Preheating o f the Briquette
As discussed previously, the silicate laterite heats more readily after it is preheated. It was
observed that preheating increases the ability of the silicate laterites to absorb
microwaves. This characteristic of the laterite may facilitate the absorption of
microwaves during the segregation process resulting in higher temperatures and higher
recoveries. The effect of preheating the briquette before the microwave segregation
process was investigated. The briquette samples were preheated at varying temperatures
in the conventional resistance furnace for one hour. The results are shown in Figure 5.25.
It was found that the recovery decreased with increasing preheat temperature. Higher
recoveries were achieved with samples that were not preheated. There are two possible
reasons for these results. Firstly, moisture can be removed from the sample at all the
preheat temperatures. Secondly, charcoal can react with oxygen in the air. Both of these
functions would be expected to reduce the amount of reagents available for the
segregation process and thus the nickel recovery would decrease.
116
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
25
Metallic Nickel Recovery (%)
20
15
10
5
0
0
100
200
300
400
500
600
700
Preheat Temperature (°C)
Figure 5.25 Nickel Recovery versus Different Conventional Preheat Temperatures
for the Microwave Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.1.10 Effect o f Water additives
Water is an important reactant in the segregation process. Thus, the method of water
addition was studied. The moisture content in the system was increased by adding more
free water in the briquette, more free water in the reactor and also more chemically
combined water as CaCl2*2H20. It can be seen from Figure 5.26 that in comparison to the
standard sample, the recovery decreased with water additions to either the sample or in
the reactor. It is likely that in both cases, the water absorbs considerable microwave
energy and this reduces the microwave energy absorbed by the laterite. In the experiment
with CaCl2*2 H2 0 , the equivalent amount of calcium chloride was added as in the
standard sample. As can be seen the recovery increased in comparison to the standard
sample. This result could not be explained.
118
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
35
2 ml water in the
sam ple
2 ml water in the
reactor
CaCI2.2H20
Standard condition
Additives
Figure 5.26 Nickel Recovery for Various Water Additions for the Microwave
Segregation Process for the Standard Conditions
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2.2 E ffect o f Temperature on the Nickel Recovery an d the Grade fo r the
Conventional Segregation Process
As discussed previously in Section 5.2.1, the results for the effects of operating variables
on the nickel recovery in the microwave process were similar to those observed by
previous researchers for the conventional process. In the microwave process it was not
possible to study the effect of temperature on the nickel recovery because of the
difficulties associated with measuring the temperature of a sample with a non-uniform
temperature distribution in a microwave system, as discussed previously in Section 2.2.1
and Section 4.2.2.1. In order to determine the effect of temperature on the nickel recovery
in the concentrate, a series of experiments were performed using the conventional system
under the standard conditions. The recovery was determined using the Davis Tube Tester,
which is a magnetic separation technique. Thus the concentrate consisted of mainly
ferronickel and magnetite. Figure 5.27 shows the effect of temperature on the nickel
recovery for the conventional segregation process. The temperature was varied in the
range of about 750°C to 1200°C. The maximum recovery of about 87.52% was achieved
at a temperature of 1212°C. In the conventional smelting process for silicate laterites, the
smelting temperature is about 1500°C to 1600°C and the nickel recovery is also about
90% (Thumeyssen et al)[541. Thus, there is considerable potential for energy savings with
the segregation process.
Figure 5.28 shows the relationship between the nickel recovery and both the grade in the
concentrate and the grade in the ferronickel for the experiments discussed above. As
expected the nickel grade in both the concentrate and the ferronickel, decreased with
increasing nickel recovery as would be expected. The grade in the concentrate is lower
120
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Metallic Nickel Recovery (%)
100 1
80 60 40 -
20
-
700
800
900
1000
1100
1200
Temperature (°C)
Figure 5.27 The Effect of Temperature on the Nickel Recovery for the
Conventional Segregation Process
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
1300
Nickel Grade (%)
60
— □— Nickel grade in
concentrate for the
conventional process j
50
I
40
Nickel grade in
!
ferronickel for the
}
conventional process j
30
20
—o — Nickel grade in
concentrate for the
microwave process
10
0
0
20
40
60
80
M etallic Nickel R eco v ery (%)
100
Nickel grade in
ferronickel for the
microwave process
Figure 5.28 Comparison of the Nickel Recoveries and Grade for the Microwave
Process with the Conventional Process
122
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
|
than the grade in the ferronickel due mainly to the presence of magnetite in the
concentrate.
5.2.3 Nickel Grade in the Ferronickel
As mentioned in Section 2.1.1.3, the metallic particles from the segregation process
contained both iron and nickel. The nickel contents of the ferronickel from the
conventional experiments at two different temperatures of 750°C and 1050°C were
measured using an electron microprobe analyser (EMPA) and the results are shown in
Figure 5.29. The standard composition was utilized. Also included in Figure 5.29 are the
results for a microwave experiment at a reaction time of 2 minutes and 5 minutes and
again the standard composition was utilized. The metallic particles at the center of the top
surface and at the two opposite edges of the top surface were analyzed. It can be seen that
for the conventional segregation process, the particle compositions were similar
throughout the briquette. At 750°C, the ferronickel contained about on average 44 % Ni
while at 1050°C the nickel concentration was on average 13 % Ni. These results are
consistent with the results of other researchers*33*, who also found that the nickel content
in the metallic phase decreased with increasing temperature due to increased reduction of
the iron oxide. It also can be seen that for the microwave segregation process, the nickel
grade is higher at the center of the sample than at the edge by about 25%. Also the
average nickel grade is higher than the average results for the conventional samples.
In the conventional nickel segregation process, the sample is usually heated in a
resistance furnace or by a flame, and thus the energy is transferred from the exterior to
the interior via radiation and conduction processes. Also, eventually the sample
temperature is relatively uniform.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Nickel Grade in Ferronickel (%)
70
■Microwave process
for 2 minutes
60
50
■ Microwave process
for 5 minutes
40
30
□Conventional
process at 1050C
□ Conventional
process at 750C
20
10
0
z
<5^
<$T
®
jF
Position in the Sample
Figure 5.29 Nickel Grade in the Ferronickel at Different Positions on the
Cross-Section of the Segregated Product
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In the microwave process, since the heat is generated in the center of the sample and
transferred to the outside surface by conduction, then the reaction temperature
distribution is not uniform throughout the sample. Usually the temperature at the center
of the sample is higher than the outside surface. This was confirmed by infrared
pyrometer measurements which showed that the interior temperature was about 695°C
while the exterior was about 605°C. Although there are substantial errors associated with
these measurements it is clear that the temperature of the interior of the sample is
considerable higher than the surface. Figure 5.30 and Figure 5.31 show the different
heating mechanisms for the two processes. In the microwave process, the steep
temperature gradient from the inside to the outside of the sample results in the
preferential promotion of the nickel reduction reaction (3.1.4) as compared to the iron
reduction reaction (3.1.5). Figure 5.32 shows the standard free energy data for reactions
(3.1.4) and (3.1.5) at the temperature range of 900°C to 1600°C. It can be seen that
nickel chloride is more easily reduced than iron chloride at all temperatures. As a result,
in the microwave experiments, the nickel grade is higher at the center of the sample than
at the edge of the sample. In some cases, the nickel grade in the microwave process was
higher than that in the conventional process. Therefore, it appears that in the microwave
process the nickel segregation reactions are being promoted in preference to the iron
segregation reactions.
5.2.4 Microwave Experiments with Sm ail Briquettes
In the microwave experiments with the small briquettes it was observed that there was a
steep temperature gradient from the inside of the briquette to the surface. Since the
125
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Heat
Heat
Briquette Sample
Heat
Heat
Temperature °C
Final Tem perature Distribution
Initial Tem perature
D istribution
Sample Length
Figure 5.30 Conventional Segregation Heating Process and Temperature
Distribution in the Sample
126
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Heat
Heat
Briquette Sample
Heat
Heat
Temperature
Sample Length
Figure 5.31 Microwave Segregation Heating Process and Temperature
Distribution in the Sample
127
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20 i
o
E
“9
JC
®
-20
-40
-60
-80
-100
800 900 1000 1100 1200 1300 1400 1500 1600 1700
Tem perature (°C)
— NiCI2(g) + H2(g) = Ni(s) + 2HCI(g)
-«-FeC I2(g) + H2(g) = Fe(s) + 2HCI(g)
Figure 5.32 Standard Free Energy Change for Reactions (3.1.4) and (3.1.5)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sample temperatures in the microwave experiments were not known then the recoveries
in the microwave experiments were superimposed on the curve for the conventional
experiments in order to estimate the bulk sample temperature. In Figure 5.33 the nickel
recoveries from the microwave experiments are compared to the recoveries for the
conventional process. The microwave results are from the studies of the effects of power
and processing time since temperature is the major variable in these experiments.
From these results it can be seen that the bulk temperature in the microwave samples is
less than about 1000°C and thus the nickel recoveries are below about 36 %. Even with
alumina insulation it was not possible to improve the recovery. Because of the high
surface area to volume ratio of the small briquettes (48 mm'1), it was not possible to
achieve high temperatures in the interior and thus the bulk sample temperature was
relatively low.
5.2.5 Microwave Experiments with Large Briquettes
As mentioned in section 5.2.4. the smaller briquettes have a higher surface area to
volume ratio (48 mm'1), thus making it impossible to achieve high temperatures during
the microwave segregation process. Therefore, in order to reduce the heat loss, a larger
briquette with a surface area to volume ratio of 29 mm'1 was used. The briquette
composition was the same as the small briquettes.
It was found that the larger briquettes disintegrated due to the rapid evolution of water
from the sample. In order to prevent this phenomenon, the silicate laterite was first dried
in a conventional drying oven at 80°C for twenty-four hours and then at 120°C for an
other hour. As shown in Figure 5.34, the free moisture content of the silicate laterite is
relatively low (about 4%). The disintegration phenomenon is likely due to the preferential
129
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
60
Ni Recovery (%)
50
40
30
20
10
0
700
800
900
1000
1100
T em p era tu re (°C)
♦ Conventional process
■ Microwave process at different powers
• Microwave process for different processing tim es
Figure 5.33 Nickel Recovery visas Temperature for the Conventional Process (The
Microwave Results are Superimposed on the Conventional Curve)
130
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
5
4
3
%2
1
0
70
80
90
100 110
120 130
Temperature (°C)
140
150
160
Figure 5.34 Effect of Temperature on the Moisture Removal for Silicate Laterite
131
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
heating of the charcoal particles in the briquettes. This results in the rapid evolution of
water vapour around the briquette. The pressures in the interior of the briquette become
too large and the briquette explodes. For the briquettes with the smaller dimensions this
was not a problem.
Figure 5.35 shows a micrograph of the reacted sample and it can be seen that
considerable melting had taken place. The predicted composition of the oxide melt from
the segregation process is shown on the phase diagram in Figure 5.36. It can be seen that
the predicted liquidus temperature of the melt is about 1520°C. This demonstrates that
very high temperatures can be achieved in the large briquette. In some cases there was
partial melting in the middle of the sample, while in others there was total melting. The
experiment was not reproducible and this demonstrates that thermal runaway was
occurring.
The nickel recovery in the melted fraction of the sample was again low at about 35%. As
shown in Figure 2.2. the recovery in the conventional segregation process decreases at
higher temperatures. Also as shown in Figure 5.32, the reduction of iron chloride is not
possible above about 1100°C. If iron chloride can be not reduced this many hinder the
segregation reactions. Thus, in the microwave segregation process it is difficult to
achieve the optimum temperature for maximum recovery.
5.2.6 Morphology o f the Segregation Product
Micrographs of the segregated ferronickel (Fe-Ni) produced by both the conventional and
the microwave processes are shown in Figures 5.37 to 5.39. Small briquettes and the
standard conditions were employed. Figures 5.37 and 5.38 are micrographs of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.35 Optical Micrograph of the Melted Segregation Sample
from Large Briquette
133
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
TaO-MgO-VOt
CiyMaflin* P tm n
utolfan
OUdi Rsrmula
SiOt
(MS.WOSiO*
2(Mf>FMO'SiOt
WfcFWO
Liquidus temperature
of the melt
UfOSiOi
fM S U t
2MgO-SIO|
ZfWSOi
no. 1 .
O y i f iuo-^tcr-ao,; IM H * »
(OddtPkaMiiB
w H k ltH ttk h a m J .
Figure 5.36 Mg0-Si02-Fe0 Phase Diagram^56'
134
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 5.37 Optical Micrograph of the Conventional Segregation Product
a. 750°C and b. 1050°C
135
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Figure 5.38 Optical Micrograph of the Microwave Segregation Sample
a. 2 minutes, b. 5 minutes
136
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Figure 5.39 SEM Micrograph of the Microwave Segregation Sample under the
Standard Conditions
137
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
segregation samples taken with an optical metallograph. The Fe-Ni particles (lighter
phase) were fine (about 1pm) and roughly spherical and thus the morphology was similar
to that observed in the literature
f22|
. Some metallic nickel is associated with the charcoal,
but a number of metallic particles are not adjacent to reductant surfaces. For the
conventional process, Figure 5.37 (a) and (b), it can be seen that the segregated Fe-Ni
particles (lighter phase) are homogeneously distributed throughout the sample and this
indicates that the sample temperature was uniform. Figure 5.38 (a) and (b) shows that the
Fe-Ni particles in the microwave product were not evenly distributed which demonstrates
that the sample temperature is not uniform, and therefore the reaction degree is different
throughout the sample. A Scanning Electron Microscope (SEM) micrograph of the
microwave segregated product is shown Figure 5.39. The microwave segregated particles
are similar in shape and size to the particles observed in the conventional segregation
process. For both processes, the ferronickel particle size increases with temperature.
138
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6. CONCLUSIONS AND RECOMMENDATIONS
1. The dielectric constants of the silicate laterite ore and the limonitic laterite ore and
associated minerals were measured as a function of temperature. The dielectric
constants of both the ores were low, with the dielectric constant of the limonitic
laterite being higher than the silicate laterite. With regards to the major minerals,
again the dielectric constants were low. In the limonitic laterite, goethite had a higher
dielectric constant than limonite. In the silicate laterite, the only major mineral is
serpentine and again this mineral had a very low dielectric constant. The dielectric
constants of all the minor minerals were very low. For the ores and most of the
minerals, the dielectric constant increased rapidly above about 600°C. The dielectric
constant of serpentine was independent of temperature.
2. The microwave heating characteristics of the two ores and associated minerals,
charcoal, and the calcium chloride reagent were measured by determining the bulk
sample temperature as a function of time. Also, the effects of particle size and
conventional preheating temperature were evaluated. The ores and most of the
minerals are difficult to heat using microwaves. However, magnetite and charcoal
exhibited excellent microwave heating characteristics. Calcium chloride is not a good
microwave absorber. Preheating improved the microwave heating characteristics of
goethite, limonite and hematite. But there was no significant effect for serpentine,
kaolin or olivine. With regards to the effects of particle size, the bulk sample
temperature of the limonitic laterite was low at both small and large particle sizes. For
the silicate laterite, the bulk sample temperature increased dramatically at particle size
sizes ranging from -100 to +150 mesh (Tyler).
139
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3. Although laterite ores are not good microwave absorbers at low temperatures, the
presence of charcoal in the reacting mixture in the microwave process results in rapid
heating and the dielectric constant of the reacting mixture increases very rapidly with
temperature. In addition, the half-power penetration depth is relatively high and thus
rapid internal heating is possible.
4. The nickel recovery from the microwave segregation process increased with
increasing microwave power and the optimum reaction conditions were: a reaction
time of 30 minutes, a silicate laterite particle size of -100+150 mesh, a calcium
chloride addition of 10%, a charcoal addition of 6%. a charcoal particle size of -150
mesh, an argon flow rate of 1000 cm3/min. The trends observed, for the effect of
these additions, were similar to those observed for the conventional process.
5. The maximum nickel recovery in the microwave segregation process was 38.3%. The
low recovery in the microwave segregation process, as compared to the conventional
process, can be attributed to the non-uniform heating in the microwave process. For
the small briquettes, the high heat losses resulted in not only low surface temperatures
but also relatively low interior temperatures and thus the overall recovery was low.
For the large briquettes, the heat loss was reduced because of the lower surface area
to volume ratio. Thus, it was possible to achieve very high temperatures in the center
of the sample. However, since the dielectric constant of the reacting mixture increases
rapidly with temperature then thermal runaway occurred which resulted in melting.
Thus since the temperatures were too high the segregation reaction was limited. Thus,
in the microwave segregation process it is difficult to achieve the optimum
temperature for maximum recovery.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6. The morphology of the reacted product in the microwave segregation process was
similar to that observed in the conventional process, but in the microwave process the
particles were not uniformly distributed. However, the nickel grade in the ferronickel
was higher than in the conventional process.
7. The results indicate that it is difficult to achieve uniform heating in the microwave
process and thus the nickel recovery is low. However, it is likely that if microwave
heating was combined with conventional heating in a hybrid process then the nickel
recovery would be equivalent to that observed in the conventional process.
Furthermore, the rapid internal heating when using microwaves would result in
improved heat transfer into the reacting mixture and thus an increase in the
throughput rate in the reactor.
141
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7. REFERENCE
1. D.G.E.KERFOOT (1997), Handbook o f Extractive Metallurgy, Part 3, No. 12, pp.
715-791.
2. J.K.WRIGHT (1994), “Nickel Laterite Treatment by Segregation”, The Minerals,
Metals & Materials Society, pp. 219-232.
3. G.A.CRAWFORD (Feb. 1972), “Segregation of Nickel in Laterites - the
Falconbridge Experience”, Proceedings o f the Panel Discussion on Nickel
Segregation, San Francisco, California, USA, pp. 219-240.
4. A.A.DOR (Feb. 1972), “The Chloro-Metallization of Lateritic Nickel Ores”,
Proceedings o f the Panel Discussion on Nickel Segregation, San Francisco,
California, USA, pp. 145-210.
5. J.H.CANTERFORD (Jan.1975), “The Treatment of Nickeliferrous Laterites”,
Minerals Science and Engineering, Vol. 7, No. 1, pp. 3-17.
6. S.P.MEHROTRA, and V.SRINIVASAN (May-Aug.1994), “Extraction of Nickel
from an Indian laterite by Segregation Roasting", Transactions. Institute o f Mining
and Metallurgy (Section C: Mineral process and Extractive Metallurgy), Vol. 103,
pp. C97-C104.
7. I.HUDYMA (May-Aug.1993), “Role of Segregation Process Atmosphere in
Formation and Composition of Ferronickel Grains”, Transactions. Institute o f Mining
and Metallurgy (Section C: Mineral process and Extractive Metallurgy), Vol. 102,
pp. C125-C129.
8. J.Y.HWANG, X.LRJ, R.S.KRAMER and S.D.MCDOWELL (1995), “Microwave
Heating Characteristics of Selected Foundry Sands and Bentonite”, The Minerals,
Metals & Materials Society, pp. 307-321.
9. B.P.BARNSLEY (1989), “Microwave Processing of Materials”, Metals and
Materials, Vol. 5, N o .ll, pp. 633-636.
10. F. POLAND and M.E. PEASE (1959/1960), “Extraction of Copper and Silver by the
Segregation Process in Peru.Trans”, Institution o f Mining and Metallurgy, Vol.69, pp.
687-697.
1 1 .1.IWASAKI (1972), “A Thermodynamic Interpretation of The Segregation Process
for Copper and Nickel Ores”, Minerals Science and Engineering, Vol. 4, No.2, pp.
14-23.
142
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12. C.M.DIAZ (1958), Mechanism o f the Segregation Process and its Potential
Application to Nickel Ores, Thesis. Columbia University, New York, USA.
13. J.K.WRIGHT, and J.E.A.GOODEN (1973), “ The Treatment of Refractory Nickel
Oxide Ores by The Segregation Process - Laboratory Testing”, Australian Mineral
Development Laboratories Bulletin, No.15, pp. 49-63.
1 4 .1.IWSAKI, Y.TAKAHASI, and H. KAHATA (1961), “Extraction of Nickel from
Iron Laterites and Oxidized Nickel Ores by a Segregation Process", Transactions.
American Institute o f Mining Engineers, Vol. 225, pp. 308-320.
1 5 .1.IWASAKI (Feb. 1972), “Segregation Process for Nickel Ores”, Proceedings of the
Proceedings o f the Panel Discussion on Nickel Segregation, San Francisco,
California, USA, pp. 1-21.
1 6 .1.IWASAKI, A.S.MALICSI, and N.C.JAGOLINO (1973), “ Segregation Process for
Copper and Nickel Ores”, Extractive Metallurgy, Vol.l, pp. 127-186.
17. A.STOJSIC and F.SER (1970) “Method for Producing Nickel Concentrate from
Lateritic Ores”, Canadian Patent 848377.
18. Y.TAKAHASHI, K.NAGANO, K.KOJIMA (1970), “ Extraction of Nickel from Low
Grade Nickel Ores by Segregation Process ”, Proceedings IXth International Mineral
Proceedings Congress, Praha, pp. 319-327.
19. T.SUGARWARA and T.AGAYA (1964), “Method of Treating Oxidized Nickel
Ores”, Japanese Patent 39-16772.
20. S.NAKA, T.SHIMATANI, G.NAKAZAWA, M.IWAMOTO, and S.OMORI (1965),
“Process for Treating Oxidized Nickel Ores Containing Iron”, Japanese Patent 4016893.
21. M.I.BRITTAN (1970), “Kinetics of Copper Segregation by the TORCO Process”,
Africa Institute o f Mining Metallurgy, Vol. 70, pp. 278-289.
22. M.REY, V.FORMANEK, J.L.ORSINI, G.LUSSIEZ, and M.PIERRE (Feb.1972),
“ Influence of Various Factors on Segregation of Oxidized Nickel Ores”, Proceedings
o f the Panel Discussion on Nickel Segregation, San Francisco, California, USA, pp.
265-285.
2 3 .1.HUDYMA (Dec.1981) “Growth of Metallic Nickel Grains in the Segregation
Process: I - Growth and Aggregation of Metallic Nickel Grains on Coal Particles”,
Mineral Processing and Extractive Metallurgy, Vol. 90, pp. C147-C151.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24. A.S.ERICSON, J.SVENSSON and K.ISHII (1987) “ The MINPRO-PAMCO Nickel
Segregation Process”, International Journal o f Mineral Processing, Vol. 19, pp. 223236.
25. J.W.WALKIEWICZ, G.KAZONICH, and S.L.MCGILL (1988), “ Microwave Heating
Characteristics of Selected Minerals and Compounds”, Minerals and Metallurgical
Processing, Vol. 5, No.l, pp. 39-42.
26. J.D.FORD, and C.T.PE1 (1967), “High Temperature Chemical Processing via
Microwave Absorption”, Journal o f Microwave Power, Vol. 2, No.2, pp. 61-64.
27. Atomic Energy of Canada Limited Research Company and Voss Associates
Engineering LTD (1990), Microwaves and Minerals, Ontario, Canada, pp. 1-77.
28. T.T.CHEN, E.HAQUE.K, W.WYSLOUZIL, and S.KASHYAP (1984), “ The
Relative Transparency of Minerals to Microwave Radiation”, Canadian
Metallurgical Quarterly, Vol. 23, No.3, pp349-351.
29. D.K.XIA and C.A.PICKLES (1997), “ Applications of Microwave Energy in
Extractive Metallurgy, a Review”, Canadian Institute o f Mining and Metallurgy
Bulletin, Vol. 90, pp. 96-107.
30. M.I.BRITTAN, and R.R.LIEBENBERG (Feb. 1972), “Kinetics and Mechanism of
Nickel Segregation”, Proceedings o f the Panel Discussion on Nickel Segregation, San
Francisco, California, USA, 1972, pp. 72-107.
31. M.REY (1967), “Notes on the Theory of the Copper Segregation Process”, L ’Ecole
Nationale Superieure des Mines de Paris
32. Y.TAKAHASHI, K.KOJIMA, and H.KAHATA (1966), “On Segregation Roasting
Reactions of Nickel ores”, Iron and Steel Institute o f Japan, Vol. 52, pp. 1310-1312.
33. M.REY (1967), “Early Development of the Copper Segregation Process”, Institution
o f Mining and Metallurgy, Transaction (Section C: Mineral Processing and
Extractive Metallurgy), Vol. 76, pp. 101-107.
34. H.MIMURA, et al. (1966), “On Application of the Segregation Process to gamierite”,
Nippon Kogyo Kaishi, Vol. 82, No. 514, pp. 514-518.
35. P.M.PERLOV, A.I.YESKIN, and N.V.ZASHICHIN (1968), “New Trends in
Combined Processes of Nickel, Copper and Cobalt Recovery from Non-responding
Concentration Ores, and from Refractory Flotation Middlings” Proceedings o f VIII
International Mineral Processing Congress, Vol. 2, Paper E-7, pp. 77-86.
36. R.V.DECAREAU (1977), “ The Amana Story”, Microwave Energy Application,
Newsletter, Vol.10.
144
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
37. P.L.JONES and A.T.ROWLEY (1996), “Dielectric Drying”, Drying Technology,
Vol.14, No.5, pp. 1063-1098.
38. R.F.SCHIFFMANN (1976), “An Update on the Applications of Microwave Power in
the Food Industry in the United States”, Journal o f Microwave Power, Vol. 2, pp.
221- 224.
39. (Nov.1968) “Microwave Drying is Becoming Practical”, Canadian Chemical
Processes, Vol.52, pp.67-71.
40. G.FREEDMAN (1972), "The Future of Microwave Power in Industrial Application”,
Journal o f Microwave Power, Vol. 7, pp. 353-365.
41. J.W.WALKIEWICZ, G.KAZONICH, and S.L.MCGILL (Feb.1998), “Microwave
Heating Characteristics of Selected Minerals and Compounds", Minerals and
Metallurgical Processing, pp. 39-42.
42. P.S.SCHMIDT, T.L.BERGMAN, and J.A.PEARCE (1972), “Heat and Mass Transfer
Considerations in Dielectrically-Enhanced Drying", International Symposium on
drying Proceedings, pp. 137-158.
43. C.A.PICKLES and D.K.XIA (1997), “Microwave Drying of Ferric Oxide Pellets”,
1997 Ironmaking conference proceedings, pp. 329-341.
44. S.O.NELSON and A.W.KARSZEWSKI (1990), “Dielectric Properties of Materials
and Measurement Techniques”, Drying Technology, Vol. 8, No.5, pp. 1123-1142.
45. R.G.FORD (1988), “ Microwave processing Symposium Report”, Materials in
Processing Report, Vol.3, No.4, pp. 1-4.
46. W.H.SUTTON (1989), “Microwave Processing of Ceramic Materials”, Ceramic
Bulletin, Vol. 68, No.2, pp. 376-386.
47. F.GARDIOL (1984), Introduction to Microwaves, pp. 394-400.
48. C.SALTIEL and A.K.DATTA (1999), “Heat and Mass Transfer in Microwave
Processing". Advances in Heat Transfer, Vol.33, pp. 1-94.
49. E.WOLSKA and U.SCHWERTMANN (1989), “ Nonstoichiometric Structures
During Dehydroxylation of Goethite”. Zeitschrift fuer Kristallographie, Vol. 189, pp.
223-237.
50. W.A.DEER, R.A.HOWIE, and J.ZUSSMAN (1996). An Introduction to the RockForming Minerals, Longman, London
145
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
51. N.STANDISH, H.K.WORNER, and D.Y.OBUCHOWSKI (1991) “ Particle Size
Effect in Microwave Heating of Granular Materials”. Powder Technology, Vol. 66,
pp. 225-230.
52. S.E.KEEGAN and R.KEMP (1997), Handbook o f Extractive Metallurgy, Part 13,
No. 12. pp. 2311-2317.
53. F.SER, and A.STOJSIC (1968), “Recovery of Nickel for Laterite Ores by a
Combined Method of Segregation and Flotation”, In 8Ih International Mineral
Processing Congress, Leningrad, Vol. 1, pp. 632-641.
54. C.G.THURNEYSSEN, J.SZCAENIOWSKI. J.MICHEL (1960),
Smelting in New Caledonia”, Journal O f Metal, Vol.12, pp.202-205.
“Ferro-nickel
55. K.OKAMOTO, Y.UEDA and F.NOGUCHI (1971), “ Extraction of Nickel from
Gamierite Ores by the Segregation-Magnetic Separation Process”, Nippon Kogyo
Kaishi 87, pp. 179-184.
56. E.F.OSBORNE and A.MUAN (1960), "Phase Equilibrium Diagrams of Oxide
Systems”, Ceramic Foundation, Plate 8, pp. 236.
146
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX A
A Standard Sizing Scale Based on the Standard 200 mesh Screen
MUUmiUn
Example
S iain s m ethod
Mioroci
ae.07
u .n
direr gnva!
13. S3
0.433
s. no
4.300
3.337
3.333
1.351
1.133
0.333
0.080
0.417
0.300
0.308
0.147
0.104
0.074
0.003
0.037
3
4
3
3
fee ■rmrel
10
14
30
33
35
48
35
100
53
87
33
18.0
13
0.30
3.0
4.83
3.30
3.33
1.83
1.13
0.81
0.08
0.41
0.30
150
330
370
400
Floe e ilt
(800)
Blood celle
(1.300)
(3.300)
Many creme
W ar* len g th o f v iaib la lia h t
0.30
0.14
0.10
0.07
0.00
0.030
0.030
0.017
0.013
0.008
Thinneet irideeoent fllme r U b b by
light iatatfaranea
0.004
0.003
0.003
0.0015
Vary large molerabe
0.003
0.001
0.0007
0.0003
A w a it unit acyatal
A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX B
The Dielectric Constants of Nickel Laterite Ores and Associated Minerals Vs
Temperature as Measured by MPN
1* : Frequency:2466 MHz ; Plot coding: x
2* : Frequency: 912 MHz; Plot coding: □
25
isi
**,1
XXX
«*b,2
a a a
10
a in
200
1200
W ’V l
Temperature (C)
Silicate Laterite Ore
B
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10
1
I
1200
200
IW l,1W.2
Temperature (C)
Limonitic Laterite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Goethite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Limonite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
0
LOO
200
300
400
500
600
700
MO
900
Temperature (C)
Serpentine
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1000
■e
MJ
100
200
300
400
500
600
700
000
Temperature (C)
Hematite
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission
900
1000
Temperature (C)
Magnetite
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
1
_Q
°
8* X
s
□
o
+
*
9 9 *
(fu
i
ta
'* .,2
*
3
d a a
200
400
600
800
1000
Tm ,l,T* ,2
Temperature (C)
Kaolin
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1200
0
100
2M
300
400
MO
T
<00
7M
MO
9M
T
B,r»,2
Temperature (C)
Olivine
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10M
VITA
Name:
Jia Ma
Place and year of birth:
Beijing, P.R.China; 1972
Education:
Nanjing University of Chemical Technology, P.R.China
1991-1995
B.Sc. (Chemical Engineering, 1995)
Experience:
Junior Engineer
Beijing Research Institute of Chemical Industry, P.R.China
1995-1999
Research Assistant
Department of Chemical Engineering
Queen’s University, 1999-2002
Teaching Assistant
Department of Mining Engineering
Queen’s University. 1999-2001
Awards:
Queen’s Graduate Award, 2000-2001
Scholarship of Nanjing University of Chemical Technology
Reproduced with permission o f the copyright owner. Further reproduction prohibited without permission.
Документ
Категория
Без категории
Просмотров
0
Размер файла
5 216 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа