Metallothermic Processes

(Self-propagating High-temperature
Synthesis Method)

Prof. Dr. Onuralp Yücel

Istanbul Technical University,

Department of Metallurgical & Materials Engineering,
Maslak, Istanbul, 34469, Turkey
1
2
3
4
5
6
7
 Main Reaction Parameters for
SHS and Metallothermic Reactions
• Specific heat;
determining the heat evolved during the reaction and to estimate if the temperature
achieved is sufficient to smelt the metal and the slag and to separate them due to the
different density.
calculated by dividing the enthalpy of the reaction at 25°C by the sum of the
molecular weights (MW) of the reaction products.
MeIO + MeII  MeI + MeIIO

GR =GMeIIO – GMeIO < 0

Specific heat < 2250 J/g, insufficient to melt the charge and to separate metal/slag
Specific heat > 4500 J/g, violent reaction and may even be explosive.
Specific heat : 2250 – 4500 J/g, a controlled and self-sustaining metallothermic
reaction

Spesific Heat = ΔH 298 K / Σ MWProduct

8
 Main Reaction Parameters for
SHS and Metallothermic Reactions
• Adiabatic combustion temperature (Tad);
It is suggested that a combustion reaction will self-propagate when Tad is higher than
1527 °C

Without Al2O3
Adiabatic Temperature, °C

Adiabatic Temperature, °C

Al addition, mole Mg addition, mole

Simulation of Tad change in Aluminothermic and Magnesiothermic Reactions

9
A Metal Joining applications in Istanbul Metro

10
In our laboratories, we have studied:

• Some metals (V, Cr, B)

• Alloys and ferroalloys (FeB, FeV, FeW, FeCr,
FeMn, FeMnSi, FeMo, CrNi, AlTiB)
• Some advanced ceramics (TiB2, B4C, WxBy)

The process parameters of each production study

and thermochemical considerations were also
discussed.

11
 Selection of Reductant Materials

Reductant/ Melting Point, Boiling Point, -∆H298 , Metal price, Density

Product °C °C kJ/mol O $/kg g/cm3
Al 660.4 2519 - 2,72 2,7
Mg 649 1088 - 3,28 1,74
Ca 839 1484 - N/A 1,54
Si 1410 2355 - 2,65 2,33
Al2O3 2050 558.1 3,96
CaO 2570 634.7 3,34
MgO 2825 601.6 3,58
SiO2 1720 303.6 2,60

12
13
14
 Aluminothermic Reduction
 FeTiO3(s) + 2Al(s) = [Fe – Ti](l) + Al2O3(slag)

 3TiO2(slag) + 2Al(s) = 3TiO(slag) + Al2O3(slag)

 3TiO + 2Al = 3Ti + Al2O3

 3TiO + 2Al = 3Ti[Fe] + Al2O3

Aluminothermic ilmenite reduction of 20-25 % Ti alloy produced by the

addition of magnetite is done in a way.

 3Fe3O4 + 8Al = 9Fe + 4Al2O3

 FeTiO3 + 2Al = [Fe – Ti](l) + 2Al2O3

 3SiO2(slag) + 4Al = 3[Si](l) + 2Al2O3(slag)

 MoO3 + Al ―› Mo + Al2O3

29.12.2020 15
 Silicothermic Reduction
 2MgO + Si = 2Mg(g) + SiO2

 2Mg(g) + O2(g) ―› 2MgO(s)

 10CaO + 5Si ―› 2CaSi2 + Ca2SiO4 + 6CaO

 10MgO + 2CaSi2 +6CaO ―› 10Mg(g) + 5Ca2SiO4

 10(CaO.MgO) + 5Si ―› 10Mg(g) + 5Ca2SiO4

29.12.2020 16
 Reduction of Uranium Compounds
 3UO3 + Ca ―› U3O8 + CaO

 U3O8 + Ca ―› 3UO2 + CaO

 UO2 + Ca ―› U(l) + CaO

 UF4(s) + 2Ca(s) ―› 2CaF2(l) + U(l) ΔH = 156 kcal/mol

 UO2 + Mg(g) ―› 2MgO + U(l)

 UF4 + 2Mg ―› U + 2MgF2 ΔH = -8,4 kcal/mol

29.12.2020 17
 Niederschlag Process
 Me1X + Me2 = Me1 + Me2X
 Sb2S3 + 3Fe(s) = 2Sb(l) + 3FeS(k)
Very high grade (over 50% Sb) or pure crudum (Sb2S3) is reacted with iron.

Deoxidation in Iron and Steel Metallurgy

 3FeO(metal) + 2Al = 3Fe(metal) + Al2O3(s,l)

Cementation
 Me1n+ + Me02(s) = Me01(s) + Me2n+
 Cu+2 + Fe0(s) = Cu0(s) + Fe+2

29.12.2020 18
19
20
PRODUCTION OF FERROCHROMIUM

21
High Carbon Ferro Chromium Production
 LC FeCr Production
Perrin Prosesi

25-27 % Cr2O3
7-8 % FeO 40-45 % Si
2-3 % SiO2 45-40 %Cr
45-48 % CaO 0.05-0.02 %C

~70 % Cr
>1.5 %Si 8-10 %
0.05-0.02 %C Cr2O3

20-25 % Si
60-55 %Cr
0.05-0.03 %C

23
24
25
26
27
28
29
30
31
 Production of LC Ferrochromium by the Perrin Process

Perrin Process based on a silicothermic reduction of a chrome ore - lime melt

(rich slag) with a ferrosilicon-chromium (Fe-Si-Cr) alloy by using two ladle
steps. In the first step, basically, the Fe-Si-Cr alloy produced in a submerged
arc furnace is cast into a ladle and reacted with a molten intermediate slag
taken from the second ladle to produce intermediate alloy and waste slag.
Meanwhile, in the second step, the rich slag produced in an open arc furnace
is cast into a basic-lined ladle together with crushed intermediate Fe-Si-Cr
alloy taken from the first ladle to form LC ferrochromium and intermediate
slag.

32
33
The optimum quantity of intermediate Fe-Si-Cr alloy (60 wt% Cr, 20 wt %Si, 19.95 wt%
Fe and 0.05 wt% C) to reduce 100 g of molten rich slag (27 wt% Cr2O3, 48 wt% CaO, 10
wt% FeO, 6 wt% MgO, 6 wt% Al2O3, and 3 wt% SiO2) was estimated simulating the
process conditions of the second ladle with “Equilib” module of FactSage 6.0. The liquid
phases occurring in the reaction process were selected as Fe-liquid (FSstel database,
includes Fe, C, Cr, Al Si, Mg and O) and as ASlag-liq (FTOxid database, includes Al2O3,
SiO2, CaO, FeO, Fe2O3, MgO, CrO,Cr2O3), whereas all gas and stoichiometric solid phases
were taken from the Fact53 database.

The reaction of the process was assumed as adiabatic (∆H=0) and the initial reaction
temperature was selected as 1700 °C. The adiabatic condition was searched for using the
temperature while adding a variable amount <A> of Fe-Si-Cr with <A> in the range
between 0 and 100 g.

Figure 3 shows that as the quantity of intermediate Fe-Si-Cr alloy added to the molten rich
slag increases, the degree of simultaneous reduction of Cr3+ to Cr2+ in the slag and a
subsequent transfer of Cr° to the alloy phase increases.

34
 31.25 g

Simulation result of effect of intermediate Fe-Si-Cr alloy on Perrin Process

for the production of LC ferrochromium
35
As the amounts of CaO and Al2O3 remain unchanged in the slag, SiO2 content
increases with increasing intermediate Fe-Si-Cr addition due to the oxidation of
silicon. The addition of the intermediate Fe-Cr-Si to the molten rich slag should be
limited to a level that does not increase the Si content in the molten LC FeCr alloy
higher than that of desirable specification limits (<1.5 wt % Si). For example, the
results of the calculation showed that in order to obtain an alloy with % 70 Cr
content, the amount of intermediate Fe-Si-Cr should be selected as 31.25 g and the
adiabatic temperature of the process then was found to be 2395 °C.

The output substances of calculated LC FeCr alloy with 70% Cr

36
37
 Self Propagating High Temperature Synthesis (SHS),
Metallothermic Reduction Applications

Slag
Slag
Greenand
Smelting
Solidification
Solidification
Combustion
Combustion
mixture
Phase
and
andSeperation
cooling
cooling
FeCr
FeCr
Liquid SlagAlloy
Alloy
Liquid Metal

• Metals, alloys, ferroalloys • Organic and inorganic

• Advanced ceramics, compounds,
• Intermetallics, • Polymers
• Oxygen free single crystals • Welding and metal joining
In these processes, after initiation, reaction becomes self-sustaining and propagates in the
reactant mixture.

A high amount of heat which is generated during the process accelerates the reaction rate and
thus it makes the process highly productive and economically feasible for different
production scales. 38
 Enthalpies of reactions and Specific heat of the products

Specific Equation
Reaction H298, kJ
heat, J/g No
Cr2O3 + 2 Al  2Cr + Al2O3 -535.573 -2600.45 (1)
Cr2O3 + 2 Al + CaO  2 Cr + CaO.Al2O3 -551.514 -2104.75 (2)
Cr2FeO4 + 8/3Al  Fe + 2Cr + 4/3Al2O3 -775.016 -2620.17 (3)
Cr2FeO4 + 8/3Al+4/3CaO  Fe+2Cr + 4/3CaO.Al2O3 -796.265 -2148.81 (4)
CrO3 + 2Al  Cr + Al2O3 -1085.748 -7052.24 (5)
CrO3 + 2Al + CaO  Cr + CaO.Al2O3 -1101.689 -5245.21 (6)
NaClO3+2Al  NaCl+Al2O3 -1728.661 -10776.9 (7)
FeO + 2/3Al  Fe+1/3Al2O3 -291.155 -3241.03 (8)

Since some oxides in the chromite concentrate such as Al2O3, SiO2, MgO etc. may decrease the
reaction heat during aluminothermic process, in the same cases, additives such as sodium
chlorate (NaClO3) and chromic acid (CrO3) may be used to increase the reaction heat as shown
in Eq 5 and 7.

When CaO and CaF2 are added to the charge to decrease the melting point of the slag, the heat of
the reaction is still sufficient to advance the reaction.
39
 FERROBORON
H298 Specific heat
Reaction
kJ J/g
B2O3 + 2 Al + 2 Fe  2FeB + Al2O3 - 402 2310
B2O3 + 4 Al + Fe2O3  2 FeB + 2 Al2O3 - 1255 4140

40
 FERROVANADIUM
H298 Specific heat
Reaction
kJ J/g
V2O5 + 11.8 Al + 0.9 Fe2O3  Fe1.8V6 + 5.9 Al2O3 - 4860 4800

41
 FERROTUNGSTEN
Specific
Reaction H298, kJ
heat, J/g
WO3 + 2 Al  W + Al2O3 - 832 2912
1.5 Fe2O3 + 7 Al + 2 WO3  Fe3W2 + 3.5 Al2O3 - 2973 3332
1.5 Fe2O3 + 7 Al + 2 WO3 + 7 CaO  Fe3W2 + 3.5
- 3017
Ca2Al2O5 2348
1.5 SiO2 + 2 Al  1.5 Si + Al2O3 - 309 2144
KClO3 + 2 Al  KCl + Al2O3 - 1501 8503

90 100 80 100
85 90 70 90
80 80
W, Fe in alloy, wt. %

80
60

W, Fe in alloy, wt. %
Recovery of W , %
75
70

Recovery of W, %
W 70
70 50
Fe 60 60
65 Recovery
60 50 40 W 50
55 Fe 40
40 30 Recovery
50 30
30
45 20
20 20
40 10
10 10
35
30 0 0 0
80 90 100 110 120 130 80 90 100 110 120 130
Stoichiometric Al, % Stoichiometric Al, %

42
calcine 26 wt.% W Calcine 40 wt.% W (KClO3/Calcine ratio: 20 %).
 Ferroniobium
• The refractory metals Nb, V, Ti, Mo, and W, when used in steelmaking, are
produced via aluminothermic reduction in the form of a ferroalloy.
• This is done for two main reasons:
– the alloy has a lower melting point, being close to one of the eutectics of the
binary system
– the energy of the iron oxide (hematite) released during the reduction process is
necessary to melt the bulk of the mixture autothermically and to permit good
metal-slag separation.
• Approxiamately 85-90 % of total niobium production is used in the steel industry
in the form of iron niobium alloy (ferroniobium) containing 40-70 % niobium.
• Ferroniobium is usually produced by aluminothermic reduction of niobium oxide
ores with the addition of iron oxides if the niobium ore used contains insufficient
iron.
• The starting materials are mainly columbites and pyrochlore concentrates.

43
 Traditional Production of Ferroniobium
• The traditional method of producing ferroniobium from pyrochlore concentrate was by
batch aluminothermic reduction. A typical batch was composed of the following:
 Pyrochlore concentrate : 18000 kg
 Hematite: 4000 kg
 Aluminium powder: 6000 kg
 Fluorspar: 750 kg
 Lime: 500 kg
• Each batch produced approximately 11 t of ferroniobium containing 66 % niobium. The
plant was capable of firing six batches per day, giving an annual capacity of 22,8 kt per
year operating 345 days. The yield was around 96 %.
• The reaction gives 10000 kg
of ferroniobium of composition: and 20000 kg of slag containing:

44
 Traditional Production of Ferroniobium
• The reactor consisted of a steel cylinder lined with one layer of magnesia bricks,
3,8 m in diameter and 1,8 m in height. Six of these cylinders were positioned over
concave pits, lined with lime and fluorspar, prepared on a sand bed.
• Externally, the reactor base was sealed with sand, and the mixture that had been
stored in the steel containers was charged to the reactors.
• The reaction was initiated with magnesium turnings, and the total reaction time
was about 20 minutes. During the reaction in open air, an enormous volume of
fumes, mainly alumina, came from the reactor.
• Metal and slag are readily separated by gravity. The metal settled in the concave
pits in the sand bed, and the slag floated above the metal and was contained by the
reactor. It was then tapped into a nearby pit when the reactor had been raised from
the sand bed with an overhead crane.
• Some 16 hours after the reaction had been completed, the slag was removed from
the sand bed and transferred to a 35 t dump truck and hauled to a disposal area.
• The ferroniobium 'button' was kept in the sand bed for about 6 hours until it
solidified, and was then also removed from the sand bed. Af ter 12 more hours of
cooling in air, the ferroniobium was crushed in a jaw crusher, sampled, and stored.
45
Traditional Production of Ferroniobium

46
 Autothermic Semi-continuous Process
• The original ferroniobium plant in Araxa was designed in the early 1960's and its
capacity was gradually increased, basically by enlarging the reactor size.
• Despite offering a high yield and smaIl investment costs, the operational and
environmental conditions demanded a new engineering concept, incorporating
modem materials handling, more safety, and fume control.
• Studies carried out indicated that the 'fed reaction process' is the best alternative.
• In this process, the reaction occurs in a fixed vessel during mixture feeding. This
enclosed controlled reaction, occurring in different cycles of a single reactor,
makes it easier to treat the fumes generated. After the batch feeding has been
completed, the slag is tapped through the upper hole, allowing water granulation,
and afterwards the metal is tapped through the bottom hole, allowing casting.
• In order to save on the aluminium powder, which is used to reduce the hematite,
the use of an electric furnace as an external source of energy was considered,
which would permit the replacement of the hematite by scrap iron.

47
Autothermic Semi-continuous Process

48
 Autothermic Semi-continuous Process
• Studies, including some exploratory tests at Mintek, a South African research
organization, and several meetings with manufacturers of electrical furnaces, led
to the following conclusions.
– Investment in an electric pilot fumace would be of the same magnitude as that
in an industrial fumace.
– Most of the pending questions such as metal-slag separa-tion and tapping, slag
granulation, and refractory life are common to both options.
• As a consequence of these conclusions, it was decided to conduct a series of pilot
tests on the autothermic 'fed reaction process'. A pilot plant with a fixed reactor of
1,6 m in internal diameter and 1,7 m in height was set up in May 1989, and a
modus operandi was gradually developed during that year by use of the present
autothermic mixture.
• Based on the data generated during the pilot phase, it was decided to design an
industrial facility with a reactor con-figuration that would, if necessary in the
future, accept an external heat source to optimize the aluminium saving.

49
Process Flow Diagram of the Autothermic Semi-Continuous Process
50
 FERROMANGANESE & SILICOMANGANESE

Products Process Mn Recovery, Energy Consumption, Coke,

% (kWh/t) (kg/t)
EAF 60-75 2600-2800 350
HC FeMn BF 80-85 - 2000
SiMn EAF 70-80 3500-4200 850-1000
LC FeMn EAF 60-85 1600-1900 -
Metallothermic 63-85 - -
Metallic Mn Electrometallurgy 70-87 9000 -

Reactions -ΔH298 , Specific Heat,

(kJ) (J/g)
Mn2O3 + 2Al = 2Mn + Al2O3 718.4 3391
Mn3O4 + 2.667Al = 3Mn + 1.333Al2O3 845.3 2810
Mn2O3 + Fe2O3 + 4Al = 2FeMn + 2Al2O3 1569.4 3688
2.5Mn2O3 + 3SiO2 + 9Al = Mn5Si3 + 4.5Al2O3 2614.8 3197
Mn2O3 + 1.5Si = 2Mn + 1.5 SiO2 409.4 2057
Mn3O4 + 2Si = 3Mn + 2 SiO2 433.9 1522
Mn2O3 + Fe2O3 + 3Si = 2FeMn + 3 SiO2 951.5 2367
Mn2O3 + Fe2O3 + 4Al + CaO = 2FeMn + CaO.2Al2O3 1609.6 3342
1.5 Mn2O3 + Fe2O3 + 3 Al + 1.5 Si + 1.5 CaO = 2Fe+ 3Mn + 1.5 CaO.Al2O3. SiO2 1735.1 2874
KClO3 + 2Al = KCl(g) + Al2O3 1500.8 8502
51
 FERROMANGANESE
90 20 20 80
80 18 18 70
Mn, Fe in alloy, wt.%

Al, Si in alloy, wt. %

70 16 16 60

Recovery of Mn, %
Mn

Mn in slag, wt. %
60 14 14
Fe 50
Si 12 (Mn)
50 12
Al 10 40
40 10 Recovery
8 30
30 8
6 20
20 6
4
4 10
10 2
0 0 2 0
50 60 70 80 90 100 110 120 130 140 50 60 70 80 90 100 110 120 130 140
Stoichiometric Al, %
Stoichiometric Al, %

SILICOMANGANESE
80 22 20 100
20 90
70 18
Si, Al, Fe in alloy, wt. %

18

Recovery of Mn, %
80

Mn in slag, wt. %
60 16 16
Mn in alloy, wt. %

Mn 70
50 Si 14 14
Fe 12 60
40 12
Al 10 50
10
30 8 40
6 8 (Mn) 30
20
4 6 Recovery 20
10
2 4 10
0 0
2 0
0 0.1 0.2 0.3 0.4 0.5
0 0.1 0.2 0.3 0.4 0.5 52
Si / Al, wt.
Si / Al, wt.
 FERROMOLYBDENUM
MoO3 + 2 Al → Mo + Al2O3 ΔH = – 915.0 kJ
Fe2O3 + 2 Al → 2 Fe + Al2O3 ΔH = – 847.8 kJ
To estimate SHS reaction process: Released Energy per g Product: 4433.65 J

Purity %
Fe2O3 96.5
FeMo Metals

Al 96.0
MoO3 98.0

Slags

ICP Spektrometer EPMA XRD

Mo % Fe % Si % Al % Cu % Cr % Ni % C% S%
62 – 69 27 – 32 1.5 – 2.1 0.5 – 1.1 0.4 – 1.0 < 0.2 < 0.2 < 0.1 < 0.05
53
July 2009

54
WO3 + 2 Al  W + Al2O3 H0298 = -832 kJ
1.5 Fe2O3 + 2 WO3 + 7 Al  Fe3W2 + 3.5 Al2O3 H0298 = -2973 kJ
1.5 Fe2O3 + 2 WO3 + 7 Al + 7 CaO  Fe3W2 + 3.5 Ca2 Al2O H0298 = -3017 kJ
1.5 SiO2 + 2 Al  1.5 Si + Al2O3 H0298 = -309 kJ
KClO3 + 2 Al  KCl (g) + Al2O3 H0298 = -1501 kJ

55
56
57
58
Akış Şeması

Magneziyotermik Redüksiyonla
Titanyum Üretiminde Kullanılan
Kroll Tipi Bir Reaktörün
Şematik Yapısı

59
60
61
 MAGNESIUM

Mg Metal

∆G°-T diagram for oxide formation indicates that MgO reduction with silicon is only possible
beyond ; 2140 °C at 1 bar and 1312 °C at 1mbar
62
 2 MgO.CaO + Si = 2 Mg (g) + Ca2SiO4

2 MgO.CaO + FeSi = 2 Mg (g) + Ca2SiO4 + Fe

2 MgO.CaO + 2 Al = 2 Mg (g) + Ca2Al2O4 63

The vapor pressure changes of the Ca, Sr, Mg and Al
at 1 mbar and 1bar by temperature

64
65
* Pidgeon Prosesi Patent No: US5.658.367 (1997)
66
Dolomitin Redüksiyonu
• Biriketler redüklenme
işleminin yapıldığı retort
bataryalarına şarj edilirler.
• Fırın içi sıcaklık 1200-
1250C’dir.
• Vakum pompalarıyla retort
içi basınç 0.13-0.06 mbar
seviyelerine getirilmektedir.
• Retortların fırın dışındaki
ucunda bulunan
kondansörlerde magnezyum
buharı birikerek yoğunlaşır ve
“taç şekilli Mg tanecikleri”
olarak adlandırılan dendiritik
yapı haline dönüşür.
67
 1- Furnace, 2-Stainless steel retort, 3-
Briquetted charge, 4-Mg condensation
section, 5-Cooling water in and out, 6-
Vacuum connection, 7-Vacuum pump,
8-Digital pressure gauge, 9-
PtRh30/PtRh6 thermocouple, 10-Ice
water box, 11-Temperature measuring
unit.

Mg
Fe Na Al Ca K Si Mn
Metal
0.60 0.005 - 0.05 0.006 0.009 0.01

68
69
Crown Magnesium

70
71
72
Eczacıbaşı ESAN, started trial production in September 2015 and reached
25% capacity as of May 2016 in Eskisehir, Turkey. Primary ingots of 99.80-
99.95% purity and weighing 8-12 kg.

73
Magnatherm Prosesi
• Reaksiyon 5.7 m yüksekliğinde,
4.3 m çapında olan silindirik
reaktörde gerçekleştirilir.
• Sistem atmosferik basınçta iken
Kalsine dolomit, Kalsine boksit
ve FeSi hammadde silolarına
doldurulur.
• Sistem vakuma alındıktan sonra
hammaddeler reaktöre verilir.
• Bir önceki safhadan kalan
curufla temas edildikten sonra
reaksiyon meydana gelir
(1600°C).
• Oluşan magnezyum buharı
650°C’de tutulan kondansöre
difüze olduktan sonra su
soğutmalı pota içerisinde sıvı
halde birikir.
• Reaktör proses çevriminin
ortasında ve sonunda olmak
üzere iki kere boşaltılır. 74
Self-propagating high-temperature
synthesis (SHS)

Self-propagating High-temperature synthesis (SHS) means the

synthesis of compounds (or materials) in a wave of chemical
reaction (combustion) that propagates over starting reactive
mixture owing to Layer-by-layer heat transfer.
75
76
 Çizelge 2. Oluşabilecek reaksiyonlar, entalpileri ve açığa çıkan birim ısılar.

Reaksiyonlar - H298 ,(kJ) Birim ısı,(J/g)

1 1.5 TiO2 + 2 Al = 1.5 Ti + Al2O3 258.1 1485.2

2 B2O3 + 2Al = 2B + Al2O3 403.3 3263.7

3 TiO2 + B2O3+ 3.3 Al= TiB2+1.6 Al2O3 899.8 3757.1

4 2TiO2 + B2O3+ 4.6 Al= 2TiB+ 2.3 Al2O3 1067.5 3004.5

5 KClO3 + 2Al = KCl(g) + Al2O3 1500.8 8502.7

77
TiB2 contains 31.10 wt. % B and 68.90 wt. % Ti.
78
The Reduction and Formation Reactions in B2O3-Mg-TiO2 System

 3TiO2 + Mg  Ti3O5 + MgO

 2Ti3O5 + 7Mg  3Ti2O + 7MgO
 Ti2O + Mg  2Ti + MgO
 B2O3 + 3Mg  2B + 3MgO
 B2O3 + 4Mg  MgB2 + 3MgO
 Ti + 2B  TiB2
 Ti + MgB2  TiB2 + Mg
 B2O3 + 3MgO  3MgO.B2O3
 2MgO +TiO2  2MgO.TiO2

79
Experimental Setup of the Metallothermic Reduction
Experiments
1. Power supply
2. Electric cable
3. Reaction crucible
4. MgO lining
5. Charge material
6. CrNi resistance wire
7. Stainless steel cover
8. Metal tube
9. Plastic gas hose
10. Flow meter
11. Regulator
12. Argon tube

80
Metallothermic Reduction Experiments

81
 Theoritical Study
B4C ZrB2 TiB2 WxBy
Spesific Heat (J/g) 3820.6 3051.8 3947.7 3482.8
Adiabatic Temperature (°C) 2323 2120 2979 2150

Simulation of SHS products

Simulation of Tad change with Mg addition
composition change with addition of C
82
 Experimental Procedure
B2O3 + 0.5 C + 3 Mg → 0.5 B4C + 3 MgO
ZrO2 + B2O3 + 5 Mg → ZrB2 + 5 MgO
TiO2 + B2O3 + 5 Mg → TiB2 + 5 MgO
a WO3 + b B2O3 + c Mg → WxBy + z MgO

B2O3+ 3 Mg + ½ C

Raw Purity, Particle

Materials wt % Size, µm
Mg 99.7 < 150
Mg
H3BO3 99.5 -
B2O3 97.0 < 53
C 98.0 -
ZrO2 99.0 < 106
TiO2 98.8 < 75
CaWO4 98.0 < 106
WO3 99.8 < 106

83
 Metallothermic Process
B2O3 + 0.5 C + 3 Mg → 0.5 B4C + 3 MgO
B2O3 (1 – 2 mole) C (0.5 – 1.1 mole)
Mg (2.5 – 3.5 mol)

ZrO2 + B2O3 + 5 Mg → ZrB2+ 5 MgO

B2O3 (1 – 3 mole) Mg (5 – 7 mole)
TiO2 + B2O3 + 5Mg → TiB2 +5MgO
1. Power Supply B2O3 (100 – 110 % sto.) Mg (90 – 110 % sto.)
2. Electricity Cable 5. W heating wire
3. Crucible 6. Steel Cover α WO3 + β B2O3 + γ Mg → WxBy + σ MgO
4. SHS Mixture 7. Argon Gas B2O3 (5 – 8 mole) Mg (0.6 – 3 mole)

ZrB2

B4C TiB2
84
 Leaching Process
Since the SHS product is a mixed solid of
B4C, or ZrB2, or TiB2, or WxBy with (MgO+Mg3B2O6)
MgO was leached out from the SHS product by HCl acid leaching process

In the leaching step, MgO + 2HCl = Mg2+ + 2Cl– + H2O

• The temperature variations were
the SHS product was leached by HCl solution monitored
• Water was added gradually when
- S/L ratio, loss of water was observed after
- Leaching temperature obtaining highest solution
- Leaching Time temperature values.
• Filter cakes were obtained by S/L
- HCl Concentration separation method.

1. 400°C contact thermometer

2. Magnetic Mixer
3. Hot-plate
4. Magnet
5. Beaker
6. Acid solution

Leaching Step Filtering Step

85
SHS Process – Results B4C
■

◊ ■ ◊ ■
◊ ◊ ◊ ◊◊ ◊◊ ■
●◦ ●◊ ◊◊

XRD analysis of the SHS product

(100 % stoic. Mg, C and B2O3 addition)
MgO(■), Mg3B2O6 (◊), B4C (●)

Possible B2O3 + 0,5C + 3Mg = 0,5B4C + 3MgO

Reactions B2O3 + 3MgO = 3MgO.B2O3

86
 B2O3 + 0,5 C + 3 Mg → 0,5 B4C + 3 MgO
MgO(■),
no clear structural changes between these products Mg3B2O6 (◊),
B4C (●)

XRD analysis of SHS products obtained by XRD analysis of SHS products obtained
different stoic. C by different stoic. Mg additions and 160
additions and 100% stoic. Mg and B2O3. % stoic. C & 100% stoic. B2O3.

Mixture XRD Identified Phases Mixture XRD Identified Phases

100 % sto. C MgO, Mg3B2O6, B4C 100 % sto. Mg MgO, Mg3B2O6, B4C
160 % sto. C MgO, Mg3B2O6, B4C 105 % sto. Mg MgO, Mg3B2O6, B4C
200 % sto. C MgO, Mg3B2O6, B4C 110 % sto. Mg MgO, Mg3B2O6, B4C
87
 SHS Process – Results ZrB2
ZrO2 + X B2O3 + 6 Mg

Initial Molar Ratios

XRD Identified Phases
(ZrO2/B2O3/Mg)

1/1/6 ZrB2, MgO, ZrO2, (Zr, Mg)O2, Zr3O

1/2/6 ZrB2, MgO, (Zr, Mg)O2, Mg3B2O6

1/2,5/6 ZrB2, MgO, (Zr, Mg)O2, Mg3B2O6, Mg2B2O5

1/3/6 ZrB2, MgO, Mg3B2O6, Mg2B2O5

88
 SHS Process – Results TiB2

TiO2 + X B2O3 + Y Mg

XRD analysis of SHS products obtained by XRD analysis of SHS products obtained by
different stoic. Mg additions and % 100 stoic. different stoic. Mg additions and % 110 stoic.
B2O3 B2O3

:TiB2 :MgO : Mg3B2O6 :Mg2TiO4

89
 SHS Process – Results WxBy

WO3 + 8 Mg + X B2O3 CaWO4 + 8 Mg + X B2O3

XRD analysis of SHS products obtained by XRD analysis of SHS products obtained by
different stoic. B2O3 additions with WO3 different stoic. B2O3 additions with CaWO4

♦ W, ■ W2B, ● WB, ▲ W2B5, ♦ W, ■ W2B, ● WB, ▲ W2B5, ○ MgO,

○ MgO, □ Mg3B2O6, + Mg □ Mg3B2O6, * CaO, ◊ Ca3(BO3)2
90
 Leaching Process – Results
Effect of HCl concentrations on B4C SHS product obtained by 100% stoic. C and Mg
solution temperature.

340

330 9.3 M HCl

Temperature, K

7.4 M HCl
320 5.6 M HCl

310
3.7 M HCl Maximum increase in temperatures were
2.6 M HCl detected within the first 3 minutes.

300
0 M HCl
TiB2 SHS product obtained by
290 100% stoic. B2O3 and Mg
0 5 10 15 20 25 30
Time, min. 91
Leaching Process – Results B4C

1/5 S/L ratio, 8,11 M HCl,

room temperature Effect of L. Time

92
Leaching Process – Results ZrB2

1/20 S/L ratio, 80°C, 60 min.

Effect of HCl Conc.

93
Leaching Process – Results TiB2

1/5 S/L ratio, 30 min. Room temperature

Effect of HCl Conc.

94
 Leaching Process – Results WxBy

■ W2B,
1/10 S/L ratio,
2.85 M HCl,
● WB,
Room temperature ▲ W 2 B5 ,
Effect of Leaching Time. ○ MgO,
□Mg3B2O6,
◊ Ca3(BO3)2

95
 Leaching Process – Results
at optimum leaching conditions
B4C = 1/10 S/L ratio, 8,93 M HCl, 60 minutes, 80 °C
MgTotal (wt. %) Fe (wt. %) CTotal (wt. %)
0.49 0.085 15.298

ZrB2 = 1/20 S/L ratio, 1,88 M HCl, 60 minutes, 80 °C

BET analysis
Leached SHS product: 2,20 m2/g Commercial ZrB2 particle: 1,04m2/g

TiB2 = 1/5 S/L ratio, 9,3 M HCl, 30 minutes, 25 °C

FCT Systeme GmbH – FAST
Density Relative D.
(Field Assisted Sintering Technique)
(g/cm3) %
Graphite Crucible
Inner diameter 40 mm
Commercial
Under Vacuum 4.22 93.4
TiB2
50 MPa pressure
2100°C Leached
2 minutes sintering time 4.44 98.2
SHS Product

WxBy = 1/10 S/L ratio, 5.7 M HCl, 60 minutes, 80 °C

96
 (■: B4C, ▲: ZrB2, ●: TiB2, □: MgO, ○: Mg3B2O6, ◊: Mg2B2O5)

TiB2 TiB2

ZrB2 ZrB2

B4C B4C

Commercial Leached Unleached powders

97
 SEM micrographs of
a) unleached B4C c) unleached ZrB2 e) unleached TiB2
b) leached B4C d) leached ZrB2 f) leached TiB2
(a) (c) (e)

(b) (d) (f)

98
 WxBy ■ W2B,
● WB,
▲ W2B5

unleached leached
W xBy W xBy

99
Bir metal, kendi oksidinden daha negatif değerli serbest enerjiye sahip oksit
oluşturan bir element tarafından indirgenebilir.
Tüm oksitlerin artan sıcaklıkla kararlılığı azalır. Bazı önemli metallerin, artan
serbest enerji değerlerine göre sıralanışı şöyledir: CaO, MgO, Al2O3, ZrO2,
TiO2, SiO2, Cr2O3, FeO, MoO3 vs. gösterilen bu oksitlerin her birinin metalik
hali, sağdaki oksidi redükler ve bu redükleme reaksiyonu sonrasında ısı açığa
çıkar. Redükleyici olarak Ca metali kullanıldığı zaman işlemin adı
kalsiyotermik,

Mg kullanıldığında magnezyotermik,

Al kullanıldığında aluminotermik,

Si kullanıldığında Silikotermik redüksiyon adını alır.

100
Bir metalotermik reaksiyonda, reaksiyon entalpisi açığa çıkan ısının
belirlenerek işlem esnasında metal ve curufun ergitilmesi ve yoğunluk
farkları yardımıyla birbirlerinden iyice ayrılmaları için yeterli sıcaklığın
oluşup oluşmadığının belirlenmesinde kullanılan önemli bir göstergedir.
Reaksiyon entalpisi reaksiyon ürünlerinin moleküler ağırlıklarının
toplamına bölünerek şarj veya ürünün gramı başına üretilecek ısı
hesaplanabilmektedir. Eğer bu değer 4500 J/g dan daha büyükse
reaksiyon patlayıcılardaki kadar şiddetlidir. 2250 J/g dan daha düşük
değerlerde ise açığa çıkan ısı ergitme ve metal-curuf ayrımı için yeterli
olmamaktadır. 2250 ile 4500 J/g arasındaki ısı değerlerinde ise
metalotermik reaksiyon kontrollü ve kendiliğinden devam eden bir seyir
gösterir.

101
Eğer ısı veren reaksiyon ile yeterli ısı sağlanamıyor ise şarj bir
ön ısıtmaya tabi tutulabilir ve/veya reaksiyona CrO3, KClO3,
NaClO3 gibi ısı vereci maddeler ilave edilebilir. Diğer bir
alternatifte, üretim için elektrik ark fırınlarının kullanılmasıdır.

Başarılı bir aluminotermik proses için, düşük ergime

sıcaklığına ve yüksek akışkanlığa sahip bir curuf geliştirmek
gereklidir. Bu işlem kireç, flourspar ve kuarsit gibi curuflaştırıcılar
kullanılarak sağlanabilir. Kireç ilavesinin faydalı etkisi yüksek
aluminalı curuflann liküdüs sıcaklığını hızla düşürmesidir. Aşırı
curuflaştırıcı ilavelerinde ise bu inert maddeler termal gereksinimi
arttırarak reaksiyonun yavaş ilerlemesine ve curuf sıcaklığının
artmasına neden olurlar.

102
Aluminotermik redüksiyon yüksek reaksiyon hızı ile karakterize edilir ve
aluminotermik redüksiyonun hız kontrolü metal oksit ve aluminyumun
değişik tane boyutlarından etkilenir. Kaba partiküllü aluminyum içeren
şarj, ince partiküllülere nazaran daha düşük reaksiyon hızına sahiptir ve
yetersiz metal/curuf ayrımına sebep olur, curuf ve metalin ergime
sıcaklıklarına güçlükle ulaşılır. İnce aluminyum taneleri ise kısa bir
zaman periyodunda yüksek sıcaklık piki verirler. Bunun anlamı kısa bir
süre sıvı kalan yüksek akışkanlıklı bir curuftur. Bu nedenle bir
aluminotermik proses için aluminyumun tane boyutu çok önemli bir
parametredir.

Aluminotermik redüksiyondan sonra herhangi bir işlem görmeden

satılan ferroalaşımlar için alaşımdaki aluminyum içeriği, çelikte
istenmeyen bir empürite olduğundan sınırlıdır. Alaşımın Al/metal
oranının artması alaşımın yoğunluğunu düşürür ve curufta kalma
şansının yükselmesine sebep olur.

103
REFERENCES
Yücel,O., Çınar,F., Addemir,O., Tekin,A., “The Preparation of Ferroboron and
Ferrovanadium by Aluminothermic Reduction”, High Temperature Materials and
Processes, Vol. 15, Nos.1-2, 103-109,1996.
O. Yücel, F. Çınar, Ç. Gözkaman, “Aluminothermic Production of Al-Ti-B Pre-alloys”,
10th Int. Metalllurgy and Materials Cong., UCTEA, Chamber of Metallurgical
Engineers, İstanbul, TURKEY, 24-28 May, 2000.
O. Yücel, F. Çınar, A.O. Keskin, “ Metalothermic Production of Manganese Alloys”,
10th Int. Metalllurgy and Materials Cong., UCTEA, Chamber of Metallurgical
Engineers, İstanbul, TURKEY, 24-28 May, 2000.
O. Yücel, A. Özçelebi, “Reduction Smelting Of Bursa-Uludağ Tungsten Concentrates
By The Aluminothermic Process”, Scandinavian J. of Metallurgy, 29, 3: 108, 2000.
O. Yücel, F. Çinar, “Production of Aluminum-Titanium-Boron Master Alloy by
Aluminothermic Process”, High Temp. Mater&Proc., 20, 2,2001.
O. Yücel, A.O. Keskin, F. Ç. Şahin, “Metallothermic Reduction Of Tavas Manganese
Ore”, INFACON 9, Proceedings of the 9th International Ferroalloys Congress, 230,
Quebec City, 2001.
Demircan U., Derin B., Yücel O., “Synthesis of TiB2 Powder by Pyro-
Hydrometallurgical Process”, XII. International Materials Research Congress, August
22-26, 2004, Cancun-Mexico
Yücel O., Yiğit S., Derin B. “Production of Magnesium Metal from Turkish Calcined
Dolomite using Vacuum Silicothermic Reduction Method” International Conference on
Magnesium, September 20–24, 2004 , Beijing, China. 104