EXTRACTION AND SEPARATION OF TANTALUM AND NIOBIUM

FROM MOZAMBICAN TANTALITE BY SOLVENT EXTRACTION IN

THE AMMONIUM BIFLUORIDE-OCTANOL SYSTEM

by

KABANGU MPINGA JOHN

Dissertation submitted in partial fulfillment of the requirements for the degree of

MSc (Applied Science): Chemical Technology

Department of Chemical Engineering

Faculty of Engineering, the Built Environment and Information Technology

Supervisor: Prof. Philip Crouse

University of Pretoria

© University of Pretoria
DECLARATION

I, Kabangu Mpinga John, student No. 28140045, hereby declare that all the work
provided in this dissertation is to the best of my knowledge original (excepted where
cited) and that neither the whole work nor any part of it has been, or is to be, submitted
for another degree at University of Pretoria or any other University or tertiary education
institution or examining body.

SIGNATURE………………………………………………………………………………………

DATE……………………………………………………………………………………………….

KABANGU MPINGA JOHN

I
SYNOPSIS

EXTRACTION AND SEPARATION OF TANTALUM AND NIOBIUM FROM

MOZAMBICAN TANTALITE BY SOLVENT EXTRACTION IN THE AMMONIUM
BIFLUORIDE-OCTANOL SYSTEM

Autor: Kabangu Mpinga John

Supervisor: Prof. Philip Crouse

Department: Chemical Engineering

Degree: MSc (Applied Science): Chemical Technology

The principal aim of this research was to determine the optimum conditions of extraction
and separation of niobium and tantalum with octanol as solvent, from Mozambican
tantalite using ammonium bifluoride as an alternative to hydrofluoric acid. The extraction
of niobium and tantalum from tantalite can be divided into three activities, viz., acid
treatment of the ore to bring the niobium and tantalum values into solution, separation of
niobium and tantalum by solvent extraction and preparation of pure niobium pentoxide
and tantalum pentoxide by precipitation followed by calcination.

An initial solution was prepared by melting a mixture of tantalite and ammonium

bifluoride followed by leaching of the soluble component with water and separation of
the solution by filtration. The solution filtered was successfully used after adjustment of
the acidity for the extraction and separation of niobium and tantalum. After liquid-liquid
extraction highly pure niobium pentoxide and tantalum pentoxide were obtained through
precipitation with ammonium hydroxide and calcination.

II
Comparative experiments were performed modifying the following variables:
decomposition temperature; decomposition time of the digestion of niobium and
tantalum; acid concentration of feed solution; solvent agent; and stripping agent.

From the experimental results it was determined that the decomposition temperature,
decomposition time of the digestion of niobium and tantalum, the acidity of the feed
solution, the solvent agent, and stripping agent, all have an important effect on the
extraction and separation of tantalum and niobium.

The optimal conditions were determined to be: tantalite-to-ammonium bifluoride 1:30;

decomposition temperature 250°C; decomposition time 3 hours; and a water leach
period of 10 minutes. Under these conditions the leach recovery of niobium and
tantalum was about 95.07% and 98.52%, respectively.

For nearly complete extraction of tantalum and niobium with 2-octanol, two and three
equilibrium stages, respectively, are required. The equilibrium data were obtained at an
aqueous-to-organic ratio (A/O) of 1:1 using: 100% 2 octanol; 6 M H2SO4; 10 minutes
contact time at room temperature for tantalum; and 100% 2 octanol, 9 M H2SO4, 10
minutes contact time at room temperature for niobium.

Different stripping agents were used and water seems to give good result for both. For
the nearly complete stripping process four equilibrium stages are required. From the
results obtained an optimum stripping efficiency is achieved after 15 minutes for a 1:1
ratio.

Niobium and tantalum were then neutralized using 28% ammonium hydroxide. The
precipitate obtained was dried and placed in a muffle furnace for 4 hours at 900°C. After
calcinations, pure tantalum pentoxide and niobium pentoxide were obtained and
characterized using XRD and SEM. Tantalum pentoxide and niobium pentoxide
synthesized contain trace impurities such as iron and titanium which can be removed by
incorporating an appropriate intermediate treatment.

III
Keywords – Digestion, ammonium hydrofluoride, solvent extraction, octanol, niobium,
tantalum, recovery.

IV
ACKNOWLEDGEMENTS

Firstly, I would like to thank my Heavenly Father for endowing me with the required
persistence and every necessary provision needed for the accomplishment of this work.

My very special thanks go to my supervisor, Prof. Philip Crouse, for his academic
guidance and support.

I would like to express my profound gratitude to the University of Pretoria and National
Research Fund for their financial assistance.

My sincere thanks also go to my colleague Onias Sitando for his encouragement.

Last, but certainly not least, I am grateful to my family for their endless love and
unconditional support while I was working on this project. Special thanks to my wife
Nathalie Kabangu and my children Chris, Jonathan and Glodice Kabangu for their loyal
support and patience during those times when I was busy working on this dissertation.

V
 TABLE OF CONTENTS

DECLARATION ............................................................................................................... I

SYNOPSIS ...................................................................................................................... II

ACKNOWLEDGEMENTS ............................................................................................... V

TABLE OF CONTENTS................................................................................................. VI

LIST OF SYMBOLS ........................................................................................................ X

LIST OF FIGURES......................................................................................................... XI

LIST OF TABLES ........................................................................................................ XIII

CHAPTER 1 INTRODUCTION ....................................................................................... 1

1.1. Introduction ........................................................................................................ 1

1.2. The objective of the study .................................................................................. 3

1.3. Methodology....................................................................................................... 4

1.4. Outline of dissertation ........................................................................................ 4

CHAPTER 2 LITERATURE SURVEY............................................................................. 5

2.1. Niobium and tantalum raw materials .................................................................. 5

VI
 2.2. Recovery of niobium and tantalum ..................................................................... 6

2.2.1. Decomposition (digestion) of niobium and tantalum .................................. 19

2.2.2. Solvent extraction theory ........................................................................... 24

2.2.3. Separation of niobium and tantalum ......................................................... 28

2.3. Preparation of pure tantalum and niobium oxides ............................................ 29

CHAPTER 3 EXPERIMENTAL ..................................................................................... 32

3.1. Apparatus and instruments .............................................................................. 32

3.2. Raw materials .................................................................................................. 33

3.3. Planning ........................................................................................................... 33

3.4. Experimental methods ..................................................................................... 33

3.4.1. The effect of time and temperature on the recovery of Nb and Ta………..34

3.4.2. Concentration of sulfuric acid ...............................................................................35

3.4.3. Selection of stripping agent ..................................................................................35

3.4.4. Precipitation .........................................................................................................36

CHAPTER 4 RESULTS AND DISCUSSION ................................................................ 37

VII
4.1. Particle size analysis ........................................................................................ 37

4.2. Mineralogy of tantalite ...................................................................................... 38

4.3. Preparation of the leach liquor ......................................................................... 40

4.4. Results of the decomposition experiments ....................................................... 41

4.4.1. Effect of time and temperature on the recovery of niobium and tantalum .. 41

4.4.2. Effect of the mass ratio of tantalite-to-ammonium bifluoride ...................... 43

4.4.3. Effect of contact time 45

4.4.4. Concentration of sulfuric acid .................................................................... 45

4.4.5. Distribution and separation coefficient ....................................................... 48

4.5. McCabe-Thiele analysis ................................................................................... 50

4.5.1. McCabe-Thiele diagram for Ta extraction.................................................. 50

4.5.2. Tantalum stripping ..................................................................................... 52

4.5.3. Effect of stripping agent ............................................................................. 52

4.5.4. Effect of contact time ................................................................................. 52

4.5.5. McCabe-Thiele diagram for Ta stripping.................................................... 53

4.5.6. Ta precipitation .......................................................................................... 54

4.6. McCabe-Thiele diagram for Nb extraction........................................................ 56

VIII
 4.6.1. Niobium stripping ....................................................................................... 57

4.6.2. Effect of stripping agent ............................................................................. 58

4.6.3. Effect of contact time ................................................................................. 58

4.6.4. McCabe-Thiele diagram for Nb stripping ................................................... 59

4.6.5. Nb precipitation.......................................................................................... 59

CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS ....................................... 63

5.1. Conclusions...................................................................................................... 63

5.2. Recommendations ........................................................................................... 65

REFERENCES .............................................................................................................. 66

APPENDICES…………………………………………………………………………………74

IX
LIST OF SYMBOLS

Distribution coefficient.

Kex Extraction coefficient.

HA Concentration of the acidic extractant in organic phase.

E Extraction factor in percentage

Separation factor

Ta Distribution coefficient for tantalum.

Nb Distribution coefficient for niobium.

HA Partition coefficient.

kA Acidity coefficient of the extractant.

kf Formation coefficient of the metal complex with the

extractant.

Man Partition coefficient of metal complex.

n Charge of metal ion.

Vo Volume of organic phase

Vaq Volume of aqueous phase

A/O Aqueous-to-organic mass ratio

X
LIST OF FIGURES

Figure 2.1 Collective extraction ..................................................................................... 9

Figure 2.2 Selective extraction .................................................................................... 10

Figure 2.3 Modified scheme of collective extraction, including selective extraction of

tantalum ........................................................................................................................ 12

Figure 2.4 Hydrometallurgical processing of niobium and tantalum ............................ 18

Figure 2.5 Possible flow chart of decomposition of tantalum- or niobium-containing raw

material by the hydrofluoride method ............................................................................ 23

Figure 2.6 Flow chart of tantalum/niobium oxide preparation-precipitation of hydroxides

by ammonia solution ..................................................................................................... 31

Figure 4.1 Particle size distribution of tantalite ore………………………………………..37

Figure 4.2 XRD pattern of tantalite…………………………………………………………40

Figure 4.3 Effect of time and temperature on the recovery of niobium ........................ 42

Figure 4.4 Effect of time and temperature on the recovery of tantalum ....................... 43

Figure 4.5 Effect of tantalite-to-ammonium bifluoride mass ratio (digestion conditions: 3

hours, 250°C, 500 rpm) ................................................................................................. 44

Figure 4.6 Effect of contact time on extraction efficiency of Nb and Ta ....................... 45

Figure 4.7 Effect of sulphuric acid concentration on Ta and Nb extraction by 1-octanol

(error ± 0.04%) (A/O 1:1, 100% 1-octanol, 10 min). ...................................................... 47

Figure 4.8 Effect of sulphuric acid concentration on Ta and Nb extraction by 2-octanol

(error ± 0.04%) (A/O 1:1, 100% 2-octanol, 10 min). ...................................................... 47

Figure 4.9 McCabe-Thiele diagram of Ta extraction (A/O 1:1, 100% 2-octanol, 10 min,
6 M). .............................................................................................................................. 51

Figure 4.10 Effect of contact time on stripping efficiency of Ta ................................... 53

Figure 4.11 McCabe-Thiele diagram of Ta stripping.................................................... 54

Figure 4.12 XRD pattern of synthesized Ta2O5 ........................................................... 55

XI
Figure 4.13 XRD pattern of purchased Ta2O5 ............................................................. 55

Figure 4.14 SEM result of synthesized Ta2O5 ............................................................. 55

Figure 4.15 SEM result of purchased Ta2O5................................................................ 55

Figure 4.16 McCabe-Thiele diagram of Nb extraction ................................................. 57

Figure 4.17 Effect of contact time on stripping efficiency of Nb ................................... 58

Figure 4.18 McCabe-Thiele diagram of Nb stripping ................................................... 59

Figure 4.19 XRD pattern of synthesized Nb2O5 ........................................................... 60

Figure 4.20 XRD pattern of purchased Nb2O5 ............................................................. 60

Figure 4.21 SEM result of synthesized Nb2O5 ............................................................. 61

Figure 4.22 SEM result of purchased Nb2O5 ............................................................... 61

Figure 4.23 Proposed technical flow sheet for the extraction of Nb/Ta ....................... 62

Figure A.1 Major components and layout of a typical ICP-OES instrument…………...75

Figure A.2 Peristaltic pump used for ICP-OES ................................. …………………..77

Figure A.3 A typical spray chambers used in ICP-OES............................................... 79

Figure A.4 A combination radial (side-on) and axial (end-on) ICP .............................. 82

Figure A.5 The Pashen-Runge mount used in a Rowland circle polychromotor ......... 83

Figure A.6 A typical PMT............................................................................................ 84

Figure A.7 X-ray fluorescence principle……………………………………………………86

Figure A.8 Scattered beam of X-rays………………………………………………………88

Figure A.9 Schematic diagram of a diffractometer ...................................................... 89

Figure A.10 Block diagram of scanning electron microscopy………………………….91

XII
LIST OF TABLES

Table 2. 1 Composition of main tantalum-niobium containing minerals ......................... 6

Table 2. 2 Applications of tantalum .............................................................................. 14

Table 2. 3 Applications of niobium ............................................................................... 16

Table 4. 1 Chemical composition of tantalite………………………………………………39

Table 4. 2 Distribution-separation coefficient and extraction factor for 1-octanol with

different H2SO4 concentrations ...................................................................................... 49

Table 4. 3 Distribution-separation coefficient and extraction factor for 2-octanol with

different H2SO4 concentrations...................................................................................... 50

XIII
 CHAPTER 1 INTRODUCTION

1.1. Introduction

The scope of this dissertation is captured in its title, viz. “Extraction and separation of
tantalum and niobium from Mozambican tantalite by solvent extraction in the ammonium
bifluoride system”. The study focused on conditions that affect the digestion of the ore,
separation and recovery of tantalum and niobium using 1−octanol and 2−octanol as
extractants and their precipitation as oxides and characterization of the final products.

The principal minerals of niobium (Nb) and tantalum (Ta) are columbite and tantalite.
This, depending on whether the concentration of niobium or tantalum is greater
corresponds to the general formula (Fe,Mn)(Nb,Ta)2O6 (Damodaran et al.,1969).

Tantalite ore is found primarily in Australia, Canada, Brazil, Southern Africa, Nigeria and
Central African countries such as the Democratic Republic of Congo-Kinshasa,
Rwanda, Uganda, Gabon and Burundi. Pegmatites continue to be the largest source of
tantalum. The Greenbushes pegmatite near Perth in Western Australia is one of the
world’s largest resources, and is mined by the Australian company Sons of Gwalia. The
firm alone accounts for over 75% of world production capacity and it is the world’s
largest producer of tantalum (Baba et al., 2005; Amuda et al., 2007).

Deposits of tantalite ore are also found in Zimbabwe, Mozambique, India and the U.S.A.
Australia has an output of about 17.3% and Canada 6.9% of tantalite. Developing
countries like Thailand, the Democratic Republic of Congo, and Nigeria, produce 3.6%
of the worldwide concentrates (Baba et al., 2005).

The particular properties of niobium and tantalum have resulted in increasing their
usage in electronic, optic, mechanical, aerospace, nuclear and other modern
applications. Tantalum and niobium are used widely, in the form of metals and alloys, in
the process industry, vacuum techniques, chemical machinery, surgical devices, and
the automotive industry (Agulyansky, 2004).

1
Chapter 1 Introduction

Niobium-tantalum ore materials are generally decomposed with alkali or acidic fusion at
elevated temperature, or leaching with mineral acids. At present, most minerals
containing niobium and tantalum are decomposed by concentrated hydrofluoric acid (El-
Hussaini and Mahdy 2002; Gupta and Suri, 1994). However, such an approach is only
appropriate for high-grade niobium-tantalum ore (Zhou et al., 2005). For a low-grade
niobium-tantalum ore, a new process for the decomposition has been proposed by Zhou
et al., (2005) and Wang et al., (2009) using a highly concentrated caustic potash (KOH)
solution under atmospheric pressure and a KOH roast-water leach solution,
respectively. The decomposition was greatly improved compared to hydrofluoric acid
leaching.

According to the analysis the tantalite from Mozambique is of high-grade and can only
be decomposed by hydrofluoric acid, HF, or mixture of hydrofluoric and sulfuric acid, HF
and H2SO4. Hydrofluoric acid is not ideal for the decomposition stage because the
process is accompanied by the loss of about 6-7% HF due to volatilization (He et al.,
1998), which is hazardous. Furthermore, a large amount of wastewater containing
fluoride is generated that needs to be treated (Wang et al., 2009).

Using ammonium bifluoride as an alternative to hydrofluoric acid, the leaching process

is performed with water and a large amount of impurities are precipitated in the form of
insoluble compounds that can be separated from the solution by filtration (Agulyansky,
2004). Leach liquor prepared in this way shows consistent concentrations of Ta and Nb.
The most important property of the solution is its low acidity. Such a solution can be
treated by liquid-liquid extraction using both the collective and selective methods after
appropriate acidity adjustment (Agulyansky, 2004). Ammonium bifluoride is also easy to
handle. In addition, in the context of South Africa, large quantities of inexpensive
ammonium bifluoride are becoming available as byproduct from local plants producing
e.g. NF3.

2
Chapter 1 Introduction

In previous work (He et al., 1998), the efficiency of extraction with octanol is shown to
be comparable with MIBK, the latter being commonly used in the industrial technologies
of Ta and Nb (Babkin et al., 1988). However, octanol is preferable since MIBK is fire
and explosion-hazardous and soluble in aqueous solutions (Mayorov et al., 2002).

The main advantages of 2-octanol compared with MIBK as extractant are: very low
solubility in water (< 0.1%); lower volatility; less dangerous; and a relatively high flash
point of 71°C (Zhu et al., 2011). It has been shown, particularly in Russia, that for the
separation and purification, the extractions of Nb and Ta with 1−octanol are very similar
to those with 2−octanol (Mayorov et al., 2002).

1.2. The objectives of the study

The specific objectives of the study were:

• To determine if ammonium bifluoride can be used as an alternative to hydrofluoric

acid;

• To determine the time and temperature for efficient recovery of niobium and tantalum
from tantalite;

• To investigate the effect of sulfuric acid concentration in the feed solution;

• To determine the mass ratio tantalite ore-to-ammonium bifluoride;

• To compare performance of extraction between 1−octanol and 2−octanol;

• To investigate the effect of different stripping agents;

• To characterize the niobium and tantalum pentoxides and compare with the
purchased product.

3
Chapter 1 Introduction

1.3. Methodology

The methodology used in this research was firstly the collection of information related to
the objective of this study. The information was collected from different books, patents,
encyclopedia and scientific journals. Secondly, a method was devised to set up and run
different experiments. Thirdly, the results obtained were evaluated, analyzed and also
compared to previous research. Finally, conclusions were presented and
recommendations were suggested.

1.4. Outline of dissertation

The outline of this research can be subdivided in four main parts:

1 Literature review: It contains the information about the extraction, separation and
precipitation of tantalum and niobium from tantalite. The information is focused on
the conditions that affect the decomposition step, separation, extraction using
solvent as extractant and using ammonium hydroxide as a precipitation agent.

2 Experimental methods: It describes the methods and materials used.

3 Results and discussions: In this chapter the results of the decomposition process,
extraction, stripping, precipitation and characterizations of the products are
presented and discussed in terms of optimum extraction parameters of niobium and
tantalum.

4 Conclusions and recommendations: The chapter contains concluding remarks

regarding decomposition, extraction, separation and precipitation conditions of
niobium oxide and tantalum oxide and recommendations with regards to further test
work.

4
 CHAPTER 2 LITERATURE SURVEY

The extraction of niobium and tantalum from tantalite of Mozambique is investigated in

this dissertation. In order to plan the relevant laboratory experiments, information was
gathered on the raw materials utilized and general conditions for digestions, extraction
and separation of niobium and tantalum from the raw materials.

2.1. Niobium and tantalum raw materials

Niobium and tantalum naturally occur in combination with oxygen and one or more other
metals as niobates and tantalite in various types of deposits. Of these, the important
ones are niobites-tantalites as pegmatite, pyrochlores as carbonites, and niobium-
tantalum-bearing cassiterite as placer deposits (Gupta and Mukerjee, 1990).

Table 2.1, shows that the most important source of tantalum and niobium is
tantalite−columbite, (Fe, Mn)(Nb, Ta)2 O6 (Agulyansky, 2004). The most important
niobium mineral is pyrochlore with the general formula (Ca, Na)2-mNb2O6(O,OH, F)1-n
.xH2O. The lattice positions of Na and Ca can also be occupied by Ba, Sr, rare earths,
Th and U. Thorium and uranium are responsible for the radioactivity of some of the
pyrochlore concentrate. The most important tantalum−containing minerals are tantalite,
wodginite, microlite (the tantalum–rich end member of the pyrochlore series), and
columbite (Fathi, 1997).

5
Chapter 2 Literature survey

Table 2. 1: Composition of main tantalum-niobium containing minerals (Gibalo, 1970, Sean et al.,
2007)

Minerals Composition Ta2O5, Wt.% Nb2O5, Wt.%

Columbite (Fe,Mn)(Nb,Ta)2O6 5-30 55-78
Tantalite (Fe,Mn)(Ta,Nb)2O6 42-84 2-40
Pyrochlore (Ce,Ca,Y)2(Nb,Ta)2O6(OH,F) 0-6 37-66
Microlite (Na,Ca)2Ta2O6(O,OH,F) 66-77 0-7
Loparite (Ce,Na,Ca)2(Ti,Nb)2O6 0.5-3 4-20
Euxenite (Y,Ca,Ce,U,Th)(Nb,Ta,Ti)2O6 0-47 4-47
Limenorutile (Ti,Nb,Fe)3O6 0-36 20-43
Simpsonite Al4(Nb,Ta)3O13(OH) 60-80 0.3-6
Thoreaulite SnTa2O6 73-77 -
3+
Strueverite (Ti,Ta,Fe )3O6 6-13 9-13
Fergusonite (Re3+,Nb,Ta)O4 4-43 14-46
Sumarskite (Fe,Ca,U,Y,Ce)2(Nb,Ta)2O6 15-30 40-45
Tapiolite (Fe,Mn)(Nb,Ta,Ti)2O6 40-85 8-15

2.2. Recovery of niobium and tantalum

Tantalum and niobium normally are found together in various minerals. The most
important of these minerals are columbite and tantalite.

The procedure for recovery of pure niobium and tantalum from their minerals requires a
complex sequence of processing operations. This sequence involves digestion or
decomposition of corresponding raw materials with hydrofluoric acid, sulfuric acid,
mixtures of these, or ammonium hydrofluoride (bifluoride), to allow a complete
dissolution of ore. The accompanying elements, such as iron, manganese, titanium,
etc., are dissolved in hydrofluoric acid solution along with the tantalum and niobium.
2− − 2− −
Two types of complex ions are formed TaF7 / TaF6 and NbOF5 / NbF6 ,

respectively (Agulyansky, 2004). The equilibrium between the complexes depends on

6
Chapter 2 Literature survey

the acidity of the solution and can be schematically displayed as follows for tantalum
and niobium, respectively

TaF72- + HF TaF6- + HF2- (1)

NbOF52- + 3HF NbF6- + HF2- + H2O (2)

The complexes can be regarded as Lewis acids in which acidity increases in going from
2− 2−
TaF7 or NbOF 5 to TaF6 − or NbF6 − . It is obvious that high acidity ions are more stable
in high acidity media. At higher acidity, the equilibria (1) and (2) move to the right,
yielding complexes of lower charge, TaF6 − or NbF6 − , that are mostly extracted into the

organic phase (Agulyansky et al., 2004). Under this condition of lower acidity, the
complexes are stripped from the organic into the aqueous media in the form of TaF7 2 −

or NbOF 5 2 − . Owing to their lower molecular weights niobium complexes are stronger

Lewis acids and need higher acidity of the solution to be converted to the NbF6 − form.

This particular difference in the behavior of complexes provides a possibility for effective
separation of tantalum and niobium by liquid-liquid extraction (Agulyansky et al., 2004).

After filtration of the insoluble residues (Al, Si, Ca, rare earth metals, etc.), the aqueous
hydrofluoric or hydrofluoric-sulfuric acid solution containing tantalum and niobium is
contacted with organic extractant such as methyl isobutyl ketone (MIBK), tributyl
phosphate (TBP) and cyclohexanone in a multistage mixer-settler plant(Joachim et al.,
sa).

There are two main schemes that exist for the separation and purification of niobium
and tantalum using liquid-liquid extraction (Agulyansky, 2004). The first method is a
collective liquid-liquid extraction that consists of extraction of tantalum and niobium

7
Chapter 2 Literature survey

together in the organic phase from an initial solution, and most of the impurities remain
in aqueous phase. Figure 2.1 shows the flow chart of the process. The second is a
selective method that consists of bringing the organic solution of tantalum and niobium
into contact with an aqueous solution. The niobium is back extracted into the aqueous
phase with tantalum remaining in the organic phase. The aqueous niobium solution is
re-extracted with a fresh solvent to remove traces of tantalum (Agulyansky, 2004). The
acidity must be adjusted before niobium extraction because it requires a high
concentration of acid. Figure 2.2 shows the flow chart of the process based on the
selective extraction scheme.

8
Chapter 2 Literature survey

Initial solution – high acidity

Extractant Collective extraction of Ta and Nb and Raffinate (impurities)

extract washing

To waste treatment
Extract containing of Ta and Nb

Diluted acid Nb stripping Extract containing Ta

Ta stripping Water or stream

Nb strip solution Ta strip solution

Figure 2.1: Collective extraction (Agulyansky, 2004)

9
Chapter 2 Literature survey

Initial solution-low acidity

Raffinate containing Nb
Extractant Selective
Selective extraction
extraction of and
of Ta Ta and
and impurities
extract
Extractwashing
washing

Acidity correction Acid

Extract containing Ta

Extraction of Nb and Raffinate

Extractant
extract washing

Water or steam Ta stripping To Waste treatment

Extract containing Nb

Nb stripping Water or steam

Ta strip solution Nb strip solution

Figure 2.2: Selective extraction (Agulyansky, 2004)

The first approach is the more popular because currently existing technology is based
on complete dissolution of the raw material in HF or HF-H2SO4 acids of high
10
Chapter 2 Literature survey

concentration (Agulyansky et al., 2004). The excessive acidity of the resulting solution
allows one to perform only collective liquid−liquid extraction. The second method seems
to be more effective and leads to a better separation of tantalum and niobium and to a
more effective purification but needs an initial solution of sufficiently low acidity. Another
known method (Eckert, 1995) is based on the collective extraction and subsequent
collective stripping of tantalum and niobium. The strip solution is used for further
selective extractive separation of tantalum followed by niobium. This way is more
complex and expensive but it allows one to obtain tantalum and niobium products of
higher purity (Agulyansky et al., 2004). Figure 2.3 shows the modified flow chart of the
process.

11
Chapter 2 Literature survey

Initial solution-high acidity

Collective extraction of Ta and Nb and

Extractant Raffinate (impurities)
extract washing

Extract containing Ta and Nb To Waste treatment

Collective stripping of Ta
Steam
and Nb

Ta and Nb strip solution

Extractant Selective extraction of Ta Extract containing Ta

Ta stripping Steam

Nb strip solution Ta strip solution

Figure 2.3: Modified scheme of collective extraction, including selective extraction of tantalum
(Agulyansky, 2004)

12
Chapter 2 Literature survey

The organic phase resulting is returned to the tantalum−niobium extraction stage. The
niobium hydrate is precipitated by the addition of gaseous ammonia or ammonia
dissolved in water. After re-extraction of the tantalum from the organic phase with water
or a dilute ammonia solution, tantalum may be precipitated as the oxide hydrate with
ammonia or, by addition of potassium salts, such as K2TaF7 (Joachim et al., sa). These
compounds are used as starting materials for the production of tantalum metals for
super alloys and tantalum powders for capacitor manufacture.

The precipitation of the oxides can be done in a batch process or in a continuously

operated facility. The oxide hydrates are collected by filtration, dried and calcined at up
to 1100°C (Fathi, 1997). The conditions of precipitation, drying and calcinations are
varied to produce different particle size distributions needed for oxides for various
applications. Depending on the quality requirements, the calcinations are carried out in
a directly or indirectly heated chamber or a rotary furnace. The nature of the furnace
lining has a considerable influence on the purity of the oxide (Joachim et al., sa).

The primary tantalum and niobium chemicals of industrial significance in addition to

K2TaF7, Ta2O5, K2NbF7 and Nb2O5, are tantalum chloride (TaCl5), lithium tantalite
(LiTaO3), tantalum carbide (TaC), niobium chloride (NbCl5), niobium carbide (NbC), or
lithium niobate (LiNbO3) (Giulio et al., 2009).

The various metallurgical products are generally produced from electron beam or
vacuum arc melting of the metal feedstock. Tables 2.2 and 2.3 show the tantalum and
niobium products, respectively, their applications and technical attributes (TIC, 2009).

13
Chapter 2 Literature survey

Table 2.2: Applications of tantalum

Tantalum Product Application Technical

Attributes/Benefits

Tantalum carbide Cutting tools. Increase high temperature

deformation, control of
grain growth.

Lithium tantalate Surface Acoustic Wave (SAW) filters in mobile Electronic signal wave
phones, hi-fi stereos and televisions. dampening provides for
clearer and crisper audio
and video output.

Tantalum oxide Lenses for spectacles, digital cameras and Ta2O5 provides a high
mobile phones. index of refraction so that
X-ray film. lenses for a given focal
Ink jet printers. strength can be thinner and
smaller.
Yttrium tantalite phosphor
reduces X-ray exposure
and enhances image
quality.
Wear resistance
characteristics. Integrated
capacitors in integrated
circuits (ICs).

Tantalum powder Tantalum capacitors for electronic circuits in: High reliability
Medical appliances such as hearing aids and characteristics and low
pacemakers; failure rates, operation over
Automotive components such as ABS, airbag a wide temperature range
activation, engine management modules, GPS; from -55 to +200˚C, can
Portable electronics e.g. laptop computers, withstand severe
cellular/mobile phones, video cameras, digital vibrational forces, small
still cameras; size per microfarad
Other equipment such as DVD players, flat rating/electrical storage
screen TVs, games consoles, battery charges, capacity.
power rectifiers, cellular/mobile phone signal
masts, oil probes.

Tantalum fabricated Chemical process equipment including lining, Superior corrosion

14
Chapter 2 Literature survey

sheets and plates cladding, tanks, valves, heat exchangers. resistance equivalent in
Cathodic protection systems for steel structures performance to glass.
such as bridges, water tanks.
Corrosion resistance fasteners, screws, nuts,
bolts.
Spinnerettes in synthetic textile manufacture.

Tantalum fabricated Prosthetic devices for humans-hip joints, skull Prosthetic devices for
sheets, plates, rods, plates, mesh to repair bone removed after humans-hip joints, skull
wires damage by cancer, suture clips, stents for blood plates, mesh to repair bone
vessels. removed after damage by
cancer, suture clips, stents
for blood vessels.

Tantalum fabricated High temperature furnace parts. Melting point is 2996˚C

sheets, plates, rods, although protective
wires atmosphere or high
vacuum required.

Tantalum ingot High temperature alloys for: Applications of tin coatings

Air and land based turbine (e.g. jet engine discs, of tantalum, tantalum oxide
blades and vanes). or nitride coatings to semi-
Rocket nozzles. conductors to prevent
copper migration.

Tantalum ingot High temperature alloys for: Alloys compositions

Air and land based turbine (e.g. jet engine discs, containing 3-11% tantalum
blades and vanes). offer resistance to
Rocket nozzles. corrosion by hot gases,
allow higher operating
temperature and thus
efficiency and fuel
economy.

Tantalum ingot Computer hard drive discs. An alloy containing 6%

tantalum has shape
memory properties.

Tantalum ingot Explosively formed projectile for TOW-2 missile. Balance of density and
formability allow for a
lighter and more efficient
system.

15
Chapter 2 Literature survey

Table 2.3: Application of niobium

Niobium product Application Technical

attributes/benefits

HSLA Ferro-Niobium Niobium additive to high strength low alloy steel Impart a doubling of
(~60%Nb) and stainless steel for oil and gas pipelines, car strength and toughness
and truck bodies, architectural requirements, due to gain refining weight
tools steel, ships, hulls, railroad tracks. reduction.

Niobium oxide Manufacturing lithium niobate for surface High index of refraction.
acoustic wave filters. High dielectric constant.
Camera lenses. Increase light
Coating on glass for computer screens. transmittance.
Ceramic capacitors.

Niobium carbide Cutting tool compositions. High temperature

deformation, control grain
growth.

Niobium powder Niobium capacitors for electronic circuits. High dielectric constant,
stability of oxide dielectric.

Niobium metals plates, Cathode protection system for large steel Corrosion resistance,
sheets, wire, rod, tubing structures. formation of oxide and
Chemical processing equipment. nitrite films. Increase in
high temperature
resistance and corrosion
resistance,
oxidation resistance,
improved creep resistance,
reduction erosion at high
temperatures.

Niobium – titanium alloy Superconducting magnetic coils in magnetic Electrical resistance of

Niobium – tin alloy resonance imagery (MRI), magneto alloy wire drops to virtually
encephalography, magnetic levitation transport zero at or below
systems, particle physics experiments. temperature of liquid
helium (-268.8 ˚C).

16
Chapter 2 Literature survey

Niobium-1% Zirconium Sodium vapor lamps. Corrosion resistance,

alloy Chemical processing equipment. fixation of oxygen,
resistance to
embrittlement.

Vacuum-grade ferro- Superalloy additions for turbine blade application Increasing in high
niobium and nickel- in jet engines and land-based turbines. Inconel temperature resistance and
niobium family of alloys, superalloys. corrosion resistance,
improved creep resistance,
reduced erosion at high
temperature.

Today niobium and tantalum compounds are processed in large tonnage by the fluoride
process, including solvent extraction, as described by the flow chart given in Figure 2.4
(Joachim et al., sa)

17
 Chapter 2 Literature survey

Ore

Tin slags
concentrate HF Digestion
H2SO4

Filtration
Residue

MIBK Solvent Raffinate

H2SO4 Extraction

Dil H2SO4
Ta/Nb
MIBK
Separation

H2O stream Ta in MIBK MIBK

Ta Strip Nb Strip

Solution Solution

NH4OH
NH4OH
Ta acid Nb acid
HCl Crystallization H2O
H2O

Filtration Filtration
Filtration

Calcination Calcination
K2TaF7

Ta2O5 Nb2O5

Figure 2.4: Hydrometallurgical processing of niobium and tantalum (Joachim et al., sa)

18
Chapter 2 Literature survey

2.2.1. Decomposition (digestion) of niobium and tantalum

After preparation and separation of the ore according to chemical composition a number
of processes follow. The next step after preparation and separation of ore in
hydrometallurgical processes is to dissolve (digestion or decomposition) the
constituents of the ore to form a solution.

The process of extraction of niobium and tantalum mainly depends on the

decomposition and leaching of the tantalite ore. Obviously the better the leaching, the
higher the quantity of tantalum and niobium to be extracted. The difficulty of tantalum
and niobium extraction from the ore is due to the fact that only a few solvents can leach
the ore and only in the presence of the fluorine ion (Maina et al., 2007).

Leaching is concerned with the extraction of a soluble constituent from a solid by means
of a solvent (Coulson and Richardson, 2002); or it is the removal of a soluble fraction, in
the form of solution, from an insoluble, permeable solid phase with which it is
associated. The separation usually involves selective dissolution, with or without
diffusion (Robert et al., 2007). The soluble constituent may be solid or liquid; and it may
be incorporated within, chemically combined with, adsorbed upon, or held mechanically
in the pore structure of the insoluble material. The process of leaching may be used
either for the production of a concentrated solution of a valuable solid material, or in
order to remove an insoluble solid. The method used for the extraction is determined by
the proportion of soluble constituent present, its distribution throughout the solid, the
nature of the solid and the particle size (Coulson and Richardson, 2002). The
mechanism of leaching may involve simple physical solution or dissolution made
possible by chemical reaction. The rate of transport of solvent into the mass to be
leached, or of soluble fraction into the solvent, or of extract solution out of the insoluble
material, or some combination of these rates may be significant. A membranous
resistance may be involved. A chemical-reaction rate may also affect the rate of

19
Chapter 2 Literature survey

leaching (Coulson and Richardson, 2002; Robert et al., 2007). The process can be
considered generally to comprise of three parts: firstly the change of phase of the solid
as it dissolves in the solvent, secondly its diffusion through the solvent in the pores of
the solid to the outside of the particle, and thirdly the transfer of the solute from the
solution in contact with the particles to the main bulk of the solution.

The rate of leaching is affected by a large number of factors. These include particle
size, the nature of solvent, temperature and agitation.

• Smaller particle sizes result in a greater interfacial area between the relevant solid
and liquid. The diffusion path of solute through the porous structure of the residual
solid is also shorter. Smaller particle may, however, impede circulation of the liquid
and separation of the particle from the liquid, making subsequent drainage of the
solid residues more difficult. Distribution of particle size is desired as this may lead to
a uniform time of extraction and minimal obstruction to solvent flow (Coulson and
Richardson, 1996).

• The solvent should be selective and its viscosity should be sufficiently low for it to
circulate freely (Coulson and Richardson, 1996).

• The increase of temperature will also increase the solubility of the materials that are
being extracted, to give a higher rate of extraction. The diffusion coefficient will be
expected to increase as well. It has been shown that tantalum and niobium extraction
seems to be unchanged at least within a temperature range of 25 50 °C (Agulyansky,
2004).

• Agitation of the fluid is important because it increases the eddy diffusion, prevents
sedimentation and allows more effective use of the interfacial surface (Coulson and
Richardson, 1996).

20
Chapter 2 Literature survey

The chemical study of niobium and tantalum elements shows that the decomposition of
their ore involves a large number of procedures. Among them, alkali fusion which is one
of the first methods that was industrially adopted to achieve simultaneous breakdown of
columbite-tantalite (Eckert, 1995). In this method, sodium hydroxide and sodium
carbonate (or the potassium salts) were used as the main reactants. The shortcoming
of this method is the high reaction temperature (typically 800 C) required. To avoid this
disadvantage, a hydrothermal method has been developed (Cardon, 1962; Zelikman
and Orekhov, 1965) which requires high pressure and expensive autoclave equipment.
Some types of minerals containing niobium and tantalum can also be processed by
different approaches, such as chlorination (Gupta and Suri, 1994), alkali fusion followed
by acid leaching (Foos, 1960), fusion with ammonium fluoride and bifluoride (Gupta and
Suri, 1994), pressure dissolution with alkaline solution (Cardon, 1962; Zelikman and
Orekhov, 1965), direct acid dissolution with H2SO4 (El-Hussaini and Mahdy, 2002) or
combination of H2SO4 and HF (Krismer and Hoppe, 1984).

Currently, most of the niobium-tantalum ores are decomposed by concentrated

hydrofluoric acid (El-Hussaini and Mahdy, 2002; Gupta and Suri, 1994). However, the
decomposition process is accompanied by the loss of a considerable amount of HF due
to volatilization (He et al., 1998), which brings great harm to human beings and
equipment. A large amount of fluorspar sludge is generated and needs to be disposed
off (Bludssus and Eckert, 1993).

In this investigation an initial solution was prepared by melting a mixture of ammonium

hydrogen fluoride (bifluoride) (NH4HF2) and tantalite. The solution was then leached
with water and the filtrate was successfully used for the precipitation of niobium
hydroxide and tantalum hydroxide or further purification by liquid-liquid extraction after
appropriate adjustment of the solution acidity. The possible interaction, based on equal
concentrations of tantalum and niobium in the concentrate, can be represented by
equation (3) (Agulyansky, 2004):
21
Chapter 2 Literature survey

(M)(Ta, Nb)2 O6 + 12NH4HF2 NH4MF3 + (NH4)2TaF7 + (NH4)3NbOF6 + 5H2O + 6NH3

+ 8HF (3)

Here M = Fe or Mn. The efficiency of the fluorination process can be increased by

milling and through mixing of the tantalum-niobium containing raw material with
ammonium hydrofluoride. It is assumed that the process temperature can be set
between 200−350°C. In general, the process flow chart of raw material decomposition
by hydrofluoride method can be represented as shown in Figure 2.5 (Agulyansky,
2004).

22
Chapter 2 Literature survey

Raw material

Raw material preparation: Milling,

blending, classification

NH4HF2 Mixing

Fluorination (Melting) Gaseous H2O, NH3, HF

Leaching with water To NH4F or NH4HF2

Filtration Residue
Separation of solution from insoluble residues

Partially Purified Initial Solution To waste

Containing Ta, Nb and impurities

To liquid-liquid extraction

Figure 2.5: Possible flow chart of decomposition of tantalum- or niobium-containing raw material
by the hydrofluoric acid method

23
Chapter 2 Literature survey

The main advantages of this method compared to the decomposition with HF can be
summarized as follows: The HF is not needed for decomposition of the raw materials;
the amount of fluorine can be calculated or adjusted. Since the leaching is performed
with water, a significant fraction of the impurities can be precipitated in the form of
insoluble compounds which can be separated by filtration. A solution prepared in this
way has high concentration of tantalum and niobium, independent of the initial raw
material composition. The most important reason for this choice is that the solution has
a low acidity level and such a solution can be treated by liquid-liquid extraction (solvent
extraction) using both the collective and selective methods after an appropriate acidity
adjustment (Agulyansky, 2004).

2.2.2. Solvent extraction theory

Liquid-liquid extraction (solvent extraction) is the only method available for the
separation of the rare earth group elements to obtain individual metals (Agulyansky,
2004). This method is very effective for the separation of dissolved components and
enables the production of high-purity products, in addition to its advantages of being a
low-cost and simple method (Agulyansky, 2004).

During this process an organic solution is contacted with the filtrate produced from the
leaching process. Intensive mixing is required to establish sufficient contact time
between the organic and the aqueous phase. For a solvent reagent to perform
satisfactorily in the recovery and purification of metals present in aqueous leach
solutions the reagent must meet a number of criteria. Sudderth and Kordosky (1986)
summarized the most important criteria as follows:

• Extraction of the desired metal selectively from the aqueous solution containing the
dissolved metal.

24
Chapter 2 Literature survey

• Be able to be stripped to produce a solution from which the desired metal can be
recovered in an acceptable form.

• Be chemically and physically stable in the solvent extraction circuit so that it can be
recycled through extraction and strip many times without experiencing undue
physical loss or chemical breakdown.

• The reagents must meet today’s stringent environmental and work-place regulations.

• The extract and strip kinetics must be sufficiently fast to allow these processes to
take place in an industrially acceptable time frame.

• The extraction must be soluble, both in the loaded and stripped form, in relatively
inexpensive diluents which also meets the environmental and workplace regulations.
Alternatively the extractant may be capable of being used at a volume concentration
of 100%, in which case the reagent forms its own diluents.

• The made-up circuit organic must phase separate from the aqueous solution at a
reasonable rate and the separated phases must have acceptable levels of
entrainment.

• The extractant must not transfer deleterious species back from the strip section to the
extraction.

• The extractant should be tolerant of crud and should not promote crud formation.

• The extractant must have a reasonable cost, which will enable it to provide an
economically attractive recovery route for the metal being treated.

It is interesting before using the currently available types of commercially accepted

extractants and those factors which govern the selectivity of these reagents to consider

25
Chapter 2 Literature survey

the properties of the metals which are recovered by solvent extraction (Sudderth et al.,
1986).

The simplest form of the extraction process with acidic extractant is expressed by the
following reaction:

Mn+ + nHA MAn nH+ (4)

Here Mn+ is the metal ion, HA is an extractant. H+ ions move from the organic to
aqueous phase and Mn+ from the organic to aqueous phase. Equation (4) indicates that
the extraction of the metal described by this equation is heavily dependent on the
equilibrium pH of the aqueous phase. Thus, as the pH is decreased (hydrogen ion
concentration increased) the metal extraction will decrease since the equilibrium will be
shifted towards the left. Conversely, as the pH is increased, metal extraction will
increase (Ritcey and Ashbrook, 1984).

The effect of various parameters involved in the solvent extraction of metals can be
quantified by the following equations:

(5)

(6)

The distribution ratio is a function of temperature, the concentration of chemical species

in the system, the phase (A/O) ratio, pH and a larger number of other parameters
depending on the system. It is also quoted as a measure of how well-extracted a
species is.

26
Chapter 2 Literature survey

For practical purposes, as in industrial applications, it is often more popular to use the
percentage extraction %E (sometimes named the extraction factor), which is given by
Rydberg et al., (2004).

(7)

The technique of solvent extraction is employed to separate the solute of interest from
substances that interfere in the ultimate quantitative determination of the material. In
order to effect a simple separation, it is essential that the distribution ratios of the
material of interest and interference be sufficiently different. The effectiveness of
separation is usually expressed by means of the separation factor or separation
coefficient (Morrison and Freiser 1957).

The separation coefficient, , is calculated as the ratio between the two distribution
coefficients. It is a measure of the ability of the system to separate two solutes. For
instance, if the distribution ratio for tantalum is Ta and the distribution ratio for niobium

is Nb, then the separation factor is equal to (8)

Separation factors of greater than one indicate that the two metals can be separated,
but this gives no indication of the ease of separation or the number of stages which may
be required (Ritcey and Ashbrook, 1984).

A plot of log versus pH at constant concentration (HA), should give a straight line of
slope n and an intercept equal to log Kex. The extraction constant Kex has the form
(Ritcey and Ashbrook, 1984):

-n n
Kex = HA K AKf Ma (9)

The relative magnitude of Kf determines the amount of metal extracted at a defined

reagent concentration and pH; therefore, metals with high Kf are extracted at lower pH
27
Chapter 2 Literature survey

than those with smaller Kf (Ullmann et al., 1988). A specific ion can be extracted by
changing the pH of the feed to the contact step between the organic and the aqueous
phase because different ions have different Kf values. Niobium is extracted at higher
acidity level while tantalum is extracted at a lower acidity level. This shows that a good
selectivity of extraction between niobium and tantalum can be obtained.

2.2.3. Separation of niobium and tantalum

Processing of various resources of tantalum and niobium as described above proceed

to a number of separation processes. The main methods of separation are fractional
crystallization, solvent extraction and chlorination.

Solvent extraction is the process generally adopted these days to separate tantalum
and niobium to requisite purity levels. However, fractional crystallization was the first
separation method adopted for the large-scale preparation of pure tantalum
intermediate. This method that originated in the classic work of Marignac is based on
the differences in the solubility of potassium fluotantalum K2TaF7 and potassium
niobium oxyfluoride K2NbOF5.H2O (Bose and Gupta, 2001). Separation of tantalum and
niobium with their mixed chloride as the starting point has not only attracted more
interest but also has resulted in achieving good separation (Gupta, 1984). In this
dissertation solvent extraction was the only method used for the extraction of niobium
and tantalum.

This process was developed about 50 years ago and has found wide application in the
hydrometallurgy of rare refractory and rare earth metals (Agulyansky, 2004). The
process is based on the selective extraction of a compound of a given element from
aqueous solution into an organic solvent which is immiscible with water. When the
aqueous solution and the organic solvent are brought into contact, aqueous-phase
species depending on their distribution coefficients are distributed between the two
liquid phases (Gupta, 1994). The commercial technology for the separation of niobium
28
Chapter 2 Literature survey

and tantalum developed from the work carried out at the US Bureau of Mines and at the
Ames Laboratory, Iowa, used two organic solvents, MIBK and TBP in contact with
aqueous solution of Nb/Ta in HF-H2SO4. The separation process involved co-extraction
of both niobium and tantalum fluoride complex from other metallic impurities in the first
stage in the organic phase. Then the solvent is contacted with dilute HF-H2SO4 to
remove niobium selectively. The resulting organic phase is rich in tantalum and the
aqueous phase rich in niobium. Re-extraction of the aqueous phase with fresh organic
solvent and the organic with dilute acid (HF-H2SO4) yield the organic phase loaded with
tantalum and aqueous phase with niobium. The tantalum-bearing organic phase is
stripped with water. The tantalum-product stream is fluotantalic acid solution from which
tantalum can be precipitated either as potassium tantalum fluoride, by addition of
potassium chloride or fluoride, or as Ta2O5, by addition of ammonia. The niobium-
product stream goes forward to a similar chemical processing scheme to obtain either
potassium niobium fluoride or Nb2O5 (Gupta 1994; Bose and Gupta, 2001).

2.3. Preparation of pure tantalum and niobium oxides

Tantalum and niobium oxides, Ta2O5 and Nb2O5, are among the final products obtained
from tantalum and niobium strip solutions following the liquid-liquid extraction processes
(Agulyansky, 2004). Pure Ta2O5 and Nb2O5 can be prepared from acidic solutions of
H2TaF7 and H2NbF7 by ammonia precipitation, filtration, washing, drying and
calcinations (Agulyansky, 2004).

The most frequently used method in the industry for the preparation of tantalum and
niobium oxides is based on the precipitation of tantalum and niobium hydroxides from a
strip solution by ammonium hydroxide and it consists of several steps (Agulyansky,
2004).

Figure 2.6 presents a flow chart of the process and the complete process can be
represented as follows (Agulyansky, 2004):
29
Chapter 2 Literature survey

In the case of the tantalum strip solution:

2 H2TaF7 + 14NH4OH Ta2O5.nH2O (solid) + 14 NH4F + (9-n)H2O (10)

In the case of the niobium strip solution:

2 H2NbOF5 + 10NH4OH Nb2O5.nH2O (solid) + 10 NH4F + (7-n) H2O (11)

Equation (10) and (11) describe the equilibrium between the hydrolysis of complex
fluoride acids and the fluorination of hydroxides. By applying an excessive amount of
ammonia solution up to pH 8 to 9 a complete precipitation of hydroxides can be
achieved (Agulyansky, 2004). It is recommended to perform the process at a pH no
lower than 10, in order to achieve precipitation of tantalum and niobium hydroxides with
minimal levels of fluorine contamination (Agulyansky, 2004).

The way in which ammonia hydroxide is added to tantalum or niobium strip solutions is
also important for the quality of the precipitated hydroxides and final oxides. Pouring
ammonium hydroxide into a container of strip solution and the mixture agitated is not
optimal. This method slows the increase of pH and can cause contamination of the
hydroxide with crystalline oxyfluoride compounds. The opposite process reduces the
concentration of fluorine in the hydroxide formed significantly (Agulyansky, 2004). It
was reported by Bludssus et al. (1993) that the introduction of tantalum or niobium
containing acid solution to the ammonia solution until achieving a pH of 9 enabled the
production of tantalum or niobium hydroxides with fluoride contents as low as 0.5% wt
with a relatively low consumption of ammonia.

30
Chapter 2 Literature survey

Ta or Nb Strip solution

NH4OH Hydroxide precipitation

Pulp

Filtration Filtrate (H2O, NH4OH, NH4F)

To liquid waste treatment

Wet cake

Water, washing Repulping (washing),

Filtrate (H2O, NH4OH, NH4F)
decantation and filtration
solution

Drying Vapor (H2O, HF, NH3)

Dried cake
To gaseous waste treatment

Calcination Vapor (H2O, HF, NH3)

Ta2O5 or Nb2O5

Figure 2.6: Flow chart of tantalum/niobium oxide preparation-precipitation of hydroxides by

ammonia solution

31
 CHAPTER 3 EXPERIMENTAL

Different experiments were performed in order to determine the optimal conditions for
the extraction and separation of tantalum and niobium from the raw material. These
optimal conditions of extraction and separation are determined by a number of factors.
They include the decomposition temperature, decomposition time, the initial acidity of
the feed solution, the nature of the solvent, and the nature of the stripping agents.
Ammonium hydroxide was used as the precipitation agent. When changing certain of
the variables, the effects on the percentage extraction of tantalum and niobium were
determined. All experiments were batch experiments. The extraction and separation of
tantalum and niobium from tantalite were performed with the apparatus, raw material
and experimental set-up described below.

3.1. Apparatus and instruments

The apparatus and instruments used are bulleted below:

• Flasks and bottles made in polytetrafluroethylene (PTFE), perfluoroalkoxy (PTA), or

polyethylene because glass apparatus cannot be used since HF dissolve silica.

• Beakers made of PTFE

• Separation funnel

• Furnace

• Stirring bar

• Magnetic stirrer

• Sampling pipette

• A measuring cylinder

32
Chapter 3 Experimental

• Thermometer

• Inductively Coupled Plasma-Optical Emission Spectrometer (ICP-OES)

• X-Ray Diffractometry (XRD)

• X-Ray Fluorescence spectrometer (XRF)

• Scanning Electron Microscopy (SEM)

3.2. Raw materials

The following raw materials were used during the experimental procedures.

The tantalite ore was mined in Mozambique, and supplied by Thuthuka Group Ltd,
South Africa. All chemical reagents employed were of analytical grade. Dionized water
was used in all aqueous preparations. 1-octanol and 2-octanol, ammonium biflouride,
ammonium hydroxide, sulfuric acids were obtained from Merck Chemicals (Pty) Ltd and
used without further purification.

3.3. Planning

The extraction and separation of tantalum and niobium from tantalite was investigated
using 1-octanol and 2-octanol as extracting agents. The extraction is influenced by the
decomposition process, the initial solution acidity, decomposition temperature, time of
decomposition, the extracting agent, the stripping agent and the effect of precipitation.
These variables were changed to see the effectiveness of the extraction.

3.4. Experimental methods

The aim of this research was to find optimum conditions for extracting and separating
tantalum and niobium from Mozambican tantalite and produce tantalum oxide and
niobium oxide containing as few impurities as possible starting with a typical tantalite
concentrate. The selected processes to accomplish this are as follows:

33
Chapter 3 Experimental

• Tantalite ore was first crushed; the particle size distribution and its corresponding
particle surface area were determined by Mastersizer 3000 analyzer using water as
dispersant.

• Digestion or decomposition of the tantalite with ammonium bifluoride. An initial

solution was prepared by melting the mixture of ammonium bifluoride and tantalite at
different temperatures and times followed by leaching of the soluble component with
water and separation of the solution by filtration. The acidity of the original solution
was adjusted by adding an equal volume of sulfuric acid to the feed solution. This
solution is used to extract the metal values of interest, i.e. niobium and tantalum,
along with iron, aluminum and titanium as the major interfering elements.

• An equal volume of the organic solvent either 1-octanol or 2-octanol, was contacted
with the filtrate from the leaching process. The solution was then stirred at different
times and then left to separate. The solutions were separated using a separation
funnel. Extraction was performed using a polypropylene beaker and a magnetic
stirrer.

• The aqueous phase was analyzed by ICP-OES (Appendix A.1) and the organic
phase was stripped by different stripping agents, e.g. water and sulfuric acid.
Tantalum and niobium stripping was carried out at room temperature by shaking
equal volumes of the loaded solvent with a suitable stripping solution.

• By neutralizing the niobium and tantalum strip solution using an ammonia solution,
niobium and tantalum were completely precipitated and after calcination, niobium
oxide and tantalum oxide were identified and quantified by XRD (Appendix A.3) and
SEM (Appendix A.4).

3.4.1. The effect of time and temperature on the recovery of Nb and Ta

The effect of duration of decomposition and temperature on the percentage recovery of

Nb and Ta was studied for a temperature between 200 C and 250 C at a stirring speed

34
Chapter 3 Experimental

of 500 rpm and reaction time from 1 to 3 hours. All other conditions were kept constant.
They were:

• 50 g of tantalite

• 1 500 kg of ammonium bifluoride

• 500 ml of distilled water.

3.4.2. Concentration of sulfuric acid

This variable was investigated by first adding to the leach solution a predetermined
concentration of sulfuric acid from 2 M to 9 M. Equal volume of organic solvent (1-
octanol or 2-octanol) and aqueous solution obtained after adjusting the acidity of leach
solution were contacted for 2 minutes at various sulfuric acid concentrations from 2 M to
9 M. Liquid liquid extraction was performed using a polypropylene beaker and magnetic
stirrer. The organic phase and aqueous phase were separated in a separating funnel
after mixing for a certain time. The aqueous phase was analyzed using the ICP-OES
while the concentration of the organic phase was inferred from a mass balance. The
distribution coefficient ( ), separation coefficient ( ) was then determined along with the
extraction factor (%E).

3.4.3. Selection of stripping agent

Different stripping agents such as distilled water and sulfuric acid were used to strip
niobium and tantalum from the loaded solvent. The stripping experiments were
conducted at different contact time from 5 to 20 minutes and at an A/O (i.e. aqueous to
organic) ratio of 1/1. The organic phase from the extraction process was washed with
distilled water and then extracted with dilute sulfuric acid to obtain niobium by selective
extraction or stripping. The aqueous phase takes up the complex fluoroniobate while the
complex fluorotantale remains dissolved in the organic phase. The acidity of the
aqueous phase niobium solution was adjusted and then contacted with 2-octanol to
extract niobium. Tantalum and niobium were stripped from the organic phase with water
35
Chapter 3 Experimental

or sulfuric acid. Aqueous ammonia was then added to the solution to precipitate niobium
and tantalum.

3.4.4. Precipitation

A 28% of ammonium hydroxide (NH4OH) was added to the pregnant solution from the
stripping process, to precipitate niobium and tantalum. The precipitate was allowed to
settle for one day and then filtered. The precipitate were washed with boiling distilled
water and dried in an oven at 100°C. The dried precipitate was put in a porcelain
crucible and then placed in the muffle furnace. Calcinations were done at 900 C for four
hours. Niobium and tantalum pentoxide were obtained. The products were then
characterized by XRD and SEM analysis.

36
 CHAPTER 4 RESULTS AND DISCUSSION

This chapter presents the results and discussion of the decomposition process,
extraction, stripping, precipitation and characterizations of the products. It also
discusses further treatment required in order to produce a pure tantalum and niobium
oxide.

4.1. Particle size analysis

The influence of mineral particle size of the raw material plays a major role in
hydrometallurgical treatment process which cannot be neglected. The typical particle
size distribution of the tantalite ore determined by a Mastersizer 3000 analyzer is shown
in Figure 4.1. The volume weighted particle size mean is 18µm and it corresponds to a
surface area of 66,070 m2/g. Particle size was not included as an experimental
variable. The particle size was kept constant, or as constant as the technique allowed.

Figure 4.1. Particle size distribution of tantalite ore

37
 Results and discussion

4.2. Mineralogy of tantalite

For better understanding of the experiments presented in this dissertation, it is important

to know the nature of the raw materials utilized, what compounds they consist of and
what the characteristics and the values of these compounds are.

The mineralogical analysis of the ore sample was carried out by X-ray fluorescence
(XRF) (Appendix A.2) and X-ray diffractometry (XRD) (Appendix A.3) analysis. The
results of the XRF analysis are shown in Table 4.1 and it indicates that the sample is
really a tantalite because it has more tantalum oxide (75.89%) than niobium oxide
(4.72%) and also rich in iron oxide (15.09%). This mineral rich in tantalum and iron is
called tantalite-(Fe) or ferrotantalite.

The mineral phase recorded by XRD analysis of the tantalite is provided in Figure 4.2.
XRD analysis confirms that the main component in the ore sample is ferrotapiolite with
small quantity of ferrocolumbite which have almost the same chemical composition as
tantalite. Furthermore the XRF and XRD analysis showed the presence of associated
minerals such as Fe2O3, TiO2, MnO, etc.

38
Chapter 4 Results and discussion

Table 4. 1: Chemical composition of tantalite (XRF analysis)

Species Percentage
SiO2 0.37
TiO2 1.26
Al2O3 0.1
Fe2O3 15.09
MnO 1.4
ZrO2 0.12
SO3 0.02
Cl 0.01
ZnO 0.03
Co3O4 0.02
HfO2 0.11
Nb2O5 4.72
MoO3 0.02
SnO2 0.04
Ta2O5 75.89
OsO4 0.04

39
Chapter 4 Results and discussion

Figure 4.2: XRD pattern of tantalite

4.3. Preparation of the leach liquor

The decomposition process was performed by melting a mixture of ammonium

bifluoride and tantalite ore.

For each experiment, ammonium bifluoride was first charged into the reactor and
heated up to the dissolution temperature (230°C) and then the tantalite ore sample was
added in the required amount. The mixture was then heated to the required temperature
and kept there for the required time. After a given decomposition period the soluble
components were leached with water and the insoluble component separated by
filtration. The liquid sample was analyzed by ICP-OES to determine the content of
40
Chapter 4 Results and discussion

niobium, tantalum, iron, titanium, etc. The produced pregnant solution has a pH of 3 and
an assay of 32.82g/l of Ta2O5, 4.86g/l of Fe2O3, 2.51g/l of Nb2O5 and 2.43 g/l of TiO2,
after a leach period of 3 hours, at a ratio of tantalite ore-to-ammonium bifluoride of 1:30
and at decomposition temperature of 250°C.

4.4. Results of the decomposition experiments

The results are presented graphically in the paragraphs below. The variable ranges
investigated were determined by values in the literature or by observations made during
the preliminary experiments. The percentage of extraction was calculated from the
concentration of niobium and tantalum present in the filtrate and the concentration
reported in the feed solution.

The mass of niobium and tantalum present in the feed solution used during
decomposition was calculated from Table 4.1. The mass of the metals present in the
filtrate was also calculated from the concentration of niobium and tantalum present in
the filtrate produced, analytically determined with the ICP-OES. The percentage
extraction was then the mass of niobium or tantalum in the filtrate divided by the mass
of the relevant metals present in the feed solution. All the experiments were repeated
three times, the process error is below 4 percent and the percentage extraction reported
are the average of three identical experiments in each case.

The following variables were investigated: time and temperature on the recovery of
niobium and tantalum and the tantalite−to−ammonium bifluoride mass ratio.

4.4.1. Effect of time and temperature on the recovery of niobium and tantalum

The experimental conditions that were used to perform and evaluate time and
temperature on the recovery of niobium and tantalum are described in section 3.4.1.
The data are presented in Figures 4.3 and 4.4 for Nb and Ta, respectively. It can be
seen that by increasing the temperature an increasing recovery of Nb and Ta was
41
Chapter 4 Results and discussion

obtained, which was attributed to increase decomposition rate and diffusion of reactants
and reaction products. The recovery of tantalum, niobium and other impurities such as
titanium and iron did not change significantly when the decomposition temperature was
increased from 250°C to 300°C. It can also be seen that the recovery of Nb and Ta
gradually improves with time. The highest extraction of 95.07% and 98.52% of Nb and
Ta, respectively, was realized at 250°C after 3 hours. A decomposition temperature of
250°C and decomposition time of 3 hours are thus recommended. It was also reported
in previous work, for all grade ore that increasing the decomposition time and
temperature would increase the recovery of Nb and Ta significantly.

Figure 4.3: Effect of time and temperature on the recovery of niobium

42
Chapter 4 Results and discussion

Figure 4.4: Effect of time and temperature on the recovery of tantalum

4.4.2. Effect of the mass ratio of tantalite-to-ammonium bifluoride

A set of experiments was carried out to examine the effect of the tantalite-to-ammonium
bifluoride ratio. The experimental conditions were kept constant (500 ml of distilled
water, 250°C, at a stirring speed of 500 rpm) as mentioned before while varying the
tantalite-to-ammonium ratio from 1.0:20 to 1.0:50. The results are illustrated in Figure
4.5. From this result about 60% of titanium (IV) was leached. This foreign metal is one
of the least separable from Ta and Nb but it can be removed by backwashing with 1.5-2
M H2SO4 at Vo : Vaq of about (3.5-4.5):1 for Ta and with 5-10 M HF and 5-6 M H2SO4 at
Vo : Vaq = (3-4): 1 for Nb (Maiorov et al., 2001).

43
Chapter 4 Results and discussion

The percentage extraction of Ta, Nb, Fe and Ti increases up to a ratio of ~1:30, beyond
which no further advantage is observed. This may be attributed to the fact that
increasing the ammonium-to-tantalite mass ratio decreases the solid pulp density, and
therefore decreases the mass transfer resistance in the liquid-solid interface. From this,
an ammonium bifluoride-to-ore mass ratio of 1:30 was consequently regarded as the
optimum value and considered as appropriate in this dissertation.

Figure 4.5: Effect of tantalite-to-ammonium bifluoride mass ratio (decomposition conditions: 3

hours, 250°C, 500 rpm)

As mentioned previously in section 3.4.2, different concentration of sulfuric acid was

added to the leach solution. An equal volume of the mixture was then contacted with the
organic solvent (1-octanol or 2-octanol). In this study the effect of contact time between
the aqueous phase and organic solvent and the effect of sulfuric acid concentration
were investigated. Consequently, the distribution and separation coefficients were
determined along with the extraction efficiency.

44
Chapter 4 Results and discussion

4.4.3. Effect of contact time

The effect of contact time on the extraction efficiency of Nb and Ta by 2-octanol was
studied over the time range of 2 to 15 minutes. A set of experiments was performed at
different contact times at fixed conditions: an aqueous-to-organic (A/O) ratio of 1:1,
100% 2-octanol and with a feed solution concentration of 6 M for Ta extraction and 9 M
for Nb extraction. As seen in Figure 4.6, optimum extraction efficiency is achieved after
5-10 minutes. Extending the time longer than 15 minutes did not significantly change
the extraction efficiency. Therefore, a preferred contact time of 10 minutes is
recommended.

Figure 4.6: Effect of contact time on extraction efficiency of Nb and Ta

4.4.4. Concentration of sulfuric acid

As described in section 3.4.2, the effect of the concentration of sulfuric acid was
evaluated by adding an equal volume of the aqueous phase to the organic phase. In
Figure 4.7 and 4.8 the extraction efficiencies obtained by varying the amount of sulfuric
acid concentration in the leach liquor solution which is contacted with the organic
45
Chapter 4 Results and discussion

solvent are presented. The extraction efficiency of tantalum is higher than for niobium,
and the extraction of both increases when the concentration of sulfuric acid increases.
The extraction of tantalum increased from 93.1% to 99.6% with increasing H2SO4
concentration in the range of 2.0 to 5.0 M and from 55.2 % to 89.1% with increasing
H2SO4 concentration in the range of 2.0 to 9.0 M H2SO4 for niobium with 1-octanol,
(Figure 4.7). From 92.8% to 99.8% for tantalum with increasing H2SO4 concentration in
the range of 2.0 to 6.0 M and from 55.5% to 89.2% for niobium with increasing H2SO4
concentration in the range of 2.0 to 9.0 M with 2-octanol, (Figure 4.8). Tantalum starts to
move into solution at much lower acid concentration while niobium starts moving into
solution only at higher concentration. As the extraction of both elements increased
because of this reason, collective extraction can be done where both elements are in
the organic phase at high concentration. However, a selective extraction was chosen in
order to produce a product of higher purities.

Extraction of Nb and Ta with octanol has increased considerably on the introduction of

sulfuric acid into the aqueous phase. It can also be seen that the extraction of tantalum
and niobium with 2-octanol are almost similar to those with 1-octanol, as the literature
also shows (Mayorov et al., 2002; Maiorov et al., 2001).The literature also show that the
introduction of H2SO4 to the feed solution improve the extracting characteristics system
with octanol considerably (Mayorov et al., 2002).

46
Chapter 4 Results and discussion

Figure 4.7: Effect of sulfuric acid concentration on Ta and Nb extraction by 1-octanol (error ±
0.04%) (A/O 1:1, 100% 1-octanol, 10 min).

Figure 4.8: Effect of sulfuric acid concentration on Ta and Nb extraction by 2-octanol (error ±
0.04%) (A/O 1:1, 100% 2-octanol, 10 min).
47
Chapter 4 Results and discussion

4.4.5. Distribution and separation coefficient

The extraction of tantalum and niobium experiments were carried out by mixing 30 mL
of the organic phase with an equal volume of aqueous phase, containing tantalum and
niobium. Each mixture was shaken at room temperature (25 C) for about 10 minutes.
The phases were then separated in a separating funnel.

Table 4.2 shows the 1-octanol results. The distribution coefficient of niobium increases
with an increase of sulfuric acid concentration in the feed solution throughout the range
investigated (from 2 M to 9 M). The distribution coefficient of tantalum on the other hand
increases up to 5 M (sulfuric acid concentration), reaching a maximum, and then again
decreases at higher acid concentrations. The separation factor also reaches a
maximum at 5 M, indicating the optimal acid concentration for separation of niobium and
tantalum.

The 2-octanol results in Table 4.3 show similar behavior, with the optimum shifted to a
higher concentration of 6 M for Ta and 9 M for Nb. But in the work of Agulyansky et al.,
(2004) it was reported that a solution containing 2.5-3.5 M H2SO4 was optimal for Ta
extraction while Nb begins to move into the organic phase at acidity level greater than
that of 5 M. The difference could be attributed to the behavior of the ore in acid solution.
The ore used in this study is different from the one used by Agulyansky and even the
impurities attached to it which can react differently under acidic solution.

48
Chapter 4 Results and discussion

Table 4.2: Distribution-separation coefficient and extraction factor for 1-octanol with different
H2SO4 concentrations

H2SO4(M) Ta Nb ETa % ENb %

2 16.952 1.460 11.614 94.429 59.343

3 52.711 1.794 29.378 98.138 64.212

4 167.753 2.421 69.378 99.407 70.765

5 250.183 2.995 83.533 99.602 74.969

6 246.944 3.612 68.349 99.577 78.322

7 216.477 4.288 50.483 99.540 81.090

8 157.333 5.722 27.498 99.368 85.123

9 126.478 7.368 17.166 99.216 88.050

49
Chapter 4 Results and discussion

Table 4. 3: Distribution-separation coefficient and extraction factor for 2-octanol with different
H2SO4 concentrations

H2SO4 Ta Nb ETa % ENb %

2 12.936 1.249 10.359 92.824 55.530

3 46.331 1.614 28.712 97.887 61.739

4 142.882 2.097 68.151 99.305 67.706

5 340.400 3.049 110.632 99.707 75.304

6 528.813 4.582 115.409 99.811 82.086

7 452.234 4.565 99.066 99.779 82.031

8 143.170 6.991 20.479 99.306 87.486

9 139.484 8.249 16.908 99.288 89.189

4.5. McCabe-Thiele analysis

4.5.1. McCabe-Thiele diagram for Ta extraction

Solvent extraction studies of niobium and tantalum with 2-octanol as a solvent from the
NH4.HF-H2O system was carried out. After filtration of the insoluble residue, the
aqueous solution of tantalum-niobium was extracted with the 2-octanol. The complex
fluorides of niobium and tantalum were extracted with 2-octanol and the impurities like
iron, manganese, titanium, etc., remained in the aqueous phase.

In order to determine the number of stages of extraction of Ta, a McCabe-Thiele

diagram was constructed by plotting the equilibrium concentration data of Ta extraction
in the organic and aqueous phases, followed by fitting a suitable operating line. The

50
Chapter 4 Results and discussion

equilibrium data in Figure 4.9 was obtained at different A/O ratios using optimum
extraction condition of 100% 2-octanol, 6M H2SO4, 10 minutes contact time at room
temperature. From the McCabe-Thiele diagram for Ta, Figure 4.9, theoretically two
stages are required for almost 99.7% Ta extraction from the feed solution. According to
Agulyansky, (2004), for the collective extraction of Ta and Nb, 5 to 7 extraction stages
are required. The difference could be attributed to the difference of concentration of
niobium oxide and tantalum oxide in the sample ore. In the study by Agulyansky,
(2004), the feed solution contain a concentration of 50-60g/l of Ta2O5 and 30 g/l of
Nb2O5 while in this research the feed solution contain only 32.82 g/l of Ta2O5 and 2.51
g/l of of Nb2O5.

Figure 4.9: McCabe-Thiele diagram of Ta extraction (A/O 1:1, 100% 2-octanol, 10 min, 6 M).

In this thesis tantalum was extracted from the tantalite-NH4.HF-H2O solution at low
sulfuric acid concentration. In order to remove traces of niobium from tantalum, distilled
water and various concentrations of sulfuric acid were tested from 0.5 M to 2 M.
According to this test, the concentration of 2 M sulfuric acid was chosen because this
51
Chapter 4 Results and discussion

condition gave the maximum amount of niobium and tantalum, respectively, in the
aqueous and organic phase. Then four stages of scrubbing of the tantalum-laden
organic phase with 2-octanol can be used to reduce the amount of niobium
contamination in tantalum. The aqueous phase takes up the complex fluoroniobate and
free hydrofluoric acid, while the complex fluorotantalate remains dissolved in the organic
phase. The pure tantalum in the form of H2TaF7 was then finally stripped.

4.5.2. Tantalum stripping

Different stripping agents were used to strip Ta from the tantalum-laden organic phase.
These include distilled water and sulfuric acid. The experiments were carried out at
room temperature by shaking equal volume of the tantalum-laden organic phase with a
suitable stripping agent for the desired time. Different factors affecting the efficiency of
stripping were studied. These include the stripping agent and its concentration and
stripping time.

4.5.3. Effect of stripping agent

Different stripping agents such as distilled water and sulfuric acid with different
concentrations (0, 5 and 1M) were used to strip Ta from the tantalum-laden organic
phase. The tests were conducted at 15 minutes contact time and 1:1 A/O ratio. This test
revealed that distilled water was the most efficient stripping agent of Ta from the
tantalum-laden organic.

4.5.4. Effect of contact time

The effect of contact time on the stripping efficiency of Ta was studied over the time
range of 2 to 15 minutes. A set of experiments were performed at different contact times
at fixed conditions: A/O ratio of 1:1. From the results obtained in Figure 4.10 an
optimum stripping efficiency is achieved after 15 minutes. A contact time of 15 minutes
was then recommended.
52
Chapter 4 Results and discussion

Figure 4.10: Effect of contact time on stripping efficiency of Ta

4.5.5. McCabe-Thiele diagram for Ta stripping

McCabe-Thiele diagrams were constructed in order to determine the number of steps

needed to reach the desired separation. The data were obtained by shaking the
tantalum-laden solvent with distilled water for 15 minutes using different A/O ratios. The
diagram was constructed assuming these optimal operating conditions, as determined
above. From the McCabe-Thiele diagram, theoretically four stripping stages are
required for complete Ta stripping from the tantalum-laden solvent. Figure 4.11 shows
the McCabe-Thiele diagram for Ta stripping. The stripping efficiency of our process is
almost similar to that of Agulyansky (2004), who suggest that tantalum could be stripped
from the loaded solution by 4 to 6 extractions stages.

53
Chapter 4 Results and discussion

Figure 4.11: McCabe-Thiele diagram of Ta stripping

4.5.6. Ta precipitation

The precipitation of the tantalum oxide was done in a batch process. The tantalum strip
solution was neutralized by adding 28% ammonium hydroxide (NH4OH) solution to the
strip solution. After filtration, the precipitates were washed with distilled water and dry in
the oven at 100°C. The dried precipitates were then put in a porcelain crucible. The
crucible was placed in the muffle furnace at 900°C for 4 hours.

After calcinations, pure tantalum pentoxide was obtained and characterized using XRD
and SEM analysis as shown in the following Figures 4.12, 4.13, 4.14 and 4.15.

54
Chapter 4 Results and discussion

Figure 4.12: XRD pattern of synthesized Ta2O5 Figure 4.13: XRD pattern of purchased Ta2O5

Figure 4.14: SEM result of synthesized Ta2O5 Figure 4.15: SEM result of purchased Ta2O5

55
Chapter 4 Results and discussion

The XRD patterns and SEM results of the synthesized and purchased products (Ta2O5)
are shown in Figures 4.12, 4.13, 4.14 and 4.15, respectively. It can be seen that the
XRD patterns of synthesized and purchased products (Ta2O5) are similar. An XRD
pattern indicates that no other elements are presents except Ta2O5. The SEM results
indicate the presence of traces of iron and titanium and this may be attributed to the
excess of this oxide in the ore which was not removed completely during the extraction
process. To obtain higher quality tantalum pentoxide, additional process may be
needed.

For the removal of the last traces of iron and titanium, an intermediate treatment of the
hydrated oxide precipitate with concentrated HCl can be incorporated (Damodaran et
al., 1969).

4.6. McCabe-Thiele diagram for Nb extraction

After separation of tantalum and niobium by scrubbing the solution with 2 M H2SO4, the
aqueous phase takes up the complex fluoroniobate. The pH of the niobium containing
solution was adjusted by mixing the solution with 9 M H2SO4 for solvent extraction. The
aqueous solution was then contacted with fresh octanol to extract niobium.

A McCabe-Thiele diagram was constructed by plotting the equilibrium concentration

data of Nb extraction in the organic and aqueous phases. The equilibrium data in Figure
4.16 was obtained at different A/O ratios using optimum extraction condition of 100% 2-
octanol, 10 minutes contact times, 9 M H2SO4 at room temperature. From the McCabe-
Thiele diagram for Nb in Figure 4.16 it can be seen that three theoretical equilibrium
stages are required for 93.6% Nb Extraction.

56
Chapter 4 Results and discussion

Figure 4.16: McCabe-Thiele diagram of Nb extraction( A/O 1:1, 100% 2-octanol, 10min, 9M)

The organic phase rich in niobium pregnant solution was obtained and stripped with
different stripping agent.

4.6.1. Niobium stripping

Different stripping agents were used to strip Nb from the niobium-laden organic phase.
These include distilled water and H2SO4. The stripping experiences were carried out at
room temperature by shaking equal volumes of the niobium-laden organic phase with a
suitable stripping agent for the desired time. Different factors affecting the efficiency of
stripping were studied. These include stripping agent and its concentration, and
stripping time.

57
Chapter 4 Results and discussion

4.6.2. Effect of stripping agent

Different stripping agents such as distilled water and sulfuric acid with different
concentrations (1 M to 5 M) were used to strip the Nb from the niobium-laden organic
phase. The tests were conducted at 10 minutes contact time and 1:1 A/O ratio.
According to this test distilled water was the most efficient stripping agent of Nb from the
niobium-laden organic.

4.6.3. Effect of contact time

The effect of contact time on the stripping efficiency of Nb was studied over the time
range of 2 to 15 minutes. A set of experiments were performed at different contact time
at fixed conditions: A/O ratio of 1:1 and H2O. From the results obtained an optimum
stripping efficiency is achieved after 15 minutes. A contact time of 15 minutes was then
recommended.

Figure 4.17: Effect of contact time on stripping efficiency of Nb

58
Chapter 4 Results and discussion

4.6.4. McCabe-Thiele diagram for Nb stripping

The data points were obtained by contacting metal-rich solvent with water for 15 min,
using different A/O ratios.

From the McCabe-Thiele diagram in Figure 4.18, theoretically four stripping stages are
required for almost complete Nb stripping from the loaded solvent.

Figure 4.18: McCabe-Thiele diagram of Nb stripping

4.6.5. Nb precipitation

Precipitation of niobium pentoxide was done in a batch process. The niobium strip
solution was neutralized by adding 28% ammonium hydroxide (NH4OH) solution to the
strip solution. After filtration, the precipitates were washed with distilled water and dry in
59
Chapter 4 Results and discussion

the oven at 100°C. The dried precipitates were then put in a porcelain crucible. The
crucible was placed in the muffle furnace at 900°C for 4 hours. After calcinations, pure
niobium pentoxide was obtained and characterized using XRD, and SEM as shown in
different Figures 4.19, 4.20, 4.21, and 4.22. A general flow sheet for treating tantalite by
the ammonium-octanol process is proposed and represented in Figure 4.23.

Figure 4.19: XRD pattern of synthesized Nb2O5 Figure 4.20: XRD pattern of purchased Nb2O5

60
Chapter 4 Results and discussion

Figure 4.21: SEM result of synthesized Nb2O5 Figure 4.22: SEM result of purchased Nb2O5

The XRD patterns of the synthesized and purchased product (Nb2O5) are almost similar.
An XRD pattern indicates traces of titanium and tantalum. According to Maiorov et al.,
(2001) titanium can be removed from niobium by backwashing. By stripping niobium
with water it is possible to remove tantalum completely from niobium by increasing the
number of steps. In order to remove the impurities completely, a HCl digestion could be
necessary.

61
 Chapter 4 Results and discussion

Tantalite
Fusion NH4HF2

S\R 1:30; 250 C, 3 Hours Decomposition

H2O

Leaching

100% octanol A/O 1:1, 15 Min Residue

Filtration

H2SO4 6 M
Extraction, Ta

H2SO4 9 M
H2SO4 2 M
Ta- Extract Raffinate 100% octanol A/O 1:1, 15 Min

H2O A/O 1:1

Ta- Stripping Raffinate Nb- Extract Raffinate

H2O A/O 1:1, 15 Min

28% NH4OH
Nb- Stripping
Ta
Raffinate

H2O 28% NH4OH

Nb
Washing

Washing H2O
Calcinations

Calcinations

Ta2O5
Nb2O5

Figure 4.23: Proposed technical flow sheet for the extraction of Nb/Ta.
62
 CHAPTER 5 CONCLUSIONS AND RECOMMENDATIONS

5.1. Conclusions

The aim of this research was to extract and separate niobium and tantalum with octanol
as solvent, from Mozambican tantalite using ammonium bifluoride as an alternative to
hydrofluoric acid. The specific objectives were: 1) To determine if ammonium bifluoride
can be used as an alternative to hydrofluoric acid in the decomposition process,
determine the time and temperature for efficient recovery of niobium and tantalum and
also the mass ratio between tantalite and ammonium bifluoride. 2) To investigate the
effect of sulphuric acid concentration in the feed solution and the effect of different
stripping agents. 3) To compare the performance of extraction between 1-octanol and 2-
octanol. 4) To characterize the niobium and tantalum pentoxide and compare to the
purchased product.

Ammonium bifluoride can be used successfully for the decomposition of Mozambican

tantalite. The leach liquor prepared in this way is of consistent concentration and Ta and
Nb can be extracted successfully using octanol as extractant. This process present
some advantage compared to HF. Firstly the cost could be reduced because of the
availability of ammonium bifluoride in South Africa and secondly could prove easier to
handle compare to aqueous hydrofluoric acid.

From the experiment results it was observed that the increase of time and temperature
was effective in increasing the leach recovery of Nb and Ta. The highest recovery was
realized at 250°C after 3 hours. By varying the tantalite-to-ammonium mass ratio, the
percentage extraction increases for both Nb and Ta up to a ratio of 1:30, beyond this no
further advantage was observed.

The introduction of sulfuric acid improved considerably the extraction efficiency of Nb

and Ta with octanol. Tantalum starts to move into solution in the range of 2.0 to 6.0 M

63
with highest percentage of extraction of 99.7% at 6.0 M with feed contacted with 2-
octanol. Nb starts moving into solution at high sulfuric acid concentration. The highest
value of extraction was 89.2% for 9.0 M sulfuric acid.

The performance of extraction of 1-octanol and 2-octanol was almost similar, as the
literature also shows.

Mc-Cabe-Thiele diagrams of Nb and Ta extraction were constructed using optimum

conditions: A/O of 1:1, 100% 2-Octanol, 10 minutes contact time and 6 M sulfuric acid
concentration and A/O of 1:1, 100% 2-octanol, 10 minutes contact time and 9 M sulfuric
acid concentration respectively for Ta and Nb extraction. The results show that two and
three theoretically numbers of stages were observed respectively for Ta and Nb.

The Mc-Cabe-Thiele diagram of Nb and Ta stripping was also constructed using the
following optimum parameters: A/O 1:1, H2O as stripping agent and 15 minutes contact
time. It can be concluded that four theoretically stripping stages are sufficient to strip Ta
and Nb from the organic phase.

It was also observed that distilled water was the most efficient agent, compared to
sulfuric acid, for stripping both Nb and Ta from the niobium and tantalum-laden organic
phase.

After precipitation and calcination tantalum and niobium pentoxide with fewer impurities
such as iron and titanium were obtained as characterized by XRD and SEM analysis. In
order to obtain a pure product it is therefore necessary to incorporate an intermediate
treatment.

64
5.2. Recommendations

The following recommendations are suggested for further investigations which could
help the extraction and separation of niobium and tantalum from Mozambican tantalite
by solvent extraction in the ammonium bifluoride-octanol system to be used efficiently.

• Further processing is necessary in order to produce a high purity tantalum and

niobium pentoxide suitable for electronics industries for manufacturing capacitors,
superconductivity and other specific applications.

• The economic aspects must be evaluated from the flow diagram.

65
 REFERENCES

• Amuda, M.O.H, Esezobor, D.E, and Lawal, G.I (2007) “Adaptable technology for life-
cycle processing of tantalum bearing minerals” Journal of Minerals & Materials
Characterization & Engineering, vol.6, no.1, pp 69-77.

• Agulyansky, A (2004) The chemistry of Tantalum and Niobium Fluoride compounds,

Elsevier, Amsterdam.

• Agulyansky, A, Ludmila, A and Viktor F. T (2004) “Liquid–liquid extraction of

tantalum with 2-octanol” Chemical Engineering and Processing: Process
Intensification, vol 43, no 10, pp. 1231-1237.

• Aprill, U and Richard L.D (2005) Methods of Soil Analysis, Madison, Wisconsin,
U.S.A.

• Baba, A.A, Adekola, F.A, and Faseki, M(2005) “ A study of the kinetics of the
dissolution of a Nigerian tantalite ore in hydrochloric acid” Ife Journal of Science, vol.
1, no 2, 221-227.

• Babkin, A.G, Mayorov, V.G, Nikolaev, A.I, (1988) “The Extraction of Niobium,
Tantalum and other Elements from Fluoride Solutions” Nauka,Leningrad 204 pp.

• Bennett H and Graham J. O (1992) XRF Analysis of Ceramics, Minerals and Allied,
John Wiley & sons, New York, U.S.A.

66
 References

• Bludssus, W and Eckert, J (1993) “Process for the recovery and separation of
tantalum and niobium”, US Patent 5209910, assigned to Hermann C. Starck Bertin,
Berlin DE.

• Bose, D.K and Gupta, C.K (2001) “Extractive metallurgy of tantalum” Mineral
Processing and Extractive Metallurgy Review, 22, pp 389-412.

• Cardon, P.B (1962)” Process for recovering niobium and tantalum from ores and
ores concentrates containing same”, U.S. Patent 3058825.

• Charles B. B and Kenneth J. F (1997) Concepts, Instrumentation and Techniques in

Inductively Coupled Plasma Optical Emission Spectrometry, Perkin Elmer second
edition, U.S.A.

• Clive, W (1987) X-ray Methods, John Wiley and sons, New York, U.S.A.

• Coulson J.M and Richardson’s (1996) Chemical Engineering, Butterworth Heineman,

vol 2, 287, Oxford.

• Coulson J.M and Richardson’s (2002) Chemical Engineering, Butterworth Heineman,

Vol.2.5th Edition, oxford.

• Damodaran, A.D, Deshpande, S.G, Majmudar, A.A and Sastri, M.S (1969)
“Extraction and utilization of pure niobium and tantalum from Indian ores” Bhabha
Atomic Research Center, Trombay, Bombay, vol. 36, No.5, pp. 306-318.

67
 References

• Eastell, J and Willis, J. P (1990) “A low dilution fusion technique for the analysis of
geological samples” X-RaySpectrometry, 19, 1–14.

• Eckert, J, (1995)” Hydrometallurgical processing of Ta/Nb compounds”, Present state

of the art. Proceeding of International Symposium on Tantalum and Niobium,
Germany. Tantalum-Niobium International Study Center, Belgium, pp. 55 – 64.

• El-Hussaini, O.M. and Mahdy, M.A. 2002. Extraction of niobium and tantalum from
nitrate and sulphate media by using MIBK. Minerals Processing and Extractive
Metallurgy Review 22, (4-6) 633-650.

• El-Hussaini, M.O and Rice, M .N (2003) “liquid-liquid extraction of Niobium and

tantalum from aqueous sulphate/fluoride solution by tertiary amine” hydrometallurgy,
vol 72, no 3-4, pp 259-267.

• Fathi H (1997) Handbook of Extractive Metallurgy, volume III, Quebec, Canada.

• Foos, R.A (1960) “Recovery of columbium values”, U.S. Patent 2953453, assigned to
Nat Distillers Chem Corp.

• Gibalo, I.M (1970) Analytical Chemistry of Niobium and Tantalum, Ann-arbor,

London.

68
 References

• Giulio, M, Pierfrancesco, R.I, Francesco, M, Luca, R, Paolo, F.M and Carlo, F (2009)
“ An overview about biomedical applications of micron and nano size” Benthan
science, vol 3, No 3, 157-165.

• Gupta C.K, Mukherjee T.K (1990) Hydrometallurgy in Extraction Processes, CRC

Press, Boca Raton, Florida, USA.

• Gupta, C.K (1994) Extractive Metallurgy of Niobium, Tantalum and Vanadium, CRC
Press, Boca Raton, Florida, USA.

• Gupta, C.K and Suri, A.K (1994) in: Extractive Metallurgy of Niobium, CRC Press,
Boca Raton, Florida, USA.

• Gupta, C.K (1984) “Extractive metallurgy of niobium, tantalum, and vanadium”

International Metals Reviews, 29, pp 405-444(40).

• He, C, Liu, Z and Zhang, H (1998) “Treatment of fluorine-containing waste gas from
hydrometallurgy of tantalum and niobium ore” Nonferrous Metals 50(4), 141-142.

• Hearle, J.W, Sparrow, J.T and Cross, P.M (1974).The Use of the Scanning Electron
Microscopy, Pergamon Press, New York.

• Hutton, J, T. and Elliott, S. M (1980) “An accurate XRF method for the analysis of
geochemical exploration samples for major and trace elements using one glass disc”
Chemical Geology, 29, 1–11.

69
 References

• James R.C (2007) “Introduction to X-ray Powder Diffraction”, available at

http://epswww.unm.edu/xrd/xrdclass/01-XRD-Intro.pfd (accessed on 08 November,
2011).

• Joachim E, Karlheinz R, Christoph S and Hady S, sa, “Processing of columbite-

tantalite ores and concentrates for niobium and niobium compounds in electronic
applications” Germany, H.C. Starck GmbH and Co. KG. Available at
www.cbmm.com.br (accessed 01.11.2011).

• Joseph, I.G, Dale. E.N, Patrick, E, David, C.J, Charles, F, Eric , L (1981) Scanning
Electron Microscopy and X-ray Microanalysis, Plenum Press, New York.

• Krismer, B, Hoppe, A (1984)”Process for the recovering niobium and /or tantalum
compounds from such ores further containing complexes of uranium, thorium,
titanium and/or rare earth metals”, U.S. Patent 4446116, assigned to Hermann C.
Starck Bertin, Berlin DE.

• Loubser, M and Verryn (2008) “Combining XRF and XRD analyses and sample
preparation to solve mineralogical problems” South African Journal of Geology,
vol.111, 2-3, pp.229-238.

• Maina, N.S., Ameh, A.O. and Orkuma, H.H (2007) “Leaching of Tantalum and
Niobium from Tantalite ore in aqueous media” Chemclass Journal vol 4, 19-28.

• Mayorov, V.G. and Nikolaev, A.I (2002) "Tantalum (V) and niobium (V) extraction by
octanol" Hydrometallurgy, vol. 66, no. 1-3, pp. 77-83.

70
 References

• Maiorov V.G, Nikolaev A.I, and Kopkov V.K (2001) “Extraction Recovery of Tantalum
(V) and Niobium (V) from hydrofluoric and hydrofluoric- sulfuric acid aqueous
solutions with Octanol” Russian Journal of Applied Chemistry vol 74, no 4, pp 357-
360.

• Moore, G.L (1989) Introduction to Inductively Coupled Plasma Atomic Emission

Spectrometry, Elsevier, New York, U.S.A.

• Morrison, G.H and Freiser, H (1957) Solvent Extraction in Analytical Chemistry, New
York, U.S.A.

• Ritcey.G.M and Ashbrook.A.W (1984) Solvent Extraction Principles and Applications

to Process Metallurgy, Part 1; Elsevier, Amsterdam.

• Robert H. Perry and Don W. G (2007) Perry’s Chemical Engineers’ Handbook,

seventh edition, McGrall-Hill, New York.

• Rydberg, J, Michael C, Claude M, Gregory R. C (2004) Solvent Extraction Principles

and Practice, second edition, revised and expanded, New York, U.S.A.

• Sean,P, Arza, S, Michalina ,B, Kellsee C, Stephanie, A (2007) “Tantalum and

tantalum compounds”, In Kirk-Othmer Encyclopedia of Chemical Technology, fourth
edition, vol 24, John Wiley & Sons, New Jersey.

71
 References

• Sudderth, R.B. and Kordosky, G.A (1986)”Some Practical Considerations in the

Evaluation and Selection of Solvent Extraction Reagent” Chemical Reagent in the
Mineral Processing Industry. Ed Malhotra and Riggs. S.M.E. Littleton Colorado. 181-
196.

• TIC (2009) “Niobium and tantalum –raw materials and processing”, Available at
<http://www.tanb.org (accessed on 28, February, 2011).

• Thompson, M. J and Nicholas W (1989) Handbook of inductively coupled plasma,

Chapman and Hall, New York, U.S.A.

• Ulman F, Gerhartz W and Elvers B (1988) Ullmann’s Encyclopedia of Industrial

Chemistry, 5th edition, B3: Unit Operations 2, 6-45.

• Wang, X., Shili, Z., Hongbin, X., Yi. Z., (2009). Leaching of niobium and tantalum
from low-grade ore using a KOH roast-water leach system. Hydrometallurgy 98,
219-223.

• Zelikman, A.N, Orekhov, M.A (1965) “Decomposition of tantalum concentrates by

sodium and potassium hydroxides at increased temperatures and pressures” Izvestia
Akademii Nauk SSSR. Metally 6, 38 -45.

• Zhu, Z., Cheng, C.Y., 2011. Solvent extraction technology for the separation and
purification of niobium and tantalum: A review. Hydrometallurgy 107 (1-2), 1-12.

72
 References

• Zhou, H., Zheng, S., Zhang, Y., 2005. Leaching of a low-grade niobium–tantalum ore
by highly concentrated caustic potash solution. Hydrometallurgy 80(1-2), 83-89.

73
 APPENDICES A

A.1 INDUCTIVELY COUPLED PLASMA OPTICAL EMISSION SPECTROMETRY

(ICP-OES)

THE THEORETICAL BACKGROUND

The inductively coupled plasma optical emission spectrometry is an effective source of

atomic and ionic emission which can be used predominantly for quantitative multi-
elements analysis of most types of samples except argon and helium. It is an ideal
technique which allows the analysis of both major and trace element from single sample
due to it high dynamic range. Certain elements cannot conveniently be determined
using an ICP-OES without additional specific modifications to the instruments (Charles
et al., 1997). .

In ICP-OES, the sample is usually transported into the instrument as a stream of liquid.
Inside the instrument, the liquid is converted into an aerosol through a process known
as nebulization. The sample aerosol is then transported to the plasma where it is
desolvated, vaporized, atomized, and excited and/or ionized by the plasma. The excited
atoms and ions emit their characteristic radiation which is collected by a device that
sorts the radiation by wavelength. The radiation is detected and turned into electronic
signals that are converted into concentration information for the analyst. A
representation of the layout of a typical ICP-OES instrument is shown in Figure A.1
(Charles et al., 1997).

An ICP-OES system is comprised of the following components (Figure A.1):

• A sample introduction system, which houses a nebulizer, peristaltic pump, spray

chamber and drains;

• A torch and radio frequency generator;

74
 Appendix A

• The transfer optics;

• Spectrometer and detectors;

• Signal processing and instrument control system;

• Additional accessories, such as autosamplers, samples introduction accessories and

hybrid generation unit.

Figure A.1: Major components and layout of a typical ICP-OES instrument (adapted from Charles
et al., 1997)

75
 Appendix A

THE VARIOUS COMPONENT OF AN ICP-OES

Sample introduction

The sample introduction system consists of a nebulizer to aspirate sample that have
been pumped through peristaltic pump, a spray chamber which removes large droplets
from the aerosol and a drain which carries the excess of sample from the spray
chamber to a waste container.

Peristaltic pump

The sample solution is transported by peristaltic pump through the tubing using a
process known as peristalsis to the nebulizer'sample inlet connection. The pump used
must be compatible with weakly acidified aqueous media. Proper tubing made of
specific materials is usually required for strongly acidic solutions or organic solvents
(Charles et al., 1997). By using a peristaltic pump it is possible to control and prevent a
steady stream of sample which is being delivered to the nebulizer. Peristaltic pump
must be frequently replaced to avoid permanent depressions in the tubing. This aspect
can compromise the performance of the instrument. Figure A.2 shows the peristaltic
pump used for ICP-OES.

76
 Appendix A

Figure A.2: Peristaltic pump used for ICP-OES (adapted from Charles et al., 1997)

Nebulizers

There is general agreement that sample introduction into the ICP is critical. Most
samples pump into the ICP is in the liquid form which is converted to a fine aerosol. The
smallest droplets (generally less than 10 µ m in diameter) are transported to the plasma
by the injector gas and the largest droplets are removed in the spray chamber (Moore et
al., 1989; Thompson et al., 1989; Charles et al., 1997)

There are two main types of nebulizers: pneumatic and ultrasonic. Pneumatic
nebulization is the most widely used method for introducing samples into the ICP. It
uses a high speed gas flow to create an adequate aerosol from the introduced sample.
Different types of pneumatic nebulizer are in use. The concentric, cross flow and
Bablington types are most commonly used (Moore et al., 1989; Thompson et al., 1989;
Charles et al., 1997).

77
 Appendix A

In an ultrasonic nebulizer, the sample is pumped into a vibrating piezoelectric

transducer (1 to 10 MHZ), which breaks the sample in a fine aerosol. The sensitivity of
this nebulizer can be up to 4 times higher than the pneumatic type. Ultrasonic
nebulizers suffer from poor long-term stability, and their performance is influenced by
the change in the operational parameters. The advantage of this type is that the
detection limits are in order of magnitude lower than those normally obtained by
pneumatic nebulizers (Moore et al., 1989).

Spray chamber

The spray chamber classifies the aerosol droplets and determines the sizes of droplets
that will pass into excitation source. Only small droplets in the aerosol are suitable for
injection into the plasma. The main purpose of a spray chamber is to remove the larger
droplets from the nebulizer, which are condensed and run to waste. An ideal spray
chamber will have the best separation capabilities and the shortest memory effect for
the particulars samples to be analyzed. The most widely used spray chamber is of the
single pass or double passes (Moore et al., 1989; Thompson et al., 1989; Charles et al.,
1997).

78
 Appendix A

Figure A.3: A typical spray chambers used in ICP-OES (adapted from Charles et al., 1997)

Drains

The drain carries the excess sample from the spray chamber to a waste container. It
can have an impact on the performance of the ICP instrument. It also provides the
backpressure necessary to force the sample aerosol-carrying nebulizer gas flow
through the torch’s injection tube and into the plasma discharge. It is important to keep
the liquid level within the drain system at a recommended position. Drains systems
come in many forms loops, blocks, U−tubes, or even tubing connected to peristaltic
pump (Charles et al., 1997).

79
 Appendix A

PRODUCTION OF EMISSION

Torches

The torch consists of three concentric quartz tube for argon flow and aerosol injection.
There are two types of torches. The first and the most popular torches are of the
demountable type. This torch can be disassembled, modified or the parts replaced
easily and inexpensively. The main advantage of this torch is the lower cost of
replacement and the ability to use a variety of injector tubes (Charles et al., 1997).

The second torch is a fixed torch that has many designs, but the most commonly used
is the Scott-Fassel type (Moore et al., 1989). This type of torch uses lower gas flow and
requires a high degree of concentricity. The drawback of this torch is that it does not
tolerate the injection of air and this may result in the plasma being extinguished
(Thompson et al., 1989).

Radio frequency generators

The radiofrequency (RF) generator is used to provide a high frequency voltage to the
ICP for the generation and sustainment of the plasma discharge. There are essentially
two classes of generators. The first type is called crystal−controlled generator. It uses a
piezoelectric quartz crystal to produce an output at constant frequency.

The second is a free-running generator, which operates at an oscillation frequency that

is not only circuitry but also on the plasma discharge. Free-running generators are
made from simpler parts and are smaller and less costly than crystal-controlled
generators. The performance of the analytical instrument is not greatly dependent on
which type of RF generator is used and there is no clear evidence which one is superior
to the other.

80
 Appendix A

COLLECTION AND DETECTION OF EMISSION

Transfer optics

The region, in which the emitted radiation from the plasma is collected, is known as the
normal analytical zone (NAZ). There are two NAZ of the plasma that are employed. The
first is a radical or side-on viewing of the plasma. The NAZ is observed from the side of
the plasma operating in vertical orientation.

The second is referred as end-on viewing. In this type, the zone is observed axially from
the end of the plasma.

Recently, a new type was introduced, which combines both radial and axial viewing,
called dual viewing. The advantage of this is the ability to optimize the appropriate
configuration for the types of the sample without the expense of two separate ICPs
(Charles et al., 1997).

A combination radial (side-on) and axial (end-on) ICP is shown in Figure A.4.

81
 Appendix A

Figure A.4: A combination radial (side-on) and axial (end-on) ICP (adapted from Charles et al.,
1997)

SPECTROMETER

The main functions of a spectrometer is to form the light into a well-defined beam and
dispersed it according to wavelength with a grating and focus the dispersed light into an
exit plane or circle (Charles et al., 1997). There are two types of spectrometers. The
simultaneous spectrometer (or polychromotor), and the sequential spectrometer (or
monochromotor).

Simultaneous systems have a higher sample throughput as more elements can be

analyzed within a short period of time. The disadvantages of this are the higher cost and
lack of flexibility. These can be improved by using an ICP-OES with more than one
polychromotors systems. The most widely used mounting for polychromotor is the
Pashen-Runge construction (Charles et al., 1997).

82
 Appendix A

Figure A.5: The Pashen-Runge mount used in a Rowland circle polychromotor (adapted from
Charles et al., 1997)

The sequential spectrometer is designed to move sequentially from one wavelength to

the next during the measuring sequences. It therefore measures the elements in the
sample in pre-defined order. Their main disadvantage is the speed. The most widely
use sequential spectrometer is the Czerny-Turner spectrometers.

Detectors

Once the proper emission line has been isolated by the spectrometer, the detector is
used to measure its intensity. The most used detector for the ICP-OES is the
photomultiplier tube or PMT. The PMT is a vacuum tube that contains a photosensitive
material called the photocathode. The photocathode ejects electrons when it is struck
by light. These ejected electrons are accelerated towards a dynode which ejects two to
five secondary electrons for every one electron which strikes its surface. The secondary
83
 Appendix A

electrons strike another dynode, and the multiplicative effect is caused along the way.
Typical PMTs contain 9 to 16 dynode stages. The final step is the collection of the
secondary electrons from the last dynode by the anode (Charles et al., 1997).

The main advantages of the PMT are that it can be used to measure light over a
relatively wide wavelength range and it can amplify very weak emission levels and its
range of response can be extended over nine orders of magnitude in light intensity
(Charles et al., 1997).

Figure A.6 : A typical PMT (adapted from Charles et al., 1997)

84
 Appendix A

Signal processing and instrument control

Signal processing

Here, the electrical current measured at the anode of the PMT is converted into
information that is used by a computer or analyst (Charles et al., 1997).

Computers and processors

The computer is an important part in any ICP-OES because it controls the function of
the instrument. All parameter of the instrument such as argon flow, generator control
and other parameters are controlled by computer.

85
 Appendix A

A.2. X-RAY FLUORESCENCE (XRF)

An X-ray fluorescence (XRF) spectrometer instrument, model ARL9400 XP using

UniQuant software was used for the analysis of the tantalite.

XRF (X-ray fluorescence spectrometry) is an analytical technique used to identify and

determine chemical elements both quantitatively and qualitatively. It has the advantage
of being non-destructive, multi-elemental, fast, cost effective and easy sample
preparation. The major drawback is that analyses are restricted to elements from
fluorine to uranium in the periodic table (Bennet and Oliver, 1992).

The main principle behind XRF spectrometry is that X-rays of characteristic wavelength
( and energy ) are emitted from a sample when the sample is ionized by a stream of X-
rays; this process is known as X-ray fluorescence, and use of it is made in X-ray
fluorescence analysis (Clive, 1987). The operating principle is illustrated in Figure A.7
below.

Figure A.7. X-ray fluorescence principle

In this method, an atom is excited by primary X-ray which originated either from a
radioisotopic source or an X-ray tube. Subsequently detection of the X-rays emitted by
the sample during the de-excitation process. The X-ray fluorescence spectrum of a
86
 Appendix A

sample reveals a number of characteristic peaks. The energy position of the peaks
identifies the atoms present in the sample (qualitative analysis), whereas the peak
intensity leads to the relative or absolute elemental composition of the sample (semi-
quantitative or quantitative analysis)

Sample preparation is the most important step in any analytical technique. Therefore
selections of the correct sample preparation technique play a major role and depend on
the goals of accuracy and precision the analyst needs to attain. Sample preparation
involves milling the sample in a tungsten carbide milling pot to particle size below 75
µm. The samples are dried at 110˚C (weight 1) and roasted at 1000 ˚C (weight 2), to
determine the percentage loss on ignition (LOI) using the following equation (Laubser
and Verryn, 2009):

weight 1 − weight 2
% LOI = * 100 (a.1)
weight 1 − weigth Crucible

There are two main sample preparation methods: one is based on pressed powder and
the other on the fusion of the sample mixed with a suitable flux, resulting in glass disc.
In the pressed powder pellet method, the X-ray emission intensity can be affected by
mineralogical structure and by the particle size. It is difficult to obtain a homogeneous
sample with this technique and contamination of the surfaces of the pellet can
frequently be observed. In the glass disc (bead) method, the powdered material is
mixed in appropriate ratios with a boric flux, melted, poured into a mould and cooled. A
great advantage of this technique is that reference materials of different natures can be
used for calibration and synthetic standards can be prepared. Other advantages are
due to the physical and chemical homogeneity of the vitreous disc (more reproducible
results), and the possibility of widening the validity ranges of the regression curves and
reducing the number of calibration (Hutton and Elliot, 1980; Eastell and Willis, 1990).
Major elements analyses are done on fused beads methods.

87
 Appendix A

A.3. X-RAY DIFFRACTION (XRD)

Tantalite, tantalum and niobium analysis was performed by an X-ray diffraction (XRD)
analysis using a PANalytical X’Pert PRO powder X-ray diffractometer with Fe filtered
CoK as a radiation source. The phases were identified using X’Pert Highscope plus
software.

X-ray diffraction is a material characterization technique that provides detailed

information about the atomic structure of crystalline substances. It is generally used in
the identification of minerals. A simple way to intuitively understand the complex
phenomenon of XRD is to consider a space plane of atoms in mineral structure Figure
A.8. The d-spacing (distance between a given set of planes), on a scale of Amstroms,
can be determined accurately using XRD. The principles underlying this determination
are expressed by the Bragg equation (Aprill and Richard, 2005).

n = 2d sin (a.2)

Where n is an integer, is wavelength of the radiation, d is d-spacing, and is the angle

between the planes and the incident X-rays beam.

Figure A.8: Scattered beam of X-rays

88
 Appendix A

The Bragg’s equation is derived by setting the path difference between incoming
beams, i.e. DE + EF, equal to an integral number of wavelengths.

Data are collected using X-rays of a known wavelength. The sample is rotated so that
the angle of diffraction changes. When the angle at which the exiting rays will be in
phase for diffraction a signal will be recorded by X-ray diffractrometers and then
converted into peaks. Below is a schematic diagram of a diffractometer (James, 2007).

Glacing angle, 2 Diffraction angle, Aperture angle.

Figure A.9: Schematic diagram of a diffractometer

89
 Appendix A

A.4. SCANNING ELECTRON MICROSCOPY

The scanning electron microscope (SEM) was used for the characterization of niobium
and tantalum pentoxide.

The SEM is one of the most versatile instruments available for the examination and
analysis of the microstructural characteristics of solids objects. The basic components
of the SEM are the lens system, electron gun, electron collector, visual and recording
cathode ray tubes (CRTs), and the electronics associated with them (Joseph et al.,
1981).

The SEM machine works as follows: The heated element causes a beam of electron to
be produced by electron gun from the top of the microscope. The electron beam travels
down toward the sample through electromagnetic lenses which focus and direct them to
the sample surface. When the beams collide with the sample, secondary electrons,
backscattered electrons and X-rays are ejected from the sample. Detectors collects
these electrons ejected and convert them to a signal in the display producing an image.
A block diagram of a scanning electron microscopy is shown on Figure A.10 (Hearle et
al., 1974)

90
 Appendix A

Electron gun

Focusing lenses

Final lens
Magnification Scan
control generator

Signal
amplifier
Specimen

Collector Viewing

tube
Vacuum System

Figure A.10: Block diagram of scanning electron microscopy

91