ICP-OES Determination of Ta, Nb, Fe, Ti, Sn, Mn, and W

in Indian Tantalite-Niobate Ore Concentrate

Using Microwave Digestion
G. Anil, M. Ramanaji, *M.R.P. Reddy, Arbind Kumar, T.L. Prakash, and V.C. Sethi a
Centre for Materials for Electronics Technology (C-MET), IDA, Phase-II, HCL Post
Cherlapalli, Hyderabad – 500 051, India
aMaterials and Components Division, Ministry of Information Technology

6 CGO Complex, Lodhi Road, New Delhi - 110 003, India

INTRODUCTION ICP system equipped with a High
ABSTRACT Dynamic Range Detector (HDD)
Niobium and tantalum occur and Image software. The HDD
together in nature in the form of Indian tantalite-niobate ore
concentrate is analyzed for Ta, helps in overcoming serial dilution
mixed oxides, corresponding to the required for the analysis of major
general formula (Fe, Mn) (Ta, Nb, Fe, Ti, Sn, Mn, and W con-
tent using ICP-OES. This paper constituents.
NbO3)2 with widely varying ratios
evaluates two different methods
of Nb to Ta. The elements Fe and EXPERIMENTAL
of dissolution of the concentrated
Mn are also encountered in this ore, namely microwave digestion
family. The minerals are known as and fusion. For microwave diges- Instrumentation
columbite when the predominant tion of the ore samples, a mix- Sample digestion was performed
component is niobium and as tanta- ture of acids (HBF4, HCl, HF, etc.) using a microwave digestion system
lite when tantalum is the more was used for complete dissolu- (MDS) (Model MLS 1200, Milestone,
abundant constituent. For niobium tion. For the fusion technique,
USA). The instrument used for the
alone, pyrochlore (NaCaNb2O6F) is the ore samples were fused with
a mixture of lithium tetraborate determination of major elements in
another important source mineral Indian tantalite-columbite ore was
found in countries like Canada, and lithium metaborate and the
results compared with the an inductively coupled plasma opti-
Brazil, and Norway. Most of the cal emission spectrometer (ICP-
reported values. In the latter
niobium and tantalum metal pro- case, the Sn value showed a 10% OES) (Model JY-24R, Jobin Yvon,
duced today is extracted from decrease. The microwave diges- France). A cross-flow type
columbite-tantalite ore obtained tion method was found to be fast, nebulizer was used for pneumatic
from Africa, Brazil, and Australia. sensitive, accurate, and more nebulization in the ICP-OES instru-
In India, small quantities of precise with a RSD of less than ment. The instrumental parameters
columbite-tantalite mineral occur in 1%. The most suitable analytical
of the ICP-OES and the microwave
pegmatites found in the mica belt lines were selected using the
Image software of the ICP-OES digestion system are given in Tables
of Bihar, Rajasthan, and Madhya I and II. The wavelength selected
Pradesh. A good stockpile of this system used for this study. The
reactivity of ore in various acids for each element to be determined
ore has accumulated mainly as a by- is listed in Table III. A schematic
and acid mixtures was studied
product of the mica mining opera- and the results are presented and diagram of the tetrafluoromethaxil
tions in Bihar. The concentrated discussed. (TFM) vessel used in MDS is shown
ore was procured from AMD, in Figure 1.
Hyderabad, India, and was propane. Rosa A. Conta (10) used a
processed to prepare tantalum Reagents
HBF4, H2SO4 and HClO4 mixture in
powder. Earlier work was done on microwave digestion in three stages A lithium tetraborate and lithium
the analysis of tantalum ore by to achieve complete dissolution of metaborate flux mixture (Spectra
Pickett, C.K. Pillai, and others (1-8) the sample with subsequent analy- flux) was used for fusion. All other
to determine Nb, Ta, and Ti by sis in a 1-meter sequential-simulta- chemicals used were of Suprapur®
flame emission spectrometry (FES) neous ICP system. (Merck) reagent grade. The multi-
and atomic absorption spectrome- elemental standard solutions were
try (AAS). Very little work has been The C-MET Hyderabad Labora- prepared from high-purity metals
done to determine other tory is concentrating mainly on the and salts (Glen Spectra Reference
constituent elements in the ore as development of viable technologies Materials). The distilled deionized
the complete dissolution of the ore in the field of high-purity metals, water used was prepared using the
is not achieved very easily. Russell’s alloys, and semiconductors. Tanta- Nanopure system (Barnstead, USA).
(9) dissolution technique involved lum powder preparation is one of
a number of repetitive chemical the major projects to be undertaken. Sample Preparation by
steps and a non-aqueous separation The present work reports on the Microwave Digestion
technique using octachloro- analysis of major constituents in ore A tantalum ore sample was
such as Ta, Nb, Fe, Sn, Ti, Mn, and ground finely on a Vibratory Disc
*Corresponding author. W by using a 0.64-meter sequential Mill (Retsch, Germany) and the
e-mail: mrpreddy@yahoo.com

Atomic Spectroscopy 32
Vol. 23(1), January/February 2002
 Vol. 23(1), January/February 2002

average particle size was less than TABLE I

25 microns. A 0.2000-g tantalum ICP-OES Instrumental Operating Conditions
ore sample was dissolved in the
Sequential spectrometer Jobin Yvon JY-24R, 0.64 m Czerny-Turner
MDS in a mixture of 1 mL HNO3,
5 mL HBF4 (4HF+H3BO3), 3 mL HCl, Grating Holographic, 3600 grooves mm-1
and 1 mL H2O2. Eight different Spectral range 160-500 nm
sequential steps of microwave heat- Slit width 20 nm for entrance and exits, adjustable
ing with varying power and time
were applied in order to control the HDD Detector Dynamic range 2.5 x 1010
reaction of the sample during disso- Image Measures complete emission spectrum
lution (Table II). Into four TFM ves- in 2 min (230,000 points)
sels, a 0.2000-g tantalum ore sample
was placed and 10 mL HCl, 10 mL Plasma torch assembly Fused quartz with capillary injection
HNO3, 10 mL aqua regia, or 10 mL Nebulizer and spray chamber Cross-flow type nebulizer and Teflon®
HF was added to check the reactiv- dual-tube spray chamber
ity of each sample with the respec- Argon coolant gas flow rate 1.5 L min-1
tive acid. After completing the Argon plasma gas flow rate 12 L min-1
digestion, the solution was made up
to 100 mL volume before aspiration TABLE II
into the plasma. Operating Conditions of Microwave Digestion System
The calibration curves for each Frequency 2450 MHz, (equivalent to a wavelength of 12.25 cm)
analysis consisted of four standard Power supply 220 Volt/50 Hz–2.2 kW
solutions of appropriate concentra- Exhaust flow 5 m3/min
tions. The standard solutions were Digestion Program
matrix-matched by adding the ana-
lytes. Blank solutions were Step No. Power Time
prepared in a similar way but with- (Watts) (min)
out the analytes. 1 200 5
Fusion Technique 2 250 5
3 300 4
A 0.2000-g tantalum ore sample
(<25 microns) was fused with 4 400 2
1.500 g of Spectra flux (lithium 5 150 5
tetraborate and lithium metaborate 6 100 5
mixture) in a platinum crucible and 7 50 5
kept in the muffle furnace (Carbo-
8 0 5

TABLE III
Wavelength Selected for the
Determination of Each Element
No. Elements Wavelength
(nm)
1. Tantalum 263.558
2. Niobium 295.088
3. Titanium 334.941
4. Iron 239.562
5. Manganese 257.610
6. Tin 242.170
7. Tungsten 207.921

Fig. 1. TFM sample container vessel used in the microwave digestion system.

33
lite, USA) for three hours at 950ºC. TABLE IV
The cooled melt was dissolved in a Comparison of MDS and Fusion Technique
Teflon® beaker with 30 mL of 20% (Results in %)
(v/v) HCl. It should be noted that No. Elements Reported valuesa MDSb Spectra fluxb
some precipitate still remained in
the solution. The supernatent 1 Niobium 17.48 17.41± 0.09 17.21±0.12
solution was decanted and 10 mL of 2 Tantalum 33.16 32.98±0.07 31.12±0.10
10% (v/v) HF and 5 mL H2O2 (50 3 Titanium 2.70 2.62±0.02 2.49±0.09
volumes) were added to achieve 4 Iron 5.89 5.77±0.02 5.63±0.08
complete dissolution; excess HF 5 Tungsten 0.91 0.86±0.032 0.77±0.04
was removed by boiling the solu- 6 Tin 7.79 7.67±0.04 6.83±0.18
tion. The two solutions were mixed 7 Manganese 4.77 4.70±0.04 4.61±0.12
and made up to 100 mL volume
a
before aspiration into the plasma. Values reported by AMD, Hyderabad, India.
b
The blank was prepared in a similar ± SD of 5 readings.
way but without the analytes. TABLE V
Results Obtained with Different Acids in Microwave Digestion
RESULTS AND DISCUSSION (Results in %)
a
Sample Elements HCl HNO3a Aqua HFa Reported
The reaction speed and No. regiaa values
efficiency of the sample decomposi-
tion can be improved with an 1 Niobium 0.064±0.04 0.064±0.02 0.0160±0.05 12.80±0.12 17.48
increase in temperature. As a rule 2 Tantalum 0.040±0.03 0.007±0.03 0.020±0.04 26.46±0.18 33.16
of thumb, a temperature increase of 3 Titanium 0.064±0.05 0.007±0.03 0.042±0.03 2.21±0.08 2.70
10ºC doubles the reaction speed. In 4 Iron 0.456±0.03 0.176±0.03 0.586±0.03 4.60±0.10 5.89
the pressure-controlled operation,
5 Tungsten 0.003±0.05 0.005±0.02 0.004±0.06 0.59±0.15 0.91
the temperature in the vessel is
dependent on the reagent mixture 6 Tin 0.972±0.06 0.095±0.04 0.983±0.09 4.26±0.18 7.79
and on the amount of gaseous prod- 7 Manganese 0.825±0.05 0.412±0.05 0.874±0.04 3.60±0.14 4.77
ucts formed during the decomposi- a ± SD of 5 readings.
tion process. It is beneficial to run
the digestions at the highest tem- columbite ore, except for Sn using Analytical Results for Ore in Dif-
perature allowed by the vessel the fusion method, agreed within ferent Acids and Acid Mixtures
design. 2% of the reported values. The The goal of this work was to
microwave digestion method was establish the reactivity of ore in
Comparison of Analytical found to be very accurate, showed
Results different acids in order to
good reproducibility, and required concentrate/purify the ore before
In order to evaluate the accuracy limited time for the complete disso- dissolution in HF medium. The
of the developed analytical method, lution of the samples. The analytical acids used were added starting with
a matrix standard was prepared and results are closely related to the HNO3, HCl, aqua regia, and ending
the tantalite-columbite ore analyzed reported values. The concentration with HF. Special care was taken to
using the fusion technique and of each element in the blank solu- rinse the sample particle from the
microwave dissolution. The wave- tion throughout all analytical proce- reaction vessel walls of the
length for each element was identi- dures was less than the ICP-OES microwave digestion system. Due
fied using the Image Software of the detection limits. to partial dissolution, repeated and
ICP-OES instrument, which scans careful washings of the filtrate were
the entire spectrum from 180 nm to The method developed was
verified by using two certified made prior to making up the sam-
495 nm in two minutes. The analyti- ple in a standard flask after filtra-
cal results of the sample, dissolved standards, 206GB34 and 206 A
TAN1, from MBH Analytical Ltd., tion. The results obtained with
by microwave digestion and by different acids are given in Table V.
fusion, are given in Table IV. The UK, and the results are presented in
Table VI. It can been seen that the Except for HF, no other acids or
analytical values and reproducibility acid mixtures showed any apprecia-
are represented by the standard experimental values are very close
to the certified values, indicating ble reactivity with the ore.
deviation from the measurements
of five readings for each sample. that the microwave digestion proce-
The analytical results of tantalite- dure can be adopted for the analysis
of tantalite ore.

34
 Vol. 23(1), January/February 2002

TABLE VI
Analysis of Certified Standards
(Results in %)
Sample No. Elements 206GB34 MDS Fusion 206 A TAN1 MDS Fusion
(certified values) (certified values)
1 Tantalum 40.88 40.70±0.08 38.60±0.12 0.236 0.232±0.07 0.205±0.18
2 Niobium 18.95 18.85±0.04 17.20±0.10 0.02 0.021±0.04 0.018±0.20
3 Titanium 0.74 0.71±0.02 0.65±0.09 NR NA NA
4 Iron NR NA NA 0.20 0.19±0.12 0.16±0.15
5 Manganese NR NA NA 0.02 0.018±0.15 0.017±0.13
6 Tin NR NA NA 0.01 0.0098±0.16 0.0065±0.28
NR = Not reported.
NA = Not analyzed.
* ± SD of 5 readings.

CONCLUSION ACKNOWLEDGMENTS REFERENCES

The microwave digestion proce- The authors would like to thank 1. E.E. Pickett and S.R. Kirttyohann,
dures generally used for tantalite- the Ministry of Information Tech- Anal. Chem. 41(14), 28A (1969).
niobate ore require a number of nology (MIT), Government of India, 2. V.S. Sastri, C.L.Chakrabarti, and D.E.
stages due to the refractory nature New Delhi, for sponsoring the Tan- Willis, Talanta 16,1093 (1969) .
of the ore and the difficulty in dis- talum Technology Mission Project. 3. A.M. Bond, Anal. Chem. 42, 932
solving all constituents in a single The authors are also thankful to Dr. (1970).
step. The present microwave diges- B.K. Das, Executive Director, C-
4. P.E. Thomas and W.F. Pickering,
tion procedure developed for the MET, Dr. S. L. Sarnot, Director Gen- Talanta 18, 127 (1971).
determination of Ta, Nb, Fe, Ti, Sn, eral, STQC, MIT, and Dr. K.S.K. Sai,
Mn, and W in tantalite-niobate ore Senior Director, MIT, for their 5. J. Husler, Talanta 19, 863 (1973).
is a single-step method using a mix- encouragement. 6. Z.A. deBenzo, M.H. deBierman, C.M.
ture of acids (HBF4 + HCl + HF). Ceccarelli, and J.J. LaBrecque, At.
This method is rapid, accurate, and Spectrosc. 5(3), 83 (1984).
superior to the fusion techniques 7. C.K. Pillai, S. Natrajan, and Ch. Ven-
employed. The results of the keteshwarulu, At. Spectrosc. 6 (2),
microwave digestion procedure are 53 (1985).
very close to the certified values 8. K. Satyanarayana and M.A. Nayeem,
with a variation of less than 1% At. Spectrosc. 14(6), 180 (1993).
RSD. This method can be success- 9. H. Russel Atkinson, Joseph Stefgman,
fully applied for the analysis of tan- and C.F. Hiskey, Anal. Chem.
talite-niobate ore. 24(3), 477 (1952).
10. A. Rosa Conte, Jean-Michel Mermet,
Received September 25, 2001. Jose De Anchieta Rodrigues, and
Jean-Louis Di Martino, JAAS 12,
1215 (1997) .

35