Determination of phosphorous in titanium bearing minerals by potentiometric titration using Pb-ion selective electrode

Talanta 51 (2000) 57 – 62 www.elsevier.com/locate/talanta

Determination of phosphorous in titanium bearing minerals by potentiometric titration using Pb-ion selective electrode K. Ramadoss *, D.S.R. Murty, P.L. Mahanta, B. Gomathy, R. Rangaswamy Atomic Minerals Directorate for Exploration and Research, Department of Atomic Energy, Nagarbha6i, Bangalore 560 072, India Received 2 April 1999; received in revised form 29 July 1999; accepted 4 August 1999

Abstract A method for phosphorous determination in titanium bearing minerals by potentiometric titration using a Pb-ion selective electrode has been developed. Sample decomposition is achieved by means of K2CO3 fusion in a platinum crucible at 800°C for 30 min in a muffle furnace, and subsequent leaching with water of the fused melt. The aqueous leachate is neutralised with HClO4 and subsequent boiling. The obtained solution is used for titration with Pb(ClO4)2, and the Pb-ion selective electrode detects the end point. The lowest concentration determinable is 0.02% P2O5 in a solid sample. The method was applied on in-house titanium bearing mineral samples and on IGS-31 ilmenite sample (British Geological Survey, UK). Synthetic samples were prepared and analysed, and phosphorous recovery is in the range 98–106%. The recovery and accuracy of the present method have been validated by spiking experiments and by comparing with the spectrophotometric values, respectively. The precision of the proposed method in terms of relative standard deviation is 2.0%. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Titanium bearing minerals; Potassium carbonate fusion; Phosphorous; Pb-ion selective electrode; Potentiometric titration

1. Introduction Titanium bearing minerals are much used for the manufacture of pigment in the paint industry around the world. Ilmenite, rutile, leucoxene, titania slag and beneficiation/upgraded ilmenite known as synthetic rutile are the main titanium bearing minerals. Rutile and synthetic rutile commands a higher price in the world market. Titanium IV oxide pigment, which is an essential component in all quality paints, is manufactured * Corresponding author. Fax: +91-80-3211511.

from these minerals. Titanium IV oxide pigments have excellent covering power, refractive index, high durability and non-toxic properties, making it superior to all other known white pigments. It is used in papers, plastics, textile and ceramics. Titanium metal is used in aerospace industries due to their strength, light weight and corrosion resistance. It is used in heat exchange tubes, cardiac pace makers, deep-sea vessels, bullet-proof vests, etc. Rutile is an important ingredient in welding electrodes. Impurities that are generally found in titanium minerals comprise V2O5, P2O5, Cr2O3, Al2O3, SiO2

0039-9140/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 3 9 - 9 1 4 0 ( 9 9 ) 0 0 2 4 7 - 7

58

K. Ramadoss et al. / Talanta 51 (2000) 57–62

and ZrO2. Placer deposits are the major source for titanium minerals and these deposits are always associated with monazite, sillimanite, garnet, zircon, etc. A fraction of these minerals associated with titanium minerals and isomorphic substitutions of elements are responsible for the impurity. A rapid method for chemical characterisation of ilmenites using inductively coupled plasma-Auger electron spectroscopy (ICP-AES) was published earlier from this laboratory [1]. The presence of V2O5 and Cr2O3 have negative impact on whiteness of paint pigments [2]. In the process of making synthetic rutile by the sulphate route [3] (Stanaway Process), more than 0.05% P2O5 is not permissible in the raw materials used. However, in the end product, synthetic rutile, the maximum tolerance limit of P2O5 is fixed as 0.9% P2O5 [2]. Although the chloride process for making synthetic rutile is eco-friendly, still many industries follow the sulphate route [2]. Phosphorous is also responsible for colour aging in titanium IV oxide pigments [4]. An accurate and reliable analytical impurity control is indispensable. The authors have not come across any certified value for P2O5 reported in the literature for standard reference ilmenite samples [5– 7]. P2O5, which is found in titanium minerals, is a component that must be rigourously quantified for a better price. The most widely used method in spectrophotometric analysis is either by molybdovanadophosphoric acid complex formation [8] or by means of molybdenum blue [9]. Titanium is a major component in titanium minerals and it interferes in the yellow molybdovanadophosphoric acid complex [10]. Therefore, determination of phosphorous in the presence of high titanium is not possible by the molybdovanadophosphoric acid method. Direct phosphorous determination by atomic absorption is not sensitive [11]. Despite the indirect method of phosphorous determination in rocks through yellow molybdophosphoric acid complex formation, organic solvent extraction, stripping to aqueous phase and determination of molybdenum by atomic absorption spectrometry is also reported [12]. This method is tedious, time consuming and not suitable in high titanium minerals. Titanium and or iron cause spectral interference

in the phosphorous determination by ICP-AES [13]. Lead forms one of the most insoluble orthophosphates known. Therefore, a potentiometric titration of phosphorous with Pb(ClO4)2 using a Pb-ion selective electrode is investigated for the determination of phosphorous in titanium-rich minerals. These findings are reported in this paper.

2. Experimental

2.1. Equipment and reagents The equipment used included a Digital Elico pH meter (LM-121), an Orion ion analyser (Model 407A), a lead-ion selective electrode (Orion Model 94-82), a reference electrode, double junction (Orion Model 90-02), and an outer chamber (salt bridge) filled with 1 M sodium nitrate (NaNO3) solution. Lead perchlorate Pb(ClO4)2 (Fluka, Swiss), 0.005 and 0.001 M, was used, with adjusted pH 4.8–5.0 with dilute perchloric acid. Potassium dihydrogen phosphate (KH2PO4) was of G.R. Grade (Merck). The buffer solution was 6 g boric acid (H3BO3) (Merck) dissolved in 500 ml distilled water, and pH adjusted to 8.6 with 10% sodium hydroxide (NaOH).

2.2. Procedure A 0.5000 g portion of the sample dried at 100°C for 2 h was thoroughly mixed with 4 g K2CO3 in a platinum crucible and fused in a muffle furnace at 800°C for 30 min. The melt was allowed to cool and the crucible transferred to a 250 ml beaker with 50 ml hot water. The beaker was heated until the melt detached from the crucible. The crucible was removed after washing. The melt was made to disintegrate, heated for 20 min and subsequently filtered through Whatman 40 filter paper. Three hot-water washings were given to the residue. The filtrate was cooled, concentrated perchloric acid was slowly added with constant stirring until pH 4 was reached, and the mixture

K. Ramadoss et al. / Talanta 51 (2000) 57–62

kept aside for 30 min. Potassium perchlorate precipitate thus formed was filtered through a What-man 541 filter paper, and four cold distilled-water washings of precipitate were given. The filtrate was boiled for 10 min and made up to 100 ml volume after cooling. A 10 ml aliquot (10 – 50 mg phosphorous) was taken in a 250 ml beaker, pH adjusted to 8.6 by 10% NaOH solution, 4 ml borate buffer was added and levelled to  150 ml volume

59

with distilled water. Titrant 0.001 M Pb(ClO4)2 was added and the millivolt change was recorded for each addition. Near the end point, smaller increments of the titrant were added. The end point was determined from the titration curve (Fig. 1). The titrant was standardised against a phosphate solution of known purity (KH2PO4) containing about 0.1 mg P/ml. The titre of 0.01 M lead perchlorate is 0.206 mg of P.

Fig. 1. Potentiometric titration of phosphorous (as ortho phosphate) with 0.001 m lead perchlorate.

K. Ramadoss et al. / Talanta 51 (2000) 57–62

60

Table 1 Major elemental analysis of titanium bearing minerals Sample number

Sample code

%TiO2

%Fe2O3

%FeO

%Cr2O3

%V2O5

1 2 3 4 5 6 7 8 9 10

AMDE-1 AMDE-2 AMDE-3 AMDW-4 AMDW-5 AMDW-6 AMDW-7 AMDW-8 AMDW-9 AMDW-10

38.39 39.49 46.44 63.03 61.72 62.23 55.97 55.43 56.33 56.12

36.17 32.89 19.96 25.81 16.86 20.48 16.03 16.27 14.71 7.69

17.96 18.68 28.74 6.89 16.65 13.66 26.24 26.49 27.28 34.50

0.12 0.11 0.06 0.43 0.20 0.19 0.02 0.03 0.04 0.06

0.29 0.27 B0.01 0.19 0.25 0.17 0.15 0.12 0.13 0.12

3. Results and discussion Titanium bearing minerals are generally brought to dissolution by KHSO4 [14], KF + K2S2O7 [15], NaF + H3BO3 [16], Na2O2 [17], Na2CO3 +NaBO2 [18], and H3PO4 heating [1]. The sample solution thus obtained is suitable for the analysis of Ti, Fe and other trace elements. Phosphorous from this solution cannot be analysed because titanium interferes in the spectrophotometric method of vanado molybdophosphoric acid [10] and the titanium/iron emission lines interference in the ICP-AES method [13]. The removal of titanium is essential before the determination of phosphorous. K2CO3 fusion and water leach is followed in the present method to remove the interfering titanium and iron. Neutralisation with HClO4 and boiling removes H2CO3, CO23 − and HCO− 3 and also the potassium salt as KClO4 precipitate. Thus, the prepared sample solution is free from major matrix (Ti and Fe) and K2CO3. The filtrate on analysis showed few milligrams titanium per litre, and traces of silicon. Chromium and iron are not detected in the filtrate. In the present method, Pb(ClO4)2 titration of phosphorous using the Pb-ion selective electrode is proposed. The Pb-ion selective electrode was used for the determination of sulphate [20] and oxalate [21] by potentiometric titration in p-dioxane solution. Orthophosphate forms one of the most insoluble lead salts [22] (−log KSP =42.1 at 25°C) and therefore can be titrated with lead perchlorate in

aqueous solution [21]. Since the insoluble phosphates are readily soluble in acids and basic lead salts precipitate in very basic solution (pH\10), controlling pH by buffering is necessary. A borate buffer adjusted to pH 8.6 has been found satisfactory. At this pH, the normal orthophosphate of lead Pb3(PO4)2 precipitates. At lower pH, formation of Pb(H2PO4)2 and PbHPO4 species are possible [23]. Pb3(PO4)2 precipitates only at high dilution [23], and in more concentrated solution, mixed salts may form and hence the titrated solution volume is maintained to 150 ml. The flux (K2CO3) introduced for fusion is also removed as KClO4 from the solution. Hence, ionic strength of the solution is lowered. Interference from carbonate and bicarbonate are eliminated by acidification using HClO4 to pH 4. Chromium [2] generally is in low amount in Indian titanium bearing minerals. In samples containing high chromium, PbCrO4 precipitates and gives positive interference. A few drops of 1% Na2SO3 reduces up to 1000 mg/l chromium to a lower valency state, thereby avoiding PbCrO4 precipitation. In order to validate the recovery of P by the present method, 10 titanium bearing minerals of varied titanium content from 38 to 63% TiO2 are taken for percent recovery studies. The major elemental compositions of these samples are given in Table 1. These samples are spiked with 200 mg phosphorous at fusion stage and the phosphorous recovery values are determined by the present method. As shown in Table 2, the percent recov-

K. Ramadoss et al. / Talanta 51 (2000) 57–62

61

Table 2 Percent recovery data of phosphorous by the present method (n = 5)a Sample number

1 2 3 4 5 6 7 8 9 10 11 12

Sample code

P (mg/g) by the present method

AMDE-1 AMDE-2 AMDE-3 AMDW-4 AMDW-5 AMDW-6 AMDW-7 AMDW-8 AMDW-9 AMDW-10 SYN-1 SYN-2

Before spiking

After spiking

305 520 915 1820 1235 1515 1810 1730 1905 2115 – –

510 725 1110 2015 1425 1718 2014 1930 2100 2313 528 558

% Recovery

102 102 98 98 98 101 102 100 98 99 100 106

Samples 1–10, 200 mg P spiked in all the samples; sample 11, 45% TiO2, 50% Fe2O3, 0.5% SiO2, 0.053% P (KH2PO4) (528 mg/g P); sample 12, 60% TiO2, 35% Fe2O3, 0.5% SiO2, 0.053% P (KH2PO4) (528 mg/g P). a

Table 3 Comparison of the P values by the present potentiometric method and spectrophotometric method (n = 5) Sample number

Sample code

IGS-31a AMD-1 AMD-2 AMD-3 AMD-4 AMD-5

1 2 3 4 5 6 a

P (mg/g) Potentiometric method

Spectrophotometric method [8,19]

440 570 1308 570 2305 1718

420 550 1310 570 2260 1740

Ilmenite, British Geological Survey (Keyworth, Nottingham, UK).

ery varies from 98 to 106%. Table 2 also gives the recovery studies on two synthetic samples. Since no ilmenite standard reference material with known phosphorous values is available, the accuracy of the method was tested by comparing the values with those obtained by spectrophotometry [8,19]. A very good agreement in the two values, as given in Table 3, is observed. The detection limit of spectrophotometric method is 10 mg phosphorous [10], whereas the same for the present method is 4 mg phosphorous with a precision of 2%. Added to this low detection limit, the present method is an absolute method of estimation of phosphorous, whereas the spectrophotometric method is a relative one.

Acknowledgements The authors are highly thankful to Sri D.C. Banerjee, Director, AMD, Sri B.M. Swarnkar, Regional Director and Dr R.K. Malhotra, Head, Chemistry Group, AMD, for their encouragement.

References [1] D.S.R. Murty, B. Gomathy, R. Bose, R. Rangaswamy, Atom. Spectrosc. 19 (1998) 14. [2] M.S. Nagar, Beach Sand Mineral Industry in India, Indian Mining — 1995 Annual Review, 1995. [3] D.C. Banerjee, Ali Mir Azam, K.K. Dwivedi, Placer

62

[4] [5]

[6] [7]

[8] [9] [10]

[11]

[12]

K. Ramadoss et al. / Talanta 51 (2000) 57–62 minerals in the economy of India, Proceedings of Chemical Analytical Techniques, March 1996, AMD, Hyderabad, India. N. Majeen, F.X. Rius, J. Zupan, Anal. Chem. Acta 348 (1997) 87. W.A. Deer, R.A. Howie, J. Zussman, Non-Silicate. In: Rock Forming Minerals, vol. Vol. 5, Longman, London, 1975 pp. 29, 41. K. Govindaraju, Geostandards newsletter, Geostandards Special Issue XVIII (1994). P.J. Potts, A.G. Tindle, P.C. Webb, Geochemical Reference Material Composition, Whittle Publishing, CRC Press, USA, 1992. H. Baadsgaard, E.B. Sandell, Anal. Chim. Acta 11 (1954) 183. J.V. Sala, V. Harmandis, A. Canals, Analyst (London) 111 (1986) 965. A.K. Babko, A.T. Philipenko, Photometric Analysis Methods of Determining Non-metals, MIR Publications, Moscow, 1976 pp. 88, 89 (English translation). Asha Varma, C.R.C. Hand book of Atomic absorption analysis, Vol.II, 4th ed., CRC Press, Inc, Boca Raton, FL, 1987 p. 311. C. Riddle, A. Turek, Anal. Chim. Acta 92 (1977) 49.

.

[13] R.K. Winge, V.A. Fassel, V.J. Peterson, M.A. Floyd, Inductively coupled plasma – atomic emission spectroscopy, Phys. Sci. Data 20 (1985) 466. [14] I.L. Marr, A Hand Book of Decomposition Methods in Analytical Chemistry, International Text Book Co, Glasgow, U.K, 1979, p. 80. [15] T.R. Ball, G.M.C.P. Smith, J.Am. Chem. Soc. 36 (1914) 1838. [16] V.S. Biskupsky, Chem. Anal. 56 (1967) 49. [17] A.J. Mukhedar, S.B. Kulkarni, Z. Anal. Chem. 207 (1965) 421. [18] S. Sahasranaman, V.G. Kulkarni, National Symposium on Analytical Application in Earth Sciences, Shillong, India, November 1988. [19] N.H. Furman, Standard Method of Chemical Analysis, Vol. I, 6th ed., Robert E. Krieger Publishing Company, New York, 1975 p. 116. [20] J.W. Ross Jr, M.S. Frant, Anal. Chem. 41 (1969) 967. [21] W. Selig, Mikrochim. Acta (Wein) (1970) 168. [22] L.G. Sillen, Stability Constants of Metal – Ion Complexes, Section I: Inorganic Ligands, 2nd ed., The Chemical Society, London, 1964. [23] W. Selig, Mikrochim. Acta (Wein) (1970) 564.