- Email: [email protected]
Phosphate rock treatment with citric acid for the rapid potentiometric determination of fluoride with ion-selective electrode
Talanta 51 (2000) 993 – 999 www.elsevier.com/locate/talanta
Phosphate rock treatment with citric acid for the rapid potentiometric determination of fluoride with ion-selective electrode Atef O. Al-Othman, Jamal A. Sweileh * Department of Chemistry, Al al-Bayt Uni6ersity, PO Box 130040, Almafraq, Jordan Received 13 August 1999; received in revised form 8 December 1999; accepted 13 December 1999
Abstract A fast method for sample treatment of phosphate rock has been developed for the purpose of quantitative leaching (98–100%) of fluoride but less of the interfering cations such as iron and aluminum. Citric acid (0.5 M) was used to extract fluoride in 15–45 min. Leaching of iron and aluminum is minimal, and these ions are complexed with citric acid. The leaching method was optimized with respect to sample size, citric acid concentration, leaching time and temperature. The analysis was completed by the rapid determination of fluoride with ion-selective electrode. The proposed treatment method was applied to phosphate rock samples from Jordan and Morocco and yielded accurate results as compared to the standard steam distillation from strong acid solution followed by thorium nitrate titration. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Fluoride; Phosphate rock; Citric acid; Potentiometry
1. Introduction The determination of fluoride is prone to several chemical interferences mainly due to joint dissolution of fluoride-complexing cations such as aluminum and iron [1 – 3], which reduce the amount of free fluoride in the sample digestate. Calcium interferes with the determination of fluoride too [4] and so does phosphate when using a spectrophotometric detection method [5]. * Corresponding author. Present address: Department of Chemistry, University of Qatar, PO Box 2713, Doha, Qatar. Tel.: + 974-8-92196; fax: + 974-8-92795. E-mail address: [email protected] (J.A. Sweileh)
Several strategies were adopted to overcome or reduce these interferences which include the addition of masking agents such as ethylene diamine tetra acetate (EDTA) [3,6–8] and related derivatives [6,7,9], sodium citrate [3,4,10–12] and tiron [1,6]. Separation of fluoride from the acid digestate by steam distillation at high temperatures (125–190°C) is an effective way for the removal of these chemical interferences [13,14]. Volatilization of fluoride as trimethyl fluorosilazine by gas diffusion through a permeable membrane is another method, which had been used in the literature [15–18].
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 3 6 1 - 6
994
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
The dissolution of solid samples for fluoride analysis involves either strong acid digestion [3,4,19] or fusion with various fluxes such as sodium carbonate and zinc oxide [7,14], sodium hydroxide pellets [3,4,14,20] and sodium peroxide [21]. The fusion step is usually followed by dissolution with water or dilute acid. All these methods lead to dissolution of aluminum and iron, when present, with the subsequent need for the addition of a suitable masking agent or a dedicated separation step. This work proposes a fast method for the treatment of phosphate rock that is capable of complete dissolution of fluoride but minimal dissolution of iron and aluminum. The method is based on the utilization of citric acid, a weak acid, which completely dissolves fluoride in fluorapatite but with little iron or aluminum. Citric acid acts as an effective complexing agent for masking iron and aluminum [22]; this obviates the need to add auxiliary complexing agents as is customary in other methods. Analysis is completed by the fluoride ion-selective electrode (ISE) after proper ionic strength and pH adjustment of the leachate solution.
2. Experimental
2.1. Chemicals and solution All chemicals used were of reagent grade quality or better. Distilled water was also used for all preparations and 10 000 mg l − 1 stock solutions of Al3 + , Fe3 + , F− were prepared by dissolution of the appropriate quantity of AlCl3 · 6H2O (Fluka) FeCl3 · 6H2O (BDH), and NaF (dried at 110°C for 2 h, Fluka) respectively, in water and completing the volume to 1 l. Working standard solutions were prepared by serial dilution of the above stocks. Citric acid stock solution (2 mol l − 1) was prepared by dissolving 210.14 g of the acid (Fluka) in water and diluting to 500 ml. Other chemicals used were sodium citrate (BDH), hydrochloric acid (Lab scan), nitric acid (Across), sulfuric acid (Across), acetic acid (Across) monochloroacetic acid (BDH), trichloroacetic acid (Fluka), alizarin red-S (Merck), thorium ni-
trate Th(NO3)4 · 4H2O (Fluka). Fluoride-containing solutions were stored in plastic bottles.
2.2. Phosphate rock samples selection A set of Jordanian phosphate rock samples from different mines were selected to represent the different grades. Samples labeled QS1–QS8 were from Ruseifa mine near Amman. Percentage Al2O3 ranges between 0.15 and 4.9% while Fe2O3 ranges between 0.06 and 3.3%; fluoride levels also vary from 1.2 to 4.2%. Low grade ore sample from Shidiyya mine (1.28% F) and a high grade shipment sample labeled E (3.7% F) were also selected. Finally, a certified reference phosphate rock material (BCR No. 32) was also included (4.04% F). All samples are powdered to 100 mesh and thoroughly homogenized.
2.3. Equipment and instruments The ion-selective electrode pH meter (Metrohm model 744) was equipped with model 629 Metrohm ion selective electrode and a matching external reference electrode. Temperature compensation and automatic calibration are some of the common features of this system. Other equipments used include a mechanical flask shaker (Stuart Scientific, model SF-1), a thermostated water bath (Heto, model DT.1) and a fully computerized atomic absorption spectrometer with a deuterium lamp background correction system (Unicam model 929). The single stage steam distillation apparatus [14] was constructed from a 2-l round bottom flask electrically heated steam generator and a 1-l electrically heated round bottom distilling flask with a side arm for vapor transport to a water-cooled double walled condenser. The condensed vapor is collected in a 250-ml volumetric receiving flask. A thermometer is inserted through a hole in the stopper of the distilling flask to monitor and adjust the distillation temperature.
2.4. Analytical procedure Weighing to the nearest 0.0001 g about 0.05 g of the powdered phosphate rock was taken and transferred to a 50 ml disposable plastic tube,
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
then 50 ml of 0.5 mol l − 1 citric acid solution was added and shaken for 45 min at 22 90.5°C (or for 15 min at 6090.5°C). The residue was filtered and washed with water, the filtrate was collected in a 100-ml volumetric flask, the volume was completed to 100 ml with distilled water and shaken thoroughly. In to a 50 ml polypropylene beaker 10.0 ml of the above solution was transferred, and 10.0 ml of sodium citrate buffer solution was added (1 mol l − 1, pH 6.0). Then the fluoride and reference electrodes were dipped in this solution and the stabilized potential difference was read (5 –7 min). If accelerated dissolution (at 60°C) is used, the digestate was filtered for 2 min, 10 ml of the filtrate was taken and diluted to 20 ml by addition of 10 ml of the citrate buffer solution. It was then cooled to room temperature and then read the potential difference as above. For instrument calibration (0.5 – 1000 mg l − 1 F), the appropriate volume of fluoride stock solution was transferred to a 50-ml plastic beaker, and diluted to 10 ml with distilled water and 10 ml of citrate buffer solution was added. Next, the fluoride and reference electrodes were dipped in each beaker and the stable potential difference was read.
Fig. 1. The effect of concentration of leaching acid on the recovery of fluoride from phosphate rock, citric acid; hydrochloric acid; monochloroacetic acid; trichloroacetic acid; × acetic acid.
995
3. Results and discussion
3.1. Selection of the fluoride leaching reagent Several organic acids of different concentrations (0.01–3.0 mol l − 1) and different acid dissociation constants were tried as fluoride-leaching reagents from phosphate rock. About 50 mg aliquots of the rock sample (E) were treated with 50 ml solution of each acid at the indicated concentration. The mixtures were shaken for 60 min and filtered; the filtrate was later collected and diluted to 100 ml and then analyzed for fluoride by ISE as indicated in Section 2.4. Hydrochloric acid was also included in this test as it was used successfully for the determination of fluoride in phosphate rock without a distillation separation step [19]. The results (Fig. 1) indicate that citric acid is superior to other weak acids and to HCl because it can recover fluoride completely from phosphate rock in the minimum acid concentration of 0.25 or 0.5 mol l − 1. A similar test for mineral acids (HCl, HNO3, H2SO4, HClO4 and H3PO4 up to 5 mol l − 1) was carried out and led to the same conclusion. Percentage fluoride recovery ranged between 95.3 for HCl and 79 for H3PO4. The success of citric acid is probably due to the combined effect of lower levels of Al and Fe extraction and the masking effect of citrate on such metal ions so that a higher fraction of F− is free. To confirm the low levels of Al3 + and Fe3 + dissolved by citric acid, triplicate aliquots of several phosphate rock samples (100 mg) were treated separately with 50 ml citric acid solution and with HCl (0.5 mol l − 1). The leachates (100 ml) were analyzed for iron and aluminum by atomic absorption spectroscopy using matching standards for each set under instrumental conditions specified by the manufacturer. A working matching blank was also included in each set. Table 1 shows the average iron and aluminum contents of each leachate. It is evident from Table 1 that HCl dissolves three to five times more iron than citric acid does. Similarly, Al leaching from phosphate rock is four to seven times higher with HCl than with citric acid.
996
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
Table 1 The concentration of Al and Fe (mg l−1) in citric acid and HCl leachates of phosphate rock (average of triplicate tests) Sample identification
Shidiyya Ruseifa QS6 QS7 E (shipment) Blank
c(Al) (mg l−1)
c(Fe) (mg l−1)
HCl
Citric acid
HCl
Citric acid
1.20 0.33 7.5 8.32 0.25 0.10
0.32 0.10 1.55 2.11 0.10 0.10
2.20 0.55 8.50 9.75 0.58 0.25
0.35 0.10 2.00 2.48 0.10 0.20
3.2. Effect of leaching time and temperature Using a flask shaker at constant speed about 50 mg of powdered phosphate rock (E) was agitated with 50 ml of 0.25 mol l − 1 citric acid solution at 229 0.1°C for different time intervals ranging from 5 to 60 min. The filtered leachate was treated and analyzed for fluoride as detailed earlier. Fig. 2 shows the change in percentage recovery of fluoride at different leaching times. At room temperature leaching for 45 min is adequate for complete liberation of fluoride from phosphate rock. Fig. 3 shows the effect of raising the temperature of the water bath from 22 to 75°C at constant agitation time of 15 min. Filtration was performed for 2 min only to avoid further leaching during filtration. Ten milliliters of the filtrate was treated and analyzed for fluoride by the batch mode ISE. It is possible to recover fluoride in 15 min of leaching at 60°C.
crease in the volume of 0.5 mol l − 1 citric acid solution.
3.4. Analytical precision Eight standard fluoride solutions and four phosphate rock samples were analyzed for fluoride (in five replicates) after treatment by the proposed citric acid leaching method and by the standard steam distillation and thorium nitrate titration method (ASTM D 3269). The results are listed in Table 2. It can be concluded from Table 2 that the analytical precision is about equal for both methods of sample treatment. This favors the simple
3.3. Effect of sample size At a fixed citric acid leaching solution conditions (50 ml, 0.5 mol l − 1) increasing the sample weight of the phosphate rock (shipment E) mass from 25 to 500 mg led to a gradual drop in fluoride recovery from 99.7 to 69.7% 92.3 (n= 3), respectively (Fig. 4). Therefore, a 50-mg sample size is ideal for the leaching conditions given in Section 2.3. Increasing citric acid concentration to 2.5 mol l − 1 did not lead to full recovery from sample sizes larger than 50 mg. However, larger samples can be handled by a proportional in-
Fig. 2. The effect of leaching time on the recovery of fluoride from phosphate rock using citric acid solution.
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
997
(n=6) for the determination of fluorine in artificial fluorapatite [23] and 2% RSD for fluoride in waste water [24].
3.5. Analytical accuracy
Fig. 3. The effect of leaching temperature on the recovery of fluoride from phosphate rock after treatment with citric acid solution.
Several phosphate rock samples were treated with citric acid (0.25 mol l − 1) at room temperature and determined by ISE as outlined in Section 2.4. The same samples were also analyzed for fluoride after a single steam distillation from sulfuric acid solution and titration of fluoride in the distillate by thorium nitrate using sodium alzarine sulfonate as indicator [14]. Five replicates of each sample were analyzed and the average concentration was recorded in Table 3. Compared to the ASTM reference method, the recovery of fluoride by the sample leaching with citric acid ranges between 98.8 and 101.6%. This is not very different from the quoted recoveries of the ASTM reference method for aqueous fluoride standards (40–160 mg l − 1) which ranged between 99.2 and 103%. Further evidence for the accuracy of the citric acid leaching method is in the proximity of the obtained fluoride concentration to the certified value for the reference material BCR No. 32, a phosphate rock standard of Moroccan origin. Table 2 Precision of fluoride determination after treatment with citric acid and with the standard steam distillation from sulfuric acid Sample identification
Fig. 4. The effect of phosphate rock sample size on the recovery of fluoride after treatment with citric acid solution for 15 min.
leaching pretreatment over distillation. Precision for aqueous standards is about the same as that of phosphate rock samples. This means that taking a small sized sample (50 mg) does not affect the analytical precision of the proposed method. These precision figures of merit are not excessively high since other workers reported 95.7% RSD
40 mg l−1 80 mg l−1 120 mg l−1 160 mg l−1 E Shidiyya QS6 BCR no. 32 5 mg l−1 10 mg l−1 20 mg l−1 40 mg l−1
RSD (%) (n = 5) Citric acid leaching
Separation by distillation
– – – – 2.15 3.15 2.82 1.75 3.6 3.0 2.4 2.0
3.7 3.0 2.3 2.2 2.5 3.5 3.3 1.9 – – – –
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
998
Table 3 Comparative determination of fluoride in phosphate rock samples after leaching with citric acid followed by ISE determination method and by the standard steam distillation — thorium nitrate titration method (ASTM) Sample identification
Shidiyya QS5 QS6 E(shipment) BCRc 32b a b
w(F) (%)a
Relative difference (%)
Citric acid-ISE
ASTM
1.30 2.49 1.97 3.51 4.09
1.28 2.52 1.99 3.53 4.07
(0.04) (0.08) (0.05) (0.07) (0.06)
(0.04) (0.06) (0.06) (0.08) (0.07)
+1.6 −1.2 −1.0 −0.6 +0.5
The figures in parenthesis are 95% confidence limits. Certified value % F =4.049 0.06.
4. Conclusions
References
The proposed leaching of fluoride from phosphate rock with 0.25 or 0.5 mol l − 1 citric acid is a simple way of sample treatment prior to fluoride ISE analysis. Many samples can be treated simultaneously. This is superior to strong acid dissolution because less iron and aluminum is co-dissolved which in turn obviates the need for a separation step as is customary with other methods. Addition of auxiliary complexing agents for aluminum and iron is unnecessary because citric acid is an excellent masking agent for such elements. Leaching could be accelerated with temperature and complete liberation of fluoride is possible in 15 min at 60°C when applied for analysis of fluoride in phosphate rocks obtained from Jordan and Morocco. Precision and accuracy of the proposed method were comparable to the more elaborate method of steam distillation from strong acid solution and thorium nitrate titration.
[1] D.E. Davey, D.E. Muggleton, G.R. O’Connell, Anal. Lett. 25 (1992) 607. [2] D.E. Davey, D.E. Mulcahy, G.R. O’Connell, Anal. Lett. 19 (1986) 1387. [3] L. Evans, R.D. Hoyle, J.B. Macaskill, New Zealand J. Sci. 13 (1970) 143. [4] W.J. Williams, Handbook of Anion Determination, Butterworths, London, 1979. [5] T.J. Cardwell, R.W. Catterall, M. Mitri, I.C. Hamilton, Anal. Chim. Acta 214 (1988) 433. [6] A. Yuchi, N. Yanai, H. Wada, G. Nakagawa, Analyst 113 (1988) 1405. [7] A. Yuchi, T. Niwa, H. Wada, Analyst 120 (1995) 167. [8] N.T. Crosby, A.L. Dennis, J.G. Stevens, Analyst 93 (1968) 643. [9] J.E. Harwood, Water Res. 3 (1969) 273. [10] M.S. Frant, J.M. Ross, Anal. Chem. 40 (1968) 1169. [11] M.S. Corbillon, M.P. Carril, J.M. Madariaga, I. Uriarte, Analyst 120 (1995) 2227. [12] B. Vickery, M.L. Vickery, Analyst 101 (1976) 445. [13] J.A. Maxwell, Rock and Mineral Analysis, second ed., Chemical Analysis Series, vol. 27, Inter-science, 1968, pp. 211 – 222. [14] Annual Book of ASTM Standards, vol. 11.03, Philadelphia, PA, 1992, pp 125 – 137. [15] J. Tusl, Anal. Chem. 44 (1972) 1693. [16] T.J. Cardwell, R.W. Catterall, M. Mitri, Talanta 41 (1994) 115. [17] I. Papaefstathiou, M.D. Luque de Castro, Anal. Chem. 67 (1995) 3916. [18] S.M.S. Carvalho, A.C.M. Polizello, C.H. Thais, S.A.C. Cesar, Anal. Lett. 30 (1997) 673.
Acknowledgements The Research Center of Jordan Phosphate Mines Company is acknowledged for providing the phosphate rock samples and standards.
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999 [19] C.R. Edmond, Anal. Chem. 41 (1969) 1327. [20] B. Fabbri, F. Donati, Analyst 106 (1981) 1338 (Chem. Abstr. 96 (1982) 96663z). [21] L. Brentrup, W. Koschei Zem-Kalk-Gips (Ed.) B 37(1984) 87 (Chem. Abstr. 100 (1984) 161032k). [22] R.J. Moetkaitis, A.E. Martell, Inorg. Chem. 23 (1984) 18.
.
999
[23] R. Fabian, I. Kotsis, P. Zimany, P. Halmos, Talanta 46 (1998) 1237. [24] S.L. Valdintet, S.P. Keyla, N.J.L. Ricardo, C.E. Cahere, D.T. Cerqueira, M.M.B. Duarte, L. Fernando, Ann. Assoc. Bras. Quim. 45(1996) 106 (Chem. Abstr. 126(1997) 6498L).
Phosphate rock treatment with citric acid for the rapid potentiometric determination of fluoride with ion-selective electrode Atef O. Al-Othman, Jamal A. Sweileh * Department of Chemistry, Al al-Bayt Uni6ersity, PO Box 130040, Almafraq, Jordan Received 13 August 1999; received in revised form 8 December 1999; accepted 13 December 1999
Abstract A fast method for sample treatment of phosphate rock has been developed for the purpose of quantitative leaching (98–100%) of fluoride but less of the interfering cations such as iron and aluminum. Citric acid (0.5 M) was used to extract fluoride in 15–45 min. Leaching of iron and aluminum is minimal, and these ions are complexed with citric acid. The leaching method was optimized with respect to sample size, citric acid concentration, leaching time and temperature. The analysis was completed by the rapid determination of fluoride with ion-selective electrode. The proposed treatment method was applied to phosphate rock samples from Jordan and Morocco and yielded accurate results as compared to the standard steam distillation from strong acid solution followed by thorium nitrate titration. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Fluoride; Phosphate rock; Citric acid; Potentiometry
1. Introduction The determination of fluoride is prone to several chemical interferences mainly due to joint dissolution of fluoride-complexing cations such as aluminum and iron [1 – 3], which reduce the amount of free fluoride in the sample digestate. Calcium interferes with the determination of fluoride too [4] and so does phosphate when using a spectrophotometric detection method [5]. * Corresponding author. Present address: Department of Chemistry, University of Qatar, PO Box 2713, Doha, Qatar. Tel.: + 974-8-92196; fax: + 974-8-92795. E-mail address: [email protected] (J.A. Sweileh)
Several strategies were adopted to overcome or reduce these interferences which include the addition of masking agents such as ethylene diamine tetra acetate (EDTA) [3,6–8] and related derivatives [6,7,9], sodium citrate [3,4,10–12] and tiron [1,6]. Separation of fluoride from the acid digestate by steam distillation at high temperatures (125–190°C) is an effective way for the removal of these chemical interferences [13,14]. Volatilization of fluoride as trimethyl fluorosilazine by gas diffusion through a permeable membrane is another method, which had been used in the literature [15–18].
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 3 6 1 - 6
994
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
The dissolution of solid samples for fluoride analysis involves either strong acid digestion [3,4,19] or fusion with various fluxes such as sodium carbonate and zinc oxide [7,14], sodium hydroxide pellets [3,4,14,20] and sodium peroxide [21]. The fusion step is usually followed by dissolution with water or dilute acid. All these methods lead to dissolution of aluminum and iron, when present, with the subsequent need for the addition of a suitable masking agent or a dedicated separation step. This work proposes a fast method for the treatment of phosphate rock that is capable of complete dissolution of fluoride but minimal dissolution of iron and aluminum. The method is based on the utilization of citric acid, a weak acid, which completely dissolves fluoride in fluorapatite but with little iron or aluminum. Citric acid acts as an effective complexing agent for masking iron and aluminum [22]; this obviates the need to add auxiliary complexing agents as is customary in other methods. Analysis is completed by the fluoride ion-selective electrode (ISE) after proper ionic strength and pH adjustment of the leachate solution.
2. Experimental
2.1. Chemicals and solution All chemicals used were of reagent grade quality or better. Distilled water was also used for all preparations and 10 000 mg l − 1 stock solutions of Al3 + , Fe3 + , F− were prepared by dissolution of the appropriate quantity of AlCl3 · 6H2O (Fluka) FeCl3 · 6H2O (BDH), and NaF (dried at 110°C for 2 h, Fluka) respectively, in water and completing the volume to 1 l. Working standard solutions were prepared by serial dilution of the above stocks. Citric acid stock solution (2 mol l − 1) was prepared by dissolving 210.14 g of the acid (Fluka) in water and diluting to 500 ml. Other chemicals used were sodium citrate (BDH), hydrochloric acid (Lab scan), nitric acid (Across), sulfuric acid (Across), acetic acid (Across) monochloroacetic acid (BDH), trichloroacetic acid (Fluka), alizarin red-S (Merck), thorium ni-
trate Th(NO3)4 · 4H2O (Fluka). Fluoride-containing solutions were stored in plastic bottles.
2.2. Phosphate rock samples selection A set of Jordanian phosphate rock samples from different mines were selected to represent the different grades. Samples labeled QS1–QS8 were from Ruseifa mine near Amman. Percentage Al2O3 ranges between 0.15 and 4.9% while Fe2O3 ranges between 0.06 and 3.3%; fluoride levels also vary from 1.2 to 4.2%. Low grade ore sample from Shidiyya mine (1.28% F) and a high grade shipment sample labeled E (3.7% F) were also selected. Finally, a certified reference phosphate rock material (BCR No. 32) was also included (4.04% F). All samples are powdered to 100 mesh and thoroughly homogenized.
2.3. Equipment and instruments The ion-selective electrode pH meter (Metrohm model 744) was equipped with model 629 Metrohm ion selective electrode and a matching external reference electrode. Temperature compensation and automatic calibration are some of the common features of this system. Other equipments used include a mechanical flask shaker (Stuart Scientific, model SF-1), a thermostated water bath (Heto, model DT.1) and a fully computerized atomic absorption spectrometer with a deuterium lamp background correction system (Unicam model 929). The single stage steam distillation apparatus [14] was constructed from a 2-l round bottom flask electrically heated steam generator and a 1-l electrically heated round bottom distilling flask with a side arm for vapor transport to a water-cooled double walled condenser. The condensed vapor is collected in a 250-ml volumetric receiving flask. A thermometer is inserted through a hole in the stopper of the distilling flask to monitor and adjust the distillation temperature.
2.4. Analytical procedure Weighing to the nearest 0.0001 g about 0.05 g of the powdered phosphate rock was taken and transferred to a 50 ml disposable plastic tube,
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
then 50 ml of 0.5 mol l − 1 citric acid solution was added and shaken for 45 min at 22 90.5°C (or for 15 min at 6090.5°C). The residue was filtered and washed with water, the filtrate was collected in a 100-ml volumetric flask, the volume was completed to 100 ml with distilled water and shaken thoroughly. In to a 50 ml polypropylene beaker 10.0 ml of the above solution was transferred, and 10.0 ml of sodium citrate buffer solution was added (1 mol l − 1, pH 6.0). Then the fluoride and reference electrodes were dipped in this solution and the stabilized potential difference was read (5 –7 min). If accelerated dissolution (at 60°C) is used, the digestate was filtered for 2 min, 10 ml of the filtrate was taken and diluted to 20 ml by addition of 10 ml of the citrate buffer solution. It was then cooled to room temperature and then read the potential difference as above. For instrument calibration (0.5 – 1000 mg l − 1 F), the appropriate volume of fluoride stock solution was transferred to a 50-ml plastic beaker, and diluted to 10 ml with distilled water and 10 ml of citrate buffer solution was added. Next, the fluoride and reference electrodes were dipped in each beaker and the stable potential difference was read.
Fig. 1. The effect of concentration of leaching acid on the recovery of fluoride from phosphate rock, citric acid; hydrochloric acid; monochloroacetic acid; trichloroacetic acid; × acetic acid.
995
3. Results and discussion
3.1. Selection of the fluoride leaching reagent Several organic acids of different concentrations (0.01–3.0 mol l − 1) and different acid dissociation constants were tried as fluoride-leaching reagents from phosphate rock. About 50 mg aliquots of the rock sample (E) were treated with 50 ml solution of each acid at the indicated concentration. The mixtures were shaken for 60 min and filtered; the filtrate was later collected and diluted to 100 ml and then analyzed for fluoride by ISE as indicated in Section 2.4. Hydrochloric acid was also included in this test as it was used successfully for the determination of fluoride in phosphate rock without a distillation separation step [19]. The results (Fig. 1) indicate that citric acid is superior to other weak acids and to HCl because it can recover fluoride completely from phosphate rock in the minimum acid concentration of 0.25 or 0.5 mol l − 1. A similar test for mineral acids (HCl, HNO3, H2SO4, HClO4 and H3PO4 up to 5 mol l − 1) was carried out and led to the same conclusion. Percentage fluoride recovery ranged between 95.3 for HCl and 79 for H3PO4. The success of citric acid is probably due to the combined effect of lower levels of Al and Fe extraction and the masking effect of citrate on such metal ions so that a higher fraction of F− is free. To confirm the low levels of Al3 + and Fe3 + dissolved by citric acid, triplicate aliquots of several phosphate rock samples (100 mg) were treated separately with 50 ml citric acid solution and with HCl (0.5 mol l − 1). The leachates (100 ml) were analyzed for iron and aluminum by atomic absorption spectroscopy using matching standards for each set under instrumental conditions specified by the manufacturer. A working matching blank was also included in each set. Table 1 shows the average iron and aluminum contents of each leachate. It is evident from Table 1 that HCl dissolves three to five times more iron than citric acid does. Similarly, Al leaching from phosphate rock is four to seven times higher with HCl than with citric acid.
996
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
Table 1 The concentration of Al and Fe (mg l−1) in citric acid and HCl leachates of phosphate rock (average of triplicate tests) Sample identification
Shidiyya Ruseifa QS6 QS7 E (shipment) Blank
c(Al) (mg l−1)
c(Fe) (mg l−1)
HCl
Citric acid
HCl
Citric acid
1.20 0.33 7.5 8.32 0.25 0.10
0.32 0.10 1.55 2.11 0.10 0.10
2.20 0.55 8.50 9.75 0.58 0.25
0.35 0.10 2.00 2.48 0.10 0.20
3.2. Effect of leaching time and temperature Using a flask shaker at constant speed about 50 mg of powdered phosphate rock (E) was agitated with 50 ml of 0.25 mol l − 1 citric acid solution at 229 0.1°C for different time intervals ranging from 5 to 60 min. The filtered leachate was treated and analyzed for fluoride as detailed earlier. Fig. 2 shows the change in percentage recovery of fluoride at different leaching times. At room temperature leaching for 45 min is adequate for complete liberation of fluoride from phosphate rock. Fig. 3 shows the effect of raising the temperature of the water bath from 22 to 75°C at constant agitation time of 15 min. Filtration was performed for 2 min only to avoid further leaching during filtration. Ten milliliters of the filtrate was treated and analyzed for fluoride by the batch mode ISE. It is possible to recover fluoride in 15 min of leaching at 60°C.
crease in the volume of 0.5 mol l − 1 citric acid solution.
3.4. Analytical precision Eight standard fluoride solutions and four phosphate rock samples were analyzed for fluoride (in five replicates) after treatment by the proposed citric acid leaching method and by the standard steam distillation and thorium nitrate titration method (ASTM D 3269). The results are listed in Table 2. It can be concluded from Table 2 that the analytical precision is about equal for both methods of sample treatment. This favors the simple
3.3. Effect of sample size At a fixed citric acid leaching solution conditions (50 ml, 0.5 mol l − 1) increasing the sample weight of the phosphate rock (shipment E) mass from 25 to 500 mg led to a gradual drop in fluoride recovery from 99.7 to 69.7% 92.3 (n= 3), respectively (Fig. 4). Therefore, a 50-mg sample size is ideal for the leaching conditions given in Section 2.3. Increasing citric acid concentration to 2.5 mol l − 1 did not lead to full recovery from sample sizes larger than 50 mg. However, larger samples can be handled by a proportional in-
Fig. 2. The effect of leaching time on the recovery of fluoride from phosphate rock using citric acid solution.
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
997
(n=6) for the determination of fluorine in artificial fluorapatite [23] and 2% RSD for fluoride in waste water [24].
3.5. Analytical accuracy
Fig. 3. The effect of leaching temperature on the recovery of fluoride from phosphate rock after treatment with citric acid solution.
Several phosphate rock samples were treated with citric acid (0.25 mol l − 1) at room temperature and determined by ISE as outlined in Section 2.4. The same samples were also analyzed for fluoride after a single steam distillation from sulfuric acid solution and titration of fluoride in the distillate by thorium nitrate using sodium alzarine sulfonate as indicator [14]. Five replicates of each sample were analyzed and the average concentration was recorded in Table 3. Compared to the ASTM reference method, the recovery of fluoride by the sample leaching with citric acid ranges between 98.8 and 101.6%. This is not very different from the quoted recoveries of the ASTM reference method for aqueous fluoride standards (40–160 mg l − 1) which ranged between 99.2 and 103%. Further evidence for the accuracy of the citric acid leaching method is in the proximity of the obtained fluoride concentration to the certified value for the reference material BCR No. 32, a phosphate rock standard of Moroccan origin. Table 2 Precision of fluoride determination after treatment with citric acid and with the standard steam distillation from sulfuric acid Sample identification
Fig. 4. The effect of phosphate rock sample size on the recovery of fluoride after treatment with citric acid solution for 15 min.
leaching pretreatment over distillation. Precision for aqueous standards is about the same as that of phosphate rock samples. This means that taking a small sized sample (50 mg) does not affect the analytical precision of the proposed method. These precision figures of merit are not excessively high since other workers reported 95.7% RSD
40 mg l−1 80 mg l−1 120 mg l−1 160 mg l−1 E Shidiyya QS6 BCR no. 32 5 mg l−1 10 mg l−1 20 mg l−1 40 mg l−1
RSD (%) (n = 5) Citric acid leaching
Separation by distillation
– – – – 2.15 3.15 2.82 1.75 3.6 3.0 2.4 2.0
3.7 3.0 2.3 2.2 2.5 3.5 3.3 1.9 – – – –
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999
998
Table 3 Comparative determination of fluoride in phosphate rock samples after leaching with citric acid followed by ISE determination method and by the standard steam distillation — thorium nitrate titration method (ASTM) Sample identification
Shidiyya QS5 QS6 E(shipment) BCRc 32b a b
w(F) (%)a
Relative difference (%)
Citric acid-ISE
ASTM
1.30 2.49 1.97 3.51 4.09
1.28 2.52 1.99 3.53 4.07
(0.04) (0.08) (0.05) (0.07) (0.06)
(0.04) (0.06) (0.06) (0.08) (0.07)
+1.6 −1.2 −1.0 −0.6 +0.5
The figures in parenthesis are 95% confidence limits. Certified value % F =4.049 0.06.
4. Conclusions
References
The proposed leaching of fluoride from phosphate rock with 0.25 or 0.5 mol l − 1 citric acid is a simple way of sample treatment prior to fluoride ISE analysis. Many samples can be treated simultaneously. This is superior to strong acid dissolution because less iron and aluminum is co-dissolved which in turn obviates the need for a separation step as is customary with other methods. Addition of auxiliary complexing agents for aluminum and iron is unnecessary because citric acid is an excellent masking agent for such elements. Leaching could be accelerated with temperature and complete liberation of fluoride is possible in 15 min at 60°C when applied for analysis of fluoride in phosphate rocks obtained from Jordan and Morocco. Precision and accuracy of the proposed method were comparable to the more elaborate method of steam distillation from strong acid solution and thorium nitrate titration.
[1] D.E. Davey, D.E. Muggleton, G.R. O’Connell, Anal. Lett. 25 (1992) 607. [2] D.E. Davey, D.E. Mulcahy, G.R. O’Connell, Anal. Lett. 19 (1986) 1387. [3] L. Evans, R.D. Hoyle, J.B. Macaskill, New Zealand J. Sci. 13 (1970) 143. [4] W.J. Williams, Handbook of Anion Determination, Butterworths, London, 1979. [5] T.J. Cardwell, R.W. Catterall, M. Mitri, I.C. Hamilton, Anal. Chim. Acta 214 (1988) 433. [6] A. Yuchi, N. Yanai, H. Wada, G. Nakagawa, Analyst 113 (1988) 1405. [7] A. Yuchi, T. Niwa, H. Wada, Analyst 120 (1995) 167. [8] N.T. Crosby, A.L. Dennis, J.G. Stevens, Analyst 93 (1968) 643. [9] J.E. Harwood, Water Res. 3 (1969) 273. [10] M.S. Frant, J.M. Ross, Anal. Chem. 40 (1968) 1169. [11] M.S. Corbillon, M.P. Carril, J.M. Madariaga, I. Uriarte, Analyst 120 (1995) 2227. [12] B. Vickery, M.L. Vickery, Analyst 101 (1976) 445. [13] J.A. Maxwell, Rock and Mineral Analysis, second ed., Chemical Analysis Series, vol. 27, Inter-science, 1968, pp. 211 – 222. [14] Annual Book of ASTM Standards, vol. 11.03, Philadelphia, PA, 1992, pp 125 – 137. [15] J. Tusl, Anal. Chem. 44 (1972) 1693. [16] T.J. Cardwell, R.W. Catterall, M. Mitri, Talanta 41 (1994) 115. [17] I. Papaefstathiou, M.D. Luque de Castro, Anal. Chem. 67 (1995) 3916. [18] S.M.S. Carvalho, A.C.M. Polizello, C.H. Thais, S.A.C. Cesar, Anal. Lett. 30 (1997) 673.
Acknowledgements The Research Center of Jordan Phosphate Mines Company is acknowledged for providing the phosphate rock samples and standards.
A.O. Al-Othman, J.A. Sweileh / Talanta 51 (2000) 993–999 [19] C.R. Edmond, Anal. Chem. 41 (1969) 1327. [20] B. Fabbri, F. Donati, Analyst 106 (1981) 1338 (Chem. Abstr. 96 (1982) 96663z). [21] L. Brentrup, W. Koschei Zem-Kalk-Gips (Ed.) B 37(1984) 87 (Chem. Abstr. 100 (1984) 161032k). [22] R.J. Moetkaitis, A.E. Martell, Inorg. Chem. 23 (1984) 18.
.
999
[23] R. Fabian, I. Kotsis, P. Zimany, P. Halmos, Talanta 46 (1998) 1237. [24] S.L. Valdintet, S.P. Keyla, N.J.L. Ricardo, C.E. Cahere, D.T. Cerqueira, M.M.B. Duarte, L. Fernando, Ann. Assoc. Bras. Quim. 45(1996) 106 (Chem. Abstr. 126(1997) 6498L).