- Email: [email protected]
Simple and selective spectrophotometric determination of ruthenium after solid phase extraction with some quinoxaline dyes into microcrystalline p-dichlorobenzene
Spectrochimica Acta Part A 58 (2002) 1831– 1837 www.elsevier.com/locate/saa
Simple and selective spectrophotometric determination of ruthenium after solid phase extraction with some quinoxaline dyes into microcrystalline p-dichlorobenzene Alaa S. Amin Chemistry Department, Faculty of Science, Benha Uni6ersity, Benha, Egypt Received 30 May 2001; accepted 10 July 2001
Abstract A simple selective and highly sensitive extraction method has been developed for the determination of ruthenium spectrophotometrically after extraction of its 2,3-dichloro-6-(3-carboxy-2-hydroxy-1-naphthylazo)quinoxaline (I), 2,3-dichloro-6-(2-hydroxy-3,5-dinitrophenylazo)quinoxaline (II) and 2,3-dichloro-6-(2,7-dihydroxy-naphthylazo)quinoxaline (III) complexes into microcrystalline p-dichlorobenzene. The optimization of experimental conditions for the procedure is studied. The solid p-dichlorobenzene containing the ruthenium-reagent (I – III) complexes is separated by filtration and dissolved in N,N-dimethylformamide. The absorbance is measured at umax 622, 518 and 542 nm against reagents I, II and III, respectively, as blank. Beer’s law is obeyed upto 2.5 mg ml − 1 of ruthenium. The molar absorptivity, Sandell sensitivity, detection and quantification limits are calculated, when compared with those parameters without using solid phase extraction method. The interference of various ions has been studied in detail and the statistical evaluation of the experimental results is reported. The proposed methods have been successfully applied for the determination of trace amount of ruthenium in seawater, ore and metallurgy products. © 2002 Published by Elsevier Science B.V. Keywords: Ruthenium determination; Quinoxaline azo dyes; Spectrophotometry; Trace analysis
1. Introduction Interest in the development of analytical techniques for determination of the noble metals is still growing as a result of rapid growth of their applications, e.g. in chemical engineering, microelectronics and medicine. Compared to most elements, ruthenium has a limited influence on the biosphere because only small quantities ever reach living organisms [1]. The amount of ruthenium readily introduced into rivers, lakes and oceans
through industrial wastes is minute [1]. Thus, sensitive, reliable and practicable methods are required for the quantitation of ruthenium at trace levels. The most common reported spectrophotometric procedures for ruthenium determination require laborious enrichment steps [2–6], e.g. flotation, precipitation and solvent extraction. However, most of these methods suffer from lack of selectivity and sensitivity [3–6]. A variety of chromogenic reagents have served as the basis for the spectrophotometric determina-
1386-1425/02/$ - see front matter © 2002 Published by Elsevier Science B.V. PII: S 1 3 8 6 - 1 4 2 5 ( 0 1 ) 0 0 6 8 1 - 3
1832
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
tion of ruthenium. Recent reference works [6– 13] list more reagents for ruthenium that is steadily increasing because of its commercial importance. The various methods differ considerably in sensitivity, tolerance to other ions, rate of reaction, useful concentration range and availability of the reagents, and there appears to be scope for the development of further procedures with readily available chromogenic reagents that give good sensitivity. Recently, certain quinoxaline azo dyes have been prepared and evaluated as high sensitive chromogenic reagents for the spectrophotometric determination of gold [14], scandium [15], mercury [16] and zinc [17]. In the present study, consideration was given to the possible reaction of ruthenium with these chromogens in the light of its position in the periodic table and the analytical resemblance of certain compounds of ruthenium to those of other studied metal ions [14– 17]. The behaviour of 2,3-dichloro-6-(3-carboxy-2hydroxy-1-naphthylazo)quinoxaline (I), 2,3-dichloro - 6 - (2 - hydroxy - 3,5 - dinitrophenylazo)quinoxaline (II) and 2,3-dichloro-6-(2,7-dihydroxynaphthylazo)quinoxaline (III) with ruthenium has now been studied and their potential as analytical reagents has been assessed. The present solid phase extraction method is more sensitive and simpler. The main advantages of this method are that the equilibrium distribution between the two phases is attained in a few seconds owing to the high temperature used, and the metal chelates are dissolved merely by contact with the molten phase. As the organic phase is solid at room temperature, phase separation is quantitative and can be achieved simply by filtration or decantation. The method is selective and simple, and the complexes are stable for a long time and much better than the other known reagents.
2. Experimental
2.1. Apparatus A Perkin –Elmer l3B double beam spectrophotometer was used with 10 mm matched quartz
cuvettes for recording absorption spectra. An Orion research model 601A/digital ionalyser pHmeter equipped with a combined glass-calomel electrode was used to check the pH of the prepared buffer solutions.
2.2. Reagents Doubly distilled water and analytical-reagent grade chemicals were used throughout, unless stated otherwise. Standard ruthenium solution. A standard solution of ruthenium(III) was prepared by dissolving ruthenium(III) chloride (Johnson Matthey, London) in doubly distilled water, containing sufficient hydrochloric acid to give a final concentration of 1.0 M HCl. Since these solutions have a tendency to turn dark brown on keeping due to oxidation, the stock solution was obtained by refluxing for 4.0 h with HCl and ethanol. The ruthenium solution was standardized gravimetrically by precipitating ruthenium as hydrated oxide, followed by careful ignition in air and then reduction to the metal in the presence of hydrogen gas and cooling in an atmosphere of carbon dioxide gas. Subsequent dilutions were made from the stock solution according to requirements. Quinoxaline dyes preparation. Reagents I–III were prepared according to the previously recommended procedures [16]. A 3× 10 − 3 M solution of the reagents was prepared by dissolving the appropriate weight of each reagent in 10 ml ethanol and then completed to the mark in 100 ml calibrated flask. Sodium acetate–acetic acid buffer. Sodium acetate solution (0.2 M) and acetic acid were mixed in suitable proportions to prepare buffer solutions of pH ranges 2.21–5.66 as reported earlier [18]. N,N-Dimethylformamide (DMF). DMF was dried by keeping the liquid over sodium or potassium hydroxide for 5.0 h. The liquid was shaken frequently during this period, then filtered directly into a flask containing freshly prepared calcium oxide, refluxed for 1.5 h and distilled at atmospheric pressure under dry conditions.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
p-Dichlorobenzene. p-Dichlorobenzene was of analytical reagent grade.
2.3. General procedure Ruthenium(III) solution (2.5–62.5 mg) is taken in a 50 ml beaker and 2.5 ml of 3×10 − 3 M reagent solution is added. Measure the pH, adjust it to 4.5, 5.5 and 4.8 for reagents I, II and III, respectively, by adding 12.5 ml of acetate buffer solution. Transfer the solution into a 50 ml round-bottom flask and heat to 60 °C in a water bath. Add 2.0 g p-dichlorobenzene, stopper the flask and continue to heat until the pdichlorobenzene has melted. Remove the flask from the water bath and shake it vigorously until the p-dichlorobenzene separates out as a solid mass. Repeat the melting and solidification procedure. Separate the p-dichlorobenzene from the aqueous phase by filtration through a filter paper. Dissolve the solid mass in DMF and dilute to 25 ml with DMF in a calibrated flask. Dry the solution by pouring onto anhydrous sodium sulfate (2.0 g) in a beaker. Place a portion of this solution in a 10 mm cell and measure the absorbance at umax 622, 518 and 542 nm against reagents I, II and III as blank, respectively. Prepare a calibration graph under similar conditions.
2.4. Determination of ruthenium in ore and metallurgy samples According to the published procedure [19], an appropriate amount of placer platinum was weighed accurately into a porcelain crucible and mixed with 4.0 g mixture of Na2CO3 and MgO (3:1, w/w). The mixture was first heated at 550 °C for 30 min, and then transferred to an iron crucible to heat with 7.0 g of Na2O2 in the muffle furnace at 700 °C for 30 min. After the residue was cooled, it was extracted with water, and treated with 10 ml of saturated KMnO4 solution and 100 ml of 9.0 M H2SO4. Ruthenium tetroxide was distilled, and absorbed in 25 ml of HCl (0.2%), ethanol (4.0%) and H2SO4 (2.0 M). An appropriate amount of NaBH4 (about half
1833
amount of the ruthenium contained in the sample) was added to the absorbed solution to reduce Ru(VIII) to Ru(III). This solution was finally transferred to a 100 ml calibrated flask and diluted to volume with water and analysed directly as recommended above. Metallurgy sample was treated the same as the placer platinum sample but the preparation in the muffle furnace was omitted. The sample solution was diluted appropriately with water and then analysed directly by the proposed method.
2.5. Determination of ruthenium in seawater samples Different water samples were collected in polyethylene containers. Aliquots of this solution were allowed to react with 5.0 ml of bromine water for 10 min, and the excess bromine was removed by boiling of the solution. The samples were then spiked with various amounts (5–40 mg) of Ru(III). Filtration of the sample solution through a 0.45 mm membrane was carried out, followed by the addition of 5.0 ml of EDTA (0.002 M) and 10 ml of NaF (1.0 M); the sample was then analysed employing the above general procedure.
3. Results and discussion
3.1. Absorption spectra Absorption spectra of reagents (I–III) and their ruthenium complex in p-dichlorobenzene-DMF solutions were recorded against reagent and DMF blank. The formation of the complex was accompanied with bathochromic shift of umax of reagent (495, 450 and 461 nm for reagents I, II and III, respectively) by 127, 68 and 81 nm. The absorption spectra of the complexes showed bands at 622, 518 and 542 nm. All absorbance measurements were made at the umax for each complex. Investigations were carried out to establish the most favourable conditions. The influence of each of the following variables on the reaction was tested.
1834
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
3.1.1. Effect of pH Extraction was carried out at different pH values (2.21–5.66), keeping the other conditions constant. The nature of the spectral curves remained constant at pH 4.04– 5.10, 5.31– 5.66 and 4.50– 5.10 for reagents I, II and III, respectively (Fig. 1). A pH of 4.5, 5.5 and 4.8 were arbitrarily selected for complete colour development using reagents I, II and III, respectively. Moreover, it was observed that, the absorbance remained con-
Fig. 1. Effect of pH on the absorbance of 1.0 mg ml − 1 Ru3+ complexed with reagents (I –III).
stant with the addition of 12.5 ml buffer in each case.
3.1.2. Effect of reagent concentration The influence of the reagent concentration was studied in the range 3× 10 − 5 –3.2×10 − 4 M. An increase of the reagent concentration resulted in an increase of the absorbance of the complex (against the reagent as blank) and was observed to be constant in the range 1.2×10 − 4 –2.1× 10 − 4 M (Fig. 2). The optimum absorbance of the complex coupled with minimum blank reading was found to be 1.5× 10 − 4 M. A higher concentration of reagent was tried, but because of increased absorbance of the reagent blank, the study was restricted to this range. In the subsequent studies, the optimum concentration of the reagent 1.5× 10 − 4 M was used. 3.1.3. Effect of p-dichlorobenzene The effect of p-dichlorobenzene was investigated by varying the amount used from 0.25 to 5.0 g. The extractions were quantitative when the amount of p-dichlorobenzene was in the range 1.5–3.0 g. Below 1.5 g, the extraction was incomplete, whereas above 3.0 g, absorbance decrease in addition to the instability of the formed complexes. 3.1.4. Effect of shaking time The extraction of the complex into pdichlorobenzene was to be very rapid and no change was observed in the extent of extraction when the shaking time was varied from 1.0 to 15 min. 3.1.5. Order of additions The order of mixing reagents seriously affects the absorbance value for each complex. However, addition in the order ruthenium-reagent–buffer– p-dichlorobenzene gives the best results in all complexes formed. 3.2. Characteristics of the analytical method
Fig. 2. Effect of 10−4 M reagents (I –III) on the absorbance of 1.0 mg ml − l Ru3+.
A straight line passing through the origin was obtained for the calibration graph at the selected wavelength for each system. Beer’s law is obeyed
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
1835
Table 1 Spectral characteristics of the ruthenium complexes Parameter
Reagents I
pH value umax (nm) Stability (h) Beer’s law limits (mg ml−1) Ringbom range (mg ml−1) Molar absorptivity (104 l mol−1 cm−1) Sandell sensitivity (ng cm−2) Detection limits (ng ml−1) Quantification limits (mg ml−1) Regression equation Intercept (a) Slope (b) Correlation coefficient (r) Standard deviation Range of error ( 9%) Student t a-test/(2.57)b F-valuea/(5.05)b
II
III
A
P
A
P
A
P
4.5 622 36 0.2–5.0 0.5–4.7 2.73 3.70 50 0.17
4.5 622 48 0.1–2.2 0.2–2.0 5.65 1.79 27 0.09
5.5 518 36 0.2–5.6 0.5–5.3 2.13 4.75 60 0.20
5.5 518 48 0.1–2.5 0.2–2.3 4.25 2.38 30 0.10
4.8 542 30 0.2–5.4 0.5–5.0 2.43 4.15 65 0.22
4.8 542 48 0.1–2.4 0.3–2.2 5.05 2.00 32 0.11
+0.009 0.27 0.9992 0.37 1.20 1.56 3.13
−0.006 0.56 0.9996 0.24 0.80 1.11 2.48
−0.005 0.21 0.9988 0.45 1.40 1.73 3.44
+0.007 0.42 0.9990 0.33 1.10 1.46 3.15
−0.008 0.24 0.9992 0.52 1.30 1.80 3.53
+0.004 0.50 0.9998 0.29 0.95 1.34 2.77
A: absence; P: present. a Compared with ferrozine method [22]. b Values in parenthesis are the theoretical t- and F-values for five degrees of freedom and 95% confidence limits.
in the concentration range of 0.2– 5.6 mg ml − 1 (Table 1), while the optimum concentration range for accurate determination, as investigated from the Ringbom plot, is 0.5–5.3 mg ml − 1 in the absence of p-dichlorobenzene. In its presence, Beer’s law is obeyed in the concentration range of 0.1–2.5 mg ml − 1 (Table 1), while the optimum concentration range for accurate determination, as investigated from the Ringbom plot, is 0.3– 2.3 mg ml − 1. The molar absorptivity and Sandell sensitivity for each system are calculated and recorded in Table 1. The standard deviation (SD) for ten independent measurements of the reagent blank absorbance was calculated (Table 1). The slope of the calibration graphs (S) was also estimated (Table 1). The theoretical limits of detection and quantification (C=K× SD/S), with K = 3 and 10 [20,21] were calculated and recorded in Table 1. The relative SD for 1.5 mg ml − 1 ruthenium without using p-dichlorobenzene was 1.13, 0.88 and 1.05%,
Table 2 Tolerance limits for ions on the determination of 1.0 mg ml−1 ruthenium as complexes with reagents I–III Foreign ions
Na+, K+, Li+, HCO− 3 Ca2+, Ba2+, Sr2+, Cl−, CO2− 3 − 3− Br−, SCN−, S2O2− 3 , I , PO4 2− 2− F−, NO− 3 , SO4 , S Tartrate, oxalate, NO− 2 Ga3+, Al3+, In3+ 2− − − WO2− 4 , MoO4 , VO3 , AsO4 Fe2+, Fe3+, Cr3+, Cr6+ Cu2+, Co2+, Ni2+, Mn2+ Sn2+, Sn4+, Zn2+, Pt4+ Pd2+, EDTA, Os8+ 4+ UO2+ , Ce4+ 2 , Th Ir3+, Au3+, Hg2+ Pb2+a a
Tolerance limits mg ml−1 I
II
III
3500 3000 2500 2000 1400 1000 700 400 200 140 80 50 30 20
3500 2800 2250 1750 1200 700 450 250 140 90 50 30 20 15
3500 2900 2300 1750 1100 750 500 300 150 100 60 35 20 18
Masking with 10 ml of 2.0 M NaF.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
1836
Table 3 Determination of ruthenium in different samples Sample
Founda (mg l−1)
Standard values (mg l−1)
I
II
III
Ferrozine [22]
Ore
1.40
1.38 t= 1.58 F =3.27
1.41 t =1.33 F= 2.81
1.43 t=1.71 F= 3.65
1.45
Metallurgy product
6.00
6.05 t= 1.46 F =3.08
5.97 t =1.25 F= 2.90
5.95 t=1.63 F= 3.17
5.90
a
Average of seven determinations.
whereas on using p-dichlorobenzene was 0.69, 0.96 and 0.83% (ten independent determinations) using reagents I, II and III, respectively. Alternatively, a significance test was applied to compare the accuracy and precision of the present and ferrozine [22] methods. The t- and F-values calculated were less than the theoretical values in all instances (Table 1).
prepared containing a 3500 molar excess of the foreign ions relative to ruthenium. For metal ions that were found to cause interference, a lower concentration of foreign ions was then prepared. Slight interference from Pb2 + was exhibited which can be eliminated using 10 ml of 1.0 M NaF. The tolerance limits are given in Table 2.
3.4. Applications 3.3. Effect of foreign ions Synthetic solutions containing 1.0 mg ml − 1 of ruthenium and various amounts of other ions were prepared and the proposed procedure for the determination of ruthenium was followed. An error of 93.0% in the absorbance reading was considered to be tolerable. Solutions were
Determining ruthenium in ore, metallurgy products and seawater samples by the proposed method tested the validity of the method. The results presented in Tables 3 and 4 were in a good agreement with those obtained with ferrozine method [22]. Alternatively, a significant test was applied to
Table 4 Determination of ruthenium in seawater samples Sample
Added (mg ml−1)
Founda (mg ml−1) I
II
III
Ferrozine [22]
Mediterranean sea
2.00 4.00 6.00
2.07 4.08 6.10 t= 1.18 F =2.73
2.08 4.10 6.08 t= 1.46 F= 3.10
2.10 4.07 6.06 t= 1.68 F =3.33
2.10 4.05 6.13
Red sea
1.50 3.00 4.50
1.55 3.07 4.60 t= 1.25 F =2.81
1.54 3.08 4.57 t= 1.42 F= 3.00
1.57 3.10 4.55 t= 1.57 F =3.18
1.69 3.06 4.64
a
Average of seven determinations.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
compare the accuracy (t-test) and precision (Fvalue) of the proposed method compared with ferrozine method [22]. The t- and F-values calculated were less than the theoretical value [23] in all instances (for five degrees of freedom and 95% confidence level). Hence, there is no significant difference between them.
[7] [8] [9] [10] [11] [12] [13] [14]
References [1] E. Merian, Metals and Their Compounds in the Environment, Occurrence, Analysis and Biological Relevance, VCH, Weinheim, 1981, p. 1143. [2] R.K. Sharma, Bull. Chem. Soc. Jpn. 66 (1993) 1084. [3] A.M. Almuaibed, A. Townshend, Microchem. J. 48 (1993) 21. [4] M.S. El-Shahawi, A.Z. Abu-Zuhri, S.M. Al-Dhaheri, Fresenius Z. Anal. Chem. 350 (1994) 874. [5] Z. Marczenko, Separation and Spectrophotometric Determination of Elements, Ellis Horwood, New York, 1986, p. 492. [6] M.S. El-Shahawi, M. Al-Mahdy, J. Chromatogr. 697 (1995) 185.
[15] [16] [17] [18] [19]
[20] [21] [22] [23]
1837
B. Morelli, Anal. Lett. 19 (1986) 503. B. Morelli, Analyst 108 (1983) 386. M.S. El-Shahawi, S.A. Barakat, Talanta 42 (1995) 1641. M. Balcerzak, E. Swiecicka, Talanta 43 (1996) 471. M. Balcerzak, E. Swiecicka, D. Bydtronska, Anal. Lett. 32 (1999) 1799. K. Malathi, M. Subbaiyan, Talanta 42 (1995) 1487. Q. Cao, Z. Li, J. Wang, Q. Xu, Anal. Lett. 33 (2000) 2305. A.S. Amin, S. Shakra, A.A. Abdalla, Bull. Chem. Soc. Jpn. 67 (1994) 1286. A.S. Amin, J. Sod. Chem. Soc. 4 (2000) 129. A.S. Amin, G.O. El-Sayed, Monatsh. 132 (2001) 587. A.S. Amin, Anal. Lett. 34 (2001) (in press). H.B.S. Britton, Hydrogen Ions, 4th ed., Chapman and Hall, London, 1952. X.S. Cai, C. Huang, Analysis of Precious Metals, Metallurgy Industrial Press, Perking, 1984, p. 162 and 233. Committee on Environmental Improvement, Anal. Chem. 52 (1980) 2242. Analytical Method Committee, Analyst 112 (1987) 199. S.K. Kundra, M. Katyal, R.P. Singh, Curr. Sci. 44 (1975) 548. J.C. Miller, J.N. Miller, Statistics in Analytical Chemistry, 3rd ed., Ellis Horwood, Chichester, 1993.
Simple and selective spectrophotometric determination of ruthenium after solid phase extraction with some quinoxaline dyes into microcrystalline p-dichlorobenzene Alaa S. Amin Chemistry Department, Faculty of Science, Benha Uni6ersity, Benha, Egypt Received 30 May 2001; accepted 10 July 2001
Abstract A simple selective and highly sensitive extraction method has been developed for the determination of ruthenium spectrophotometrically after extraction of its 2,3-dichloro-6-(3-carboxy-2-hydroxy-1-naphthylazo)quinoxaline (I), 2,3-dichloro-6-(2-hydroxy-3,5-dinitrophenylazo)quinoxaline (II) and 2,3-dichloro-6-(2,7-dihydroxy-naphthylazo)quinoxaline (III) complexes into microcrystalline p-dichlorobenzene. The optimization of experimental conditions for the procedure is studied. The solid p-dichlorobenzene containing the ruthenium-reagent (I – III) complexes is separated by filtration and dissolved in N,N-dimethylformamide. The absorbance is measured at umax 622, 518 and 542 nm against reagents I, II and III, respectively, as blank. Beer’s law is obeyed upto 2.5 mg ml − 1 of ruthenium. The molar absorptivity, Sandell sensitivity, detection and quantification limits are calculated, when compared with those parameters without using solid phase extraction method. The interference of various ions has been studied in detail and the statistical evaluation of the experimental results is reported. The proposed methods have been successfully applied for the determination of trace amount of ruthenium in seawater, ore and metallurgy products. © 2002 Published by Elsevier Science B.V. Keywords: Ruthenium determination; Quinoxaline azo dyes; Spectrophotometry; Trace analysis
1. Introduction Interest in the development of analytical techniques for determination of the noble metals is still growing as a result of rapid growth of their applications, e.g. in chemical engineering, microelectronics and medicine. Compared to most elements, ruthenium has a limited influence on the biosphere because only small quantities ever reach living organisms [1]. The amount of ruthenium readily introduced into rivers, lakes and oceans
through industrial wastes is minute [1]. Thus, sensitive, reliable and practicable methods are required for the quantitation of ruthenium at trace levels. The most common reported spectrophotometric procedures for ruthenium determination require laborious enrichment steps [2–6], e.g. flotation, precipitation and solvent extraction. However, most of these methods suffer from lack of selectivity and sensitivity [3–6]. A variety of chromogenic reagents have served as the basis for the spectrophotometric determina-
1386-1425/02/$ - see front matter © 2002 Published by Elsevier Science B.V. PII: S 1 3 8 6 - 1 4 2 5 ( 0 1 ) 0 0 6 8 1 - 3
1832
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
tion of ruthenium. Recent reference works [6– 13] list more reagents for ruthenium that is steadily increasing because of its commercial importance. The various methods differ considerably in sensitivity, tolerance to other ions, rate of reaction, useful concentration range and availability of the reagents, and there appears to be scope for the development of further procedures with readily available chromogenic reagents that give good sensitivity. Recently, certain quinoxaline azo dyes have been prepared and evaluated as high sensitive chromogenic reagents for the spectrophotometric determination of gold [14], scandium [15], mercury [16] and zinc [17]. In the present study, consideration was given to the possible reaction of ruthenium with these chromogens in the light of its position in the periodic table and the analytical resemblance of certain compounds of ruthenium to those of other studied metal ions [14– 17]. The behaviour of 2,3-dichloro-6-(3-carboxy-2hydroxy-1-naphthylazo)quinoxaline (I), 2,3-dichloro - 6 - (2 - hydroxy - 3,5 - dinitrophenylazo)quinoxaline (II) and 2,3-dichloro-6-(2,7-dihydroxynaphthylazo)quinoxaline (III) with ruthenium has now been studied and their potential as analytical reagents has been assessed. The present solid phase extraction method is more sensitive and simpler. The main advantages of this method are that the equilibrium distribution between the two phases is attained in a few seconds owing to the high temperature used, and the metal chelates are dissolved merely by contact with the molten phase. As the organic phase is solid at room temperature, phase separation is quantitative and can be achieved simply by filtration or decantation. The method is selective and simple, and the complexes are stable for a long time and much better than the other known reagents.
2. Experimental
2.1. Apparatus A Perkin –Elmer l3B double beam spectrophotometer was used with 10 mm matched quartz
cuvettes for recording absorption spectra. An Orion research model 601A/digital ionalyser pHmeter equipped with a combined glass-calomel electrode was used to check the pH of the prepared buffer solutions.
2.2. Reagents Doubly distilled water and analytical-reagent grade chemicals were used throughout, unless stated otherwise. Standard ruthenium solution. A standard solution of ruthenium(III) was prepared by dissolving ruthenium(III) chloride (Johnson Matthey, London) in doubly distilled water, containing sufficient hydrochloric acid to give a final concentration of 1.0 M HCl. Since these solutions have a tendency to turn dark brown on keeping due to oxidation, the stock solution was obtained by refluxing for 4.0 h with HCl and ethanol. The ruthenium solution was standardized gravimetrically by precipitating ruthenium as hydrated oxide, followed by careful ignition in air and then reduction to the metal in the presence of hydrogen gas and cooling in an atmosphere of carbon dioxide gas. Subsequent dilutions were made from the stock solution according to requirements. Quinoxaline dyes preparation. Reagents I–III were prepared according to the previously recommended procedures [16]. A 3× 10 − 3 M solution of the reagents was prepared by dissolving the appropriate weight of each reagent in 10 ml ethanol and then completed to the mark in 100 ml calibrated flask. Sodium acetate–acetic acid buffer. Sodium acetate solution (0.2 M) and acetic acid were mixed in suitable proportions to prepare buffer solutions of pH ranges 2.21–5.66 as reported earlier [18]. N,N-Dimethylformamide (DMF). DMF was dried by keeping the liquid over sodium or potassium hydroxide for 5.0 h. The liquid was shaken frequently during this period, then filtered directly into a flask containing freshly prepared calcium oxide, refluxed for 1.5 h and distilled at atmospheric pressure under dry conditions.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
p-Dichlorobenzene. p-Dichlorobenzene was of analytical reagent grade.
2.3. General procedure Ruthenium(III) solution (2.5–62.5 mg) is taken in a 50 ml beaker and 2.5 ml of 3×10 − 3 M reagent solution is added. Measure the pH, adjust it to 4.5, 5.5 and 4.8 for reagents I, II and III, respectively, by adding 12.5 ml of acetate buffer solution. Transfer the solution into a 50 ml round-bottom flask and heat to 60 °C in a water bath. Add 2.0 g p-dichlorobenzene, stopper the flask and continue to heat until the pdichlorobenzene has melted. Remove the flask from the water bath and shake it vigorously until the p-dichlorobenzene separates out as a solid mass. Repeat the melting and solidification procedure. Separate the p-dichlorobenzene from the aqueous phase by filtration through a filter paper. Dissolve the solid mass in DMF and dilute to 25 ml with DMF in a calibrated flask. Dry the solution by pouring onto anhydrous sodium sulfate (2.0 g) in a beaker. Place a portion of this solution in a 10 mm cell and measure the absorbance at umax 622, 518 and 542 nm against reagents I, II and III as blank, respectively. Prepare a calibration graph under similar conditions.
2.4. Determination of ruthenium in ore and metallurgy samples According to the published procedure [19], an appropriate amount of placer platinum was weighed accurately into a porcelain crucible and mixed with 4.0 g mixture of Na2CO3 and MgO (3:1, w/w). The mixture was first heated at 550 °C for 30 min, and then transferred to an iron crucible to heat with 7.0 g of Na2O2 in the muffle furnace at 700 °C for 30 min. After the residue was cooled, it was extracted with water, and treated with 10 ml of saturated KMnO4 solution and 100 ml of 9.0 M H2SO4. Ruthenium tetroxide was distilled, and absorbed in 25 ml of HCl (0.2%), ethanol (4.0%) and H2SO4 (2.0 M). An appropriate amount of NaBH4 (about half
1833
amount of the ruthenium contained in the sample) was added to the absorbed solution to reduce Ru(VIII) to Ru(III). This solution was finally transferred to a 100 ml calibrated flask and diluted to volume with water and analysed directly as recommended above. Metallurgy sample was treated the same as the placer platinum sample but the preparation in the muffle furnace was omitted. The sample solution was diluted appropriately with water and then analysed directly by the proposed method.
2.5. Determination of ruthenium in seawater samples Different water samples were collected in polyethylene containers. Aliquots of this solution were allowed to react with 5.0 ml of bromine water for 10 min, and the excess bromine was removed by boiling of the solution. The samples were then spiked with various amounts (5–40 mg) of Ru(III). Filtration of the sample solution through a 0.45 mm membrane was carried out, followed by the addition of 5.0 ml of EDTA (0.002 M) and 10 ml of NaF (1.0 M); the sample was then analysed employing the above general procedure.
3. Results and discussion
3.1. Absorption spectra Absorption spectra of reagents (I–III) and their ruthenium complex in p-dichlorobenzene-DMF solutions were recorded against reagent and DMF blank. The formation of the complex was accompanied with bathochromic shift of umax of reagent (495, 450 and 461 nm for reagents I, II and III, respectively) by 127, 68 and 81 nm. The absorption spectra of the complexes showed bands at 622, 518 and 542 nm. All absorbance measurements were made at the umax for each complex. Investigations were carried out to establish the most favourable conditions. The influence of each of the following variables on the reaction was tested.
1834
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
3.1.1. Effect of pH Extraction was carried out at different pH values (2.21–5.66), keeping the other conditions constant. The nature of the spectral curves remained constant at pH 4.04– 5.10, 5.31– 5.66 and 4.50– 5.10 for reagents I, II and III, respectively (Fig. 1). A pH of 4.5, 5.5 and 4.8 were arbitrarily selected for complete colour development using reagents I, II and III, respectively. Moreover, it was observed that, the absorbance remained con-
Fig. 1. Effect of pH on the absorbance of 1.0 mg ml − 1 Ru3+ complexed with reagents (I –III).
stant with the addition of 12.5 ml buffer in each case.
3.1.2. Effect of reagent concentration The influence of the reagent concentration was studied in the range 3× 10 − 5 –3.2×10 − 4 M. An increase of the reagent concentration resulted in an increase of the absorbance of the complex (against the reagent as blank) and was observed to be constant in the range 1.2×10 − 4 –2.1× 10 − 4 M (Fig. 2). The optimum absorbance of the complex coupled with minimum blank reading was found to be 1.5× 10 − 4 M. A higher concentration of reagent was tried, but because of increased absorbance of the reagent blank, the study was restricted to this range. In the subsequent studies, the optimum concentration of the reagent 1.5× 10 − 4 M was used. 3.1.3. Effect of p-dichlorobenzene The effect of p-dichlorobenzene was investigated by varying the amount used from 0.25 to 5.0 g. The extractions were quantitative when the amount of p-dichlorobenzene was in the range 1.5–3.0 g. Below 1.5 g, the extraction was incomplete, whereas above 3.0 g, absorbance decrease in addition to the instability of the formed complexes. 3.1.4. Effect of shaking time The extraction of the complex into pdichlorobenzene was to be very rapid and no change was observed in the extent of extraction when the shaking time was varied from 1.0 to 15 min. 3.1.5. Order of additions The order of mixing reagents seriously affects the absorbance value for each complex. However, addition in the order ruthenium-reagent–buffer– p-dichlorobenzene gives the best results in all complexes formed. 3.2. Characteristics of the analytical method
Fig. 2. Effect of 10−4 M reagents (I –III) on the absorbance of 1.0 mg ml − l Ru3+.
A straight line passing through the origin was obtained for the calibration graph at the selected wavelength for each system. Beer’s law is obeyed
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
1835
Table 1 Spectral characteristics of the ruthenium complexes Parameter
Reagents I
pH value umax (nm) Stability (h) Beer’s law limits (mg ml−1) Ringbom range (mg ml−1) Molar absorptivity (104 l mol−1 cm−1) Sandell sensitivity (ng cm−2) Detection limits (ng ml−1) Quantification limits (mg ml−1) Regression equation Intercept (a) Slope (b) Correlation coefficient (r) Standard deviation Range of error ( 9%) Student t a-test/(2.57)b F-valuea/(5.05)b
II
III
A
P
A
P
A
P
4.5 622 36 0.2–5.0 0.5–4.7 2.73 3.70 50 0.17
4.5 622 48 0.1–2.2 0.2–2.0 5.65 1.79 27 0.09
5.5 518 36 0.2–5.6 0.5–5.3 2.13 4.75 60 0.20
5.5 518 48 0.1–2.5 0.2–2.3 4.25 2.38 30 0.10
4.8 542 30 0.2–5.4 0.5–5.0 2.43 4.15 65 0.22
4.8 542 48 0.1–2.4 0.3–2.2 5.05 2.00 32 0.11
+0.009 0.27 0.9992 0.37 1.20 1.56 3.13
−0.006 0.56 0.9996 0.24 0.80 1.11 2.48
−0.005 0.21 0.9988 0.45 1.40 1.73 3.44
+0.007 0.42 0.9990 0.33 1.10 1.46 3.15
−0.008 0.24 0.9992 0.52 1.30 1.80 3.53
+0.004 0.50 0.9998 0.29 0.95 1.34 2.77
A: absence; P: present. a Compared with ferrozine method [22]. b Values in parenthesis are the theoretical t- and F-values for five degrees of freedom and 95% confidence limits.
in the concentration range of 0.2– 5.6 mg ml − 1 (Table 1), while the optimum concentration range for accurate determination, as investigated from the Ringbom plot, is 0.5–5.3 mg ml − 1 in the absence of p-dichlorobenzene. In its presence, Beer’s law is obeyed in the concentration range of 0.1–2.5 mg ml − 1 (Table 1), while the optimum concentration range for accurate determination, as investigated from the Ringbom plot, is 0.3– 2.3 mg ml − 1. The molar absorptivity and Sandell sensitivity for each system are calculated and recorded in Table 1. The standard deviation (SD) for ten independent measurements of the reagent blank absorbance was calculated (Table 1). The slope of the calibration graphs (S) was also estimated (Table 1). The theoretical limits of detection and quantification (C=K× SD/S), with K = 3 and 10 [20,21] were calculated and recorded in Table 1. The relative SD for 1.5 mg ml − 1 ruthenium without using p-dichlorobenzene was 1.13, 0.88 and 1.05%,
Table 2 Tolerance limits for ions on the determination of 1.0 mg ml−1 ruthenium as complexes with reagents I–III Foreign ions
Na+, K+, Li+, HCO− 3 Ca2+, Ba2+, Sr2+, Cl−, CO2− 3 − 3− Br−, SCN−, S2O2− 3 , I , PO4 2− 2− F−, NO− 3 , SO4 , S Tartrate, oxalate, NO− 2 Ga3+, Al3+, In3+ 2− − − WO2− 4 , MoO4 , VO3 , AsO4 Fe2+, Fe3+, Cr3+, Cr6+ Cu2+, Co2+, Ni2+, Mn2+ Sn2+, Sn4+, Zn2+, Pt4+ Pd2+, EDTA, Os8+ 4+ UO2+ , Ce4+ 2 , Th Ir3+, Au3+, Hg2+ Pb2+a a
Tolerance limits mg ml−1 I
II
III
3500 3000 2500 2000 1400 1000 700 400 200 140 80 50 30 20
3500 2800 2250 1750 1200 700 450 250 140 90 50 30 20 15
3500 2900 2300 1750 1100 750 500 300 150 100 60 35 20 18
Masking with 10 ml of 2.0 M NaF.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
1836
Table 3 Determination of ruthenium in different samples Sample
Founda (mg l−1)
Standard values (mg l−1)
I
II
III
Ferrozine [22]
Ore
1.40
1.38 t= 1.58 F =3.27
1.41 t =1.33 F= 2.81
1.43 t=1.71 F= 3.65
1.45
Metallurgy product
6.00
6.05 t= 1.46 F =3.08
5.97 t =1.25 F= 2.90
5.95 t=1.63 F= 3.17
5.90
a
Average of seven determinations.
whereas on using p-dichlorobenzene was 0.69, 0.96 and 0.83% (ten independent determinations) using reagents I, II and III, respectively. Alternatively, a significance test was applied to compare the accuracy and precision of the present and ferrozine [22] methods. The t- and F-values calculated were less than the theoretical values in all instances (Table 1).
prepared containing a 3500 molar excess of the foreign ions relative to ruthenium. For metal ions that were found to cause interference, a lower concentration of foreign ions was then prepared. Slight interference from Pb2 + was exhibited which can be eliminated using 10 ml of 1.0 M NaF. The tolerance limits are given in Table 2.
3.4. Applications 3.3. Effect of foreign ions Synthetic solutions containing 1.0 mg ml − 1 of ruthenium and various amounts of other ions were prepared and the proposed procedure for the determination of ruthenium was followed. An error of 93.0% in the absorbance reading was considered to be tolerable. Solutions were
Determining ruthenium in ore, metallurgy products and seawater samples by the proposed method tested the validity of the method. The results presented in Tables 3 and 4 were in a good agreement with those obtained with ferrozine method [22]. Alternatively, a significant test was applied to
Table 4 Determination of ruthenium in seawater samples Sample
Added (mg ml−1)
Founda (mg ml−1) I
II
III
Ferrozine [22]
Mediterranean sea
2.00 4.00 6.00
2.07 4.08 6.10 t= 1.18 F =2.73
2.08 4.10 6.08 t= 1.46 F= 3.10
2.10 4.07 6.06 t= 1.68 F =3.33
2.10 4.05 6.13
Red sea
1.50 3.00 4.50
1.55 3.07 4.60 t= 1.25 F =2.81
1.54 3.08 4.57 t= 1.42 F= 3.00
1.57 3.10 4.55 t= 1.57 F =3.18
1.69 3.06 4.64
a
Average of seven determinations.
A.S. Amin / Spectrochimica Acta Part A 58 (2002) 1831–1837
compare the accuracy (t-test) and precision (Fvalue) of the proposed method compared with ferrozine method [22]. The t- and F-values calculated were less than the theoretical value [23] in all instances (for five degrees of freedom and 95% confidence level). Hence, there is no significant difference between them.
[7] [8] [9] [10] [11] [12] [13] [14]
References [1] E. Merian, Metals and Their Compounds in the Environment, Occurrence, Analysis and Biological Relevance, VCH, Weinheim, 1981, p. 1143. [2] R.K. Sharma, Bull. Chem. Soc. Jpn. 66 (1993) 1084. [3] A.M. Almuaibed, A. Townshend, Microchem. J. 48 (1993) 21. [4] M.S. El-Shahawi, A.Z. Abu-Zuhri, S.M. Al-Dhaheri, Fresenius Z. Anal. Chem. 350 (1994) 874. [5] Z. Marczenko, Separation and Spectrophotometric Determination of Elements, Ellis Horwood, New York, 1986, p. 492. [6] M.S. El-Shahawi, M. Al-Mahdy, J. Chromatogr. 697 (1995) 185.
[15] [16] [17] [18] [19]
[20] [21] [22] [23]
1837
B. Morelli, Anal. Lett. 19 (1986) 503. B. Morelli, Analyst 108 (1983) 386. M.S. El-Shahawi, S.A. Barakat, Talanta 42 (1995) 1641. M. Balcerzak, E. Swiecicka, Talanta 43 (1996) 471. M. Balcerzak, E. Swiecicka, D. Bydtronska, Anal. Lett. 32 (1999) 1799. K. Malathi, M. Subbaiyan, Talanta 42 (1995) 1487. Q. Cao, Z. Li, J. Wang, Q. Xu, Anal. Lett. 33 (2000) 2305. A.S. Amin, S. Shakra, A.A. Abdalla, Bull. Chem. Soc. Jpn. 67 (1994) 1286. A.S. Amin, J. Sod. Chem. Soc. 4 (2000) 129. A.S. Amin, G.O. El-Sayed, Monatsh. 132 (2001) 587. A.S. Amin, Anal. Lett. 34 (2001) (in press). H.B.S. Britton, Hydrogen Ions, 4th ed., Chapman and Hall, London, 1952. X.S. Cai, C. Huang, Analysis of Precious Metals, Metallurgy Industrial Press, Perking, 1984, p. 162 and 233. Committee on Environmental Improvement, Anal. Chem. 52 (1980) 2242. Analytical Method Committee, Analyst 112 (1987) 199. S.K. Kundra, M. Katyal, R.P. Singh, Curr. Sci. 44 (1975) 548. J.C. Miller, J.N. Miller, Statistics in Analytical Chemistry, 3rd ed., Ellis Horwood, Chichester, 1993.