摘要:

ANALYST, DECEMBER 1986, VOL. 111 1379 Determination of Dissolved Aluminium by the Micelle-enhanced Fluorescence of its Lumogallion Complex A. G. Howard,* A. J. Coxhead, I. A. Potter and A. P. Watt Department of Chemistry, The University, Southampton, Hampshire SO9 5NH, UK A sensitive modification of the lumogallion fluorescence assay for aluminium is presented that exploits the 5-6-fold increase in the fluorescence intensity of the lumogallion - aluminium complex in the presence of the non-ionic detergent Triton X-100. The sensitised procedure has a detection limit of 0.02 yg 1-1 of Al (30) and a relative standard deviation of 5% at the 0.1 yg 1-1 of Al level. The sensitivity of the method is independent of salinityand it can be used for the determination of aluminium in both fresh and saline waters.The interference from iron experienced in the conventional lumogallion procedure is also present in the modified method, but can be masked by the addition of 1,lO-phenanthroline. Keywords: Aluminium determination; water analysis; lumogallion; micelle-enhanced fluorescence; surfactants Aluminium was, until recently, believed to be relatively harmless, but evidence has grown to implicate it in a variety of disorders ranging from bone softening (dialysis osteomalacia) to dementia (dialysis encephalopathy syndrome) . I Alumi- nium is a minor component of natural waters, its concentra- tion being largely governed by its low solubility; typical levels range from 0.5 to 5 yg 1-1 of A1 in sea water2-5 to 10-140 yg 1-l of A1 in fresh waters.6.7 Sensitive analytical procedures are therefore required to study its geochemical cycling in the environment and its toxicology.There are comparatively few techniques that are suitable for such analyses, and the lumogallion fluorimetric analysis has gained widespread acceptance owing to its high sensitivity and relative freedom from interferences. The lumogallion reagent was initially applied to aluminium complexation in 19688 and methods were then proposed for its application to the determination of aluminium in sea water.439 The intensity of molecular fluorecence is greatly influenced by solvent effects and the presence of quenching impurities in that solvent. In some instances, however, it is possible to enhance the fluorescence and to minimise interferences by the incorporation of surfactants into the solution to be studied.The factors that are responsible for the increased fluorescence intensity in micellar solution are still poorly understood. A number of mechanisms can be proposed, ranging from the shielding of the complex by the micelle to quenching by dissolved oxygenlo or the water." The effect of the micelle appears to result from the protection of the metal chelate from quenching in the bulk solvent, The possibility of more subtle effects due to molecular ordering in the micellar phase has yet to be investigated. A six-fold enhancement of the fluorescence of the aluminium - lumogallion complex has been shown to occur in the presence of the detergent Antarox C0-890,1* but although paving the way to a more sensitive analytical procedure, the full potential of the technique was not explored.Recently, a ten-fold increase in the sensitivity of the aluminium - morin fluorescence has been reported ,using the non-ionic detergent Tergitol XD, taking the detection limit down to 0.2 pg 1-1 of Al.13 This paper reports studies of the micelle-enhanced fluores- cence of the aluminium - lumogallion complex. A comparison of a variety of detergents has been carried out and a technique has been developed for the determination of aluminium at the very low concentrations found in most natural waters. * To whom correspondence should be addressed. Experimental General Unless stated otherwise , all chemicals were of analytical- reagent grade. All detergents were used without further purification.Glassware and plasticware were cleaned by soaking in dilute hydrochloric acid (ca. 10% VlV) and were rinsed with distilled water prior to use. Buffer solutions were prepared by mixing sodium acetate and acetic acid solutions (2 M) in the proportions required to give the desired pH. A standard aluminium solution was prepared by dissolving aluminium potassium sulphate decahydrate (2.747 g) in distilled water (250 ml). This solution was diluted by a factor of 250. Iron(II1) chloride (0.121 g) was dissolved in distilled water (1 1). This was then diluted by a factor of 1000. Lumogallion was prepared as a 0.02% mlVsolution in distilled water. 1 ,lo-Phenanthroline, recrystallised from a 1 mg 1-1 lumogallion solution, was prepared as a 1% mlV solution in distilled water.Fluorescence spectra were obtained using a Perkin-Elmer LS-5 spectrometer, operated with either 2.5 or 5.0 nm excitation and emission slits. Some initial measurements were made on an uncorrected Farrand Mark 1 spectrofluorimeter. Five-fold replicates of all measurements were carried out unless stated otherwise. Recommended Procedure To 25 ml of sample solution in a 50-ml polyethylene bottle, add 2 ml of acetate buffer, pH 4.5, and 0.3 ml of the lumogallion solution. If required, 1 ml of 1 ,lo-phenanthroline solution should also be added to mask any interference from iron. Heat the sample for 90 min at 80 "C, cool, add 0.15 ml of neat surfactant and shake. The sample should also be shaken vigorously immediately prior to measuring the fluorescence. With our instrumentation, excitation at 485 nm resulted in an emission maximum at 590 nm.These wavelengths are to a certain extent machine-dependent and should be optimised for each instrument. Excitation and emission slit widths of 2.5 or 5.0 nm were employed for this work, but wider slits could result in further sensitivity improvements. Results and Discussion Surfactant Choice Preliminary experiments were carried out to identify the magnitude of the fluorescence enhancement obtained from a1380 ANALYST, DECEMBER 1986, VOL. 111 variety of detergents. The surfactants studied were sodium lauryl sulphate (anionic), Triton X-100, Nonidet P42, NOPCO and Tergitol XD (non-ionic) and cetyltrimethylam- monium bromide (cationic) (Table 1). Although the cationic detergent initially gave a significant enhancement, the fluores- cence rapidly faded.Sustained enhancement was experienced only with the non-ionic detergents. In this preliminary screening it became evident that certain operational difficul- ties could be encountered with some detergents. High-viscos- ity materials were difficult to dissolve and, if added prior to a heated development stage, phase separation occurred. Based on the degree of enhancement and ease of use, Triton X-100 was selected for further study. Initial experiments were carried out to establish the importance of development time and the appropriate time of detergent addition. Aluminium standards containing 100 pg 1-l of A1 were analysed with and without surfactant addition. Each analysis was carried out using a heated development (80 "C for 90 min), or with the samples being left overnight at room temperature.Triton X-100 (0.5% VlV) was either added before or after development. Results from this experiment demonstrated that the combination of heated development and the post-development addition of Triton X-100 resulted in the highest fluorescent yield (Table 2). Addition of the detergent prior to the high-temperature development resulted in the formation of two phases above the cloud temperature, apparently impeding complex forma- tion. Optimisation of Conditions The fluorescence intensity of the aluminium - lumogallion complex in the presence of Triton-X100 was measured over a variety of pH conditions ranging from 3.2 to 6.2 (Fig. 1). In order to account for slight (4 nm) shifts in the excitation wavelength owing to pH changes, intensity measurements were recorded at the excitation and emission wavelengths giving the maximum intensity.In addition, these results were blank-corrected to minimise contributions from the variable composition buffer components. All further investigations were carried out using the optimum pH of 4.7. Having fixed the pH at 4.7, solutions of similar composition were prepared containing various concentrations of detergent. The fluorescence intensity increased with increasing concen- tration of added detergent until a plateau was reached at 0.6% Table 1. Fluorescence of the aluminium - lumogallion complex in detergent solutions. Conditions: Al, 40 pg 1-l; surfactant concentra- tion, 0.5% (VIV for liquids, m/V for solids); fluorescence intensity in arbitrary units Fluorescence Detergent intensity None .. . . . . . . . . 24 Tergitol XD . . . . . . . . 47 Nonidet P42 . . . . . . . . 67 NOPCO . . . . . . . . . . 69 Triton X-100 . . . . . . . . 80 Cetyltrimethylammonium bromide 68 Sodiumlaurylsulphate . . . . 13 Table 2. Development procedure and time of surfactant addition. Fluorescence intensity in arbitrary units Fluorescence intensity at development temperature ca. 20 "C 80 "C Nosurfactant . . . . 29 46 Surfactant added: Pre-developmen t 24 165 Post-development 206 245 V/V (Fig. 2). This concentration was therefore employed for further studies. In order to demonstrate the magnitude of the enhancement and spectral features, typical emission spectra are shown in Fig.3. It is interesting that not only does the intensity increase in the presence of the detergent, but the emission becomes split into two peaks in the presence of the Triton X-100. Care should be taken in transferring the optimum wavelengths directly to other instruments owing to the variability of excitation and detection characteristics. Wavelengths should be re-optimised for each instrument used. Interferences Iron is known to interfere in the standard lumogallion procedure4 and experiments were therefore carried out to assess whether this remained so in the modified method. 800 I - 600 400 200 4 5 PH 1 6 Fig. 1. Effect of pH on the fluorescence intensity obtained from a 100 pg 1-1 aluminium solution. Lumogallion, 2.4 mg 1-1; Triton X-100, 0.5% V/V.Excitation and emission wavelengths selected for maximum fluorescence; readings corrected for reagent blank Surfactant, O/O Fig. 2. Variation of fluorescence intensity with surfactant concentra- tion. Conditions as in Fig. 1 40 -- 20 540 560 580 600 620 642 Wavelengt hin m Fig. 3. Typical aluminium - lumogallion emission spectra in (A) the absence and (B) the presence of Triton X-100. Aluminium, 1 pg 1-l; lumoeallion. 2.4 me 1-1: Triton X-100. 0.6% V/V (where amro- I I priatg); pH,'4.7; exstation and emission slit widths, 2:5 nmANALYST, DECEMBER 1986, VOL. 11 1 I 1381 ^.- I r e , B - 0 I I 1 I I I 2 4 6 8 Concentration of Feiyg 1-1 Fig. 4. fluorescence. Conditions as in Fig. 3 Influence of iron(II1) on the aluminium - 10- lumogallion 0.2 0.4 0.6 0.8 0.9 Concentration of Fimg I-' Fig.5. in Fig. 3, except A1 = 100 pg I-' Influence of fluoride on fluorescence intensity. Conditions as Concentration of Alikg I-' Fig. 6. Low concentration range calibration graph. Lumogallion, 2.4 mg 1-I; Triton X-100, 0.6% V/V; pH, 4.7; excitation and emission slit width, 5 nm. Blank corrected Standard solutions containing 1 pg 1-1 of A1 were prepared containing between 0 and 10 yg 1-1 of FeIII and between 1 and 10 mg 1-1 of FeIII. These were analysed both with and without the addition of 1,lO-phenanthroline. Without the masking agent the fluorescence intensity decreased with increasing iron concentration (Fig. 4). In the presence of the 1,lO-phenan- throline, however, restoration of the fluorescence was essen- tially complete, even at the higher iron levels.A second known interferent in the conventional assay is that of fluoride. Aluminium standards (100 pg I-I) were prepared containing from 0 to 1.0 mg 1-1 of F- and were analysed for aluminium using the modified lumogallion procedure. The fluorescence was decreased by the presence of fluoride, with a 50% signal suppression being obtained with 1.0 mg 1-1 of F- (Fig. 5). In many natural waters free fluoride levels might be expected to be well below this level owing to the participation of the fluoride in metal complexation and ion pairing. This is believed to occur in sea water, but it is recommended that standard additions experiments be carried out to allow for this and any other as yet undocumented interference. In the event of any discernible fluoride interference it is recommended that large amounts of Ca2+ as calcium chloride should be tried.14 I I I I 0.5 1 1.5 2 Al added/pg 1-l Fig.7. Standard addition of aluminium to (A) distilled water and (B) sea water samples. Lumogallion, 2.4 mg 1 - 1 ; Triton X-100, 0.6% V/V; pH, 4.7; excitation and emission slit widths, 5 nm; 1,10- phenanthroline treated Performance Characteristics In common with all fluorescence methods, the sensitivity is largely determined by the choice of instrument and operating conditions, and the lowest determinable levels by reagent purity and handling conditions. In order to investigate the full potential of the system it became necessary to investigate in greater detail the sources of contamination. Contributions to the blank values were found to originate from acetic acid, sodium acetate and distilled water.Significant improvements in the blank values were obtained by replacing analytical- reagent grade chemicals with AnalaR chemicals and using doubly distilled water. In addition, the lumogallion reactions were subsequently carried out in acid-leached polyethylene vessels. The experimental precision was found to be highly dependent on reagent mixing and it is therefore recommended that solutions should be vigorously mixed immediately prior to measuring the fluorescence. Using this new regime it was possible to calibrate the procedure over the range 0-100 ng 1-1 of A1 (Fig. 6) and to obtain a precision of 5% at 100 ng 1-1 of Al. Application to Fresh Water and Sea Water Samples In order to determine the influence of sample salinity on the sensitivity of the fluorescence, calibration graphs were con- structed using samples to which known amounts of aluminium had been added.Doubly distilled water and filtered (What- man GF/C) sea water obtained along a track from 49" 36' N, 5" 30' W to 49" 43' N, 5" 27' W were employed as repre- sentative fresh water and sea water samples. The gradients of the two standard addition lines (Fig. 7) were identical, indicating that distilled water could be used for the calibration of standards to be used for sea water analyses. It must be pointed out, however, that this is not always so, and that in more heavily contaminated estuarine waters matrix-depen- dent sensitivity changes have been observed. Standard addi- tion checks are therefore recommended for all studies to assess the extent of such problems.Conclusions The procedure described in this paper offers a significant improvement in the concentrations at which the lumogallion complexation can be used for the fluorescent determination of aluminium. It has been shown to be applicable to the analysis of both fresh and sea water samples, but in common with the unmodified procedure the method must be applied cautiously (using standard additions and masking agents) to highly complex samples. References McClure, J., and Smith, P. S . , J. Pathol., 1984, 142, 293. Hydes, D. J . , and Liss, P. S . , Estuar. Coast. Mar. Sci., 1977, 5 , 755. Sackett, W. M., and Arrenhius, G. 0. S . , Geochim. Cosmo- chim. Acta, 1962, 26, 955. 1. 2. 3.1382 4. 5. 6. 7. 8. 9. 10. Shigematsu, T., Nishikawa, Y., Hiraki. K., and Nagano, N., Bunseki Kagaku, 1970, 19, 551. Alberts, J. J., Leyden, D. E., and Patterson, T. A., Mar. Chem., 1976, 4, 51. Hem, J. D., Roberson, C. E., Lind, C. J., and Polzer, W. L., US Geological Survey, Water Supply Paper, 1827E, 1973. Gibbs, R. J., Geochim. Cosrnochim. Acta, 1972,36, 1061. Babko, A. K., Volkova, A. I., and Get’man, T. E., Zh. Anal. Khim., 1968, 23, 39. Hydes, D. J., and Liss, P. S., Analyst, 1976, 101, 922. Hinze, W. L., and Singh, H., Anal. Lett., 1982, 15, 221. ANALYST, DECEMBER 1986, VOL. 111 11. Atkins, P. W., “Physical Chemistry,” Second Edition, Oxford University Press, London, 1982, p. 614. 12. Ishibashi, N., and Kina, K . , Anal. Lett., 1972, 5, 637. 13. Medina Escriche, J . , de la Guardia Cirugeda, M., and Hernandez Hernandez, F., Analyst, 1983, 108, 1386. 14. Makin, J . E., and Aller, R. C., Geochirn. Cosrnochirn. Acta, 1984, 48, 299, Paper A61157 Received May 22nd, 1986 Accepted July 7th, I986

ISSN:0003-2654    

DOI:10.1039/AN9861101379

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST DECEMBER 1986 VOL. 11 1 1383 Synthesis and Chromogenic Properties of Water-soluble Substituted 2-Pyridyl-3’-sulphophenylmethanone 2-Pyridylhydrazones and Spectrophotometric and Analogue Derivative Spectrophotometric Determination of Trace Amounts of Cobalt with 2-Pyridyl-3’-sulphophenylmethanone 2-(5=Nitro)pyridylhydrazone Tsugikatsu Odashima Takemitsu Kikuchi Wataru Ohtani and Hajime lshii Chemical Research Institute of Non-Aqueous Solutions Tohoku University Katahira Sendai 980 Japan Three water-soluble hydrazones in which the 5-position of the pyridine ring in the hydrazine moiety of 2-pyridyl-3’-sulphophenylmethanone 2-pyridylhydrazone (PSPH) was substituted a methyl chloro or nitro group have been synthesised and their chromogenic properties and reactivities with metal ions investigated.The acid dissociation constants (K,) of the synthesised hydrazones were determined spectrophotometrically. These values obeyed Hammett’s law the inductive effect of the substituents being found to reflect the pK, values. Of the synthesised hydrazones 2-pyridyl-3’-sulphophenylmethanone 2-(5-nitro)pyridylhydrazone (PSNPH) is the most sensitive for cobalt(l1) determination. It reacts with cobalt(1l) to form a 1 2 (metal to ligand) complex. This complex is rapidly oxidised above pH 2 to give a stable cobalt(ll1) complex with an absorption maximum at 496 nm. Beer’s law is obeyed over the range 0.05-1.0 pg mi-I of cobalt and the apparent molar absorptivity of the complex is 5.69 x lo4 I mol-1 cm-1 at 496 nm. A sensitive and practical spectrophotometric method for the determination of trace amounts of cobalt with PSNPH is proposed and has been successfully applied to the determination of cobalt in steel samples.The method may be made more sensitive by employing an analogue derivative spectrophotometric technique. Keywords Cobalt determination; 2-p yridyl-3 ’-sulphophen ylmethanone 2-(5-nitro)p yridylh ydrazone; spectrophotometry; analogue derivative spectrophotometry Many kinds of water-insoluble hydrazones have been de-veloped as chromogenic reagents over the past 20 years and these have been reviewed by Katyal et al.1 and Singh et aZ.2 However few water-soluble hydrazones have been syn-thesised with the exception of 2-pyridyl-3’-sulphophenyl-methanone 2-pyridylhydrazone (PSPH).3 Recently we synthesised some water-soluble hydrazones derived from 2-(3’-sulphobenzoy1)pyridine and studied their chromo-genic properties as spectrophotometric reagents.2-Pyridyl-3’-sulphophenylmethanone 2-pyrimidylhydrazone was shown to be a highly sensitive and useful reagent for the determination of iron.4 In the work reported here three water-soluble hydrazones, in which the 5-position of the pyridine ring in the hydrazine moiety of PSPH was substituted by a methyl chloro or nitro group (PSMPH PSCPH and PSNPH respectively) were synthesised and the inductive effect of the substituents and their chromogenic properties and reactivities with metal ions were investigated and compared. PSNPH was found to be the best chromogenic reagent for cobalt(I1) in terms of both sensitivity and selectivity.A sensitive and practical spectro-photometric method for the determination of trace amounts of cobalt with PSNPH has been developed and successfully applied to real samples. The proposed method can be made more sensitive by the introduction of an analogue derivative spectrophotometric technique. Experimental Synthesis of Hydrazones Equimolar amounts (0.01 mol) of 2-(3’-sulphobenzoy1)pyri-dine prepared by the sulphonation of 2-benzoylpyridine by the procedure of Bradsher et a1.,5 and the corresponding 5-substituted 2-pyridylhydrazine were dissolved in 30 ml of 50% aqueous ethanol and heated under reflux for 5 h. After cooling to room temperature the precipitated compound was filtered off recrystallised twice from an ethanol - water mixture and dried in vucuo.Reagents All reagents were of analytical-reagent grade unless stated otherwise. All solutions were prepared with distilled de-ionised water. Hydrazone solutions 2 x 10-3 M. Prepared by dissolving the required mass of each synthesised hydrazone in 0.01 M sodium hydroxide solution. These solutions were further diluted with water if necessary. Standard cobalt(Z1) solution. Prepared by dissolving 0.5 g of metallic cobalt (99.99% pure) in 20 ml of nitric acid (1 + 1) and diluting to 500 ml with water. Buffer solutions. Buffers consisting of 1 M chloroacetic acid -1 M sodium chloroacetate 1 M acetic acid - 1 M sodium acetate, ~ / 1 5 potassium dihydrogen phosphate - ~ / 1 5 sodium hydrogen phosphate (0.2 M boric acid + 0.05 M sodium chloride) - 0.05 M sodium borate and 1 M aqueous ammonia solution - 1 M ammonium chloride systems were used according to the pH values required.Apparatus For measurements of the absorbance and the absorption spectrum a Hitachi 139 and a Hitachi 200-10 spectropho-tometer respectively were used. To obtain the derivative spectrum a modified Hitachi 200-0576 derivative unit com-posed of two analogue differentiation circuits (each having six different time constants from 0.07 to 2.0 s) was connected in series between the output of a Hitachi 556 dual-wavelength spectrophotometer and the input of a Hitachi 057 X - Y recorder the former being used as an ordinary double-beam spectrophotometer. The details of this apparatus and the principle and characteristics of analogue-derivative spectro-photometry have already been d e ~ c r i b e d .~ . ~ Procedure Ordinary spectrophotometry To an aliquot containing up to 25 pg of cobalt(I1) in a 25-1111 calibrated flask add 1 ml of 0.1 M thiourea solution 2 ml o 1384 ANALYST DECEMBER 1986 VOL. 111 Table 1. Structures and physical properties of hydrazones Elemental analysis,* O/O Hydrazone R VC,&m-l C PSPH . . . . H 1610 57.6 PSMPH . . . . CH3 1580 54.6 PSCPH . . . . C1 1610 52.7 PSNPH . . . . NO2 1600 47.2 (57.61) (58.69) (52.5 1) (51.12) * Values in parentheses are calculated values. H 4.0 4.5 (4.38) 3.5 2.8 (3.28) (3.99) (3.37) N S 15.1 8.8 (15.81) (9.05) 13.5 7.7 (15.21) (8.70) 14.3 (14.41) 16.4 8.4 (17.54) (8.03) -2 x 10-3 M PSNPH solution and 2 ml of 1 M chloroacetate buffer solution (pH 2.8) and dilute to the mark with water.Measure the absorbance of the resultant solution at 496 nm against a reagent blank. Second-derivative spectrophotometry When the cobalt content of the solution prepared by the procedure described under Ordinary spectrophotometry is too low to give a measurable absorbance record the second-derivative spectrum from 600 to 350 nm against a reagent blank by using a combination of both the first- and second-order differentiation circuits of No. 6 and a scan speed of 600 nm min-1 and measure the second-derivative value (the vertical distance from a peak to a trough or that from the base line to a trough of the peak). Dissolution and pre-treatment of iron and steel samples Prepare 50 ml of the sample solution by decomposing about 0.2 g of the sample with aqua regia and perchloric acid then remove the iron by extraction with 4-methylpentan-2-one in the same way as described previously.8 Use an appropriate aliquot of the resultant solution (ca.1 ml) for the determina-tion of cobalt. Results and Discussion Properties and Characteristics of the Hydrazones The infrared spectra of the synthesised hydrazones were measured with potassium bromide discs to confirm their structures. The spectra had absorption bands assigned to the stretching vibration of an azomethine bond (-N=C<) around 1600 cm-1. On the basis of these results and those of the elemental analysis shown in Table 1 the synthesised hydra-zones are considered to have the structures shown in Table 1.All the synthesised hydrazones are soluble in water and very soluble in alkaline solution. As is apparent from the structure each hydrazone is a dibasic acid as well as a triacidic base having five dissociable hydrogens in acidic solutions above 1 M. However in less acidic solutions the de-protonations from the C=N nitrogen and the sulphonic acid group are complete. Hence each hydrazone exists in solution in any of the following forms, depending on pH: Ka3 Kaz Ka 1 H3L+ H2L F HL- G L2-where L denotes the undissociable part of the hydrazone and K K and K~~ are the acid dissociation constants a i a2 Table 2. Acid dissociation constants of hydrazones. Ionic strength = 0.2; and temperature = 25 k 0.1 "C Acid dissociation constant Hydrazone PKa PKa PK3 PSPH .. . . . . 14.93 5.88 3.41 PSMPH . . . . . . 15.55* 6.24 3.75 PSCPH . . . . . . 14.09* 4.77 2.62 PSNPH . . . . . . 11.49 4.37 0.92 personal computer. * Ionic strength = 3.0 calculated by the simplex method with a I 1 I 0.0 0.4 0.8 a Fig. 1. pK value. 0 PSMPH; . PSPH; x PSCPH; and @ PSNPH Plots of substituent constants for the Hammett equation vs. The acid dissociation constants were determined spectro-photometrically at an ionic strength of 0.2 at 25 ? 0.1 "C. The results are given in Table 2. Ka3 and Ka2 may be assigned to the protonation of the pyridine-nitrogen in the hydrazine and the ketone moieties respectively whereas Kal is due to the de-protonation of the imino group. The Hammett equation, log (KaIKao) = ap was applied to the acid dissociation constants obtained where Kao and K refer to the acid dissociation constants of PSPH or its substituted derivatives (PSMPH PSCPH and PSNPIj) respectively and o and p are the substituent and the proportionality constants respec-tively.9 As can be seen from Fig.1 a logarithmic graph of the acid dissociation constant against the Hammett's substituent constant gives straight lines. This indicates that the inductive effect of the substituents on the pyridine ring reflects the pKa values of protons according to Hammett's law ANALYST DECEMBER 1986 VOL. 111 1385 Table 3. Reactivity of PSMPH with different metal ions PH4 PH7 pH 10 ~*/1 mol-1 cm-1 &*/I mol-1 cm-1 ~ * / 1 mol-* cm-* Metal ion .. . . . . . . 450 488 19 000 32 000 -488 -31 000 -488 L 32 000 CO'I . . . . . . Cr"' . . . . . . CrV' . . . . . . CU" . . . . . . 485 392 17 000 14 000 480 388 20 000 42 000 -17 000 18 000 Fe'I . . . . . . 392 395 Fe"' . . . . . . HgII . . . . . . Mn" . . . . . . -455 450 500 ---4000 13 000 9 000 ---450 510 --37 000 10 000 -NiT' . . . . . . Pb" . . . . . . Pd" . . . . . . Ti'V . . . . . . VIV . . . . . . -450 21 000 vv . . . . . . Zn" . . . . . . -448 -35 000 * E = Molar absorptivity. Table 4. Reactivity of PSCPH with different metal ions PH 4 PH7 pH 10 h,,,,,,/nm ~ / 1 mol-1 cm-1 Metal ion Agl . . . . . . Cd" . . . . . . CO" . . . . . . ~ / 1 mol-1 cm-1 ~ / 1 mol-1 cm-1 450 22 000 490 31 000 -490 --34 000 -490 -34 000 Cr"' .. . . . . CrVT . . . . . . CU" . . . . . . Fe" . . . . . . -475 388 388 -21 000 32 000 30 000 - -470 23 000 388 43 000 - -- -455 5 000 450 44 OOO 530 14 000 - --525 420 -13 000 23 000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . -455 540 495 448 ----17 000 10 OOO 10 000 9 000 ----540 -8 000 . . . . . . . . . -5 000 -- -448 45 000 . . Table 5. Reactivity of PSNPH with different metal ions PH4 PH7 pH 10 Metal ion ~ / 1 mol-1 cm-1 hrnax./nm 480 475 496 ~ / 1 mol-1 cm-1 3 000 41 000 55 000 hmax.Jnm 480 475 496 ~ / 1 mol-1 cm-1 23 000 84 000 55 000 Agl .. . . . . Cd'I . . . . . . CoI1 . . . . . . -496 --480 470 470 --480 520 493 500 475 ---56 000 Crl" . . . . . . CrV' . . . . . . CUT' . . . . . . Fe" . . . . . . Ferrx . . . . . . -35 000 48 000 42 000 --22 000 24 000 21 000 29 000 1000 ---480 470 430 480 480 480 480 520 493 500 475 --53 000 57 000 42 000 65 000 1000 77 000 3 000 23 000 30 000 14 000 83 OOO --480 470 430 480 480 480 480 500 -72 000 54 000 44 000 33 000 74 000 78 000 11 000 30 OOO HgII . . . . . . Mn" . . . . . . Ni" . . . . . . Pbl' . . . . . . Pd" . . . . . . Ti1v . . . . . . VIV .. . . . . vv . . . . . . Zn" . . . . . . 500 475 1000 87 OOO coloured complexes with high molar absorptivities in solutions of relatively low pH. The chromogenic reaction of PSNPH, which has a strong electron-withdrawing nitro group is highly sensitive for the metal ions and the molar absorptivity of its cobalt(II1) complex is 5.69 X lo4 1 mol-1 cm-1 considerably larger than those of the other metal complexes. This suggests Reactivity of the Hydrazones with Metal Ions The reactivities of the synthesised hydrazones with metal ions at pH 4 7 and 10 were investigated and the results are summarised in Tables 3 4 and 5. Every hydrazone reacts with cobalt(II) copper(II) iron(II) palladium(II) etc. to giv 1386 0.6 E c 0.4 (D m d c ANALYST DECEMBER 1986 VOL.111 ( a ) --0.6 Q p 0.4 -e Q 0.2 v) B 0.0 400 500 Wavelengthhrn 600 Fig. 2. Absorption spectra of (A-D) PSNPH and (E) its cobalt(II1) complex. PSNPH 2.0 X 7; D 13; and E 3. CoIII 1.1 x 10-4 M M; reference water. pH A 0; I I I I I I 0 2 4 6 8 10 0.4 PH Fig. 3. Effect of uH on (a) the complex formation and stibility of the complex. CoiI,' 1.1 X 10-5 M; PSNPH 1.6 x thiourea 4 x 10-3 M; reference reagent blank. (a) pH adjusted before the complexation; (b) pH readjusted after the complexation was complete at pH 3.8 B 3; C', (b) the 10-4 M; that PSNPH may be useful for the sensitive determination of cobalt and the conditions for this complexation were exam-ined in detail. Absorption Spectra PSNPH reacts with cobalt(I1) in weakly acidic to alkaline media to give a red complex.Qualitative tests10 on the complex indicate that the cobalt in the complex is oxidised rapidly to the tervalent species by dissolved oxygen. Fig. 2 shows the absorption spectra of PSNPH and its cobalt(II1) complex. The spectrum of PSNPH varies with pH but that of the complex remains unchanged between pH 0 and 10. In acidic and neutral solutions the spectrum of the complex the absorption maximum of which is at 496 nm is easily distinguishable from that of PSNPH. Effect of pH Fig. 3 shows the effect of pH on the complex formation and the stability of the cobalt(II1) complex. The complex is formed above pH 1.9 and a maximum and constant absorbance is obtained in the pH range 2-6. The complex once formed is very stable and remains unchanged even on varying the pH value between 0 and 10 after the complexation is complete.Effect of PSNPH Concentration and Stoicheiometry A molar-ratio study showed that a 3-fold molar excess of PSNPH is necessary for complete complexation to occur and that the excess of PSNPH does not interfere. The continuous variation method revealed the formation of a 1 2 (metal to ligand) complex as shown in Fig. 4. 1 .o E 0.8 C (D m d (0 0, C 0.6 0.4 + g n Q 0.2 0 0.2 0.4 0.6 0.8 1.0 [ C O ~ ~ ] / { [Co1l1+ [PSNPHI} Fig. 4. Continuous variation graph for the cobalt - PSNPH system. [Co] + [PSNPH] = 5.3 x 10-5 M; pH 3.9; reference reagent blank r l B I k 0.6 (D m d 4-0.4 u C (D + 0.2 a 13 d I I I I I 1 0 10 20 30 40 50 Timeirn i n Fig.5. PSNPH 1.6 X 10-4 M; reference reagent blank. A In the absence of thiourea; and B in the presence of 4 x 10-3 M of thiourea Rate of complex formation at pH 2.8. Co" 1.1 x Rate of Complex Formation The formation of the cobalt(II1) - PSNPH complex was not as rapid as predicted. As shown in Fig. 5 more than 30 min was required before a maximum and constant absorbance was obtained at room temperature in an acidic solution. An accelerator for the complexation reaction was therefore sought and thiourea was found to be the most effective and suitable reagent. In the presence of a small amount of thiourea the complexation is complete within a few minutes, as is also seen in Fig. 5. Calibration Graph The calibration graph for the determination of cobalt was prepared by the recommended procedure.The graph confor-med to Beer's law and gave a straight line in the range 0.05-1.0 pg ml-1 of cobalt. The equation of the line obtained by a least-squares treatment was Co (pg ml-1) = 1.04 A where A is the absorbance. The molar absorptivity and Sandell's sensitiv-ity calculated from this equation were 5.69 X 104 1 mol-1 cm-1 and 1.04 ng cm-2 of Co respectively. The precision of the proposed method was determined for the standard solutions containing 15.7 pg of cobalt(II) the coefficient of variation being 0.13 70. Effect of Foreign Ions The interference caused by various other ions was studied. The recovery of 15.7 pg of cobalt was examined in 25 ml aliquots of solution containing about 100 mg of foreign anion or 1000 pg of foreign cation.The results are given in Table 6. No anion interferes except EDTA a strong chelating agent but cations such as copper(II) palladium(II) iron(II), iron(II1) and nickel(I1) interfere when no masking agent is added as they form coloured complexes with PSNPH with spectra similar to that of the cobalt(II1) complex. In the recommended procedure however copper( 11) and a smal ANALYST DECEMBER 1986 VOL. 111 1387 Table 6. Effect of various ions on the determination of cobalt and elimination of interferences by adding masking agents. Cobalt(I1) taken 15.7 pg Relative Ion Addedlpg error YO F- C1- Br- C104- NO3- SCN-, ascorbate citrate oxalate tartrate, thiourea . . . . . . . . . . 4000 0 I- Po43- so,2- .. . . . . . . ' 4000 -2 EDTA . . . . . . . . . . . . 800 - 93 A P Call Mg" Mn" VIV Vv Znrl . . 40 0 Hg I1 40 -2 Cd" Pb" 40 2 . . . . . . . . . . . . . . . . . . . . . . . . . . Agl Cr"1 40 4 . . . . . . . . . . . . CrV1 4 17 CU" 40 0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pd" 40 0" Fe" 40 0t Ferrl . . . . . . . . . . . . . . 40 0$ Ni" . . . . . . . . . . . . . . 40 37 . . . . . . . . . . . . . . . . . . . . . . . . . . . . * 1 ml of 5% ascorbic acid was added precipitates formed were filtered off. t 2 ml of 1 M sodium fluoride were added after the Ferr had been oxidised to FeIII. $ 2 ml of 1 M sodium fluoride were added. 7 10 ml of 5 x 10-3 M PSNPH were used and 5 ml of 1 M hydrochloric acid were added after complexation was complete.Table 7. Determination of cobalt in standard iron and steel samples Cobalt content % Sample Proposed Certified Sample takenlg method? value JSS-611-5* . . 0.2000 0.36 2 0.01 0.36 0.1505 0.36 k 0.01 JSS-606-5* . . 0.2000 0.45 k 0.01 0.46 0.1510 0.45 f 0.01 * JSS Japan Standard Sample. i Average of triplicate determinations. amount of palladium(I1) are masked by the thiourea added as an accelerator for the cobalt(II1) complex formation although a pale yellow colour of a palladium(I1) - thiourea complex remains. The interference due to large amounts of palla-dium(I1) can be avoided by adding ascorbic acid which precipitates it as metallic palladium that can then be filtered off. Sodium fluoride masks iron(III) and iron(II) when present is easily oxidised to iron(II1) by heating the solution after adding a small amount of nitric acid.Although nickel(I1) interferes seriously this interference is minimised by adding excess of PSNPH. Hence the proposed procedure can be made essentially free from interferences from other ions. Application to Steel Samples To confirm the practicability of the proposed method it was applied to the determination of cobalt in steel samples after removing iron as described. The results summarised in Table 7 are in good agreement with certified values. The accuracy and reproducibility of the results are satisfactory. Use of Analogue Derivative Spectrophotometry to Improve Sensitivity Derivative spectrophotometry using the analogue differentia-tion circuit enhances the sensitivity of ordinary spectro-photometry,6-7 and therefore second-derivative spectropho-tometry was applied here to the determination of cobalt.In second-derivative spectrophotometry both the time constant of the analogue differentiation circuit and the scan speed of the spectrophotometer considerably affect the 0 20 40 60 Cobalting ml-' Fig. 6. Calibration graph for cobalt in second-derivative spectropho-tometry. A Peak-to-trough values; and B base line-to-trough values. Circuits all No. 6; scan speed 600 nm min-I; slit width 1 nm; recorder sensitivity X 1; reference reagent blank; PSNPH 1.5 x 10-5 M. Co a 3.1; b 6.3; c 12.5; d 25.0; e 43.8; and f 62.6 ng ml-1. pH 2.8 sensitivity and selectivity so both need to be optimised to give a well resolved large peak.According to an experimental result concerning this a combination of circuit No. 6 (which has the largest time constant 2.0 s the time constant increasing with circuit number in our apparatus) and a scan speed of 600 nm min-1 was found to give the best sensitivity and resolution (i. e. selectivity) for the cobalt determination. The calibration graph prepared under the recommended conditions by plotting the second-derivative value versus the cobalt concentration is linear and passes through the origin when either the peak to trough values or the base line to trough values are plotted as shown in Fig. 6. The equations for the lines were Co (ng ml-1) = 31.40 and Co (ng ml-1) = 44.60 where 0 is the second-derivative value converted into absorbance. It will be seen that cobalt can easily be deter-mined in this way at levels as low as 3 ng ml-1, Conclusions Three water-soluble hydrazones were synthesised and the complexation of one of them PSNPH with cobalt(I1) was investigated in detail. On the basis of the results spectropho-tometric and analogue derivative spectrophotometric methods for the determination of trace amounts of cobalt have been proposed and applied successfully to the analysis of steel samples. The proposed methods offer the advantages of simplicity rapidity reasonable selectivity and high sensitivity. Very high sensitivity was attained on employing second-derivative spectrophotometry. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. References Katyal M. and Dutt Y. Talanta 1975 22 151. Singh R. B. Jain P. and Singh R. P. Talanta 1982,29,77. Going J . E. and Sykora C. Anal. Chim. Acta 1974,70,127. Aita T. Odashima T. and Ishii H. Analyst 1984,109,1139. Bradsher C. K . Parham J. C. and Turner J. D., J. Heterocycl. Chem. 1965 2 228. Ishii H. and Koh H . Nippon Kagaku Kaishi 1980 203. Ishii H. and Satoh K. Fresenius Z. Anal. Chem. 1982,312, 114. Singh R. B. Odashima T. and Ishii H. Analyst 1984 109, 43. Taft R. W. J. Am. Chem. SOC. 1953 75 4231. Odashima T. and Ishii H. Bunseki Kagaku 1977 26 698. Paper A61154 Received May 20th 1986 Accepted July 3rd 198

ISSN:0003-2654    

DOI:10.1039/AN9861101383

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1389 Chromatographic 1,IO-Phenanthroli Naphthalene Pre-concentration of Iron(ll1) with ne and Tetraphenylborate Supported on Tohru Nagahiro" and Katsuya Uesugi Department of Chemistry, Himeji Institute of Technology, Himeji 671-22, Japan and Masatada Satake Faculty of Engineering, Fukui University, Fukui 910, Japan A method has been established for the determination of trace amounts of iron employing chromatographic pre-concentration of iron(l1) with 1'10-phenanthroline and tetraphenylborate supported on naphthalene. The optimum pH range for the adsorption is 3.2-6.2. Naphthalene containing iron(ll) complexes is dissolved in acetonitrile and the absorbance is measured at 508 nm. Beer's law is obeyed over the concentration range 2.0-32.0 pg of iron in 10 ml of acetonitrile solution.The molar absorptivity was calculated to be 1.13 x lo4 I m0I-l cm-1 at 508 nm. Seven replicate analyses of a sampCe solution containing 16.0 yg of iron gave a mean absorbance of 0.325 with a relative standard deviation of 0.31%. The method has been successfully applied to the determination of iron in aluminium alloy and powder and sea water samples. Keywords : Iron determination; spectrop ho tometry; naphthalene adsorption; I , 10-p h enan th roline; tetraphen ylborate A method of solid - liquid separation after liquid - liquid extraction using naphthalene as an extractant1-7 can be useful for the extraction of metal complexes that have a low solubility in non-aqueous organic solvents. However, the method has the disadvantage that the operation is carried out at high temperature and therefore cannot be applied to the extraction of metal ions which form thermally unstable complexes. In order to overcome this drawback, a second method of solid - liquid separation after the adsorption of metal com- plexes on microcrystalline naphthalene has been devel- oped.Sl4 These methods are especially useful for metal complexes that have a low solubility in non-aqueous organic solvents, but is more complicated than the usual liquid - liquid extraction method.In this work, a chromatographic method has been devel- oped for the selective pre-concentration of iron from an aqueous phase using 1,lO-phenanthroline - tetraphenylborate (TPB) - naphthalene as an adsorbent. This method is simple and trace amounts of iron can be pre-concentrated from a large volume of aqueous phase.It is assumed that the iron(I1) reacts with the ion pair, which consists of protonated 1,lO-phenanthroline and the TPB anion adsorbed on naph- thalene. The adsorbed iron(I1) can be dissolved with aceto- nitrile from the column together with the adsorbent and is determined spectrophotometrically . Various parameters have been evaluated and the method has been successfully applied to the determination of iron in a standard reference material and practical samples. Experimental Apparatus A Hitachi Model 624 spectrophotometer was used for all absorbance measurements. pH measurements were made with a Hitachi-Horiba Model F-7LC pH meter equipped with a combination calomel and glass electrode assembly.A 150 X 8 mm i.d. glass tube was used as a chromatographic column. * To whom correspondence should be addressed. Reagents All reagents were of analytical-reagent grade and de-ionised, distilled water was used throughout. Standard iron(III) solution, 4.0 pg ml-1. Prepared by diluting 4 ml of a standard stock solution of iron(II1) (1000 pg ml-1) to 1000 ml with distilled water and 16.6 ml of 20% hydrochloric acid. Adsorbent. A 2.92 g amount of TPB and 20.0 g of naph- thalene were dissolved in 100 ml of acetone. The solution was then transferred into 300 ml of 0.1 M hydrochloric acid containing 2.00 g of 1 ,lo-phenanthroline hydrochloride in a fast stream with continuous stirring. Naphthalene co-precipi- tated with 1 ,lo-phenanthroline and TPB was filtered through a filter-paper by suction, dried in a desiccator containing calcium chloride for a day and then stored in a refrigerator.Hydroxylammonium chloride solution, 10%. Prepared in distilled water. Buffer solution, pH 4.0. Prepared by mixing 1 M acetic acid and 1 M sodium acetate. General Procedure A 1 ml aliquot of 10% hydroxylammonium chloride and 5 ml of buffer solution (pH 4.0) are added to about 14 ml of sample solution containing 2-32 pg of iron(II1) in a 50-ml separating funnel and the pH is adjusted to 4.0 with 1 M sodium hydroxide solution. A 10 ml aliquot of buffer solution (pH 4.0) is passed through the chromatographic column loaded with 0.5 g of adsorbent at a flow-rate of 1 ml min-1. The sample solution is then added and the column is washed with distilled water.The coloured adsorbent is dissolved with acetonitrile and diluted to 10 ml. A portion of the solution is transferred into a 1 cm cell and measured at 508 nm against a reagent blank. Results and Discussion Absorption Spectra The absorption spectra of the reagent and the iron(I1) - 1 , l O - phenanthroline - TPB complex in naphthalene - acetonitrile solution were measured against water (Fig. 1). The complex1390 ANALYST, DECEMBER 1986, VOL. 111 has a maximum absorption at 508 nm, where the absorption of Effect of Flow-rate and Volume of Aqueous Phase the reagent is very low. All absorbance measurements were therefore made at 508 nm in subsequent studies. Effect of pH The effect of pH on the adsorption of iron (16.0 pg) was studied by the general procedure.The optimum pH range is 3.2-6.2. Subsequent studies were conducted at pH 4.0. 0.4 1 1 0 L 450 Reagent blank I I 500 550 Wavelengthlnm Fig. 1. Absorption spectra of reagent and FelI complex in naph- thalene - acetonitrile solution. FeITT taken, 16.0 pg; reference, water Table 1. Effect of foreign ions and salts. Fe"1 taken, 16.0 pg Ion or salt Tolerance limit Al"' . . . . . . . . . . 9.5 mg Ca" 500 mg . . . . . . . . . . Cd" . . . . . . . . . . 55 pg COT' . . . . . . . . . . 13 pg CUT' . . . . . . . . . . 20 pg Ni" . . . . . . . . . . 500 pg Pb" . . . . . . . . . . 3 mg Pd" . . . . . . . . . . 2mg Snl' . . . . . . . . . . 200 Pg WV' . . . . . . . . . . 500 pg Zn" . . . . . . . . 12 pg NaCI04.H20 log NaI 2g NaSCN 5g Cr"' . . . . . . . . . . 10 mg MgII .. . . . . . . . . 160 mg Mn" 15 mg . . . . . . . . . . . . . . . . . . Na2C204 150 pg Sodiumcitrate.2H20 . . . . 1.5 mg NaH2P04.2H20 . . . . 50 mg . . . . . . . . . . . . . . . . . . . . . . . . Sodium tartrate.2H20 . . 75 mg The effect of flow-rate on the adsorption of iron (16.0 pg) was studied by the general procedure. The flow-rate was varied from 0.5 to 7.0 ml min-1. The column was aspirated when the flow-rate was more than 2 ml min-1. The flow-rate did not affect the adsorption in this range. The volume of aqueous phase containing 16.0 pg of iron was varied from 10 to 1500 ml and the adsorption remained maximum and constant up to 1500 ml. Subsequent studies were carried out at about 20 ml for convenience. Calibration Graph and Sensitivity A calibration graph for iron was constructed by the general procedure. It was linear over the concentration range 2.0- 32.0 pg of iron in 10 ml of acetonitrile solution.The molar absorptivity was calculated to be 1.13 x 104 1 mol-1 cm-1 at 508 nm. Seven replicate analyses of a sample solution containing 16.0 pg of iron gave a mean absorbance of 0.325 with a relative standard deviation of 0.31%. This method is more sensitive than the method using 2,2'-bipyridyl (molar absorptivity, E = 8650),15 but less sensitive than the methods using 1 ,lo-phenanthroline (E = 20 500),16 4,7-diphenyl-l,10- phenanthroline (E = 22 400),17 2,4,6-tris(2'-pyridyl)-s-triazine (E = 24 100)1* or 3-(4-phenyl-2-pyridyl)-5,6-diphenyl-1,2,4- triazine (E = 28700).19 However, this method has some advantages as it is simple and trace amounts of iron can be pre-concentrated from a large volume of aqueous solution.Effect of Foreign Ions Sample solutions containing 16.0 pg of iron(II1) and various amounts of alkali metal salts or metal ions were prepared and the determination of iron was carried out using the general procedure. The tolerance limits of foreign ions (error <3%) are given in Table 1. The method is fairly selective and may be applied to the determination of iron in various materials. Application of the Method to an Aluminium Alloy and Metallic Aluminium Powder The proposed method has been successfully applied to the determination of iron in a standard aluminium alloy (NBS, SRM-85B) and a metallic aluminium powder (Wako Chemi- cals, 99.5%). The samples were the same as were prepared by Nagahiro et a1.20 The procedure for the determination of iron is as follows.To about 14 ml of solution containing 6.09 x g of aluminium alloy or 8.78 x 10-3 g of metallic aluminium powder in a 50-ml separating funnel, add 1 ml of 10% hydroxylammonium chloride, 5 ml of buffer solution Table 2. Analysis of samples of iron Iron content,% Iron certified Present 1 ,10-Phenanthroline Sample Composition, % value,% method method NBS SRM-85B (aluminium alloy) . . Cu 3.99, Mg 1.49, 0.24 0.235 k 0.001* Mn0.61, Cr0.211, Si 0.18, Ni 0.084, Zn 0.030, Ti 0.022, Pb 0.021, Ga 0.019, V 0.006 Aluminium powder Sea water (coastal sea water in (Wako Chemicals, 99.5%) . . . . - - 0.165 -t 0.0011- Himeji, Japan) . . . . . . . . - - 12.5 p.p.b. 0.161 k 0.001 * Average of five determinations.t Average of six determinations,ANALYST, DECEMBER 1986, VOL. 111 1391 (pH 4.0) and 1.0 g of thiourea to mask copper(1) (for aluminium alloy). It is not necessary to add thiourea to the metallic aluminium powder sample. From this point, the general procedure was applied. The results are given in Table 2. The result for metallic aluminium powder is in reasonable agreement with that obtained by the 1 ,lo-phenanthroline method .20 Determination of Iron in Sea Water A 3.4 ml aliquot of 20% hydrochloric acid was added to 2000 ml of sea water and the solution filtered through a filter- paper. Iron in sea water was determined by a standard additions method. A 150 ml aliquot of sea water was pipetted into a series of 300-ml separating funnels.Volumes of 1 , 2 and 4 ml of iron(II1) solution (4.0 yg ml-1) were then pipetted into all but one of these separating funnels. This was followed by the addition of 1 ml of 10% hydroxylammonium chloride and 10 ml of buffer solution (pH 4.0) and the mixture was passed through the chromatographic column loaded with 0.5 g of adsorbent by suction at a flow-rate of 3 ml min-1. The general procedure was then applied. The results are given in Table 2. References Fujinaga, T., Satake, M., and Yonekubo, T. , Bull. Chem. SOC. Jpn. , 1973, 46, 2090. Fujinaga, T., Satake, M., and Yonekubo, T., Bull. Chem. SOC. Jpn., 1975, 48, 899. Satake, M., Anal. Chim. Acta, 1977, 92, 423. Kumar, A., Hussain, M. F., Satake, M., and Puri, B. K., Bull. Chem. SOC. Jpn., 1982, 55,3455. 5 .Wasey, A., Bansal, R . K., Satake, M., and Puri, B. K., Bunseki Kagaku, 1983, 32, E 211. 6. Satake, M., Nagahiro, T., and Puri, B. K., Analyst, 1984,109, 31. 7. Nagahiro, T., Uesugi, K., Satake, M., and Puri, B. K., Bull. Chem. SOC. Jpn., 1985, 58, 1115. 8. Satake, M., Matsumura, Y., and Fujinaga, T., Talanta, 1978, 25, 718. 9. Fujinaga, T., Takagi, Y., and Satake, M., Bull. Chem. SOC. Jpn., 1979, 52, 2556. 10. Satake, M., Matsumura, Y . , and Mehra, M. C., Mikrochim. Acta, 1980, I, 455. 11. Satake, M., Mehra, M. C., and Fujinaga, T., Bull. Chem. SOC. Jpn., 1982, 55, 2079. 12. Satake, M., Mehra, M. C., Singh, H. B., and Fujinaga, T., Bunseki Kagaku, 1983, 32, E 165. 13. Nagahiro, T., Uesugi, K., Mehra, M. C., and Satake, M., Talanta, 1984, 31, 1112. 14. Nagahiro, T. , Satake, M., Lin, J. L. , and Puri, B. K. , Analyst, 1984, 109, 163. 15. Moss, M. L., and Mellon, M. G., Znd. Eng. Chem. Anal. Ed., 1942, 14, 862. 16. Rao, V. P. R., Rao, K. V., and Sarma, P. V. R. V., Talanta, 1969, 16, 277. 17. Smith, G. F., McCurdy, W. H . , Jr., and Diehl, H., Analyst, 1952, 77,418. 18. Collins, P. F., Diehl, H. , and Smith, G. F. , Anal. Chem. , 1959, 31, 1862. 19. Schilt, A . A . , Talanta, 1966, 13, 895. 20. Nagahiro, T., Uesugi, K., and Satake, M., Anal. Sci., 1985,1, 359. Paper A61196 Received June 16th, 1986 Accepted July 29th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101389

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

PAGE MISSINGPAGE MISSINGPAGE MISSINGPAGE MISSING

ISSN:0003-2654    

DOI:10.1039/AN9861101393

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

PAGE MISSINGPAGE MISSINGPAGE MISSINGPAGE MISSING

ISSN:0003-2654    

DOI:10.1039/AN9861101397

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

PAGE MISSINGPAGE MISSINGPAGE MISSINGPAGE MISSING

ISSN:0003-2654    

DOI:10.1039/AN9861101401

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

PAGE MISSINGPAGE MISSINGPAGE MISSINGPAGE MISSING

ISSN:0003-2654    

DOI:10.1039/AN9861101405

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1409 Applications of Xanthene Derivatives in Analytical Chemistry Part LVII.* Spectrophotometric Determination of Minocycline Using Gallium and Eosint ltsuo Mori,S Yoshikadzu Fujita, Hiroshi Kawabe, Kinuko Fujita, Takeshi Tanaka and Atsuko Kishimoto Osaka University of Pharmaceutical Sciences, 580, Ma tsu ba ra-s h i, Osaka, Japan A spectrophotometric method for the determination of micro-amounts of minocycline (MINO) as a tetracycline derivative in weakly acidic media is proposed. It is based on the formation of a ternary complex between Eosin (sodium 2,4,5,7-tetrabromofluorescein), gallium and MINO in the presence of poly(viny1 alcohol). Under optimum conditions, minocycline hydrochloride (MINO.HCI) was determined in the range 0-40 pg in I 0 ml of solution at 545 nm.The apparent molar absorptivity of MINO.HCI was 1.1 x lo5 I mol-1 cm-1 with a Sandell sensitivity of 0.0047 pg cm-2 MINO.HCI. The method was applied to the determination of MINO.HCI in pharmaceutical preparations. Keywords: Spectrophotometry; minocycline determination; gallium(IIl) - Eosin - MINO ternary complex; tetracycline derivative Tetracycline (TC) derivatives such as minocycline (MINO) are bacteriostatic antibiotics and various methods for the determination of TC derivatives have been reported,2-11 including fluorometric and spectrophotometric methods. Many spectrophotometric methods based on the complexa- tion of TC derivatives with metal ions have been reporteds-ll for TC antibiotics containing an enolysed hydroxy group of a 1,3-diketone, such as alizarin derivatives.Metal ions such as thorium, zirconium and aluminium have been used. We have already reported11 the determination of mino- cycline (MINO) using o-hydroxyhydroquinonephthalein (Qnph) , zirconium and fluoride ions in the presence of sodium dodecyl sulphate (SDS). Although this proposed method11 was highly sensitive, the colour reaction between Qnph, zirconium ions, fluoride ions and MINO was subject to severe interferences from coexisting metal ions. The colour reaction between Eosin (sodium 2,4,5,7-tetra- bromofluorescein,) as a dye, gallium as the metal ion and minocycline hydrochloride (MINO .HC1) as the TC derivative was studied in the presence of various surfactants, added to reduce the influence of metal ions. The essential conditions for a simple, rapid and sensitive spectrophotometric determina- tion of MINO were established by using the ternary complex in the presence of a non-ionic surfactant and the organic extraction step was eliminated.The proposed method was applied to the determination of MINO.HC1 in pharmaceutical preparations. Experimental Materials and Reagents All materials and reagents were of analytical-reagent grade unless specified otherwise. De-ionised water was used throughout. Standard MINO. HCl solution. A stock solution (1.0 X loF3 M) of MINO.HC1 was prepared according to Fujita et aLl1 * For Part LVI, see reference 1. t Presented at the 106th Annual Meeting of the Pharmaceutical Society of Japan, Chiba, April 1986. $ To whom correspondence should be addressed.Working solutions were prepared by dilution of this stock solution. Galliurn(ZZI) solution. A 1 .O x 10-2 M gallium stock solution was prepared by dissolving gallium metal (99.999%, Mitsuwa Chemicals, 69.0 mg) in 2 ml of concentrated hydrochloric acid by heating. The solution was cooled, diluted to 100 ml with de-ionised water and further diluted to give working solutions (5.0 x 10-4 M). Eosin solution, 1.0 X 10-3 M. Prepared as described previously. 12 Buffer solution. A 2.0 X 10-1 M sodium acetate - hydro- chloric acid buffer solution (Walpole acetate buffer) was used for the adjustment of pH. PoZy(viny1 alcohol)(PVA). A 0.5% aqueous solution of PVA (n = 500) was prepared without purification. Apparatus A Shimadzu UV-260 recording spectrophotometer with 1 .O- cm silica cells was used to measure the absorbance.A Hitachi-Horiba F-7 glass electrode pH meter was used. Standard Procedure A 1-ml volume of 0.5% PVA solution, 2.5 ml of sodium acetate - hydrochloric acid buffer solution (pH 3.2), 1 ml of 5.0 X M gallium solution and 1.5 ml of 1.0 X 10-3 M Eosin solution were added to a series of solutions containing 0-40 pg of MINO.HCI. The mixture was diluted in several 10-ml calibrated flasks to produce suitable working solutions. The absorbance of the Eosin - gallium - MINO solution (solution A) was measured at 545 nm against the Eosin - gallium solution (solution C) after 15 min. The concentration of MINO.HC1 was determined from a calibration graph. Results and Discussion Colour Reaction and Absorption Spectra Lines A, B and C in Fig.1 show the absorption spectra of Eosin - gallium - MINO solution (solution A), Eosin - (MINO) (Eosin) solution (solution B) and Eosin - gallium solution (solution C) in the presence of PVA at pH 3.2, respectively. The absorbance of solution A is larger than that of solution C and these solutions are stable. The absorbance of solution A1410 ANALYST, DECEMBER 1986, VOL. 111 0' I 1 I 1 400 450 500 550 Wavelengthinm Fig. 1. Absorption spectra of: A, Eosin - gallium - MINO solution; B; Eosin - MINO (Eosin) solution; and C, Eosin - gallium solution at pH 3.2. Conditions: Eosin, 6.0 X 10-5 M; gallium, 2.0 x 10-5 M; MINO 2.0 X 10-5 M; PVA, 0.05%; and reference, water Table 1. Effect of xanthene derivatives as reagents for the determina- tion of MINO.Reagents, 1.5 X 10-4 M; gallium, 5.0 x 10-5 M; MINO.HC1, 21.2 pg per 10 ml; PVA, 0.05%, pH, 3.2; reference, solution C Reagent Absorbance h,,,,/nm Fluorescein . . . . . . . . . . . . 2,7-Dichlorofluorescein . . . . . . Eosin . . . . . . . . . . . . . . 2,4,5,7-Trichlorofluorescein . . . . . . 3',4',5',6'-Tetrachlorofluorescein . . . . o-Hydroxyhydroquinonephthalein (Qnph) Pyrogallol Red . . . . . . . . . . Pyrocatechol Violet . . . . . . . . Gallein . . . . . . . . . . . . - 0 0 0.450 545 0.172 540 0 0.099 525 0 0 0 - - - - - Table 2. Effect of metal ions on the reaction system. Metal ions, 5.0 X 10-5 M; Eosin, 1.5 x 10-4 M; MINO.HC1, 21.2 pg per 10 ml; PVA, 0.05%; pH, 3.2: reference, solution C Metal ion Ga"' . . . . . . Al"' . . . . . . In111 SC1" .. . . . . La"' . . . . . . Fe"' . . . . . . ZrIV . . . . . . . . . . . . ThIV . . . . . . GeIV . . . . . . Mo04*-. . . . . . wo42- . . . . . . uo2+ . . . . . . PdT1 . . . . . . CU" . . . . . . Zn" . . . . . . Added as Chloride Nitrate Chloride Nitrate Chloride Sulphate Nitrate Nitrate Chloride Sodium Sodium Nitrate Chloride Nitrate Nitrate Absorbance 0.450 0.341 0.185 0.161 0.145 0.242 0.144 0.140 0.068 0.023 0.025 0.291 0.118 0.226 0.108 hmax./nm 545 545 545 540 540 540 545 545 540 540 545 540 545 540 540 The effect of metal ions on the reaction system was also investigated. Gallium was superior to various metal ions in terms of sensitivity, as shown in Table 2. Hence subsequent determinations were carried out with Eosin and gallium at 545 nm. Effect of pH and Surfactants The effect of pH was examined.Maximum and constant absorbance was seen in the pH range 3.0-3.6, obtained by adding 1.5-3.0 ml of Walpole buffer solution (sodium acetate - hydrochloric acid solution) in the final 10 ml. PVA (n = 500) was the most effective dispersing agent among the surfactants investigated, which included, 0.5% PVA (n = 500, 2000), poly(N-vinylpyrrolidone) (PVP), gum arabic, Brij-35, Triton X-100 and poly(oxyethy1ene) sorbitan monolaurate (Tween 20). Maximum and constant absorbance was obtained with 1.0 ml of 0.5% PVA (n = 500) solution in the final 10 ml. Effect of Eosin and Gallium Concentrations and Composition of the Complex The effect of Eosin volume on the absorbance of solution A was examined at various Eosin concentrations, the amounts of gallium and MINO.HC1 being kept constant.The maximum absorbance of solution A was obtained when the molar ratio of gallium to Eosin was 1 : 3 . The molar ratio and Job's methods showed that the molar ratio of gallium to Eosin in the reaction system is 1 : 3 in the presence of MINO. However, the molar ratio of MINO to gallium was found to be 1 : 1 by the molar ratio and Job's methods. It was therefore concluded that the coloured ternary complex in this reaction system can be expressed as (Eosin)3(gallium)(MINO). Consequently, all further work was carried out with 5.0 x 10-5 M gallium and 1.5 X 10-4 M Eosin in the final solution. Stability and Calibration Graph Although solutions B and C were very stable, the colour reaction systems (solution A) between Eosin, gallium and MINO were completed in a period of 10 min at 1-30 "C and its absorbance at 545 nm was constant for at least 2 h.The calibration graph for the determination of MINO was obtained by the standard procedure. Beer's law was obeyed for up to 40 pg of MINO.HC1 in the final solution. The sensitivity according to Sandell's scale was 0.0047 pg cm-2 for MINO.HC1 and the apparent molar absorptivity was 1.1 x 105 1 mol-1 cm-1 for MINO.HC1 at 545 nm. The reproducibility for 21.2 pg of MINO.HC1 (5 experiments) was 1.1%. at about 545 nm is proportional to the concentration of MINO.HC1. A similar relationship was observed when TC derivatives such as Ledermycin or Achromycin were used instead of MINO. The effect of dyes in this reaction system was studied by measuring the difference of absorbance between a dye - gallium - MINO solution and that of a dye - gallium solution in the presence of PVA.Eosin is the best reagent in terms of sensitivity and reproducibility among various xanthene dyes such as Gallein (pyrogallolphthalein) , Pyrogallol Red (pyro- gallol sulphonphthalein) , Pyrocatechol Violet (pyrocatechol sulphonphthalein) , Qnph, fluorescein, 2,7-dichlorofluores- cein, Eosin, 2,4,5,7-tetrachlorofluorescein and 3',4',5',6'- tetrachlorofluorescein. Interference of Foreign Substances The effect of various substances was examined under the standard conditions and the results are summarised in Table 3. Several metal ions (in 3-5-fold excess over MINO.HC1) such as copper, iron and thorium and several anions such as citrate and phosphate interfered with the determination, but most other foreign ions did not interfere in 100-200-fold excess over MINO.HC1.The influence of large amounts of iron and copper ions was eliminated by adding 1.0 x 10-2 M sodium cyanide solution to the final solution to form cyano complexes. Organic compounds, including caffeine, taurine, glucose, lactose, urea, starch, ascorbic acid, glycine and ampicillin, did not interfere in 50-1000-fold excess over MINO.HC1. Thiam- ine and creatine were tolerated in a 10- to 50-fold excess.ANALYST, DECEMBER 1986, VOL. 111 141 1 Table 3. Effect of foreign substances on reaction system, MINO.HCI taken, 21.2 pg per 10 ml; Eosin, 1.5 X lop5 M; PVA, 0.5%; pH, 3.2; reference, solution C M; gallium, 5.0 X Foreign substance FelI1 .. . . . . Th'" . . . . . . CU" . . . . . . Zn" . . . . . . MgII . . . . . . c1- . . . . . . H2P04- . . . . . . Citrate . . . . . . Taurine . . . . . . Ascorbicacid . . . . Caffeine . . . . . . Thiamine . . . . Lactose . . . . . . Urea . . . . . . Trichloroacetic acid . . Sulphosalicylic acid . . - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Added as - Sulphate Nitrate Nitrate Nitrate Nitrate Sodium Potassium Sodium Hydrochloride Amount added/ pg per 10 rnl - 4.5 9.3 12.7 261.5 194.5 709.1 388.0 42.3 500.6 1410.0 424.4 134.9 3600.0 2400.0 65.4 10.2 Molar ratio, foreign substance: MINO - 2 1 5 100 200 500 100 5 100 200 50 10 5000 1000 10 1 Absorbance at 545 nm 0.450 0.450 0.450 0.410 0.464 0.459 0.457 0.414 0.434 0.450 0.450 0.456 0.480 0.450 0.458 0.450 0.450 Table 4.Colour reaction between Eosin, gallium and various antibiotics. Eosin, 1.5 pH, 3.2; reference, solution C Calibration range/ Antibiotic pg per 10 ml Minocycline hydrochloride (MINO .HCl) . . . . . . . . . . 0-40 . Demethylchlortetracycline hydrochloride (Ledermycin) . . . . 0-50 Doxycycline hydrochloride (Vibramycin) . . . . . . . . . . 0-50 Tetracycline (Achromycin) . . . . . . . . . . . . . . 0-60 Doxorubicin hydrochloride (Adriamycin) . . . . . . . . . . - Daunorubicin hydrochloride (daunomycin) . . . . . . . . - X M; gallium, 5.0 X M; PVA, 0.05%; Apparent molar absorptivity/ Sensitivity1 1 mol-1 cm-1 pg cm-2 1.1 x 105 5.8 x 104 5.0 x 104 3.6 x 104 0.0047 0.0080 0.0085 0.0125 Tabie 5. Results for determination of MINO.HC1 in pharmaceutical preparations MINO.HCI found*/mg Sample Calculated/mg Proposed method Previous method" proposed method, YO variation, YO Recovery* of Coefficient of Capsule (270 mg) .. . . 100 102.1 100.0 98.2 1.5 Granule (1000 mg) . . . . 20 19.6 19.6 98.5 1.7 * Average recoveries from five determinations. 0 0.5 1 .o Molar ratio, [galliumieosin + gallium] Fig. 2. Molar ratio of gallium to Eosin in the ternary complex by Job's method of continuous variations in the presence of MINO and PVA. Conditions: MINO, 1.0 X 10-4 M; PVA, 0.05%; pH, 3.2; reference, reagent blank Reaction between Eosin - Gallium and Various Other Anti- biotics The colour reaction between Eosin, gallium and various other antibiotics was studied by the standard procedure described above.Antibiotics such as lincomycin, erythromycin cephalexin and ampicillin did not produce coloured com- pounds. The colour reaction of TC derivatives with Eosin - gallium was selective and the results as calibration range, apparent molar absorptivity and Sandell sensitivity are shown in Table 4. The application of the proposed method to the assay of daunomycin or Adriamycin was unsuccessful. E UJ d (0 0) C (0 0.4 e z n a 0 2.0 4.0 Molar ratio, [MI N Oig a I I i u m I Fig. 3. Molar ratio of MINO to gallium in the ternary complex by molar ratio method. Conditions: gallium, 1.5 x 10-5 M; Eosin, 4.5 X 10-5 M; PVA, 0.05%; pH, 3.2; reference, reagent blank Application to Pharmaceutical Preparations This method was applied to the determination of MINO.HC1 in capsules and granules and was compared with the previous procedure.11 The recovery of MINO.HC1 added to the samples was about 98.2-98.5%. The reproducibility of the proposed method compares favourably with that of other methods .&lo The proposed method is simple, rapid and selective for MINO as TC derivatives.The sensitivity (Sandell senistivity, 0.0047 pg cm-2 MINO.HC1) was about twice that of the previous method11 that used the Qnph - zirconium - fluoride ternary complex. The influences of foreign ions have been1412 ANALYST, DECEMBER 1986, VOL. 111 reduced by between 1 : 10 and 1 : 100. This method may be useful for the determination of TC derivatives such as MINO 'in pharmaceutical preparations such as capusles, granules and syrups, and in biological samples such as urine. References 1. Mori, I., Fujita, Y . , Fujita, K., Tanaka, T., Kawabe, H., and Koshiyama, Y . , Bull. Chem. SOC. Jpn., 1986, 34, 2585. 2. Poiger, H., and Shlatter, Ch., Analyst, 1976, 101, 808. 3. Regosz, A., Pharmazie, 1977, 32, 681. 4. Kohn, K. W., Anal. Chem., 1961,33, 862. 5. Mazor, L., and Papay, M., Fresenius Z . Anal. Chem., 1960, 175, 355. 6. Abdel-Khalek, M. M., and Mahrous, M. S . , Talanta, 1983,30, 792. 7. Mahrous, M. S., and Abdel-Khalek, M. M., Talanta, 1984,31, 289. 8 . Sakaguchi, T., Toma, M., Yoshida, T., Omura, H . , and Takasu, H., Chem. Pharm. Bull., 1958, 6 , 1. 9. Sakaguchi, T., Bunseki Kagaku, 1957, 6 , 662. 10. Hughes, D. W., and Wilson, W. L., Can. J. Pharm. Sci., 1973, 8, 67. 11. Fujita, Y., Mori, I., andKitano, S . , Chem. Pharm. Bull., 1983, 31, 4016. 12. Fujita, Y., Mori, I., Kitano, S., Kawabe, H., and Kamada, Y., Bull. Chem. SOC. Jpn., 1984, 57, 1828. Paper A611 77 Received June 6th, 1986 Accepted July 29th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101409

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 Use of a Silver - Gelatin Complex for the Determination Micro-amounts of Hydrazine in Water Tarasankar Pal, Durga S. Maity and Ashes Ganguly Department of Chemistry, Indian Institute of Technology, Kharagpur 72 1302, India 1413 Gelatin forms a weak complex with silver(1) in alkaline media. This complex was studied and was found to be suitable for the spectrophotometric determination of 0.65-2.62 pg mi-1 of hydrazine. The method has been successfully applied to the determination of hydrazine in water using simple, non-toxic and inexpensive chemicals. The molar absorptivity of the complex is 4.2 x lo4 I mol-1 cm-1 for the spectrophotometric procedure. The proposed method has a relative standard deviation of 0.5I0/o, a confidence limit (95%) for 13.000 pg of hydrazine (20 determinations) in solution of 12.985 _+ 0.031 and a Sandell sensitivity of 7.63 x 10-4 pg cm-2.The determination is free from interferences from a large number of foreign substances. Keywords: Hydrazine determination; water; spectrophotometry Hydrazine salts are familiar as antioxidants, photographic developers, preservatives, welding fluxes, oxygen scavengers, propellents, insecticides and blowing agents for plastics. Hydrazine is an irritant and is suspected of being carcinogenic and hence its determination at micro-levels is of interest. The determination of hydrazine is based on its oxidation to nitrogen and water. The reaction is not quantitative in acidic solutions, even in the presence of oxidising agents such as permanganate. Fluoride is an effective catalyst for following the hydrazine - permanganate reaction on a quantitative basis.1 Microgram amounts of hydrazine can be determined by several spectrophotometic methods.2-9 Other methods that have been proposed for the determination of hydrazine are based on oxidimetry followed by titrimetry,l&lg poten- tiometry,l"21 coulometry,22723 amperometry,10924 conduc- timetry's and gasometry.26-31 Some of these methods require special equipment, whereas others demand carefully con- trolled conditions,1+12 particularly the gasometric method. Inorganic oxidants can give rise to several side products,32 leading to variable results from the determination. Recently we reported a method for the determination of hydrogen sulphide33 in alkaline media and observed that even trace amounts of hydrazine interfered in this determination. This observation has led to the development of a spectrophoto- metric method for the determination of trace amounts (0.65-2.62 pg ml-1) of hydrazine by a one-step reduction technique.In this paper a simple, non-toxic, alkaline active reagent is proposed for the determination of micro-amounts of hydrazine salts. As the reduction of silver(1) by hydrazine proceeds rapidly through a one-step process to produce a silver sol stabilised by gelatin,33 it is easy to determine the hydrazine concentration in test solutions spectrophotometric- ally. Experimental Apparatus All absorbance measurements were made with a Cary-17D Varian digital spectrophotometer with 1-cm quartz cells.Reagents All reagents were of analytical-reagent grade. A fresh stock solution of hydrazine (1 X 10-2 M) was prepared by dissolving hydrazine sulphate in distilled water and was standardised by the KMn04 method.l The solution was then diluted to 10-3 M. Silver nitrate solution (1 X 10-1 M) was prepared by dissolving silver nitrate in distilled water. Gelatin solution (1%) was prepared by dissolving 1 g of powdered gelatin (0x0, Lon- don) in 100 ml of warm distilled water. Owing to the microbiological degradation of gelatin and the aerial oxidation of hydrazine in water, both solutions were prepared fresh daily. The hydrazine solution was standardisedl before use. Standard Procedure A 0.2-ml volume of 10-1 M silver nitrate solution and 2 ml of 1% gelatin solution were mixed in a 10-ml stoppered flask.The pH was adjusted to ca. 8 with 0.2 M NaOH and then 0.05-0.2 ml of hydrazine solution (10-3 M) was introduced and the flask was stoppered. After 30 min the mixture was diluted to 10.00 ml with distilled water and the absorbance was measured at 415 nm against a reagent blank. Results and Discussion Absorption Spectrum The absorption spectrum of the solution shows a maximum at 415 nm, without any scattering. The absorption due to the reagent alone is negligible at 415 nm when the gelatin concentration is kept at a minimum.34 Influence of pH The effect of pH on the formation of the silver sol was studied within the pH range 5.S11.0. The reduction of the silver - gelatin complex by hydrazine at room temperature (30 "C) is influenced by pH and goes quickly to completion between pH 7.5 and 11.0.It is therefore necessary to work under alkaline conditions using NaOH solution. pHs greater than 11.0 are not recommended as the gelatin solution is hydrolysed and measurements were therefore carried out within the pH range 7.5-1 1 .O. Effect of Silver(1) Concentration A series of solutions of the silver - gelatin complex were prepared containing 1 ml of 1% gelatin solution and various amounts (0.8493-5.0950 mg) of AgN03. The solutions were reduced with 0.1 ml of 10-3 M hydrazine solution (3.2032 pg of hydrazine) (see reaction below). The solution pH was adjusted to about 8 with NaOH solution. After 30 min the diluted (10 ml) solutions gave constant absorbance values if the solution contained 1-3 pg of silver nitrate for 0.0085 g of hydrazine.Masses below this range resulted in a non-stoi- cheiometric reduction of the gelatin with hydrazine and masses above this range de-stabilised the silver sol. 4Ag+ + N2H4 + 40H- = 4Ag" + Nz + H201414 ANALYST, DECEMBER 1986, VOL. 111 Effect of Gelatin Concentration A 2-ml volume of 1% gelatin solution was required for 1-3 mg of silver nitrate for the stabilisation of the sol and the stoicheiometric reduction of silver nitrate by hydrazine. An increase in the amount of gelatin did not hinder the determination, although the absorbance value due to the reagent blank increased. The reduction of silver ions in the absence of gelatin produced a black precipitate of silver metal, which is of no use for the determination of hydrazine.Effect of Time, Temperature and Colour Stability The time dependence of the reduction of the silver - gelatin complex with hydrazine was studied at temperatures between 25 and 40 "C, keeping all other factors constant. Constant absorbance values were seen only after 30 rnin of reaction. At temperatures below 25 "C the reduction rate was very slow; above 40 "C, fast reduction produced black precipitates of silver that gave irreproducible absorbance values. The yellow silver sol appeared in the aqueous phase and its intensity remained stable for 2-3 h at room temperature (30 "C). Direct sunlight and higher temperatures (>50 "C) decompose the yellow silver sol into a black precipitate, producing a decrease in the absorbance values. Influence of Reducing Atmosphere Man-made pollutants such as carbon monoxide, hydrogen sulphide and sulphur dioxide cause serious interferences, although instantaneous reactions do not take place with these pollutants; stoppered flasks should preferably be used for the absorbance studies.The determination using stoppered flasks required no other precaution if carried out within 45 min, but Table 1. Calibration graph of absorbance versus hydrazine concentration in solution Concentration of hydrazine in Absorbance at Sample No. solution/l0-5 M L a x . 1 0.500 0.210 2 1 .ooo 0.420 3 1 SO0 0.625 4 2.000 0.830 5 2.500 1.040 determination after a longer time period required the storage of the stabilised sol suspension in an inert (nitrogen) atmos- phere. The reduction potential of the N2 - N2H4 couple is +0.23 V, whereas that of the Ag - Ag+ couple is +0.80 V.Hence the instantaneous reduction of the AgI ion by hydrazine is possible, giving a silver sol with a A,,,, at 415 nm in gelatin solution. Calibration Graph A number of solutions of the silver - gelatin complex were prepared and reduced with various amounts (0.05 - 0.25 ml) of 10-3 M hydrazine at a pH of about 8. After 30 min the absorbance was recorded after dilution as described under Procedure. The results obtained are shown in Table 1. The absorbance was plotted against the corresponding amount of hydrazine. Beer's law was obeyed in the concentration range 0.65-2.62 pg ml-1 of hydrazine in aqueous solution. Effect of Foreign Substances The method detailed above using an alkaline medium is interesting as it does not give the usual intense yellow colour with most of the metal ions that are supposed to interfere with methods reported in the literature.These ions are tolerated in 250 330 410 490 570 650 Wavelengthinrn Fig. 1. Absorption spectra at pH 8.0 of: A, the silver sol; silver - gelatin complex (reagent blank) and B, Table 2. Effect of foreign substances on the determination of hydrazine. Hydrazine taken, 13.0 pg ml-1 Foreign Amount substance tolerated/pg Added as Corl* . . . . . . 1000 Nitrate Ni"* . . . . . . 1000 Nitrate Mo"' . . . . . . 100 Ammonium salt Fe"* . . . . . . 500 Sulphate Mn" . . . . . . 1000 Acetate CUII* . . . . . . 500 Nitrate Ag' . . . . . . 300 Nitrate Zn" . . . . . . 1000 Sulphate PbT' . . . . . . 1000 Nitrate Cd" .. . , . . 1000 Nitrate s2- . . . . . . 1000 Nasalt c1- . . . . . . 1000 Nasalt N03- . . . . . . 1000 Nasalt p a 3 - . . . . . . 1000 KH,salt NH3 . . . . . . 500 CH3NH2 . . . . 500 (CH&NH . . . . 500 Glucose . . . . 500 Sucrose . . . . 700 Fructose . . . . 500 * 3 ml of 0.2 M Na2EDTA solutions were used as masking agent. Absorbance 0.416 0.415 0.418 0.420 0.422 0.420 0.421 0.417 0.420 0.421 0.420 0.418 0.420 0.416 0.421 0.420 0.421 0.422 0.419 0.423 H ydrazine found/pg ml-1 12.88 12.85 12.94 13.00 13.06 13.00 13.03 12.90 13.00 13.03 13.00 12.94 13.00 12.88 13.03 13.00 13.03 13.06 12.97 13.10 Relative error, % -0.92 -1.15 -0.46 Nil +0.46 Nil +0.23 Nil +0.23 Nil Nil +0.23 Nil +0.23 +0.46 +0.76 -0.76 -0.46 -0.92 -0.23ANALYST, DECEMBER 1986, VOL. 111 1415 this reaction.MnII, FeII, FeIII, CoII, Ni” and Cu” can be effectively masked by Na2EDTA solution. The results of the investigation of foreign ions are given in Table 2. The common alkali and alkaline earth metal ions can be tolerated when present in a 150-fold excess. Ascorbic acid, H2S and thiols interfere at all concentrations. This observation further supports the Raschig’s process for the preparation of hydra- zine in the presence of gelatin. Gelatin retards the catalytic decompostion of N2H4 by complexing metal ions present in trace amounts. Conclusions The results obtained show that the method can be reliably applied to the determination of hydrazine in various samples as the method is free from a large number of interferences. The proposed spectrophotometric method gives a solution with a molar absorptivity of 4.2 X 104 1 mol-1 cm-1.The Sandell sensitivity for an absorbance of 0.001 is 7.63 x pg cm-2, with a relative standard deviation of 0.51% and a confidence limit (95%) of 12.985 k 0.031 for 13.000 pg of silver (20 replicates). In addition, biological samples contain- ing silver ions may be treated with hydrazine to reclaim silver metal. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Issa, I. M., Issa, R. M., Hamman, A. M., and Mahmoud, M. R., J. Indian Chem. Soc., 1976, 53, 698. James, T. H., J. Am. Chem. SOC., 1942, 64, 731. Csaky, T. Z., Acta. Chem. Scand., 1948, 2, 450. Watt, G. W., and Chrisp, J. D., Anal. Chem., 1952, 24, 2006. Riley, J. P., Analyst, 1954, 79, 76. Frear, D. S . , and Burrell, R. C., Anal.Chem., 1955,27, 1664. Schilt, A. A., and Creswell, A. M., Talanta, 1966, 13, 911. Asmus, E., Gasnske, J., and Schwarz, W., Fresenius Z. Anal. Chem., 1971, 213, 102. Dias, F., Olojola, A. S . , and Jaselskis, B., Talanta, 1979, 26, 47. Khadeev, V. A., and Mukhamedzhanova, D., Zavod. Lab., 1970,36, 1443. Gawargious, Y. A., and Besada, A., Talanta, 1975, 22, 757. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22, 23. 24. 25. 26. 27. 28, 29. 30. 31. 32. 33. 34, Kolthoff, I. M., and Belcher, R. “Volumetric Analysis,” Volume 111, Interscience, New York, 1957, pp. 233 and 308-310. Kolthoff, I. M., J. Am. Chem. SOC., 1921, 46, 2009. Browne, A. W., and Shetterley, F. F., J. Am. Chem. SOC., 1908, 30, 59; 1909, 31, 783. Bray, W. C., and Cuy, E. J., J. Am. Chem. SOC., 1924,46,858.Dernbach, C., and Mehlig, J . , Znd. Eng. Chem., Anal. Ed., 1942, 14, 58. Nair, C. G. R., Lalithakumari, R., and Senan, P. I., Talanta, 1978, 25,525. Krishna Verma, K. K., Srivastava, A., Ahmed, J., and Bose, A . , Talanta, 1979, 25, 469. Singh, B., and Rehmann, A., J. Indian Chem. SOC., 1940, 17, 169. Britton, H., and Clissold, E., J. Chem. SOC., 1942, 528. McBride, W. R., Henry, R. A., and Skolnik, S., Anal. Chem., 1951,23, 890. Szebelledy, L., and Somogyi, Z . , 2. Anal. Chem., 1938, 112, 391. Fogg, A. G., Chamsi, A. Y., Barros, A. A , , and Cabral, J. O., Analyst, 1984, 109, 901. Ikeda, S . , Satake, H., andKohri, Y., Chem. Lett., 1984,6,873. Karlik, M., and Jirounkova, H., Sb. Vys. Sk. Chem.-Technol. Praze. Anal. Chem., 1969, 4, 25; Chem. Abstr., 1970, 73, 116070j. Ray, P., and Sen, H., 2. Anorg. Chem., 1912. 76, 380. Siggia, S., and Lohr, L., Anal. Chem., 1949, 21, 1202. McKennis, M., Weatherby, J., and Dellis, E., Anal. Chem., 1958, 30, 499; Fresenius 2. Anal. Chem., 1959, 165, 457. Lloyd, C. P., and Pickering, W. F., Talanta, 1969, 16, 532. Hassan, S. S. M., andZaki, M. T. M., Mikrochem. J., 1970,15, 470. Hassan, S. S. M., Anal. Chim. Acta, 1971,54,185; Fresenius 2. Anal. Chem., 1971, 255, 364. Audrieth, L., and Ogg, B., “The Chemistry of Hydrazine,” Wiley, New York, 1951, pp. 153-162. Pal, T., Ganguly, A., and Maity, D. S., Analyst, 1986, 111, 691. Pal, T., and Maity, D. S., Analyst, 1986, 111, 49. Paper A61127 Received April 25th, 1986 Accepted July 16th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101413

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST DECEMBER 1986 VOL. 111 1417 Study of the Silver-catalysed Pyrogallol Red - Peroxodisulphate Reaction with 1,lO-Phenanthroline as Activator by a Kinetic = Spectrophotometric Method A. Sevillano-Cabeza J. Medina-Escriche M. Llobat-Estelles and M. Martin-Penella Department of Analytical Chemistry Faculty of Chemical Sciences Valencia University Burjasot, Valencia Spain The kinetics of the uncatalysed and silver-catalysed oxidation of pyrogallol red (PGR) by peroxodisulphate in the presence of 1 ,lo-phenanthroline (phen) as an activator were investigated. Rate equations for these reactions were derived from the kinetic dependences on hydrogen ion and reagent concentrations. The rate constants and the orders of reaction were also calculated. A kinetic model is proposed.The optimum conditionsfor silver(1) determination are asfollows pH 2.0 (in a sulphuric acid medium); S20& 2 x 10-2 rnol dm-3; phen 8 x 10-3 rnol dm-3; PGR 3 x 10-5 rnol dm-3; and temperature 25 "C. Under these conditions and using the fixed time and initial rate methods silver can be determined in the range 0.85-21.36 ng ml-1. The regions of analyte measurement are asfollows analyte not detected <0.26 ng ml-1 of silver(1); region of detection 0.26-0.88 ng ml-1 of silver(1); and region of determination >0.88 ng ml-1 of silver(1). The relative standard deviation of the method at a fixed time of 300 s was 3.9-1.4% for silver concentrations from 0.85 to 15.10 ng ml-1. The influence of 25 foreign species was investigated. Keywords Silver determination; silver-catalysed oxidation; p yrogallol red - peroxodisulphate reaction; I 10-phenanthroline activator; kinetic - spectrophotometric method Many workers have reported the use of catalysed reactions for the determination of silver.The silver-catalysed oxidation of organic reagents by S2082- in the presence of an activator has been widely utilised. For example piperazine acts as an activator in oxidation of p-nitrodiazoaminobenzenel; a,&'-dipyridyl has been used for the oxidation of pyrocatechol violet2; and for the oxidation of Cadion S M-Cadion, sulfarzazen3 and sulphanilic acid,4 ethylenediamine was used as an activator. The oxidation of certain azo dyes using ethylenediamine and triethylenetetramine as activators has also been studied.516 The application of the silver-catalysed oxidation of bromopyrogallol red by peroxodisulphate for the determination of silver has been discussed; 0.5-1 ng ml-1 of silver can be determined without an activator and in the presence of 1,lO-phenanthroline as an activator.7 This paper describes a study of the kinetics of the uncatalysed and the silver-catalysed oxidation of pyrogallol red (PGR) with peroxodisulphate using 1,lO-phenanthroline (phen) as an activator and describes the optimum conditions and analytical characteristics for determining trace amounts of silver using the PGR - S20& - phen system.Experimental Reagents Analytical-reagent grade chemicals and distilled water were used to prepare the reagent solutions. Standard silver(I) solution 10-2 mol dm-3. A 1.6987-g mass of silver(1) nitrate (Merck) was dissolved in distilled water and diluted to 1 1.A 10-6 mol dm-3 working solution was prepared by appropriate dilution of the stock solution. Pyrogallol red solution 8 x 10-4 mol dm-3. A 0.160-g mass of PGR (Merck) was dissolved in 500 ml of methanol. Potassium persulphate solution 10-1 mol dm-3. Prepared by dissolving the appropriate amount of reagent (Panreac) in 250 ml of water. 1,lO-Phenanthroline solution 10-1 mol dm-3. Prepared by dissolving 1.982 g of reagent (Merck) in 100 ml of methanol. Apparatus All spectrophotometric measurements were performed on a Shimadzu UV-240 spectrophotometer coupled with an optional programme unit (OPI-2) using 1-cm cells. A Thermo-tronic Selecta-389 immersion thermostat which measures temperature to f0.05 "C and a Frigedor Selecta-398 refriger-ator unit for use in a water-bath were used.All the solutions were previously heated to a working temperature of 25 _t 0.05 "C in a thermostat and this temperature was maintained in the reaction cell during the experiment. The pH was measured using a Crison-501 pH meter equipped with a Metrohm EA-121 Ag - AgCl electrode system. The pH of the solutions should be measured to an accuracy of f O . O 1 pH unit. Procedure To a series of 25-ml calibrated flasks different volumes of sulphuric acid and/or sodium hydroxide solutions were added to obtain solutions in the pH range 0.35-8.30. Various volumes of potassium persulphate and 1,lO-phenanthroline (0.1 ml dm-3) were also added to obtain final concentrations in the range 1 x 10-2-3 x 10-2 and 2 x 10-3-1.2 x 10-2 rnol dm-3 respectively.For the catalytic reaction volumes of silver(1) solution (10-6 mol dm-3) were added to obtain a final concentration in the range 8 x 10-9-2 x 10-7 mol dm-3. The solution was then diluted to ca. 20 ml with distilled water. This solution was shaken gently while various volumes of PGR (8 X 10-4 mol dm-3) were added to obtain an initial concentration in the range 1.6 x 10-5-7 x 10-5 mol dm-3. The zero time was taken as the moment at which the last drop of PGR had been added and the solution diluted to the mark with distilled water. The decrease in absorption as a function of time was measured at the A,,, of PGR for the correspond-ing pH. This absorbance measurement was made during the first 300 s in 1-cm cells at 25 "C against water as a reference blank.Results and Discussion The PGR - peroxodisulphate oxidation was almost negligible at pH values lower than 7 and no catalytic effect of silver(1) on this oxidation was observed. It is known that an activato 1418 * 8 - z X u 6 -c 4 - r-" 2 -ANALYST DECEMBER 1986 VOL. 111 increases the sensitivity of methods based on catalytic reactions by increasing the reaction rate. For the silver(1) -PGR - peroxodisulphate system 1 ,lo-phenanthroline was found to be a suitable activator and the increase in reaction rate that it caused allowed the kinetic study of the system to be carried out. The most sensitive catalytic reactions are based on redox mechanisms in which the metal catalyst acts via a redox cycle, according to the following scheme*? 0 x 1 + Mn+ $ Redl + M(n + I ) + Red2 + M(n + I ) + Ox2 + Mnf where Oxl and Redl represent the oxidised and reduced forms of the added oxidant Red2 and Ox2 are those of the reductant and M is the metal ion catalyst.High catalytic activity may be expected if the catalyst participates in the redox reactions in one-electron steps (as above) i.e. if the formal redox potential of the catalyst is between that of the added oxidant and reductant. It is also known that the over-all reaction rate is dependent on the rate of the reaction step AgI - e + Ag" and an acceleration of this step causes an increase in the over-all reaction rate. 1 ,lo-Phenanthroline is an organic reagent that accelerates the formation of AgII,10 and for our system 1,lO-phenanthroline was found to be an activator of this step.Kinetic Data Effect of variation of the acidity of the medium The influence of pH in the range 0.35-8.30 on the uncatalysed and catalysed reactions was studied. Solutions with a pH higher than 8 were not tested because the rate of the uncatalysed reaction is too fast at these pHs. The graph of the decrease in absorption (Amax 471 nm) as a function of time (during the first 300 s) was linear and the slope (tan a = -dA/dt) was used as a measure of the initial reaction rate. The results are expressed graphically in Fig. 1. According to the differential method the slope of the graph of the logarithm of initial rate (first 300 s) vs. the logarithm of the hydrogen ion concentration (log cH) gives the reaction order.In both reactions it was observed that the reaction order varies with pH. This variation can be attributed to a possible change in the mechanism of the oxidation process. The optimum pH range is that for which the reaction order with respect to pH is zero i.e. small fluctuations in the hydrogen ion concentration will not affect the initial reaction rates. For this reason a working pH range of 1.80-2.04 was chosen as the order of both reactions with respect to pH is zero in this range. 10 I n 0 1 2 3 4 5 6 7 8 PH Fig. 1. Influence of pH on reaction rate. A Uncatalysed reaction; B in the presence of the catalyst; and C catalytic reaction. Conditions PGR 3.2 X 10-5 mol dm-3; Agr 10-7 rnol dm-3; K2S20s 2 x 10-2 mol dm-3; phen 8 x 10-3 mol dm-3; Amax.of PGR 471 nm at pH 0.35-2.44; 476 nm at pH 3.06; 495 nm at pH 4.20; 538 nm at pH 6.10; and 517 rim at pH 8.30. Temperature 25 "C Catalytic effect of siZver(I) The effect of the addition of silver(1) in the concentration range 8 x 10-9-2 x 10-7 mol dm-3 on the reagents was determined with the following initial reagent concentrations: PGR 3.2 x 10-5 mol dm-3; S2082- 2 x 10-2 mol dm-3; and phen 8 x 10-3 mol dm-3 at pH 2.04 (sulphuric acid medium). The initial reaction rate (voz = tana = -d[PGR]/dt) increases proportionally to the silver concentration in the range 8 x 10-9-4 x 10-8 mol dm-3 and over the range 4 X 10-8-2 x 10-7 mol dm-3 (Fig. 2). The log - log plot shows a first-order dependence on the silver concentration for the two linear sections.When plotting initial rates vs. the initial silver concentration two straight lines were obtained. From the slopes of these apparent rate constants (kAg = vo2/[AgI]) of 0.56 s-1 in the range 8 x 10-9-4 X 10-8mol dm-3 and 0.35 s-1 in the range 4 x 10-8-2 x 10-7 mol dm-3 of silver were found. Effect of reagent concentration The effect of the concentration of each reagent on the initial rate of both reactions (vol uncatalysed reaction and vo2, catalysed reaction) was studied. The initial reaction rate increases steadily with the peroxo-disulphate concentration (Fig. 3). In both reactions the log -log plot shows a reaction order of 1 with respect to peroxodisulphate concentration. The apparent rate constants (kl(~~0~2-1 = vO1/[S2Os2-] and kz(s20 2-) = vo2/[S2082-]) were 7.8 X 10-7 and 1.8 x 10-6 s-1 for the uncatalysed and catalysed reactions respectively.The influence of the 1 ,lo-phenanthroline concentration on the initial reaction rate was studied (Fig. 4). For the uncatalysed reaction the initial rate increases with increasing phen concentration in the range 2 X 10-3-8 X 10-3 mol dm-3, and decreases when the phen concentration is higher than 8 X 10-3 mol dm-3 showing an inhibitory effect. The log - log plot shows that the reaction is first order with respect to phen in the range 2 x 10-3-8 x 10-3 mol dm-3. The apparent rate constant (kl(phen) = vol/[phen]) was 1.9 x 10-6 s-1. In the catalysed reaction the initial rate increases with increasing phen concentration in the range 2 X 10-3-6 X 10-3 mol dm-3; however it was almost constant in the range 6 X 10-3-8 X 10-3 mol dm-3 of phen.The log - log plot shows a first-order and a zero-order reaction respectively. For these reasons the optimum range of phen concentration was 6 X 10-3-8 x mol dm-3. Finally the effect of PGR concentration on the initial reaction rate was determined (Fig. 5). The concentration of PGR was varied from 1.6 x 10-5 to 7.0 x 10-5 mol dm-3 a limited range owing to the high molar absorptivity of the dye. 9 8 7 m X Z 6 u 5 i 4 3 2 1 0 2 4 6 8 10 12 14 16 18 20 IAg'l x 1O8/mol d r r 3 Fig. 2. Catalytic effect of AgI on the initial reaction rate of the PGR -peroxodisulphate - phen reaction. A In the presence of the catalyst and B catalytic reaction.Conditions PGR 3.2 X mol drn-'; S20R2- 2 x 10-2 mol dm-3; phen 8 X 10-3 mol d r r 3 ; pH 2.0; A,, , 471 nm; and temperature 25 " ANALYST DECEMBER 1986 VOL. 111 8 5 6 4 X r-" 2 , 1419 6 -5 -m 4 -X u 3 -z C r-" 2 -1 -u 1 2 3 [s20E2-1 x 102/mol dm-3 Fig. 3. Dependence of initial reaction rate of the uncatalysed and AgI-catalysed reaction on peroxodisulphatre concentration. A Un-catalysed reaction; B in the presence of the catalyst; and C cata-lytic reaction. Conditions PGR 3.2 x 10-5 mol dm-3; phen 8 X 10-3 mol dm-3; without or with Ag' mol dm-3; pH2.0; Amax. 471 nm; and temperature 25 "C I I I I 1 I 1 0 0.2 0.4 0.6 0.8 1.0 1.2 [phen] x 102/mol dm-3 Fig 4. Activator effect of 1 ,lo-phenanthroline on initial reaction rate of PGR - &OR2- and PGR - Ag* - S2OS2- systems.A, Uncatalysed reaction; B in the presence of the catalyst; and C, catalytic reaction. Conditions PGR 3.2 x mol dm-3; S2082- 2 x rnol dm-3; pH 2.0; h,, 471 nm; and temperature 25 "C The log - log plot shows reaction orders of zero and 0.5 with respect to PGR over the concentration ranges 1.6 X 10-5-7.0 X 10-5 and 3.2 x 10-5-6.4 x 10-5 mol dm-3 for the uncatalysed and catalysed reactions respectively. From the slope of the plot of initial rate (vo2) vs. the square root of the vo2/[PGR]t) of 5.5 x 10-6 (mol dm-3)-+ s-1 was determined. mol dm-3; AgI, PGR concentration an apparent rate constant (k2(P~R) -Effect of temperature A study of the influence of temperature on both reactions was performed in the temperature range 15-40 "C.The increase in reaction rate (for times between 0 and 300 s) with increasing temperature for the uncatalysed reaction is shown in Fig. 6. For the catalysed reaction the effect of temperature was greater than for the uncatalysed reaction. The initial reaction rate increased with increasing temperature and silver(1) concentration; the highest rate was obtained at 40 "C and 2 x 10-7 mol dm-3 of silver(1) (Fig. 7). The log - log plot shows a reaction order of 1 with respect to silver concentration over the temperature range 15-40 "C. The Arrhenius equation was followed over the temperature range 15-30 "C and a value for the activation energy of 8.75 kcal mol-1 and a frequency factor of 9.5 X l o 5 s-1 were calculated.Kinetic Model From these results the rate equation for the uncatalysed reaction was as follows: vol = K1 [S20&] [phen] . . . . (1) 7 6 5 a, 0 v x 4 u 2 t 0' I 1 1 I I I 1 2 3 4 5 6 7 [PGRI X lO5imol dm-3 Fig. 5. Variation of initial rate of the uncatalysed and the Agr-catalysed reaction with PGR concentration. A Uncatalysed system; B in the presence of the catalyst; and C catalytic reaction. Conditions Ag I 10-7 mol dm-3; S2082- 2 x mol dm-3; phen, 8 x 10-3 mol dm-3; pH 2.0; Lmax. 471 nm; and temperature 25 "C 15 20 25 30 40 Temperature/"C Fig. 6. reaction. Conditions PGR 3.2 X 10-5 mol dm-3; S2082- 2 X rnol dm-3; phen 8 x 10-3 rnol dm-3; pH 2.0; and A,,,, 471 nm Influence of temperature on initial rate of the uncatalysed 20 15 10 5 [Agl] x lO8/mol dm-3 Fig.7. Response surface of initial rate of the AgI-catalysed reaction for variation of AgI concentration and temperature. Conditions: PGR 3.2 x 10-5 rnol dm-3; S2082- 2 X 10-2 mol dm-3; phen 8 X 10-3 mol dm-3; pH 2.0; and A,,,, 471 nm Using this equation the mean value found for K1 was 9.6 X 10-5 mol-1 dm3 s-1 with a standard deviation of 4 X 10-6 mol-1 dm-3 s-1 calculated for nine duplicates 1420 ANALYST DECEMBER 1986 VOL. 111 The rate equation for the catalysed reaction can be written as follows: v02 = K2[Ag11 [S2OS2-l [PGRl . . . . (2) Using this equation the mean value found for K2 was 4.1 x l o 3 (mol dm~3)- s-1 with a standard deviation of 5 x 102 (mol dm-3)- 2 s-1 for twelve duplicates.The rate equation for the over-all process can be written as vol = K1[S2082-] [phen] + K2[Ag1] [S2082-] [PGRIt can be calculated by means of the following equations: (3) Alternatively for the uncatalysed reaction the value of K1 The values of Kl calculated from these equations are 9.8 x 10-5 mol-1 dm3 s-1 from equation (4) and 9.5 x 10-5 mol-1 dm3 s-1 from equation (5). For the catalysed reaction the K2 value can be calculated from the equations (7) v02 - k2 s20$-> K2 = [AgI] [S2082-] [PGR]b - [AgI] [PGR]t (8) v02 - k2(PGR) K2 = [ A ~ I I 1 ~ ~ 0 ~ 2 - 1 [PGRI~ - PWI [s2o82-1 The values of K2 calculated from these equations are 3.0 X lo3 (mol dm-3)-; s-1 from equation (6) 3.0 X 103 (mol dm-3)- $s-1 from equation (7) and 2.8 X 103 (mol dm-3)-;s-l from equation (8).By comparison of the values of Kl and K2 obtained from equations (1) and (2) with those calculated using equations (3)-(8) the average values of KI3= 9.6 X 10-5 mol-1 dm3 s-1 and K2 = 4.1 X 103 (mol dm-3)- s-1 were chosen as the most probable values. Optimum Conditions The optimum concentration is that for which the reaction order with respect to the reagent is zero or as close to zero as possible because small fluctuations in concentration do not affect the initial rate of a zero-order reaction. Our kinetic study of the uncatalysed and Ag-catalysed oxidations of PGR by S2082- in the presence of phen as an activator showed that the optimum conditions are a pH range of 1.80-2.04 (in a sulphuric acid medium) for both reactions and a phen concentration in the range 6 x 10-3-8 x 10-3 mol dm-3 for the catalysed reaction.The optimum conditions should be selected such that the maximum sensitivity and the largest linear range are obtained together with the maximum correlation coefficient and precision. The kinetic data obtained for the dependence of peroxodi-sulphate concentration on the initial rate for both reactions show that the analytical sensitivity of the catalysed reaction increases with increasing peroxodisulphate concentration in the range 1 x 10-2-3 x 10-2 mol dm-3. However the reproducibility of solutions with peroxodisulphate concentra-tions higher than 2 x 10-2 mol dm-3 was not good. Therefore, 2 X 10-2 mol dm-3 of S20& was chosen as the recommended concentration as a compromise between reproducibility and sensitivity.A PGR concentration should be used that provides an absorbance in the range of minimum photometric error and for this purpose 3 X 10-5-5 X 10-5 mol dm-3 of PGR was chosen as the most suitable concentration range. The kinetic data also show that for the uncatalysed reaction, no dependence on temperature in the range 15-25 "C was noticeable; however the initial reaction rate shows an increase between 30 and 40 "C. For the catalytic reaction the Arrhenius equation was followed over the temperature range 15-30 "C; also the graph of initial rate (during the first 300 s) vs. silver concentration shows that the slope increases with increasing temperature up to 30 "C and the best correlation coefficient was obtained at 25 "C (Table 1).Therefore 25 "C was chosen as the working temperature. To summarise the optimum experimental conditions for the catalytic determination of silver are as follows pH 2.0 in a sulphuric acid medium; S2082- 2 X 10-2 mol dm-3; phen, 8 x 10-3 mol dm-3; PGR 3 X 10-5 mol dm-3; and temperature 25 "C. Calibration Graphs Calibration graphs were obtained by applying the fixed time, initial rate and rate constant methods. Fixed time method The absorbance value was measured at 60,90,180 and 300 s. The calibration graph of absorbance decrease versus silver concentration was linear for silver concentrations from 0.85 to 4.27 and 4.27 to 21.36 ng ml-1 (Table 2). In the first range the sensitivity increases with time and the best correlation coefficients were obtained for 180 and 300 s.In the second range the best sensitivity and correlation coefficient were obtained at a time of 300 s. Therefore 300 s was chosen as the most suitable measuring time for both ranges of silver concentration. Table 1. Calibration graphs of initial rate (first 300 s) vs. silver concentration in the range 4.0-20.0 ng ml-1 at different temperat-ures. Conditions PGR 3.2 X 10-5 mol dm-3; S 2 0 g 2 - 2 X mol dm-3; phen 8 x 10-3 rnol dm-3; pH 2.0; and A,,,. 471 nm Temperature/ "C 15 20 25 30 35 40 Sensitivity/ S-1 0.880 1.040 1.521 1.801 1.761 1.841 Correlation coefficient (r) 0.997 0.996 0.9998 0.9994 0.997 0.969 Table 2. Calibration graphs for the fixed time method. Conditions: PGR 3.2 x 10-5 mol dm-3; S 2 0 8 2 - 2 x 10-2 rnol dm-3; phen 8 X 10-3 rnol dm-3; pH 2.0; A,,,.471 nm; and temperature 25 "C Silver concentration/ Equation of Correlation Time/s ng ml-1 calibration graph coefficient 60 0.85-4.27 A = 0.6137 - 0.0066~ -0.993 4.27-21.36 A = 0.6076 - 0.0048~ -0.9986 90 0.85-4.27 A = 0.6069 - 0.0095~ -0.9975 4.27-21.36 A = 0.5991 - 0.0071~ -0.9987 180 0.85-4.27 A = 0.5872 - 0.0183~ -0.9996 4.27-21.36 A = 0.5678 - 0.0128~ -0.9993 300 0.85-4.27 A = 0.5600 - 0.0292~ -0.9995 4.27-21.36 A = 0.5143 - 0.0182~ -0.9999 ANALYST DECEMBER 1986 VOL. 111 1421 Initial rate method A plot of reaction rate (first 300 s) versus silver concentration yields the equations vO2 = 0.118 + 2.088~ in the range 0.85-4.27 ng ml-1 with r = 0.9996 and v02 = 3.711 + 1.320~ in the range 4.27-21.36 ng ml-1 with r = 0.9997 (Table 3).Rate constant method The plots of log A versus time (first 300 s) for silver concentrations in the range 0.85-21.36 ng ml-1 were straight lines. The slopes of these straight lines ( K ) were plotted against silver concentration. Calibration graphs with K = 0.0002 - 0.0008~ and K = -0.0001 - 0.0001~ in the ranges 0.85-4.27 and 4.27-21.36 ng ml-1 of silver respectively were obtained. The best methods for determining calibration graphs for the catalysed reaction were the initial rate and fixed time methods chosen on the criteria of sensitivity (i.e. the slope of the calibration graph) and correlation coefficient. Precision The standard deviation s for the analysis of five duplicates of a sample containing various concentrations of silver and fixed concentrations of PGR S 2 0 8 2 - and phen are shown in Table 4.The silver concentration was calculated by substituting the absorbance values at a fixed time of 300 s in the ranges 0.85-3.88 and 3.88-15.10 ng ml-1 of silver. The relative standard deviation s, was in the range 3.9-1.4% for silver concentrations from 0.85 to 15.10 ng ml-1. Limit of Detection and Limit of Determination The theoretical limit of detection ( C L ~ = &sb/s)11312 for a numerical factor Kd = 3 (confidence level) is 0.26 ng ml-1 of silver at a fixed time of 300 s. The standard deviation s b for eleven duplicates of the blank (PGR - S20& - phen) was 2.6 Table 3. Dependence of the initial reaction rate on silver concentra-tion.voT = Initial reaction rate (first 300 s) in presence of the catalyst; vO2 = initial reaction rate (first 300 s) for the catalytic reaction. Conditions PGR 3.2 X 10-5 mol dm-3; S 2 0 8 2 - 2 X 10-2 mol dm-3; phen 8 x 10-3; pH 2.0; A,,,, 471 nm; and temperature 25 "C. vol, Initial rate (first 300 s) for the uncatalysed reaction = 6.00 ng ml-1 s-1 Silver concentration/ VOT/ V o d ng ml-1 ng ml-1 s-l ng ml-1 s-1 0.85 8.40 2.40 1.71 10.40 4.40 2.14 11.20 5.20 4.27 15.61 9.60 8.54 20.81 14.81 10.68 23.61 17.61 21.36 38.02 32.02 Table 4. Precision data for the method. Equation of calibration graph: A = 0.6294 - 0.0294~ for silver concentration in the range 0.85-3.88 ng ml-1 and A = 0.5840 - 0.0182~ for silver concentrations in the range 3.88-15.10 ng ml-1.Conditions PGR 3.6 x mol dm-3; S 2 0 8 - 2 2 x 10-2 mol dm-3; phen 8 x rnol dm-3; pH 2.0; A,,,., 471 nm; fixed time 300 s; and temperature 25 "C Concentration of silver/ ng ml-1 0.85 2.14 3.88 3.88 12.08 15.10 Mean concentration of silver foundlng ml-1 0.81 2.19 3.89 3.8 12.3 15.0 Standard deviation1 ng ml-1 0.03 0.06 0.08 0.1 0.2 0.2 X 10-3 absorbance unit (A). The slope of the calibration graph is 0.0294 A ng-1 ml. The experimental limit of detection is 0.85 ng ml-1 of silver. The limit of determination (c+ = K,sdS) for a numerical factor Kq = 10 (confidence level) is 0.88 ng ml-1 of silver. The regions of analyte measurement are as follows analyte not detected < 0.26 ng rnl-1; region of detection 0.26-0.88 ng ml-1; and region of determination > 0.88 ng ml-1 of silver.Interference Study The effect of foreign ions on the Ag - PGR - S2082- - phen system was studied for 3.88 ng ml-1 of silver. The results are shown in Figs. 8 and 9 where AcAg is the difference in silver concentration between the catalysed reaction with and with-out the possible interfering ion at a fixed time of 300 s. c I -E P cn . a" 2 -2 -0 0.5 1 .o 1.5 2.0 Concentration/pg ml-1 Fig. 8. Influence of foreign ions on the silver-catalysed peroxodisul-phate oxidation of PGR in the presence of phen as activator. Conditions PGR 3.2 x 10-5 mol dm-3; silver 3.88 ng ml-l; S 2 0 g 2 - , 2 X 10-2mol dm-3; phen 8 x 10-3 rnol dm-3; pH 2.0; A,,,, 471 nm; and temperature 25 "C 3 1 / Zn" 2 - 2 s .2 -2 -3 I -I/ -5 ,y ,"".\I -6 CU" 0 10 20 30 40 50 60 70 80 90 100 110 120 Con cent rat ion/pg m 1-1 Fig. 9. Influence of foreign ions on the silver-catalysed peroxodisul-phate oxidation of PGR in the presence of phen as activator. Conditions PGR 3.2 X 10-5 mol dm-3; silver 3.88 ng ml-l; S20g2-, 2 x 10-2 rnol dm-3; hen 8 x 10-3 mol dm-3; pH 2.0; A ,471 nm; and temperature 29° 1422 ANALYST DECEMBER 1986 VOL. 111 A species is defined as causing interference when it changes the standard deviation of the concentration of Ag determined in the Ag - PGR - S20& - phen system more than two-fold (s = 0.12 ng ml-1 calculated for five independent solutions with a silver concentration of 3.88 ng ml-1). The interfering processes may be classified as follows the values in parentheses being the maximum tolerable concentra-tion ratios of foreign ion to silver species that accelerate the catalytic activity of silver on the oxidation of PGR such as Fe 111 (5) MnVII (lo) VV (lo) BiI" (20) CeIV (30) CrVI (30), IO3-(30) IO4-(60) ZrIV (60) TiIV (230) A P (2300) and ZnII (3600); species that inhibit the catalytic oxidation of PGR such as WVI (40) CoI1 (45) SbV (120) MoVI (250) Pb" (3500) NiII (4600) CuII (5000) and Cd" (8400); and species that do not interfere such as l o 4 ng ml-1 of C103- C104-, As" CrI" and MnII.Recommended Procedure To 25-ml calibrated flasks add 1.5 ml of H2S04 (0.1 mol dm-3)) 5 ml of potassium persulphate (0.1 mol dm-3) and 2 ml of 1,lO-phenanthroline (0.1 mol dm-3).For the catalytic reaction samples containing between 21.25 and 534 ng of silver were added and the solution was diluted to ca. 20 ml with distilled water. Each solution was shaken gently while 1 ml of 8 X 10-4 mol dm-3 PGR was added and the volume was diluted to 25 ml. The zero time was taken as the moment at which the last drop of PGR solution was added. The absorbance was measured against water at the maximum wavelength of 471 nm at 25 "C. The silver concentration was calculated from the corresponding equation for the calibration graph by using the fixed time (300 s) or initial rate method (first 300 s). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. References Jankauskiene E. and Jasinskiene E. Liet. TSR Aukst. Mokykly Mokslo Darb. Chem. Chem. Technol. 1967 8 31. Jasinskiene E. Jankauskiene E. Liet. TSR Aukst. Mokyklu Mokslo Darb. Chem. Chem. Technol. 1968 9 41. Jasinskiene E. Jankauskiene E. Polukordaite G. and Raseviliute N. Elem. Mikrokiekiu Nustatymas Fiz. - Chem. Metod. Liet. TSR Chem.-Anal. Mokslines Konf. Darb. 1969, 2nd 210. Bontchev P. R. Aleksiev A. A. and Dimitrova I. Mikro-chim. Acta 1970 1104. Jasinskiene E. and Rasevichute N. Zh. Anal. Khim. 1970, 25 458. Jasinskiene E. and Rauckiene N. Nauchn. Tr. Vyssh. Ucheb. Zaved. Lit. SSR Khim. Khim. Tekhnol. 1973,15,29. Mueller H. Schuring H. and Werner G. Talanta 1974 21, 581. Bontchev P. R. Talanta 1970 17 499. Ochiai E. I. J. Chem. Educ. 1978 55 631. Bontchev P. R. and Aleksiev A. A. Teor. Eksp. Khim., 1973 9 191. IUPAC "Compendium of Analytical Nomenclature," Per-gamon Press Oxford 1978. ACS Committee on Environmental Improvement Anal. Chem. 1980,52 2242. Paper A6131 Received January 31st 1986 Accepted July Ist I98

ISSN:0003-2654    

DOI:10.1039/AN9861101417

出版商:RSC    

年代:1986

数据来源: RSC