摘要:

1086 Analyst September 1983 Vol. 108 $9. 1086-1090 Automatic Titration Stepwise Addition of Equal Volumes of Titrant Part VIII." Sea Water Determination of Alkalinity and Total Carbonate in Axel Johansson Sten Johansson and Gunnar Gran Department of Analytical Chemistry The Royal Institute of Technology 5-100 44 Stockholm 70 Sweden Previous parts of this series have presented methods for the evaluation of titration data. This paper shows the utility of some of these methods and extensions of them for solving a practical problem i.e. the determina-tion of alkalinity and total carbonate in sea water. The procedure is based on the solution of a set of linear equations. The method presented has been tested on theoretical and experimental titration data and has given results that compare extremely well with those obtained by using non-linear curve-fitting methods.Keywords Sea water analysis ; fiotentionzetric titration ; linear multiple regression ; alkalinity ; total carbonate In previous parts of this series a number of methods for the evaluation of titration data have been described. In order to confirm the reliability of the methods solutions of known composition were analysed. The aim of this paper is to show the utility of some of the methods on a special practical problem Le. the determination of the alkalinity and total carbonate in sea water. These two parameters usually designed At and Ct are used, together with the pH value for characterising a sea water and are of great interest in investi-gations on the carbon dioxide cycle.The variations in At and Ct are usually small and much work has been spent on refining the determinations. The analysis is usually performed as a titration of a sea water sample with standard hydrochloric acid solution. For the evaluation of the titrations Granl plots have been used by Dyrssen and Sil16n2 and modified Gran plots by Hansson and Jagne~-.~ Recently, Johansson and Wedborg4 have published a non-linear curve-fitting method for the evaluation of potentiometric titrations of sea water with hydrochloric acid. Their evaluation method4 is complicated and we considered it worthwhile to investigate the possibilities of using the program EKVOL~ for the evaluations. The program TITRA6s7 was also investigated and correct results were obtained as could be expected as this is also a non-linear curve-fitting method.The EKVOL program is less complicated and only involves the solution of a set of linear equations by the least-squares method. The original program can calculate only one equivalence volume but a modified version EKVOLS can calculate several equivalence volumes or concent rat ions. The basic principles on which the programs were worked out were as follows: A. The measurements of the titrant additions should be so accurate that the volume error will be negligible in comparison with e.g. the error in the e.m.f. measurements. B. All polyprotic acids are treated as a mixture of monoprotic acids. This means that, e.g. a diprotic acid can be exactly replaced by two monoprotic acids each of the same molarity as the diprotic acid but with different equilibrium constants commonly called titration con~tants.8~~ These titration constants can easily be calculated from the normal stability constants.The use of a mixture of monoprotic acids instead of polyprotic acids does not mean that approximations are introduced; the results will be exactly the same. With sea water the polyprotic acids are carbonic acid phosphoric acid and sulphuric acid. Their consecutive stability constants differ by at least a factor of 1000. Therefore the titration constants differ by less than 0.1% from the corresponding stability constants. As the * For Part VII of this series see Analyst 1981 106 1109 JOHANSSON JOHANSSON AND GRAN 1087 titration constants and the stability constants are so similar the titration constants have been set equal to the stability constants.This applies also to hydrogen ions which implies that meters have to be calibrated at known hydrogen ion concentrations. Further all equilibrium constants must be concentration constants. A further consequence is that the titrations have to be performed at a high and constant ionic strength so that activity factors remain constant during the entire course of the titration. D. The titrations are followed by determining the hydrogen ion concentration after each addition of titrant. The electrode couple consists of a glass electrode and a reference electrode and the e.m.f. is measured with a digital voltmeter. The hydrogen ion concentration is calculated with the Nernst equation. In practice one has to take into account that an error may appear after the calibration as the conditional normal potential may change by up to a few tenths of a millivolt when the electrode system is transferred from one solution to another.C. In all calculations ionic concentrations and not activities are used. Titration Curve Equation The stability constants for the acid - base equilibria in sea water have been determined by a number of workers and may be considered to be well established. The constants vary with ionic strength but the variations are small as the ionic strength is high about0.7. Thus if the salinity is known the equilibrium constants are also known. The data used in this paper were taken from the work of Johansson and Wedborg. Further the concentrations of the major components except carbonate and hydrogen carbonate are usually known for a given source and salinity of sea water.When a sample of sea water is titrated with hydrochloric acid the shape of the titration curve is influenced not only by carbonate and hydrogen carbonate but also by other com-ponents which react with the titrant. These components are sulphate borate fluoride, silicate and phosphate. Conceptually the hydrogen carbonate can be looked upon as having been formed from carbonate by the addition of a certain amount of hydrochloric acid before the start of the titration. Instead of calculating the concentrations of carbonate and hydrogen carbonate, one can calculate the total carbonate concentration Ct and the total initial hydrogen ion concentration Htin at the start of the titration.Then VoHti is the amount of hydro-chloric acid that conceptually had been added. Vo is the volume of the sample solution at the start of the titration. A general equation for a titration of a mixture of monoprotic acids is given in equation (5) in reference 7. An analogous equation is valid for the titration of bases if [H’] is exchanged for [OH’] and acid stability constants are exchanged for the equivalent base constants. With these assumptions the titration equation can be written: -VoHt in + voct/(l + Kic[OH’lf) + voct/(l + Kc[OH’lf) - ( v o + V ) [OH’If + ( V O + V)Kw/([OH’lf) - VC€i + VO(SU + B + si + p + F ) = 0 SU = [so*lt/(l + Ks,[OH’lf) B = [B(OH)Blt/(l + KB[OH’If) F = [Flt/(l + KF[OH’If) ‘ (1) where Si = [SiO(OH)Jt/(l + KBi[OH’]f) p = [PO,lt(l/(l + KlP[OH’If) + 1/(1 + fLP[OH’]f) + 1/(1 + K3P[OH’If)) In these expressions the subscript t indicates total concentrations.V is the added volume of titrant. [OH’] values are experimentally found hydroxyl ion concentrations. f is an error factor such that [OH’If = [OH]. This factor is constant during a titration and originates from an incorrect calibration of the electrode system. The error factor influences mainly the value of the term (V + V)Kw/([OH’]f) especially towards the end of the titration. Ct [SO,]t [B(OH),]t, [SiO(OH) Jt [PO,]t and [F]t are the total concentrations of carbonate sulphate borate, silicate phosphate and fluoride respectively. In equation (l) Ht in Ct and f are unknown and will be calculated; all other parameters are known or measured during the titration.The error factor f is approximately equal to 1 The stability constants Klc etc. are defined in Table I. The constants are base stability constants 1088 JOHANSSON et d. AUTOMATIC TITRATION BY STEPWISE Analyst VOl. 108 and may be given a preliminary value of 1 except in the term (V + V)K,/([OH’]f). The function F(Ht in Ct f) can after a slight rearrangement of equation (l) be written as where A(1,i)Ht in + A(2,i)Ct + A(3,i)F + A(4 i) = 0 . . . . * (2) A(1,i) = -vo A(2,i) = Vo/(l + Kl,[OH’lif) + Vo/(l + K,c[OH’lif) 43,;) = (Vo + ~i)Kw/[OH’l, A(4,i) = V o ( S U ~ + Bi + Sii + Pi + Fi) - (Vo + Vi)[OH’]if - ViCB [0H’li Sui etc. are the values of [OH’] Su etc. after the addition of Vi ml of titrant.The function F(Htin Ct,f) is a linear function and the set of equations (2) is first order with three unknowns so that at least three simultaneous sets of values of A(l,i) A ( 2 i ) , A (3,;) and A ( 4 4 are needed. With more than three measurements the set of equations (2) is solved by using a least-squares treatment. In this way preliminary values for the three parameters Ht in Ct and f are obtained. All [OH‘] values except the ones in A ( 3 i ) are then corrected by the value obtained for f and the calculations are repeated. Normally two or three iterations are sufficient. As mentioned above it may be assumed that all stability constants are known with sufficient accuracy. However Johansson and Wedborg4 also calculated the constant Kzc for the reaction HCO,- + H+ + H,CO, as an error in the value of this constant has a greater influence on the results of the calculations than a corresponding error in any of the other constants.The constant K, and the error factor f are interdependent as K, is multi-plied by f in the titration equation. This implies that an erroneous value of K, can be partly compensated for through the calculation off. It is however also possible to calculate K2, as follows. If equation (1) is solved for K2, we obtain After preliminary values for Ht in Ct and f have been obtained by using equation (2) the constant K, can be calculated with the aid of equation (3). In using this equation a few values for V are chosen in such a way that the corresponding values pOH’~logK,,. In this range of the titration curve many of the terms in equation (4) become so small that they have virtually no influence on the value of Kzc and they could thus be neglected.If the calculator is suitably programmed it is simpler to use all the terms however. Calculation Procedure Synthetic Titration For a typical surface sea water Dyrssen et aZ.10 have given the data reproduced in Table I. TABLE I CONCENTRATIONS AND PROTONATION CONSTANTS IN SEA WATER Total concentration/ Component mmol kg-1 Protonation reaction . . 28.230 SO,2- + Hf + HS0,-C032- . . 2.000 C032- + H+ $ HC03-F- . . SiO(OH),- . . 0.005 SiO(OHj,- + H+ + Si(OH), HCO,- + H+ + H2C03 0.412 B(OH)4- + H+ + B(OH)3 + 0.073 F- + Hf + HF B(OW4-Pod3- . a . 0.000 1 pop + H+ + HPO,~-HP042- + Hf + H2P04-H2PO4- + H+ + H3PO4 Htin 2.1172 H20 + H+ + OH-Log(stabi1ity constant) 7- Acid Base 1.060 12.260 = log Ksu 9.078 4.242 = log Klc 5.993 7.327 == log K2c H,C) 8.742 4.578 = log KB 2.60 10.72 = log KF 9.30 4.02 = log K B ~ 9.080 4.240 = log KIP 6.079 7.241 = log K2p 1.740 11.580 log K,p -13.320 = log K September 1983 ADDITION OF EQUAL VOLUMES OF TITRANT.PART VIII 1089 Alkalinity as defined by Dickson,ll for the system given in Table I is A t = 2[C0,2-]t + [B(OH)4-]t + [SiO(OH),-]t + 2[P043-]t - Ht in = 2.300 mmol kg-l Using the data in Table I and the program HALTAFALL,~~ Dyrssen et aZ.1° calculated pH values for the addition of 0.01-ml increments of 0.7000 M hydrochloric acid up to 0.50 ml to a 100 ml sample solution. These values have been used here as an example of a synthetic titrat ion.If the true value of the constant K, is used but 0.1 is subtracted from all pOH values (equivalent to an error of 6 mV in the e.m.f. values) the equation system (2) gives At = 2.368 Ct = 1.994 and f = 0.812 in the first run and after two iterations the correct values 2.300 2.000 and 0.792 respectively were obtained. An error of 6 mV is very great indeed, but in spite of that the correct result was obtained very quickly. Normally a difference of 0.01 between pOH’ and pOH equal to 0.6 mV in the e.m.f. values can be expected. If an erroneous value for K, was also introduced (logK’, = lo@, - 0.01) the errors were about 0.3%. If K, is to be determined the set of equations (2) is iterated three times, then equation (3) and the set of equations (2) are iterated alternatively until the results are consistent.Experimental Titrations The suggested method of calculation was tested on several titrations one of which was a titration of sea water conducted at the University of Gothenburg. Titration data had kindly been provided by Dr. Margareta Wedborg. About 150 ml of sea water with a salinity of 32.5O/oo had been titrated with 0.5 M hydrochloric acid added in 0.03-ml increments. After each addition the e.m.f. was measured with a glass electrode a reference electrode and a digital voltmeter. For the calculations we have assumed EIO = 375.0 mV. E’O is the conditional standard potential defined in Part IIL5 From the titration data and using the set of equations (2) and equation (3) we have calculated At = 2.3056 Ct = 2.0644 log K, = 7.343 and f = 1.062.All these values agree very well with those calculated by Wedborg vix. At = 2.3055 Ct = 2.0645, Synthetic sea water cannot be obtained commercially. It is also difficult to prepare water samples for which the composition is exactly known. The components that occur in high concentrations may contain unknown impurities which will influence the results. We therefore tested the titration method on an aqueous solution containing only sodium chloride (0.7 M) and sodium hydrogen carbonate (1.9955 mM). The results were At = 1.995 & 0.002 and Ct = 1.999 0.003. A set of titrations of standard sea water (IAPSO Standard Sea Water Service batch P 91-800510 chlorinity 19.374) was performed. The results were At = 2.370 & 0.003, Ct = 2.215 From the f value EfO = 376.5 can be calculated.= 7.344 and El, = 376.5. Concentrations are given in millimoles per litre. 0.002 and logK, = 7.30 & 0.01. Experimental A Mettler MemoTitrator DL40RC was used for the titrations. The stirrer the burette, the glass vessel and the registration unit were the standard equipment provided with the titrator. The electrode was a micro-size combination glass silver - silver chloride electrode (HA-401 M5 Dr. ing. chem. W. Ingold Zurich Switzerland). The titration vessel was thermostated at 25.0 & 0.1 “C. At the start of a titration the glass vessel was filled to about 95% with sample. To avoid losses of carbon dioxide the lid was sealed except for a capillary tube through which the overpressure caused by the addition of titrant could be released.The titrant was 0.1 M hydrochloric acid prepared from Merck Titrisol ampoules. The ionic strength of the titrant was adjusted to 0.7 using sodium chloride. The e.m.f. readings were started after an initial delay of 30 s. After every addition of titrant the emf. value was recorded when the change during 10 s was less than 0.2 mV. Discussion In such instances polyprotic acids have to be treated as a mixture of monoprotic acids and A set of linear equations can be used for solving even complicated titration problems 1090 JOHANSSON JOHANSSON AND GRAN approximate values for the equilibrium constants for the acid - base pairs involved have to be known. On the other hand no approximate values are needed for the concentrations to be determined.With sea water this is not important as such values are easily available. The calibration of the electrode couple can be rough. Even a difference of 10 mV in the value has no influence on the final results. It is also possible to calculate the constant K,, from the first data pairs by using a simple transformation of equations (3) and (4). This possibility may be of importance if the electrode couple changes its E value during the titration. The value for log K, calculated in this way was 4.21 instead of 4.24 according to reference 4. However this difference caused a change in the Ct values of only about 0.1%. We thank Dr. D. Jagner and Dr. M. Wedborg for valuable discussions and for providing us with samples of standard sea water.We are also grateful to Dr. Wedborg for making some of her experimental results available to us. Thanks are due to Prof. F. Ingman for valuable comments on the manuscript. References 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. Gran G. Analyst 1952 77 661. Dyrssen D. and Sill& L.-G. Tellus 1967 19 113. Hansson I. and Jagner D. Anal. Chim. Acta 1973 65 363. Johansson O. and Wedborg M. Oceanol. Acta 1982 5 209. Johansson A. and Johansson S. Analyst 1978 103 305. Ingman F. Johansson A. Johansson S. and Karlsson R. Anal. Chim. Ada 1973 64 113. Johansson A. and Johansson S. Analyst 1979 104 601. Simms H. S. J . Am. Chem. SOC. 1926 48 1251. Johansson S. Analyst 1979 104 593. Dyrssen D. Jagner D. and Johansson O. in Ostlund G. and Dyrssen D. Editors “Workshop Dickson A. G. Deep-sea Res. 1981 28 609. Ingri N. Kakolowicz W. SillCn L. G. and Warnquist B. Talanta 1967 14 1261. on Oceanic CO Standardization La Jolla California,” 1980. NOTE-References 5 and 7 are to Parts 111 and IV of this series respectively. Received February 2 1 4 1983 Accepted Apvil Sth 198

ISSN:0003-2654    

DOI:10.1039/AN9830801086

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Analyst September 1983 Vol. 108 +p. 1091-1095 1091 Determination of Dimetridazole Residues in Poultry Tissues by H igh-performance Liquid Chromatography Anthony Hobson-Frohock and Jayne A. Reader ARC Food Research Institute Colney Lane Norwich NR4 7 UA A high-performance liquid chromatographic method has been developed for the determination of the anti-histomoniasis drug dimetridazole. Recoveries of the drug down to 0.01 mg kg-l “spiked” on to poultry tissues are 80% or above and the method has been applied to the determination of residues in tissues from broiler chickens and turkeys. Keywords Dimetridazole determination ; poultry tissue ; high-performance liquid chromatography Infection of poultry with the protozoal flagellate Histornonas meleagridis gives rise to the disease known variously as histomoniasis infectious enterohepatitis or more commonly, blackhead.It is reputed to be one of the most destructive infections to which turkeys are susceptible and has been responsible for large losses in the turkey industry. Effective control can be obtained by continuous oral medication with the drug dimetridazole (1,2-dimethy1-5-nitroimidazole) at levels around 75 mg kg-l in the feed. The establishment of a National Meat Monitoring Programme by the Ministry of Agriculture Fisheries and Food requires reproducible and sensitive methods for the surveillance of therapeutic drug residues in poultry meat supply. Several different techniques have been described for dimetridazole but none appears suitable for this purpose. A simple colorimetric method for application to feed analysis1 has been recommended by the Association of Official Analytical Chemists2 but suffers from considerable interference from other therapeutic agents.A polarographic method3 has been applied with various modifications to residue studies in turkeys,*S5 pigs6 and guinea pigs,7 with a variable lower detection limit of 0.05,5 0 . 1 4 y 6 and 2 mg kg-l.’ Polaro-graphy has also been suggested by the Analytical Methods Committee as the best method for feed analysk8 A gas-chromatographic method for feed concentrations of dimetridazole between 20 and 200 mg kg-l has been reportedeg As an alternative to polarography the use of high-performance liquid chromatography (HPLC) was investigated to develop a method capable of routine application in the surveil-lance programme for residue levels down to 0.05 mg kg-l.Liquid chromatography on a silica column with chloroform-methanol as the mobile phase has been used to determine dimetridazole in pig and poultry feed.10 The detection limit was high at 10 mg kg-l although the authors claimed that a lower limit could be achieved by making a few (unspecified) modifications. The use of nitroimidazoles as antimicrobial agents in human drug therapy is well known and of these metronidazole [l-(2-hydroxyethyl)-2-methyl-5-nitroimidazole ; Flagyl] has found wide acceptance. Consequently several chromatographic methods are available for its assay at the micrograms per millilitre level in biological fluidsl1J2 and the separation described by Gulaid et aZ.lf was found most suitable for our purpose.In general dimetridazole has been isolated from feed and tissues by maceration with chlor~forml~~ or b e n ~ e n e ~ - ~ followed by extraction of the organic phase with dilute acid or, alternatively by maceration in methanol followed by removal of the solvent and dissolution of the residue in acid.8 Further purification of the acid solution has been achieved by extraction with benzene1 or dichloromethane.1° We preferred the less toxic dichloromethane for extraction and clean-up by phase distribution to provide a sensitive and reproducible method for the determination of dimetridazole in poultry tissues. Dimetridazole was fed at 75 mg kg-l to groups of broiler chickens and turkeys from 1 day old for 8 and 14 weeks, respectively and the method was used to determine residues of the drug in leg and breast muscle and liver.Experimental Reagents and Materials All reagents and solvents were of analytical-reagent grade except those used to prepare the chromatographic mobile phase which were of HPLC grade (Fisons Scientific Apparatus) 1092 HOBSON-FROHOCK AND READER DETERMINATION OF Analyst VoZ. 108 Dimetridazole. Pure (May & Baker). Metronidaxole. Sigma. Feed. Broiler and turkey feeds were mixed to typical commercial specifications. Care was taken to exclude all known therapeutic drug additives but the normal vitamin - mineral supplement was included. Groups of ten female Ross 1 broiler chickens were reared in cages from 1 day old to 8 weeks of age. One group received the control i.e.unmedicated feed and the other received feed containing 75 mg kg-l of dimetridazole. Six days before the end of the experi-ment (the “withdrawal period”) five broilers were removed at random from the medicated feed group and placed on control feed. Two groups of ten female BUT 5 turkeys were reared from 1 day old to 14 weeks of age on deep litter and were treated similarly. A t the end of the respective growing periods all birds were slaughtered by electrical stunning and exsanguination. Leg and breast muscle and the liver were immediately removed and stored in cans a t -40 “C until required. When required the samples were thawed overnight at 2 “C minced twice through a domestic mincer mixed and representative samples taken for analysis. Birds. Apparatus The liquid chromatograph consisted of a Series 2/2 pump module with an LC75 detector (Perkin-Elmer) at 315nm.The column (250 x 4.5mm i.d.) was packed with Spherisorb 6 ODS (Perkin-Elmer) and a Brownless RP-18 guard column was fitted between the column inlet and the Rheodyne 7125 valve loop injector (20-p1 loop). No column heater was used but the room was air-conditioned at 22 “C. The isocratic mobile phase consisted of 0.01 M potassium dihydrogen orthophosphate - methanol (7 + 3) de-gassed after mixing by sonica-tion under vacuum at a flow-rate of 1.0 ml min-l. All data were processed on a Sigma 15 console (Perkin-Elmer) . Procedure Weigh a tissue sample (10 g) into a centrifuge tube (150 x 61 mm 250 ml) and add 50 m 1 of dichloromethane. Homogenise the mixture using an Ultra Turrax homogeniser for 1 min.Centrifuge the extract for 10 min at 2000 rev niin-l and remove the dichloromethane layer. Repeat the extraction of the residue with two further 50-ml volumes of solvent. Combine the dichloromethane solutions and extract with two 50-ml volumes of 1.0 M hydro-chloric acid taking care to prevent emulsion formation. Combine the acid layers and centri-fuge for 2 min at 2000revmin-1. Decant the acid layer rinse the centrifuge tube with 20 ml of the same acid and add the rinsings to the initial acid extract. Repeat the extraction of the dichloromethane phase and combine the acid extracts. Adjust the pH to 9.0 with 2 M sodium hydroxide solution and extract with five 30-ml volumes of dichloromethane. Reduce the volume of the dichloromethane solution to approximately 100 p1 on a rotary film evaporator at 40 “C taking care not to reduce the solution to dryness.Add metro-nidazole standard (100 pl 5 pg ml-I) mix and inject 16-p1 aliquots on to the Spherisorb ODS column using the “partial loop filling’’ technique. Calculate the volume of final extract from the formula Area of metronidazole peak in 15 pl of standard Area of metronidazole peak in 15 pl of extract __ ._________ ____ x loop1 Recovery Experiments The recovery of dimetridazole from “spiked” tissue samples was obtained by the addition of 1-ml solutions of dimetridazole in methanol to the macerated tissue allowing the mixture to stand for approximately 30 min in the dark before extraction. Calibration standards were chromatographed with each group of tissue samples and a calibration graph was drawn for each run using a least-squares BASIC program (Perkin-Elmer).Such “spiked” control samples were analysed concurrently with each group of samples from experimentally treated birds. Results and Discussion The major evaluation of the method was performed on “spiked” samples of control turkey liver ; further recovery values were obtained on “spiked” control tissues analysed concurrentl Se$tember 1983 DIMETRIDAZOLE RESIDUES IN POULTRY TISSUES BY HPLC 1093 with experimental samples from treated birds. All recovery values are given in Table I, which shows that the method is satisfactory and capable of achieving the required sensitivity. A typical separation of dimetridazole and metronidazole from other tissue components is shown in the chromatogram of a control turkey liver “spiked” with dimetridazole at 0.05 mg kg-l (Fig.1). No interfering peaks were seen in extracts of control leg breast or TABLE I RECOVERY OF DIMETRIDAZOLE ADDED IN METHANOL SOLUTION TO TISSUE Tissue Turkey liver Broiler breast* Broiler liver* Broiler liver* Broiler leg* Broiler breast* Turkey liver Turkey liver Dimetridazole added/mg kg-l . . . . 0.01 . . 0.05 0.05 1 . . . 0.05 0.05 0.05 0.10 1 .o No. of replicates 6 4 4 2 2 2 6 6 Recovery Coefficient of variation yo 80 f 8.9 11.4 (mean f s.d.) yo 94 f 7.5 9.0 91 f 4.0 4.7 98 87 84 80 f 2.0 2.5 89 f 1.9 2.1 * Recovery values obtained on “spiked” controls analysed concurrently with experimental samples.liver samples from either chickens or turkeys. The lower limit of sensitivity was calculated to be 0.005 mg kg-l based on a peak height of 1 cm at the lowest usable detector attenuation equivalent to 0.02 a.u.f.s. During the development of the method large and irreproducible losses were experienced if the final dichloromethane extract was evaporated to dryness on the rotary film evaporator even at a water-bath temperature as low as 30 “C. No loss of dimetridazole occurred if the extract was reduced to approximately 100 p1; the final volume could then be calculated by the addition of loop1 of the metronidazole solution before chromatography. Residues in Tissues No dimetridazole was detected in any tissue from either broilers or turkeys even when no withdrawal period had been applied.This was in contrast to the work of Condren et aZ.,5 C .- c Q) .- - IL 2 L I I 1 I I 0 4 8 12 16 20 24 Ti me/m i n Typical chromatogram of control turkey liver “spiked” with dimetridazole a t 0.05 mg kg-l on Spherisorb 5 ODS column showing (1) metronidazole and (2) dimetri-dazole. Flow-rate 1.0 ml min-’; detector 315 nm; attenuation 0.02 a.u.f.s.; injection volume 15 pl. Fig. 1 1094 HOBSON-FROHOCK AND READER DETERMINATION OF Analyst VOZ. 108 who found dimetridazole residues in all tissues after feeding trials with turkeys although the concentration of the drug in the feed between 250 and 1000 mg kg-l was considerably higher than the level used in this experiment. One of the difficulties in drug residue analysis is that “spiked” control samples do not exactly reproduce the physical or chemical relation-ships that would exist between the drug and the tissue matrix after administration of the drug to the biological system.Good recovery values merely show that there is no significant loss of the test substance during the manipulative process. They give no real indication that extraction of the drug from the tissue is complete; this can only be properly evaluated by treatment of the animal with the radioactively labelled compound. In the absence of such facilities the effect of the proteolytic enzyme subtilisin A13 was investigated to deter-mine if dimetridazole was bound to protein and therefore unextractable at the concentrations likely to be encountered.Samples of “no-withdrawal” turkey liver (10 g) were macerated with the enzyme (1 mg g-l) in a borate buffer at pH 9 (10 ml) and incubated for 30 min at 60 “C. No dimetridazole was detected although 91% of the drug was recovered when added at 0.05 mg kg-1 to control liver treated similarly. Comparison of resultant tissue residue concentrations with the level of dimetridazole fed to turkeys in the study reported by Condren et aL5 suggests that the amount of dimetridazole fed in these experiments would have produced detectable residues in the “no-withdrawal” samples. It is not clear whether the absence of dimetridazole in such tissues was due to no deposition of the drug (Le, (0.005 mg kg-l) or to losses during storage at -40 “C which, in this instance had been for 12 months.Information on the effect of storage would also be of value to the surveillance programme and to attempt to clarify these points the feeding trial was repeated. A third group of five broiler chickens was reared to 8 weeks of age as before on feed con-taining 75 mg kg-l of dimetridazole and slaughtered without application of a withdrawal period. The livers and muscle tissues (leg and breast) of each bird were removed immediately after death pooled and the extraction of duplicate samples started within a few minutes. The levels of dimetridazole are given in Table 11. The values are low and for leg and breast muscle approach the limit of sensitivity of the method. TABLE I1 DIMETRIDAZOLE RESIDUES IN BROILER TISSUES ANALYSED IMMMEDIATELY AFTER SLAUGHTER Dimetridazole Tissue concentration*/mg kg-l Meanlmg kg-l Liver .. 0.12 0.08 0.10 Breast muscle . . 0.03 0.03 0.03 Leg muscle . . 0.03 0.03 0.03 * Duplicate analyses on pooled tissues from five birds. Interference from Other Additives Solutions of six common poultry drugs amprolium arprinocid clopidol sulphaquinoxaline, chlortetracycline and txytetracycline were each chromatographed on the Spherisorb column with dimetridazole at equivalent concentrations. No interference with the dimetridazole peak was seen and it is unlikely therefore that the presence of these drugs would interfere with the determination of dimetridazole residues in poultry tissues. Conclusion An HPLC method has been developed for the determination of dimetridazole in poultry tissues.Recoveries were 80% or above at levels down to 0.01 mg kg-l with a detection limit of 0.005 mg kg-l. Dimetridazole was only detected in tissues analysed immediately after slaughter when broilers had received 75 mg kg-l of the drug for 8 weeks. Concentra-tions were then liver 0.10 leg muscle 0.03 and breast muscle 0.03 mg kg-l. No detectable amounts were found in any tissues from broilers or turkeys stored for 12 months at -40 “C irrespective of the withdrawal period of the drug before slaughter. Incubation with th September 1983 DIMETRIDAZOLE RESIDUES IN POULTRY TISSUES BY HPLC 1095 proteolytic enzyme subtilisin A13 did not increase the dimetridazole content of stored “no-withdrawal” turkey liver. The authors thank Mav & Baker Ltd. for the samde of dimetridazole and Mr.A. Machin (MAFF) for helpful discksions. MAFF. The work was fiAanced by the Food Science Division, 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References Stone L. R. and Hobson D. L. J . Assoc. OH. Anal. Chem. 1974 57 343. “Dimetridazole-Official First Action,” in Horwitz W. Editor “Official Methods of Analysis of the Association of Official Analytical Chemists,” Thirteenth Edition Association of Official Analytical Chemists Washington DC 1980 p. 702. Kane P. O. J . Polarogr. Soc. 1961 8 58. Law G. L. Mansfield G. P. Muggleton D. F. and Parnell E. W. Nature (London) 1963 197, Condren H. B. Davies R. E. Deyoe C. W. Zavala M. A. Creger C. R. and Couch J. R. Poult. Parnell M. J. Pestic. Sci. 1973 4 647. Allen P. C. and McLoughlin D. K. J . Assoc. OH. Anal. Chem. 1972 55 1159. Analytical Methods Committee Analyst 1971 96 746. Di Simone L. Ponti F. Settimi G. and Martillotti F. Farmaco Ed. Prat. 1981 36 440. Buizer F. G. and Severijnen M. Analyst 1975 100 854. Gulaid A. Houghton G. W. Lewellen 0. R. W. Smith J. and Thorne P. S. Br. J . Clin. Pharmacol., Kaye C. M. Sankey M. G. and Thomas L. A. Br. J . Clin. Pharmacol. 1980 9 528. Osselton M. D. J . Forensic Sci. SOC. 1977 17 189. 1024. Sci. 1963 42 586. 1978 6 430. Received February 3rd 1983 Accepted April 14th 198

ISSN:0003-2654    

DOI:10.1039/AN9830801091

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1096 Analyst September 1983 Vol. 108 PP. 1096-1101 Determination of Water in Ethanol and in Moist Air lrshad M. Pirzada and John H. Hills Department of Chemical Engineering University of Nottingham University Park Nottingham NG7 2RD Analyses of aqueous alcohols containing as little as O.Olyo of water in both liquid and gaseous phases were performed by gas chromatography on a Porapak Q column using a thermal conductivity detector. The method is simple quick reproducible and only a 0.1-pl volume of the liquid sample is needed. It has many uses including the determination of alcohol in a fermenter as i t is formed and the same column can be used to measure the absolute humidity (0.001-0.18 or more) of atmospheric air and other gases, using a 1.0-ml sample. Selected examples of experimental results are presented and the method is compared with other published methods.Keywords Water determination ; ethanol ; moist air; thermal conductivity detector ; gas chromatography The determination of the water content in alcohol - water mixtures has become more important owing to the increasing importance of alcohols as an alternative to fossil fuel and chemical feedstock. Little work has been done recently on the determination of the water content of aqueous alcohols. The following techniques have so far been reported. Kakabadsel determined trace amounts of water in alcohol - water mixtures on the basis of the change in electrode potential with variation of concentration of organic solvents. Bhaskare2 used the solvato-chromatic properties of nickel complexes to develop a spectro-photometric method for the determination of the water content in aqueous alcohols.In addition to the above other techniques such as reverse o~mosis,~ chromatography,* permeation through a membrane,5 photometric determination of water in ethanol by near-infrared spectrophotometrys and colorimetric determination of a trace amount of alcohol in ethanol - water mixtures' have also been tried. In the gas phase trace amounts of water vapour in different gases have been measured both by Aubeau8 and Andrawes.9 They first converted water vapour into acetylene by reaction with calcium carbidelo and then analysed it by gas chromatography. Weinbergll detected and analysed water contents in pulp by gas chromatography. Although the Karl Fischer technique for trace-water analysis is well known,12 it was reported by Aubeau8 that the delicacy of operation of this method particularly the need to keep the reactants in an anhydrous state is a large obstacle for a laboratory that rarely uses chemical methods of analysis.The following methods have been used in this department: firstly liquid-phase analysis including specific gravity spectrophotometry and chromato-graphy ; and secondly gas-phase analyses in which chromatography alone was suitable. Experimental Liquid-phase Analysis SpeciJic gravity method The specific gravity (density) of different standard solutions of aqueous ethanol was measured by a specific gravity bottle. Without the adoption of refinements of technique, which are time consuming and require expertise the results are not reliable as alcohol evaporates during weighing causing an error.Moreover any dissolved impurity in the alcohol may cause scatter in the results. Spectrophotometric method Ethanol was oxidised by Agulhons' solution13 [potassium dichromate in the presence of strongly acidic media (nitric acid)]. Different aqueous standard solutions of alcohol (ethanol + water O - l O O ~ o m/m) were prepared. A 7.0-p1 volume of this solution was oxidised by 5.0 ml of normal Agulhons' solution. After diluting it by a factor of two th PIRZADA AND HILLS 1097 absorbance of light at a particular wavelength (350nm) was measured by a Pye Unicam SP600 spectrophotometer using a 1.0-mm cell at 20.0 "C. Distilled water was used as reference. A standard calibration graph was plotted of absorbance veYsw alcohol concentra-tion (Fig.1). A comparison of the results obtained by this method and the specific gravity method using data from Perry1* is shown in Fig. 2. 1.6 1.2 al C n $ 0.8 Q P 0.4 10 30 50 70 90 [CzH50Hl =YO m/m 1 2 3 5 10 20 40 80100 True value of [CzH50H] Ol0 m/m Fig. 2. Graph showing a comparison of results obtained by 0 specific gravity Fig. 1. Standard calibration graph for the method (using Perry data) and 0 spectro-photometric method. spectrophotometric analysis of aqueous alcohol. The chemical reaction involved is the oxidation of aqueous ethanol by strongly acidic potassium dichromate. The reaction can be followed in principle by the change of colour from the yellow dichromate ion to the green chromium(II1) ion.2Cr2072- + 16H+ + 3CH3CH20H = 4Cr3+ + 11H20 + 3CH3COOH (Yellow) (Green) In fact a very strong Cr20,2- absorption band in the ultraviolet region at 350 nm was used, because this band is absent from chromium(II1). Fig. 3 produced on a Pye Unicam SPSOO spectrophotometer shows the discrimination available and Fig. 1 gives the calibration graph, obtained on a Pye Unicam SP600 spectrophotometer at this wavelength. The system has 2.0 1.6 0) F 1.2 +! $ 2 0.8 0.4 325 350 400 450 500 550 Wavelengthlnm Fig. 3. Spectra showing the movement of the absorption peak with increase in alcohol concentra-tion (Le. a decrease in Cr,O,z- ions) 1098 PIRZADA AND HILLS DETERMINATION Analyst Vol. 108 the following disadvantages the colour-forming reaction varies with temperature ; and if there is any sugar present in the alcohol it is oxidised by the dichromate solution which gives inaccurate absorbance values.Chromatographic method Standard samples for calibration were made by injecting separately known amounts of "superdry" alcohol (99.95%) prepared by the Lund and Bjerrum method (reference 15 p. 269) and water (liquid phase) into a pre-weighed bottle closed with a serum cap and weighing after each injection on a sensitive balance (accurate to 0.1 mg). The detector was of the thermal conductivity type with a WX filament (Gow-Mac Company). Both column and detector were held at 120 "C in the same oven. To permit accurate work at very low water contents a sensitive power supply unit was specially constructed to give a constant 100 mA and the output was fed to a Servoscribe potentiometric recorder.Helium was used as the carrier gas. The ratio of the compositions (mass of water to mass of ethanol) was plotted against the ratio of the peak heights on logarithmic co-ordinates to give a calibration graph as shown in Fig. 4. The resulting graph had a gradient of 1.13 (C) suggesting a non-linear relationship as follows: Absolute ethanol was used for experimental purposes. The column was 2.9 m long and supplied ready-packed with Porapak Q (Varian). Samples of 0.1-0.2 p1 (ethanol + water) were injected with a Hamilton syringe. water peak height Mass of ethanol = ( ethanol peak height 1 Mass of water To investigate the source of this non-linearity we estimated the peak areas by cutting and weighing and plotted the mass ratio against the area ratio; the results are also shown in Fig.4 and although more scattered they appear to lie on a line of gradient 1.0 i.e. a linear relationship. The non-linearity is thus due to the non-proportionality between peak height and peak area. Use of a higher column temperature of 150 "C improves this proportionality, giving a gradient of 1.03 for the graph of mass ratio against peak height ratio as shown in Fig. 4. 1 I 0.01 0.1 1 .o 10.0 100.0 Ratio Fig. 4. Graphs showing the ratio of the composition mass of water to mass of alcohol versus the ratio of A water to ethanol peak area; and B and C water to ethanol peak height. Measured a t a column temperature of x 150 "C; 0 120 "C; and 0 gas-phase analysis a t a column temperature of 120 "C September 1983 OF WATER I N ETHANOL AND I N MOIST AIR 1099 Without automatic integration facilities the peak height ratio is obviously more convenient and provided a calibration graph is constructed carefully slight non-linearities will not affect the accuracy.The higher temperature of 150 "C is to be preferred and the use of a constant sample volume will minimise any remaining errors. The svstem is sufficiently sensitive to detect the water in analytical-reagent grade ethanol :a1 output is shown in Fig. 5(b). [Fig. S(i)]. A tyI CZHSOH i Fig. 5. Spectra for the detection of water in (a) analytical-reagent grade ethanol ethanol 99.9% m/m and water 0.1 m/m and (b) aqueous ethanol ethanol 92.874% m/m and water 7.126% m/m.Liquid-phase analyses. Gas-phase Analysis An apparatus was constructed to produce a stream of air containing known amounts of alcohol and water vapours and chromatography was used to analyse such mixtures. The original equipment used two flasks pure water and pure alcohol respectively and air was bubbled separately through each of them before mixing the air streams in a known ratio. Control problems especially at higher temperatures led us to the modification described below. A measured flow-rate of water was flash evaporated by injecting it into a copper tube kept at a temperature above 600 "C in a tube furnace (Fig. 6). The water was supplied by a peristaltic pump from a calibrated water container of constant head.Air at 2.0 lmin-l was circulated to carry the water vapours through the tube. Liquid alcohol was heated in a separate flask to known accurately controlled temperatures. The said flask rested on a sensitive balance. The desired alcohol vapour stream was produced by bubbling a known amount of air through it using a fine sparger. The alcohol - vapour stream thus produced and the super heated water vapour from the tube furnace were mixed with a known flow of air in a round-bottomed flask which was held in a water-bath well above the dew-point of the resultant humid air. This gives a clear warning in the event of any condensation occurring. The mixture of air water and alcohol vapours was passed through the sample point and analysed by gas chromatography. Pre-dried air (passed through a column packed with activated silica gel) was used in all instances.The tubing used after the water-bath was thoroughly lagged with thermal tape. Air of desired absolute humidity was attained by varying the pump speed or air feed rate. The alcohol vapour feed rate was controlled by carrier air flow or by the temperature of the flask. The loss in mass of the alcohol flask was a measure of the amount of alcohol fed into the system. Free movement of the flask was ensured before weighing. The experiment was repeated for different water - alcohol con-centrations. Three to five vapour samples (1-2 ml) were taken each time from the sample point with a pre-heated Hamilton syringe and injected into the same chromatograph as used for liquid-phase analysis.Again the ratio of the two peak heights depends only on the ratio of the components in the vapour phase. The calibration graphs plotted for liquid-and gas-phase analyses are shown in Fig. 4 1100 PIRZADA AND HILLS DETERMINATION Analyst Vol. 108 Fig. 6. Apparatus for producing humid air of the desired alcohol - water concentrations. 1, Sample point 2 copper coil (dipped in water-bath); 3 resultant (air + ethanol + water) vapour temperature sensor; 4 copper tube; 5 tube furnace; 6 temperature controller and indicator; 7 air and water spray nozzle; 8 peristaltic pump; 9 water constant head with fine filter; 10 air measure-ment rotameter; 11 carrier air measurement rotameter for alcohol stream; 12 carrier air measurement for water vapour; 13 silica gel packed column for dry air; 14 condensate knockout system for com-pressed air; 15 water manometer and safety against an excess of pressure; 16 digital temperature indicator; 17 temperature controller for alcohol flask; 18 sensitive balance; 19 sparger ; 20 alcohol vapour temperature sensor; 21 liquid alcohol temperature sensor; 22 alcohol flask 23 water-bath ; 24 glass flask (air water and alcohol vapour mixing zone) and 25 temperature controller and indicator for water-bath.Water Vapour Analysis The high sensitivity of the thermal conductivity detector (TCD) to water vapour makes the system described under Gaseous Analysis ideal for the measurement of humidity of air or other gases. For the analysis of water vapours only the alcohol stream in the previously mentioned apparatus was disconnected from the rest of the system.Air of the desired absolute humidity was obtained by varying the pump speed or air feed rate. To allow for small variations in the sample volume a peak height ratio method was again used for calibration by comparing the water peak with the air peak. In order to keep the two peaks in the measurable range the sensitivity of the Servoscribe millivoltmeter was adjusted between the two peaks. The retention time was long enough (1.5min) to do this adjustment. The air peak which appears before the water peak was recorded at low sensitivity (200 mV) and the water peak was recorded at higher sensitivity (5.0 mV). The Hamilton syringe was pre-heated at 125-130 "C before every injection in order to avoid any condensation in the syringe.Although there is a risk of air oxidising the filament of the TCD we experienced no problem with the 1-2 ml sample used after hundreds of analyses of moist air. The results obtained with the apparatus in Fig. 6 were compared with some values obtained by using saturated air as follows. A certain amount of de-ionised water was taken in September 1983 OF WATER IN ETHANOL AND IN MOIST AIR 1101 glass flask. The neck of the flask was closed by a rubber bung with a small hole in it and the flask was left overnight in an oven at a particular temperature. Next morning the temperature of the water in the flask was read with a thermometer. A 2.0-ml volume of sample (gas phase) was taken by a Hamilton syringe through the small hole from the neck of the flask and analysed by gas chromatography as above.The experiment was repeated at different oven temperatures. Vapour pressure data were taken from reference 16 and the absolute humidity was calculated assuming saturation of the air. A comparison of results is shown in Fig. 7. Analysis of humid air samples was performed with the same column as described previously. 0.01 0.04 0.08 0.12 0.16 Wate r/a i r Fig. 7. Standard calibration graph for humid air analysis. Based on 9 peristaltic pump data; and 0 vapour pressure data. Conclusion Chromatography is a very suitable method for the analysis of alcohol - water mixtures It is capable of measuring water contents below With the Porapak Q packing used the method also proved reliable for measuring both in the liquid and vapour phases.0.1 yo mlm. the humidity of air down to 0.00015. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. References Kakabadse G. J. Abdulahed Maleila H. Khayat M. Tassopoulos G. and Vahdati A. Analyst, Bhaskare C. K. J . Shivaji Univ. 1974 7 165. Matsuura T. Baxter A. G. and Sourirajan S. Ind. Eng. Chem. Process Des. Dev. 1977 16(1), Morozov L. G. Khim. Khim. Tekhnol. 1974 16 137. Perry E. (Monsanto Co.) US Pat. Appl. 636,400 P1 Dec. 1975. Ferrer P. and Condal-Bosch L. Afinidad 1960 17 280. Reid V. W. Analyst 1955 80 704. Aubeau R. L. J . Chromatogr. 1964 16 7. Andrawes F. F. Am. Mineral. 1979 64 453. Sundberg 0. E. and Maresh C. Anal. Chem. 1966 38 1657. Weinberg B. B. J . Chromatogr. 1964 16 40. Vogel A. I. “A Textbook of Quantitative Inorganic Analysis,” Third Edition Longmans Group, Agulhons H. Bull Soc. Chem. Fr. 1911 9 July. Perry J . H. Editor “Chemical Engineering Handbook,” Fourth Edition McGraw-Hill New York, “Vogel’s Textbook of Practical Organic Chemistry,” Fourth Edition Longmans Group London, Weast R. C. Editor “Handbook of Chemistry and Physics,” Fifty-eighth Edition CRC Press, Received February 2nd 1983 Accepted April 21st 1983 1978 103 1046. 82. London 1972 p. 944. 1963 p. 3-84. 1978 p. 269. Cleveland OH 1976 p. D-232

ISSN:0003-2654    

DOI:10.1039/AN9830801096

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1102 Analyst September 1983 Vol. 108 pp. 1102-1107 Reversed-phase Paper Chromatographic Studies of Some Rare Earth Elements C. G. Yeole and V. M. Shinde Analytical Laboratory Department of Chemistry Shivaji University Kolhapur 41 6 004 India Reversed-phase paper chromatographic studies of rare earth elements such as scandium yttrium lanthanum neodymium and cerium have been carried out on Whatman No. 1 filter-paper impregnated with trioctylamine triiso-octylamine and Aliquat 336 S as stationary phases and using organic com-plexing agents such as sodium acetate sodium malonate and sodium succinate solutions as the active mobile phase. Results for the separation of ternary and quaternary mixtures are reported. Keywords Rare earth element separation ; reversed-phase paper chromato-graphy ; acid eluents Cation exchangers such as Dowex 50,l Zeocarb 225,2 Amberlite IR 120,394 Bio-Rad AG 50W-X€15 and Dowex AG 50W6 have been used for the separation of cerium scandium and yttrium from alkali metals and other elements such as lanthanum samarium erbium and ytterbium.The eluents used in these studies were prominently hydrochloric sulphuric or nitric acid. However hydrochloric acid solutions cannot be employed for the effective separation of adjacent rare earth elements because of the small difference in distribution coefficients. Hydrochloric acid can however be used for the separation of rare earth elements as a group from several other metal ions. In sulphuric acid solutions the rare earth elements show similar adsorption characteristics to those in hydrochloric acid solutions.The cation-exchange separation of lanthanides scandium and yttrium can also be performed in media containing fluoride,' thiocyanate,*sg pyrophosphate and tri(po1y)phosphatelO but these media do not offer any advantages. Ion-exchange chromatography utilising organic complexing agents is now the most efficient technique available for the separation of the rare earth elements. A number of complexing agents can be used such as a-hydroxyisobutyrate,ll lactate,12 citrate,13 glycollate and acetate.l* Some of these reagents are mediocre some require closely controlled con-ditions for success and some require high temperature The separation of yttrium-90 from strontium-90 has been performed on paper impregnated with ammonium t~ngstophosphate,~~ zirconium(1V) phosphate,16 zirconium tungstatel' and tin(1V) phosphatela; anion-exchange methods using Dowex 1 as the nitrate form can be used for the separation of scandium and yttrium from other rare earth elements as they do not adsorb on the resin.The mutual separation of the rare earth elements can also be performed at elevated temperatures by the use of gradient elution te~hniques.1~ Rare earth elements are much more strongly adsorbed from mixed aqueous - organic systems of hydrochloric nitric and sulphuric acids than from aqueous solutions of these acids20 These systems have not been used for analytical purposes. The adsorption of rare earth elements from thiocyanate media has been investigated by Pressly.21 Partition chromatography on cellulose columns has not found broad application for the separation of rare earth elements because many other metal ions are retained by the cellulose when diethyl ether containing nitric acid is employed as the mobile phase.This method can be employed however to separate the rare earth elements as a group from large amounts of uranium( VI) and thorium.22 Similarly many methods employing paper chro-matograph~~~-~' have been described for the separation of rare earth elements from one another or from other elements but a method that allows all the rare earth elements to be separated by paper chromatography is not yet available. Methods for the separation of rare earth elements by means of partition chromatography have also been described in which Kieselguhr silica gel or cation- or anion-exchange resins were used as supports for the stationary aqueous phase.%-32 A separation of the rare earth elements from scandium can be effected using buffer solutions of ammonium acetate and acetic acid of pH 5-6 at an elevated tem~erature.~O Attempts to use bis-(2-ethylhexy1) YEOLE AND SHINDE 1103 orthophosphoric acid (HDEHP) or tributyl phosphate (TBP) as the mobile phase have also been made.29,30-32 Cerium and yttrium are separated on a column of silica gel or on the cationite KU-2 using TBP saturated with 12 N nitric acid as the mobile phase.32 Scandium, yttrium and other tervalent lanthanides are not extractable from aqueous nitric acid solutions of any normality by diethyl ether hexone or alcohols.In the presence of large amounts of a salting-out agent e.g.lithium nitrate scandium can be extracted with diethyl ether to the extent of about 84%.33 Rare earth nitrates of cerium(III) lanthanum and neodymium can be extracted with alcohols e.g. hexanol but the aqueous phase must be about 90% saturated with ~alts.3~935 No analytical applications of such extractions have been reported. Because cerium( IV) can be reduced by h e ~ o n e ~ ~ to the non-extractable tervalent oxidation state the temperature during extraction should be low and the extraction should be carried out rapidly and in the presence of a suitable oxidising agent such as sodium bromate. The thiocyanates of the rare earth metals can be extracted from solutions containing large amounts of thiocyanates with butan01,~~ pentan0138 and ketones39 but no analytical applica-tion of this has been reported.Using TPB yttrium and cerium are extracted as Y(NO3)3.-3TBP and Ce(N03)3.3TRP.40 An extraction of scandium yttrium and the lanthanides is effected with TBP from 0.1 N nitric acid saturated with calcium nitrate,41 but TBP - nitrate systems have not been employed frequently for the analytical separation of rare earth elements because of low selectivity. Similarly the extraction of lanthanum cerium and neodymium from thiocyanate with TBP has not found analytical application^.^^ Among the acid esters of phosphoric acid proposed as extractants for the rare earths HDEHP proved the most selective.43 For the large-scale separation of adjacent rare earth elements by means of extraction with HDEHP the technique of counter-current extraction has been s u g g e ~ t e d .~ ~ ~ ~ This separation on an analytical basis is best performed using reversed-phase partition chromatography which has the disadvantage of forming hydrogen-bonded dimers in organic solvents. A separation of yttrium and lanthanum groups when present in carrier concentration can be effected by extraction with a 0 . 6 ~ solution of dibutyl phosphate in dibutyl ether.46 Neodymium is separated from praseodymium on a column containing methyltrialkyl-ammonium chloride47 as the stationary phase; the recovery of neodymium is 92%. Separa-tion of lanthanum cerium Praseodymium neodymium and samarium using a column of Kel-F with bis[ (2-ethylhexy1)phosphatel as the stationary phase has been reported48 and recoveries of the elements range from 92 to 100%.Papers impregnated with either tri-~ctylamine~~ or dinonylnaphthalenesulphonic acid50 have been used for the separation of uranium lanthanum samarium thorium and scandium. The mobile phase in these systems is either ammonium nitrate nitric acid or hydrochloric acid. Sodium sulphate hydrochloric acid and nitric acid are used as the mobile phase for the separation of rare earth elements on filter-paper (Whatman No. 1) impregnated with bis[(2-ethylhexyl)hydrogen p h ~ s p h a t e l ~ l - ~ ~ or bis [ (dihexylphosphinyl)methane] .54 In some of these methods separation is achieved at a high temperature. Cerium samarium and europium are separated on Whatman filter-paper impregnated with a mixture of 2-thenoyltrifluoroacetone and trioctylphosphine oxide55 with hydrochloric acid as the developing agent but the development period is 5 h.How-ever thin-layer chromatography on silica gel56 and paper ~hromatography~~ have been used for the separation studies of rare earth elements. Amongst the separation techniques procedures using reversed-phase chromatography have become increasingly important in recent years especially for the separation of rare earth elements. This is because the separation factors between adjacent rare earth elements are, in several instances better when extracting these elements with organic phosphorus com-pounds than when eluting from cation-exchange resins in the presence of organic complexing agents. A systematic technique for the separation of rare earth elements with an organic complexing agent has not been developed.In this work organic complexing agents such as sodium acetate sodium malonate and sodium succinate were used as the active mobile phase and liquid ion exchangers such as trioctylamine (TOA) triisooctylamine (TIOA) and Aliquat 336 were used as stationary phases on Whatman filter-paper mainly to exploit their anion-exchange properties. The effects of pH concentration of mobile phase and stationary phase on R values have been critically studied and optimum conditions have been developed to separate rare earth element 1104 YEOLE AND SHINDE REVERSED-PHASE PAPER Analyst VoZ. 108 such as scandium lanthanum yttrium neodymium and cerium from ternary and quaternary mixtures. Experimental Preparation of Stationary Phase High relative molecular mass amines (HMWA) such as TOA TIOA or Aliquat 336 (Koch-Light or Fluka) are dissolved in benzene to give a 2 or 5% m/V solution and are subsequently equilibrated with equal volumes of 1 M sodium acetate sodium malonate or sodium succinate solutions for 1 h (this converts the amines into the salt forms i.e.[R,NH+ acetate-], [R,NH+ malonate-] and [R,NH+ succinate-1). The solution so obtained is then sprayed on Whatman No. 1 filter-paper. The paper is dried with warm air and used for ascending reversed-phase chromatography which is performed in glass jars. An aqueous solution of sodium acetate sodium malonate or sodium succinate adjusted to pH 4 with dilute hydrochloric acid and sodium hydroxide solution (using a precision type Philips pH meter) is used as the active mobile phase.The stock solutions of lanthanum and yttrium are prepared by dissolving their oxides in nitric acid; solutions of scandium and neodymium are prepared by dissolving their oxides in hydrochloric and perchloric acid respectively. Cerium( IV) solution is prepared by dissolving ammonium cerium(1V) sulphate in sulphuric acid. The concentration of the metal ion in these solutions was found to be 2.5 mg ml-l. TABLE I RF VALUES OF SCANDIUM(III) YTTRIUM(III) LANTHANUM(III) NEODYMIUM(III) AND CERIUM(IV) ON PAPER IMPREGNATED WITH HMWA AS THE STATIONARY PHASE USING ACETATE MALONATE AND SUCCINATE SOLUTIONS (PH 4) AS THE MOBILE PHASE Mobile phase f A \ Stationary phase Sodium acetate -7+ Metal ion 'Amine Concentration O i 0.05 M Sc(II1) .. Y(II1) . . La(II1) . . Nd(II1) Ce(1V) . . TOA TIOA Aliquat 336 TOA TIOA Aliquat 336 TOA TIOA Aliquat 336 TOA TIOA Aliquat 336 TOA TIOA Aliquat 336 2 0.70 5 0.76 2 0.69 5 0.73 2 0.75 5 0.86 2 0.89 5 0.90 2 0.89 5 0.90 2 0.94 5 0.94 2 0.15 5 0.11 2 0.08 5 0.04 2 0.10 5 0.05 2 0.18 5 0.10 2 0.30 5 0.12 2 0.09 5 0.09 2 0.45 5 0.47 2 0.50 5 0.43 2 0.53 5 0.53 0.1 M 0.73 0.77 0.76 0.80 0.83 0.83 0.89 0.92 0.89 0.92 0.99 0.97 0.20 0.10 0.09 0.12 0.10 0.11 0.16 0.12 0.12 0.10 0.07 0.10 0.62 0.52 0.30 0.46 0.52 0.55 Sodium malonate & 0.05 M 0.69 0.69 0.70 0.70 0.95 0.96 0.95 0.93 0.93 0.93 0.71 0.74 0.13 0.08 0.14 0.10 0.20 0.12 0.19 0.10 0.10 0.01 0.01 0.02 0.40 0.44 0.61 0.55 0.58 0.60 0.1 M-0.70 0.73 0.73 0.73 0.93 0.94 0.95 0.92 0.95 0.92 0.81 0.84 0.16 0.10 0.09 0.08 Tailing 0.14 0.13 0.10 0.10 0.01 0.01 0.02 0.50 0.46 0.69 0.57 0.57 0.58 Sodium succinate r-+ 0.05 M 0.75 0.75 0.60 0.65 0.70 0.41 0.94 0.92 0.93 0.92 0.95 0.86 0.08 0.02 0.08 0.12 0.24 0.10 0.11 0.14 0.11 0.07 0.10 0.03 0.54 0.57 0.60 0.48 0.60 0.58 0.1m 0.76 0.82 0.76 0.78 0.66 0.44 0.94 0.95 0.93 0.95 0.96 0.92 0.08 0.02 0.10 0.10 0.25 0.24 0.12 0.10 0.10 0.07 0.10 0.03 0.56 0.50 0.58 0.50 0.65 0.6 September 1983 CHROMATOGRAPHY OF SOME RARE EARTH ELEMENTS 1105 Procedure Take Whatman No.1 filter-paper strips (15 x 4 cm) spray with amine (TOA TIOA or Aliquat 336) solutions dissolved in benzene and pre-equilibrated with either sodium acetate, sodium malonate or sodium succinate solutions as described earlier and dry the strips with warm air thus leaving the amines on the paper as the stationary phase. Spot the desired metal ion solution or the metal ion solution mixture on the strip using capillaries of a few micrometres diameter and subsequently keep these strips in glass jars (21.5 x 5.2 cm). The jars should contain an aqueous solution of either sodium acetate sodium malonate or sodium succinate (0.05 or 0.1 M concentration) adjusted to pH 4.0 with dilute hydrochloric acid and sodium hydroxide solutions.Allow the solvent to run (by the ascending technique) for 15 cm and after drying the papers detect the cation by spraying the appropriate reagent solution in a sequential order. Calculate the R value for the individual cation by the usual procedure. Caution-Benzene is highly toxic and appropriate precautions should be taken. Detection Scandium and yttrium are detected with an aqueous solution of 0.1% 4-(2-pyridylazo)-resorcin01~~ and 0.1% Arsenazo 11159 solution to give red and green spots respectively. Lanthanum neodymium and cerium are detected with an aqueous solution of 0.1% Alizarin Red S60y61 to give yellow red and yellow spots respectively. TABLE I1 OPTIMUM CONDITIONS FOR THE SEPARATION OF TERNARY AND QUATERNARY MIXTURES Mixture Y,Sc La Y Sc Nd Sc Ce Nd .. Y Ce Nd Y,Ce,La Sc Y La Sc Y Nd Y Ce Nd . . Sc Ce Nd . . Y Ce La Y,Sc,La Y,Sc,Nd Y Ce Nd Y Sc La Y,Sc Nd Sc Ce Nd SC ce La . . Y Ce La Y Ce Nd Sc Ce Nd . . Sc Ce La . . Y Ce Nd . . Y Ce La Y Sc La Sc Ce La . . Y Ce La Y,Sc,Ce Sc Ce Nd . . Y Ce Nd Y,Sc,La Y,Sc,Nd Sc Ce Nd SC ce La Y Ce La Y Ce,Nd Y,Sc La Y,Sc Ce Y,Sc Nd Y Ce Nd Y ce La Sc Ce Nd . . Y Sc Ce La Y Sc Ce Nd Y Sc Ce La Y Sc Ce Nd Stationary phase Amine Concentration /o Mobile phase - Aliquat 336 Aliquat 336 TOA TIOA J 0.1 M sodium succinate 1 0.1 M sodium malonate 0.1 M sodium succinate } 0.05 M sodium acetate 0.1 M sodium malonate 1 } 0.1 M sodium acetate Time required for 15-cm developmen t/min RF value 100 100 100 100 100 100 100 100 100 100 80 80 80 85 85 85 85 85 85 85 85 85 85 85 85 85 85 85 85 100 100 65 65 65 65 95 95 95 95 96 95 95 95 95 95 Y; 0.921 Ce; 0.531 La,*O.lO Sc 0.94; Y 0.84; La 0.14 Sc 094.Y 084.Nd 002 Y,'O.84)Cd 0.58'; Nd 0.02 Sc 0.88; Ce 0.56; Nd 0.02 Y 0.85; Ce 0.60; La 0.14 Y 0 96- Sc 0 72.La 0 25 Y' 0'96 Sc) 0'66; Nd 0 10 Y O.*96f Ce 0:65 Nd 0:02 Y 0.90; Sc 0.76; La 0.11 Y 090. Sc 076.Nd 010 Sd 0.74; C; 0.4;; Nd 0.10 Sc 0 76- Ce 047. La 0 11 Y '0 bo*'ce'o 4 7 - ' ~ a ' 0 ii Y 0:90 Ce 0:45f Nd 6.10 Sc 073.Ce 046.Nd 010 Sc) 0'73 C< 0'46 Nd' 0'10 Y'090~Ce'0'45~Nd'OiO Y 0:92 Ce 0:44 La,'O.O8 Y 0 92- Sc 0 82.La 0 02 sd 0 s i - d 0 5 i . ~ i ; do2 Y ' 0 '95 -' ~e ' 0 51 :La '0 02 Y' 0'95; Sc' 0'82 Ce'050 Sd 0 83 Cd 0 56. Nh 0 10 Y ,'0.95 :Ce,'0.50 ;"d,'o.io Y 09O.Sc 0 7 3 . b 0 0 4 Y 0:90 Sc 0:73f Nd 6.12 Sc O73-Ce 057.Nd 001 S ; 0:731 Cg 6.57 La,' 0.08 Y. 0.92 Ce. 0.52 La. 0.08 Y; 0.90; Ce; 0.55; Nd 0.01 Y 0 9 5 . S ~ 074- La 010 Y' 0'97; Sc) 0'78 Ce'Ok5 Y' 0'95 Sc) 079 Nd 0 07 YT 0:951 CL 0:BOI Nd 0 0 7 Y' 0 95 C& 0 50 L a - 0 10 Sd 079'.Cg d5i.N; 0 0 7 Y '0 94.kc b Sb-ke 0'4i. La o 10 Y 0:94 Sc 0:81 f Ce O h f Nd 0:07 Y 0 8 8 . S ~ 078.Ce 0 4 3 - L a 010 Y:O.89 ;'Sc,b.80; b16.&; Ad 6.0 1106 YEOLE AND SHINDE REVERSED-PHASE PAPER Analyst VOZ.108 Results and Discussion Various experiments have been carried out to study the effect of the pH (2-7) of the mobile phase concentration of the mobile phase (0.01-0.1 M) and concentration of the stationary phase (1-5y0 using benzene as a diluent) on the R values of individual cations. The extrac-tion studies on scandium yttrium neodymium and cerium carried out in our laboratory with trioctylamine have shown that the optimum extraction of these metal ions occurs using a 0.05-0.1 M sodium succinate solution with a 5% solution of extractant. For the reversed-phase paper chromatographic study also it has been observed that for sequential development and subsequent separation of metal ions on paper the best results are obtained when the concentration of mobile phase (sodium acetate sodium malonate and sodium succinate) was taken as 0.05-0.1 M.The variation in concentration of mobile phase between 0.05 and 0.1 M showed only a slight effect on the RF values (Table I). Similarly variation in concentration of stationary phase (TOA TIOA and Aliquat 336) has shown that either a 2 or 5% solution of amine sprayed on paper gives good results and the variation in concentration of stationary phase between 2 and 5% had very little effect on the RF values. Hence for further studies and mutual separation of rare earth elements a 2 and 5% solution of amine (as stationary phase) and a 0.05-0.1 M solution of mobile phase were fixed. The results for a 15-cm run are reported in Table I. The variation in pH (2-7) of the mobile phase showed that the pH should be maintained between 4 and 7.The RF values of individual cations are almost constant between pH 4 and 7. Irreproducible results were obtained at pH <4 and >7. Although some separations are possible at pH 7 in most instances the pH of the mobile phase was maintained at 4 because it facilitates the subsequent detection of metal ions. The chromatograms were developed at room teperature only; an increase in temperature showed no significant effect on the RF values. Liquid ion exchangers used as the stationary phase on paper are in the salt form such as acetates malonates and succinates. The subsequent development of the metal ion occurs as a result of complex formation. The metal acetates succinates or malonates undergo an anion-exchange process on the paper thus causing the differential rates of migration.For instance trioctylamine in the succinate form [R3NH+(C4H40J2-] undergoes an exchange reaction with scandium succinate to give [R,NH+Sc(C,H,O,),-]. The proposed method is suitable for the separation of ternary and quaternary mixtures (Fig. 1). The results are reported in Table 11. The precision data for scandium yttrium, lanthanum cerium and neodymium given in Table I11 show that the method is reproducible. The time required for 15-cm development ranges between 65 and 100 min. TABLE I11 PRECISION DATA RF values* 7-sodium sodium sodium Metal ion acetate malonate succinate 0.1 M 0.1 M 0.1 M s c . . 0.80 0.73 0.78 Y . . . . 0.92 0.92 0.95 La . . . . 0.12 0.08 0.10 Nd .. . . 0.10 0.01 0.07 Ce . . . . 0.46 0.57 0.50 Standard deviation h f > 0.1 M 0.1 M 0.1 M sodium sodium sodium acetate malonate succinate 0.014 0.014 0.009 0.014 0.014 0.014 0.014 0.009 0.009 0.009 - 0.014 0.014 0.014 0.014 Coefficient of variation % sodium sodium sodium acetate malonate succinate 1.76 1.93 1.14 1.53 1.53 1.48 11.79 11.18 8.94 8.94 - 20.2 3.07 2.48 2.82 A f \ 0.1 M 0.1 M 0.1 M * Mean RF values of six chromatographic runs on a stationary phase containing 5% TIOA solution in benzene. References 1. 2. 3. 4. Seim H. J. Johnson J . L. Stever K. R. and Heady H. H. US. Bur. Mines Rep. Invest. 1962, Cabell M. J. U.K. At. Energy Res. Establ. Rep. 1957 223. Grant C. L. Anal. Chem. 1961 33 401.Schubert J. Anal. Chem. 1950 22 1359. 6097 Fig. 1. (1) Mobile phase 0.1 M sodium acetate (pH 4) ; stationary phase Aliquat 5y0 ; time of development 100 min; RF values Y (0.92), Ce (0.53) and La (0.10). (2) Mobile phase 0.1 M sodium acetate (pH 4) ; stationary phase Aliquat 5% ; time of development 100 min; RF values, Sc (0.85) Ce (0.52) and Nd (0.07). (3) Mobile phase 0.1 M sodium succinate (pH 4) ; stationary phase TIOA 5% ; time of development, 95 min; RF values Y (0.95) Sc (0.80) and Nd (0.07). (4) Mobile phase, 0.1 M sodium succinate (pH 4); stationary phase TIOA 5%; time of development 95 min; RF values Y (0.95) Ce (0.50) and Nd (0.07). (5) Mobile phase 0.1 M sodium succinate (pH 4); stationary phase, TIOA 5%; time of development 95 min; RF values Y (0.95) Sc (0.74) and La (0.10).(6) Mobile phase 0.1 sodium succinate (pH 4); stationary phase TIOA 5% ; time of development 95 min; RF values, Y (0.97) Sc (0.78) and Ce (0.45). (7) Mobile phase 0.1 M sodium succinate (pH 4) ; stationary phase TIOA 5% ; time of development, 95 m h ; RF values Y (0.94) Sc (0.80) Ce (0.41) and La (0.10). (8) Mobile phase 0.1 M sodium succinate (pH 4) ; stationary phase TIOA 5 % ; time of development 95 min; RF values Y (0.94) Sc (0.81) Ce (0.41) and Nd (0.07). (9) Mobile phase 0.1 M sodium acetate (pH 4); stationary phase TIOA 5% ; time of development 95 min; RF values, Y (0.88) Sc (0.78) Ce (0.43) and La (0.10). (10) Mobile phase 0.1 M sodium acetate (pH 4) ; stationary phase TIOA 5% ; time of develop-ment 95 min; RF values Y (0.89) Sc (0.80) Ce (0.46) and Nd (0.07).[to face page 110 September 1983 CHROMATOGRAPHY OF SOME RARE EARTH ELEMENTS 1107 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. Strelow F. W. E. and Bothma C. J. C. Anal. Chem. 1964 36 1217. Strelow F. W. E. Rethemeyer R. and Bothma C. J . C. Anal. Chem. 1965 37 106. Hettel H. J. and Fassel V. A. Anal. Chem. 1955 27 1311. Hamaguchi H. Kuroda R. Aoki K. Sugisita R. and Onuma N. Talanta 1963 10 153. Pietrzyk D. J. and Kiser D. L. Anal. Chem.1965 37 233. Subbaraman P. R. Rajan K. S. and Gupta J. Curr. Sci. 1959 28 63; 2. Anorg. Chem. 1960, Seyb K. E. and Hermann G. 2. Elektrochem. 1960 64 1065. Herrmann G. and Strassmann F. 2. Naturf. Part A 1956 11 946. Marathe E. V. J . Sci. Ind. Res. Sect. B 1955 14 354. Minami E. Honda M. and Saseki Y. Bull. Chem. SOC. Jpn. 1958 31 372. Prasilova J. and Sebesta F. J . Chromatogr. 1964 14 555. Sastri M. N. and Rao A. P. J . Chromatogr. 1962 9 250. Riedel H. J. Nukleonik 1963 5 48. Chu-Chun Cheng A d a Chim. Sin. 1965 31 547. Marcus Y. and Nelson F. J . Phys. Chem. 1959 63 77. Edge R. A. J . Chromatogr. 1961 6 452. Pressly R. S. USAEC Rep. ORNL 2843 1960. Feldman C. and Ellengburg J. Y. Anal. Chern. 1958 30 418. Johnson 0. H. and Krousa H. H. Anal. Chim. Acta 1954 11 128.Danon J. and Levi M. C. J . Chromatogr. 1960 3 193. Shu-Wei Pang Chein-Chuan Lei and Shu Chuan Liang Acta Chim. Sin. 1964 30 160. Nagai H. and Kamo M. Bunseki Kagaku 1965 14 33. Gupta S. S. and Mukerjee D. 2. Anal. Chem. 1965 213 38. Moghissi A. J. Chromatogr. 1964 14 542. Daneels A. Massert D. L. and Hoste J. J . Chromatogr. 1965 18 144. Kuteinikov A. G. and Brodskaya V. M. 2h. Anal. Khim. 1962 17 305. Smell H. J. Inorg. Nucl. Chem. 1961 19 160. Chuveleva E. A. Nazarov P. P. and Chmutov K. V. Zh. Fiz. Khim. 1962 36 1022. Bork R. and Bock E. 2. Anorg. Chem. 1950 263 146. Templeton C. C. J . Am. Chem. Soc. 1949 71 287. Templeton C. C. and Petersen J . A. J . Am. Chem. Soc. 1948 70 3967. Maeck W. J. Boomen G. L. Kussy M. E. and Rein J. E. Anal.Chem. 1961 37 1775. Appleton D. B. and Selwood P. W. J . Am. Chem. Soc. 1941 63 2029. Asschin E. F. Aurieth L. F. and Comings E. W. J . Phys. Colloid Chem. 1950 54 690. Fischer W. Bramekamp K. J. Klinge M. and Pohlmann H. P. 2. Anorg. Chem. 1964 329 44. Scargill D. Alcock K. Fletcher J . M. Hesford E. and Mckay H. A. C. J . Inorg. Nucl. Chem., Peppard D. F. Faris J. P. Gray P. R. and Mason G. W. J . Am. Chem. Soc. 1953 75 4576. Kiesl W. Sorantin H. and Pfeifer V. Mikrochim. Acta 1963 496. Wischow R. P. and Horner D. E. USAEC Rep. ORNL 3204 1962. Peppard D. F. Mason G. W. and Moline S. W. J . Inorg. Nucl. Chem. 1957 5 141. Peppard D. F. Mason G. W. Maier J. C. and Driscoll W. J. J . Inorg. Nucl. Chem. 1957 4 334. Scedden E. M. and Bellon N. E. Anal. Chem. 1953 25 1602.Li Lingying and Sun Yuanming Fen Hsi Hua Hsueh 1981 9 94. Li Lingying Sun Yuanming and Zhong Fenglin Fen Hsi Hua Hsueh 1981 9 48. Shu-Wei Pang and Shu-Chuan Liang Acta Chim. Sin. 1964 30 401. Werner G. 2. Chem. 1965 5 147; 1965 5 31. Shu Wei Pang and Shu-Chuan Liang Acta Chim. Sin. 1963 29 319. Cerrai A. and Ghersini G. J . Chromatogr. 1966 24 383. Tiao-Hsu Chang and Tsan-Lun Chow Hua Hsueh 1969 1-2 18. O’Laughlin J. W. Ferguson J. W. Richard J . J. and Banks C. V. J . Clzromatogv. 1966 24 376 Cvjeticanin N. J. Chromatogr. 1972 74 99. Rafizadeh M. and Specker A. Naturzvissenschaften 1976 63 483. Nagei H. Deguchi T. and Nakai K. Bunseki Kagaku 1972 21 788. Busev A. I. and Fan Chang Talanta 1962 9 101. Bark L. S. Duncan G. and Graham R. J. T. Analyst 1967 92 347. Belcher R. and Frieser H. “Indicators,” First Edition Pergamon Press Oxford 1972 p. 362. Brunisholz G. and Coheen R. Helv. Chim. Acta 1956 39 324. 304 191. 1957 4 304. Received August 13th 1982 Accepted April 18th 198

ISSN:0003-2654    

DOI:10.1039/AN9830801102

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1108 Analyst September 1983 Vol. 108 pp. 1108-1113 Solvent Extraction of Rare Earth Metals With Crown Ethers Lin-Mei Tsay Jeng-Shang Shih* and Shaw-Chii Wu* Department of Chemistry National Taiwan Normal University of Taipei Taiwan 117 Republic of China Crown ethers such as 15-crown-5 12-crown-4 and dibenzo- 18-crown-6 were used to extract individual rare earth ions from aqueous solutions containing picrate into nitrobenzene solution. The rare earth metal ion europium(II1) is extracted as a 2 1 crown - ion sandwich complex with 12-crown-4 but as a 1 1 complex with both 15-crown-5 and dibenzo-18-crown-6. From studies of picric acid concentration effects on complexation the extracted species of Eu (NO,) with 15-crown-5 and dibenzo- 18-crown-6 are Eu [ ( 15-crown-5)-(picrate),(NO,)] and Eu[(dibenzo-l8-crown-6) (picrate) ,(NO,)] respectively, but Eu[( 12-crown-4) (picrate),] is formed with 12-crown-4.The extraction of rare earth ions showed that Tb3+ Eu3+ Gd3+ Nd3+ and Yb3+ can be easily extracted using 15-crown-5; however the extraction of Ce4+ Sm3+ Dy3+ and Lu3+ is more difficult. Keywords Rare earth metals ; solvent extraction; crown ethers It has been demonstrated that macrocyclic polyethers can behave as highly selective complex-ing agents for cati~nsl-~ and are of potential significance in the separation of ions using solvent extraction procedures. Solvent extraction methods can not only be applied in the separation of ions but can also provide information concerning the nature of the complex formed between the ion and the ligand.Extractions of alkali metal ions with crown ethers have been exten-ively ~tudied,~-~ but there are very few reports on the extraction with crown ethers of transition metal ions and rare earths The complexation of rare earth ions by crown ethers has been little studied until recently10-20 and rare earth - crown ether complexes have been i ~ ~ l a t e d . ~ ~ $ ~ ~ ~ ~ ~ ~ ~ ~ In general rare earth ions can form 1 1 or 2 1 crown to ion complexes with 1 2 - c r o ~ n - 4 ~ ~ J ~ ~ ~ ~ but only 1 1 complexes with 15-crown-5 and 18-cr0wn-6.~~ Rare earth metals have similar chemical properties and are difficult to separate from each other. Because crown ethers showed unusual sensitivity to the size of the metal ion rare earth ions with reasonable difference in their sizes may be separated from each other or from other transition-metal ions by extraction with crown ethers.In this work rare earth metals were extracted with 15-crown-5 (I) dibenzo-18-crown-6 (11) and 12-crown-4 (111) ethers. This work has provided information about the relative stabilities of rare earth - crown ether complexes and presented evidence for the formation of a rare earth - 12-crown-4 ether sandwich complex. I II 111 Experimental Reagents Crown ethers 15-crown-5 and 12-crown-4 were obtained from E. Merck. Dibenzo-18-crown-6 was synthesised All reagents except the following were of analytical-reagent grade. * To whom correspondence should be addressed TSAY SHIH AND WU 1109 as described by Pedersen-l Rare earth oxides were obtained from Fluka and picric acid and nitrobenzene from E.Merck. The mixture of radioisotopes 152E~ and 154E~ was prepared from the irradiation of Eu,03 with 1 x 1013 n cm-2 thermal neutrons in a nuclear reactor at the Nuclear Research Institute in Taiwan. All rare earth nitrates were prepared from the dissolution of rare earth oxides in nitric acid. Apparatus with a sodium iodide scintillation counter. determined with a Spectrametrics SMI-I11 DC plasma emission spectrometer. The total radioactivity of the mixture of europium isotopes 152E~ and 154E~ was measured Concentrations of other rare earth ions were Procedure extracted with 0.1 g of a given crown ether in 10 ml of nitrobenzene. the rare earth ions used was in the range 0.640 p.p.m. room temperature for 30 min for extraction.The rare earth ions in 10 ml of aqueous solution containing 10 mg ml-1 of picric acid were The concentration of The mixed solution was shaken at Theory The distribution coefficient (D) of the extraction is given by the following equation : where M3+ L and A are the rare earth ion crown ether and picrate ion respectively For the extraction of alkali metal ions with crown ether^,^ we can assume that for a large excess of A-, most of the MLn3+ ions readily form the complex [MLn3+] [A-1 which can be extracted into the organic phase thus [MLn3+]aq may be considered much smaller than [M3+]aq and D becomes under equilibrium conditions. The partition of L between organic solvent and water is given by Complex formation in aqueous solution is expressed by Extraction of the complex from the aqueous phase into the organic phase is given by By combining equations (2) (3) (4) and (5) we obtain KexKf [Llnorg Kdn D = When the volumes of organic and aqueous solutions are the same the initial concentration of crown ethers [Lo] can be expressed by EL01 = [Llorg + [Llaq (7 1110 TSAY et al.SOLVENT EXTRACTION OF Analyst Vol. 108 Further by insertion of equation (3) into equation (7) the following relationship can be obtained : Hence by substitution of equation (8) into equation (6) : Kex Kf [&I" [A-I (1 + Kd)" D = In the extraction picric acid (HA) dissociates according to the equation (9) Under the conditions HA 9 M3+ and [H+] = [A-] [A-I = (Ka [HA]), and equation (9) becomes and (13) 0 + c -0.5 .-0 5 8 c -1.0 a $ -1.5 CR .- t l c --2.0 -2.5 3 3.5 4 4.5 Log(concentrati0n of picric acid/pg mi-Fig.2. Effect of picric acid con-centration on the extraction of Eu3+ ions with various crown ethers [0.1 g of crown ether and 0.588 pg ml-l of Eu[NO,),]. A 15-Crown-5 slope = 1.16; B 12-crown-4 slope = 1.43; and C dibenzo-8-crown-6 slope = 1.10. I I I -2 -1 0 1 Log(concentration of crown ether M) Fig. 1. Effect of crown ether con-centration on the extraction of Eu3+ ions [0.588 pg ml-l of Eu(NO,) and 10 rng ml-l of picric acid]. A 15-Crown-5 slope = 1.12; B 12-crown-4, slope = 1.83; and C dibenzo-&crown-6 slope = 1.07 September 1983 RARE EARTH METALS WITH CROWN ETHERS 1111 Results and Discussion It is reasonable to predict that rare earth ions can form stronger complexes and be easily extracted by 15-crown-5 ether because 15-crown-5 has the most suitable cavity size (0.85-1.1 A)21 to fit a rare earth ion with the size 0.85-1.06 A,22 while dibenzo-18-crown-6 (1.3-1.6 A or 12-crown-4 (0.65-0.70 A) ethers are either too big or too small for rare earth ions.This is shown in Fig. 1 where for a given concentration of crown ether Eu3+ is more readily extracted with 15-crown-5 than 12-crown-4 or dibenzo-18-crown-6. Fig. 1 also shows the linear relation-ship between distribution coefficient (lo@) and crown ether concentration (log [L,]) which is TABLE I EFFECTS OF VARIOUS ORGANIC ACIDS ON THE EXTRACTION OF Eu3+ IONS WITH 15-CROWN-5 A 0.1-g mass of 15-crown-5 in 10 ml of nitro-benzene was used and each acid contained 10 mg ml-1 of the individual acid and 0.588 pg ml-l of Eu(NOB), in 10 ml of aqueous solution.Acid D LogD Picric 1.29 0.11 Tartaric . . . . 3.00 x -2.52 Citric . . 2.96 x -1.53 Acetic . . 3.46 x -2.46 Oxalic . . 3.38 x -1.47 Sebacic . . . . 2.99 x -2.52 in agreement with equation (13). Equation (13) shows that when the concentration of picric acid (HA) remains constant the graph of logD against log[L,] must give a straight line with slope n. From the n value the structure of extracted species (e.g. EuLn3+) can be elucidated. The slopes of the graphs in Fig. 1 seem to indicate that the Eu3+ ion can form 1 1 (crown ether - Eu3+) complexes with 15-crown-5 and dibenzo-18-crown-6 but forms a 2 1 sandwich complex with 12-crown-4.La Ce Pr Nd PmSm Eu Gd Tb Dy Ho Er Tm Yb Lu Atomic number Fig. 3. Graph of distribution coefficients (D) v m u s atomic numbers of rare earth ions extracted by 15-crown-5 (0.1 g of 15-crown-5 in 10 ml of nitrobenzene was used to extract the individual rare earth 40 pg ml-l in an aqueous solution containing 10 mg ml of picric acid) 1112 Analyst Vol. 108 In addition according to equation (13) the picric acid concentration [HA] should affect the extraction. As shown in Fig. 2 at constant concentration of crown ether the graph of picric acid concentration (log [HA]) veysus the distribution coefficient (logD) gives a straight line for each ether. The m values obtained from the slopes of the graphs ( i e . slope = m/2) seem to indicate that the extraction species of Eu3+ as Eu[(crown ether) n(picrate),] with 15-crown-5 and dibenzo-18-crown-6 may be Eu[(l5-~rown-5)(picrate)~(NO~)] and Eu[(dibenzo-l8-crown-6) (picrate),(NO,)] but Eu[( 12-crown-4),(picrate) 3] with 12-crown-4.Besides picric acid, some other organic acids were also tried for the extraction. However as shown in Table I, picric acid is the best which may be due to its higher dissociation constant (0.42 M) or the stronger tendency of the picrate anion to form an ion pair with the complexed cation than other organic acids of interest. The extraction of individual rare earth ions with 15-crown-5 was investigated. The graph of D VCYSUS the atomic number of rare earth ions is shown in Fig. 3. The extractions of rare earth ions showed that Tb3+ Eu3+ Gd3+ Nd3+ and Yb3+ can be most easily extracted with 15-crown-5 but the extraction is difficult for Ce4+ Sm3+ Dy3+ and Lu3+.A similar trend was found in the study of the adsorption of the rare earth ion with a synthetic adsorbent of 15-crown-5 with phosphomolybdic acid - p~lyacrylamide.~~ In general heavier rare earth ions always have stronger complexing abilities with ligands owing to their greater ionic potentials (charge/radius). However the complexing behaviour of an ion with a crown ether depends not only on the ionic potential but also the relative size of the ion and cavity of the crown ether ring. As shown in Fig. 3 the Tb3+ ion is the easiest rare earth ion to be extracted. There is not however enough evidence to show whether this is due to the best fit of the Tb3+ ion into the cavity of the 15-crown-5 ether; the actual reason is still not understood and further study is necessary.TSAY et al. SOLVENT EXTRACTION OF TABLE I1 EFFECT OF THE PRESENCE OF OTHER TRANSITION METAL IONS ON THE EXTRACTION OF THE Eu3+ ION Added cation (B) Concentration 10 p.p.m. of B/mM D AD (EU3+)t . . * . - 1.29(0,) -Ni2+ . . 0.169 1.28 0.01 co2+ . . 0.169 1.20 0.09 . . 0.093 1.23 0.06 0.157 1.17 0.12 0.178 1.11 0.18 Fe3+ . . Zn2+ . . 0.154 1.02 0.27 Cd2+ . . 0.089 1.02 0.27 Th4+ . . 0.043 1.15 0.14 UO2Z+ . . 0.042 1.07 0.22 Hg2+ . . 0.055 0.84 0.45 Pb2+ . . 0.048 0.87 0.42 :::+ . . IR * -0.0002 0.001 6 0.001 9 0.002 3 0.003 1 0.005 3 0.009 2 0.009 8 0.0158 0.024 7 0.026 5 * [Eu3+] = 0.003 9 M.t No added cation. The presence of other metal ions always affects the extraction of a metal ion. Table II shows the effect of the addition of 10 p.p.m. of some transition metal ions on the extraction of the Eu3+ ion with 15-crown-5. The interference of other metal ions in the extraction of Eu3+ ions can be expressed by I, which is defined as I = (AD/Do)/([B]/[Eu3+]) where AD = Do - D and D and Do are the distribution coefficients in aqueous solutions of Eu3+ ions with and without other metal ions (B) respectively; [B] and [Eu3+] are concentrations of ion B and Eu3+. From the definition of I, an I of >0.5 means that crown ether prefers the added cation to the Eu3+ ion; an I < 0.5 indicates that the Eu3+ ion is the most easily extracted.In Table I1 it can be seen that transition metal ions only have small effects; only Pb2+ and Hg2+ show slightly larger effects ( i e . larger I, I > 0.02) which may be the result of the similar size of Pb2+ (0.94 A) and of Hg2+ (0.93 A) to the Eu3+ ion (0.95 A). The authors express their appreciation to the Institute of Nuclear Energy Research in Taiwan and the National Science Council of the Republic of China for its support in this work September 1983 RARE EARTH METALS WITH CROWN ETHERS References 1113 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Pedersen C. J. J . Am. Chem. SOC. 1967 89 7017. Christensen J. J. Eatough D. J. and Izatt R. M. Chem. Rev. 1974 74 351. Izatt R.M. and Christensen J . J. Editors “Synthetic Multidentate Macrocyclic Compounds,” Kolthoff I. M. Anal. Chem. 1979 51 1R. Izatt R. M. and Christensen J. J . Editors “Progress in Macrocyclic Chemistry,” Wiley-Interscience, Frensdroff H. K. J . Am. Chem. SOC. 1971 93 4684. Kopolow I. and Smid J. Macromolecules 1973 6 135. Fernado L. A. Miles M. L. and Bown L. H. Anal. Chem. 1980 52 1115. Sekine T. and Hasegawa Y . Bull. Chem. SOC. Jpn. 1978 51 645. King R. B. and Heckley P. R. J . Am. Chem. Soc. 1974 96 3118. Deserrenx J. F. Renard A. and Duyckaerts G. 1977 39 1587. Gansow 0. A. Kansar A. R. Triplett K. M. Weaver M. J andYee E. L. J. Am. Chem. Soc. 1977, Desereux J. F. Bull. Cl. Sci. Acad. R . Belg. 1978 64 814. Caton G. A. Harman M. E. Hart F. A. Hawkes G. E. and Moss G. P. J . Chem. SOC. Dalton Bunzli J. C. and Wessner D. Helv. Chim. Ada 1978 61 1454. Desereux J . F. and Duyckaerts G. Inorg. Chim. Acta 1979 35 L313. Oanh H. T. Inorg. Chim. Acta 1979 32 L33. Ciampolini M. and Nardi N. Inorg. Chim. Ada 1979 32 L9. Bunzli J . C. and Wessner D. Inorg. Chim. Acta 1980 44 L55. Bunzli J . G. Oanh H. T. and Gillet B. Inwg. Chim. Acta 1981 53 L219. Pedersen C. J. and Frensdroff H. K. Angew. Chem. Int. Ed. Engl. 1972 11 16. Templeton D. H. and Dauben C . H. J . Am. Chem. SOC. 1954 76 5237. Tsai M. L. Shih J. S. and Wu S. C. unpublished work. Academic Press New York 1978. New York 1979. 99 7087. Trans. 1978 181. Received February 2nd 1983 Accepted Afiril 26tk 198

ISSN:0003-2654    

DOI:10.1039/AN9830801108

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1114 Analyst September 1983 Vol. 108 $q5. 1114-1119 Simultaneous Determination of Choline and Betaine in Some Fish Materials Sana E. Valdes Martinez* Food Science Division Defiartment of Bioscience and Biotechnology University of Strathclyde 13 1 Albion Street Glasgow G1 1SD The assay of choline and betaine as water-insoluble reineckate derivatives, has been adapted to the simultaneous determination of these two quaternary nitrogenous compounds. No preliminary column fractionation is required. The solution resulting from the initial extraction - hydrolysis stage is divided into two portions; one portion is adjusted to pH 1.0 and the other to pH 8-9 before the addition of ammonium reineckate. Under acidic conditions, choline and betaine reineckate are precipitated and at an alkaline pH only choline reineckate precipitates.The reineckate salts are re-dissolved in acetone to give pink solutions for spectrophotometric measurement at 526 nm. Choline reineckate has a higher linear absorptivity than betaine reineckate and this feature is allowed for by reference to the respective calibration graphs; the choline content of the extract is determined directly and the betaine content by difference. Model systems containing 2-3 mg of pure choline or betaine can be assayed but sample masses of 3-4 g of biological material are recommended for the analysis. Keywords Choline determination ; betaine determination ; ammonium Yeineckate ; fish materials Aquatic species contain significant amounts of both choline and betaine in contrast to other bio-logical materials where one or other usually predominates.Similarly both these water-soluble compounds are usually present in the diets of artificially reared fish-choline as a nutrient and betaine as a flavour attractant. In fish farming young larvae are fed natural marine products, such as Artemia (brine shrimp) and Lumbrucillus or manufactured diets that may or may not, be based on fish material; the compounded diets can also be fortified with synthetic choline and betaine if required. Diffusion of betaine but not choline from the feedstuff into the surround-ing water is desirable and the ability to monitor the levels of both choline and betaine is there-fore important. Analytical techniques in the past have centred on choline or betaine with the aim of minimis-ing interference from the presence of small amounts of the other or from related quaternary nitrogenous compounds e.g.choline in lecithin preparations1 and betaine in sugar beet.2 Existing methods depend mainly on the precipitation of the reineckate or periodide derivative under specified conditions to eliminate interference or by pyrolysis to trimethylamine. The difference in yields of the periodides of choline and betaine at different pHs was reported by StanekS in 1906 and over 50 years later this principle was applied in a time-consuming tech-nique to determine one in the presence of the other.4 The lengthy extraction procedure for choline determination followed by acid or alkali hydrolysis has now been overcome by the introduction of a one-stage extraction - hydrolysis pr~cedure.~ Recently determinations of choline and betaine by lH nuclear magnetic resonance spectroscopy6 and high-performance liquid chromatography (HPLC)5 have been described but both methods require expensive equipment.This study has been undertaken to show that the reineckate procedure can be adapted to provide a simple and relatively speedy method for the simultaneous determination of both choline and betaine in a single extract of complex biological material. Experimental Reagents Fish materials. The following were used larvae of Artemia salina 20 h after hatching from dehydrated cysts ; Lumbrucillus rivalis collected from the local seashore ; Experimental Diet * Present address Food Science Department UNAM Ciudad Universitaria Mexico D.F.04560 Mexico VALDES MARTINEZ 1115 ND formulated at the Institute of Marine Biochemistry Aberdeen ; and Commercial Larvit fish diets marketed by Trou & Co. The Netherlands. Ammonium reineckate general purpose reagent was from BDH Chemicals Ltd. The solution was stored in a light-resistant bottle at 10 "C and prepared weekly. Propan-1-01. BDH Chemicals Ltd. AnalaR. Acetone. BDH Chemicals Ltd. general purpose reagent. Diethyl ether. Extraction - hydrolysis mixture. hydrochloric acid. Sand. Ammonium reineckate solution 5% m/V in methanol. BDH Chemicals Ltd. general purpose reagent. Prepared from 175 ml of absolute ethanol + 75 ml of 1 N Preparation of Standards Dry the choline chloride crystals (BDH Chemicals Ltd.) for 5 h at 100 "C and dissolve lOOmg in50ml of de-ionised water.Prepare daily. Dry the betaine hydrochloride crystals (BDH Chemicals Ltd.) for 5 h at 100 "C and dissolve 150 mg in 50 ml of de-ionised water. Prepare daily. Choline chloride standard solution 2 mg ml-l. Betaine hydrochloride standard solution 3 mg ml-1. Apparatus The following apparatus was used Soxhlet extraction equipment including cellulose double-thickness thimbles 26 x 80 mm (Whatman); a pH meter Model PW 9410 digital (Pye Unicam) ; separating funnels 250-ml capacity ; glass-stoppered test-tubes approximately 17-ml capacity; Gooch crucibles Pyrex (30 mm diameter) with sintered glass disc porosity 2; Thunberg tubes of 20-ml capacity with side-arms ; Buchner flasks; spectrophotometer Model SP8-100 (Pye Unicam) ; and a Buchi Rotavapor rotary evaporator.General Procedure Preparation of extract A sample (m) of fish material weighing 3 4 g is mixed with 12 g of sand to avoid caking and transferred into a cellulose thimble which is then placed in the Soxhlet condenser. Two glass beads are put in the flask and 100 ml of exctraction - hydrolysis mixture are added. The sample is refluxed for 8 h. The extract is then reduced in volume to a few millilitres by rotary evaporation. The concentrate is transferred quantitiatively through Whatman No. 1 filter-paper into a 250-ml separating funnel with the aid of 50 ml of de-ionised water. The solution is extracted twice with 70-ml portions of diethyl ether the aqueous phase being retained; the pH is adjusted to 1.0 by the addition of a small amount of sodium hydroxide solution.This extract is again reduced to a few millilitres by rotary evaporation transferred quantitatively into a 25-ml calibrated flask then made up to volume (V,) with de-ionised water. Determination of choline chloride plus betaine hydrochloride Duplicate 3-ml samples (V,) of the extract Vl are transferred into separate glass-stoppered tubes; chilled 5% ammonium reineckate solution (3 ml) is added to each tube the contents are thoroughly mixed and the tubes placed in a refrigerator at 4 "C for 2 h to allow complete reaction.' The crystals when formed are filtered through individual sintered glass crucibles of medium porosity aided by suction. Portions of propan-1-01 (4 x 1 ml) are taken for the washing of each tube and the washings are transferred into the crucible.The crystals in the crucible are washed with three more 2-ml portions of propan-1-01 and dried by suction. During filtration washing and drying the use of a black cardboard box8 over the crucible and dimmed light in the laboratory are recommended to avoid destruction by light. The dried crystals are dissolved in the crucible in a few millilitres of acetone and the coloured solution is sucked into a tube with a side-arm and then transferred quantitatively into a 10-ml calibrated flask and diluted to volume with acetone. Within the next 30 min the absorbance is measured at 526 nm by a spectrophotometer using 1-cm cuvettes fitted with caps 1116 VALDES MARTINEZ SIMULTANEOUS DETERMINATION Analyst Vol. 108 Determination of choline chloride only A 15-ml sample (V,) of the extract V is adjusted to pH 8-9 and diluted to 25 ml with de-ionised water (V,).Duplicate 5-ml samples (V,) of the V solution are then transferred into separate glass-stoppered tubes and analysed as described under Determination of choline chloride plus betaine IzydrochZoride. The concentration of choline chloride is determined by reference to the choline calibration graph. Determination of betaine hydrochloride only described are identical thus The dilutions leading to the measurement of choline plus betaine and choline only as -- V Vl v 4 v - v3 v, Therefore the difference in absorbance of the acetone solutions of the crystals precipitated under acidic and alkaline conditions provides the absorbance attributable to betaine.The concentration of betaine hydrochloride is determined by reference to the betaine calibration graph. Calculation The calculation for the determination of choline plus betaine is performed as follows: grams of sample taken for analysis = m; absorbance for choline + betaine = A absorbance for choline only = B (subsequently equated to milligrams of choline chloride from the choline calibration graph) and absorbance for betaine = C where C = A-B (subsequently equated to milligrams of betaine hydrochloride from the betaine calibration graph) ; volume of the original extract = V = 25 ml volume of extract V analysed for choline plus betaine = V = 3 ml volume of extract V adjusted to pH 8-9 for choline only = V = 15 ml final volume to which V 3 is adjusted = V = 25 ml and volume of V analysed for choline only = V = 5 ml.Then. V V 100 Choline chloride (mg per 100 g) = B (mg equiv.) x 2 x -3 x -v3 v5 m and v 100 Betaine hydrochloride (mg per 100 g) = C (mg equiv.) x -> x - v72 Results Choline Chloride and Betaine Hydrochloride Calibration Graphs For an equal mass the pink colour of the reineckate salt of choline chloride had a higher linear absorptivity than the reineckate derivative of betaine hydrochloride but both com-CALIBRATION REINECKATE Amount of choline chloride/mg 2 4 6 8 10 TABLE I DATA FOR THE ABSORBANCE AT 526 nm OF CHOLINE AND BETAINE REIN'ECKATE I N ACETONE SOLUTIONS Sample volumes used 10 ml. Amount of Mean* f s.d./mg betaine hydrochloride/mg Meant f s.d./mg 0.162 f 0.001 3 0.116 f 0.005 0.310 f 0.002 6 0.266 0.005 0.476 f 0.013 9 0.328 & 0.030 0.638 f 0.013 12 0.531 f 0.020 0.791 & 0.014 15 0.678 f 0.024 Correlation coefficient Y = 0.9999 Regression y = 0.0793% -0.0002 Correhtion coefficient Y = 0.992 0 Regression y = 0.0452% -0.0105 * Mean of three replicates.t Mean of four replicates September 1983 OF CHOLINE AND BETAINE I N SOME FISH MATERIALS 1117 pounds showed the same Amax. at 526 nm. The data obtained in the preparation of the calibra-tion graphs are summarised in Table I and demonstrate good reproducibility and linearity; the aqueous betaine standards were acidified with 1 ml of 1 N hydrochloride before reacting with ammonium reineckate. The choline and betaine contents of the fish materials were calculated on the basis of these calibration graphs.Precipitation of Reineckate Derivatives at Different pH Values Working with individual standard solutions and using the procedure already described, choline reineckate was found to be insoluble over the pH range 1.0-10.2; at higher pH values choline reineckate is susceptible to decompo~ition.~ Betaine reineckate was almost completely soluble at all pHs except for the acid zone of pH 1.0 (Table 11). TABLE I1 PRECIPITATION OF REINECKATE DERIVATIVES OF CHOLINE AND BETAIN‘E AT DIFFERING pH Amount of Amount of betaine hydrochloride/ PH choline chloride/mg lCecovery,* yo mg Recovery,* % 1.04 6 99.4 8 102.1 2.75 6 99.3 8 8.8 4.99 6 99.3 8 7.7 7.00 6 99.3 8 7.7 10.20 6 99.6 8 4.3 * Means of two replicates.When a 0.2 M phosphate buffer at pH 7.58 containing 6 mg of choline chloride and 6 mg of betaine hydrochloride was reacted with ammonium reineckate the choline was precipitated. The filtrate was acidified to pH 1 .OO an extra 1 ml of 5% ammonium reineckate was added and the solution was kept for a further 2 h at 4 “C during which time the betaine reineckate pre-cipitated for subsequent quantification. Over 98% of each component was recovered (Table 111); no increase in recovery for the chlorine fraction with a corresponding decrease in the betaine fraction which might have been expected from the slight precipitation of betaine at pH 7.58 (Table II) was found with these mixed solutions. However because of the similar absorbance characteristics of choline reineckate and betaine reineckate this satisfactory recovery did not prove that the two crystal fractions were “clean.” Descending paper chromatography was used as a qualitative method to establish the purity of the two fractions.Separate samples of choline and betaine reineckates were prepared to act as standards of comparison. The acid - solvent system of butanol- glacial acetic acid - water (100+30 saturated with water) demonstrated that betaine reineckate had an RF value of 0.72 while choline reineckate remained at the base line; ammonium reineckate which might be a trace contaminant in any sample but was never detected moved ahead of betaine reineckate. The finding for choline reineckate does not agree with Ackerman and Salmons who reported an R F value of 0.44 with this solvent system.The fraction precipitated at pH 7.58 was found to be only choline reineckate and the fraction precipitated at pH 1.00 yielded only a betaine reineckate spot. TABLE 111 TWO-STAGE PRECIPITATION OF REINECKATE DERIVATIVES FROM A SOLUTION CONTAINING 6 mg OF CHOLINE CHLORIDE AND 6 mg OF BETAINE HYDROCHLORIDE First ppt. (pH 7.58) : Sample choline recoverylmg 1 5.87 2 5.78 3 5.95 4 5.91 Mean = 5.91 f 0.073 Recovery = 98.5% Second ppt. (pH 1.00) : betaine recoverylmg 5.91 5.82 5.96 5.91 Mean = 5.90 f 0.058 Recovery = 98.3 1118 VALDES MARTINEZ SIMULTANEOUS DETERMINATION Analyst Vol. 108 An alkaline medium of pH 10.0 for the precipitation of choline reineckate might seem advisable to ensure maximum separation from betaine. However at higher pH values metallic ions which are present in buffer solutions or extracts of biological material form insoluble reineckate salts which may lead to filtering difficulties during analysis ; therefore the pH range 8-9 was selected for the assay of test materials which allowed almost complete differentiation between choline and betaine and reduced any potential physical problem from metallic contamination.Choline and Betaine Contents of Fish Materials The analysis of the natural marine products Artemia and Lumbvucillus shows that they are rich sources of choline and betaine (Table IV). Similarly the fish-based experimental diet ND had a high natural content of these two chemical compounds even after allowing for fortification with synthetic choline and betaine.The formulae of the commercial Larvit diets are not declared but the finding of low levels for choline and betaine indicates that the protein source is not derived primarily from marine products nor is large-scale fortification with synthetic choline and betaine included. Amino acid analysis of the Larvit diets also showed a high glutamic acid content per gram of nitrogen relative to fish protein and it is suggested that these commercial diets may be based on casein which was used by HalverlO in his studies on the vitamin requirements of chinook salmon. The similarity in the choline and protein con-tents of Larvit A B and C indicates a constant chemical composition for the diets with the difference depending on particle size as defined by the manufacturer.TABLE IV CHOLINE AND BETAINE CONTENTS OF FISH FEEDS AND DIETS WITH POTENTIAL USE FOR THE REARING OF LARVAE UN'DER FISH FARMING CONDITIONS Fish material Natural products-Experimental diet-Commercial diet-Artemia salina (freeze-dried) . . . . . . Lumbrucillus rivalis (freeze-dried) . . . . Diet NDS * . . . . . Larvit A . . . . Larvit B . . . . Larvit C . . . . . . . . Choline chloride */ mg per 100 g 64 1 765 1197 137 158 143 Betaine hydrochloride */ mg per 100 g 1115 N.a. t 1632 N.a.t 106 N.a.t Protein, % 54.7 61.1 54.1 62.9 62.0 64.4 * Average of at least four replicates. t Not analysed. 1 Based on freeze-dried cod muscle (24.6%) and freeze-dried Nephrops waste (20.0%) and fortified with SO0 mg of choline chloride per 100 g of diet ND and 1000 mg of betaine hydrochloride per 100 g of diet ND.Recovery of Added Choline Chloride and Betaine Hydrochloride Diet ND was used for recovery determinations of added choline chloride and betaine hydro-chloride to establish the reliability of the method and satisfactory results were obtained (Table V) * TABLE V RECOVERY OF CHOLIN'E AND BETAINE ADDED TO DIET ND Additive Choline chloride . . Betaine hydrochloride Additive in diet ND/mg 30.07 30.53 30.71 40.89 41.62 41.76 Amount added/mg 25.52 25.52 25.52 25.30 26.30 25.30 Total additive determined/mg 58.01 68.39 57.18 69.17 66.89 67.17 Recovery yo 109.5 109.2 103.7 Mean = 107.6 111.8 100.3 104.4 Mean = 105.September 1983 OF CHOLINE AND BETAINE IN SOME FISH MATERIALS 1119 Discussion It is probably only in fish materials that both choline and betaine are present in significant amounts and their respective quantification is required. In other biological samples one or other of these two compounds provides the major component and analytical techniques have been designed to remove the contaminating minor alternative. For example the difference in the yields of the periodides of choline and betaine at different pHs has been used by Wall et aL4 to determine one in the presence of the other. In this study both components have been assayed in the same extract as the reineckate derivatives. The described procedure shows that choline and betaine in a mixed solution can be recovered specifically and quantitatively.However having prepared the extract of a test sample the procedure is considerably shortened if a total choline plus betaine determination is run con-currently with a choline-only determination instead of waiting for the filtrate from the choline determination in order to adjust the pH and precipitate the betaine reineckate with a further holding for 2 h at 4 "C. The betaine content is determined by difference using the recom-mended procedure. Early studies on the determination of choline or betaine in natural products required lengthy periods of extraction followed by acid or alkali hydrolysis and fractionation by column chromatography; the one-stage extraction - hydrolysis procedure of Dorsey et aZ.,5 with minor modifications has proved satisfactory in the present study.Similarly the observation of More and Kennerll that reineckate derivatives should be dissolved in 100% acetone rather than 75% aqueous acetone has been applied; also the concentration of choline or betaine reineckate in the acetone solutions is read without undue delay. Both these practices contri-bute to obtaining the highest absorbance values. It is not surprising that HPLC has been suggested5 for the determination of choline free from contamination by chemically related compounds. This chromatographic technique could probably be developed for the simultaneous determination of choline and betaine in a single extract but at present a complicated specifically designed detector is required for the deter-mination of choline by HPLC.A chemical method requiring only basic equipment will always be needed and the foregoing procedure for the simultaneous determination of choline and betaine has proved valuable in a study of the leaching of water-soluble constituents when diets used in fish farming are immersed in water. The author thanks the National Autonomous University of Mexico City for financial support during the course of this work and Dr. K. Mary Clegg of the Food Science Division of the University of Strathclyde for assistance in the preparation of the manuscript. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Illingworth D. R. and Portman 0. W. J . Chromatogr. 1972 73 262. Walker H. G. and Erlandsen R. Anal. Chem. 1951 23 1309. Stanek V. V. 2. Physiol. Chem. 1906 47 83. Wall J . S. Christianson D. D. Dimler K. J. and Senti F. R. Anal. Chem. 1960 32 870. Dorsey J. G. Hansen L. C. and Gilbert T. W. J . Agric. Food Chem. 1980 28 28. Chastellain F. and Hirschbrunner P. 2. Anal. Chem. 1976 278 207. Glick D. J . Biol. Chem. 1944 156 643. Ackerman C. J. and Salmon W. D. Anal. Biochem. 1960 1 327. Engel R. W. Salmon W. D. and Ackerman C. J. Methods Biochem. Anal. 1954 1 265. Halver J. E. J . Nutr. 1957 62 225. More C. A. and Kenner C. T. J . Assoc. Off. Anal. Chew. 1970 53 588. Received February 17th 1983 Accepted April 22nd 198

ISSN:0003-2654    

DOI:10.1039/AN9830801114

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1120 Analyst September 1983 Vol 108 pp. 1120-1127 Spectrophotometric and Analogue Derivative Spectrophotometric Determination of Micro-amounts of Iron with 2,2'-Dipyridyl-2- benzothiazolyl hydrazone Raj Bhushan Singh," Tsugikatsu Odashima and Hajime lshii Chemical Reseavch .Institute of Non-Aqueous Solutions Tohoku Ugziversity Katahira Sendai 980 Japan The synthesis of 2,2'-dipyridyl-2-benzothiazolylliydrazone (DI'BH) charac-teristics reactions with various metal ions and its application in the selective determination of iron are presented. DPBH reacts with iron(I1) to form a stable 1 3 metal to ligand complex having absorption maxima a t 427 and 615 nm in 0.4% Triton X-100 solution. A solution of the complex gives a constant absorbance in the pH range 4.5-8.4 at 427nm and 3.0-9.6 a t 615 nm.The molar absorptivity sensitivity for an absorbance of 0.001 and relative standard deviation for 10.3 p g of iron (10 replicates) are 3.41 x lo4 and 1.23 x lo4 1 mol-1 cm-l 1.64 and 4.64 ng cm-2 and 0.7 and 0.5%) a t 427 and 615 nm respectively. Application of the proposed method to the deter-mination of iron in water samples and further sensitisation of the method by employing analogue derivative spectrophotometry are also described. Keywords Iron determination ; 2,2'-dipyridyl-2-benzothiazolylh3,drazone ; analogue derivative spectrophotometry ; water analy.cis In recent years many nitrogen-containing heterocyclic hydrazones have been synthesised and investigated for their potential as analytical reagents.1,2 In continuation of our studies on the analytical application of benzotl~iaz~lylhydrazones,~-~~ 2,2'-dipyridyl-Z-benzo-thiazolylhydrazone (DPBH or HL) was synthesised and applied to the spectrophotometric and analogue derivative spectrophotometric determination of iron.Sensitive and selective methods have been proposed one of which was applied successfully to the determination of iron in water samples. Experimental Reagents All solutions were prepared with distilled de-ionised water. DPBH was synthesised by refluxing equimolar amounts of 2,2'-dipyridyl ketone and 2-hydrazinobenzothiazole in an ethanolic medium for about 3 h. The crude compound which separated out on cooling was recrystallised twice from ethanol to give yellow needles melting-point 106-107 "C (yield 60%). The results of the elemental analysis of DPBH were as follows found C 64.6 H 4.0 N 20.2 S 10.1%; calculated for C,,H,,N,S C 65.2 H 3.9 N 21.1 S 9.7%.Prepared by dissolving the required mass of DPBH in ethanol by heating on a water-bath and cooling to room temperature. The ethanolic solution was stable and could be kept for at least 1 month in an amber-glass bottle. This solution was diluted further with ethanol if necessary. Standard iron(1ll) solzition. Prepared by dissolving iron(II1) ammonium sulphate [Fe,(SO,) ,(NH4),S0,.?4H,O] in water containing a small amount of sulphuric acid. This solution was standardised by EDTA titration with Variamine Blue B hydrochloride as an indicator. All reagents used were of analytical-reagent grade unless stated otherwise. Synthesis of DPBH.DPBH solution 2.5 x 1 0 - 3 ~ . Apparatus For measurements of the absorbance and the absorption spectrum a Hitachi 139 spectro-photometer and a Hitachi 556 dual-wavelength spectrophotometer were used respectively, * On leave from Bareilly College Bareilly-243005 India SINGH ODASHIMA AND ISHII 1121 the latter being used as an ordinary double-beam spectrophotometer throughout. To obtain the derivative spectrum a modified Hitaclii 200-0576 derivative unit composed of two analogue differentiation circuits (each having six different time constants) was connected between the latter spectrophotonieter’s output and the Hitachi 057 X - Y recorder’s input. The details of this apparatus and the principle and characteristics of analogue derivative spectrophotometry have been described previously.l1Sl2 Procedure Ordinary spectrophotometry Place a sample or standard solution containing less than 40 pg (or 110 pg) of iron(I1) and/or iron(II1) in a 25-ml calibrated flask and add 0.5 ml of 10% hydroxylarnmonium chloride solution 2 ml of 2.5 x M DPBH solution and 2 ml of 1 M acetate buffer (pH 4.9), followed by 1 ml of 10% Triton X-100 solution.Dilute to the mark with water and measure the absorbance of the resultant solution at 427 nm (or 615 nm) against a reagent blank using 10-mm cells. Second-derivative spectrophotometry When the iron content of the coloured solution prepared by the procedure described above, is too low to give a measurable absorbance record the second-derivative spectrum from 370 to 650 nm on a chart recorder against a reagent blank by using a combination of both first-and second-order differentiation circuits of No.6 and a scan speed of 300 nm min-l 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). Pre-treatment of samples centrated hydrochloric acid. heating with 5 ml of 1 M hydrochloric acid and dilute to 50 ml with water. suitable aliquot of this solution for the determination. To 500 ml of the water sample add 1 ml of concentrated sulphuric acid and 3 ml of con-Dissolve the resulting residue by Use 5 ml or a Evaporate to dryness. Results and Discussion Properties and Characteristics of DPBH The infrared spectrum of DPBH was measured with a potassium bromide disk in order to confirm the structure of the synthesised DPBH.The spectrum had absorption peaks at 3400 and 1602 cm-l which were assigned to the stretching vibrations of the -N-H bond and the -C=N- bond re~pective1y.l~ Considering these results and those of the elemental analysis already shown the synthesised DPBH was presumed to have the structure I. H 9 I DPBH which is believed to exist as the (2)-isomer is a yellow crystalline material insoluble in water unless in the presence of a surfactant such as Triton X-100 but soluble in most organic solvents. The acid dissociation constants were determined spectrophotometrically. DPBH exists in solution in any of the following forms depending on the pH 1122 SINGH et d. SPECTROPHOTOMETRY AND ANALOGUE Analyst VOZ.108 TABLE I REACTIVITY OF DPBH WITH METAL IONS Metal ion t@) : Co(I1) . . Cr(II1) . . Cr(V1) . . Cu(I1) . . Fe(I1) . . Fe(II1) . . Hg(I1) Mn(I1) Ni(I1) . . Pb(I1) . . Pd(I1) . . Ti(1V) . . Zn(I1) . . V(1V) V(V) . . . . . . . . PH 7 pH 10 -7 r-f-hmax./nm r*/l mol-1 cm-1 hmax./nm ~ * / 1 mol-l ctn-l Xmax./nm ~ * / 1 mol-' cm-' PH 4 - - 415 11 600 415 4 400 - - 455 54 000 455 53 800 495 31 900 495 30 900 495 30 900 - - - - - -- - - -465 18 900 455 34 200 34 100 427 34 100 12 300 615 12 300 452 32 300 452 42 900 468 42 500 468 45 500 550 10400 515 8 100 505 14200 - -- - - { :: -- - - -- - - -- - - -- - - -- 456 56 600 --455 427 615 452 464 468 422 515 ----456 -33 100 31 100 12 100 30 700 46 100 43 800 15 500 10900 ----55 300 * c is the molar absorptivity.The PKa values were determined by extrapolating the linear ethanol concentration (20-50y0) veysus apparent pKa plot to the intercept and the values were found to be pKas = 1.56, pKa2 = 4.03 and pKal = 9.96. The Ka3 and Ka2 may be caused by the protonation of the benzothiazolyl group and the pyridine nitrogen atom respectively whilst Ko is due to the deprotonation of the imino group. Reactivity of DPBH with Metal Ions pH values is summarised in Table I. The reactivity of DPBH with various metal ions in 0.4% Triton X-100 solution at various DPBH was found to react with cadmium(II) cobalt(II), P) lu P) m P) -0 0 -.- c .- L 8 fn 0.3 8 5 0.2 9 s 2 0.1 Wavelengthtnm Fig.1. (a) Absorption spectra of DPBH and its iron(I1) complex and (b) second-derivative spectrum of iron(I1) - DPBH complex in 0.4% Triton X-100 solution. Iron(II) 410 p.p.b.; DPBH 2 x lo-* M; pH 4.9; A and C iron(I1) - DPBH complex against reagent blank; and B reagent blank against water. 0.3 I 1 I I I 1 2 4 6 8 10 PH Fig. 2. Effect of pH. Iron(II) 410 p.p.b.; DPBH 2 x ~O-*M; Triton X-100 0.4%; wavelength (A) 427nm and (B) 615nm; reference reagent blank September 1983 DERIVATIVE SPECTROPHOTOMETRY OF IRON WITH DPBH 1123 copper(II) iron(II) mercury(II) nickel(II) palladium(II) vanadium(1V) and zinc(I1) to give coloured complexes that have large molar absorptivity values.Absorption Spectra and Effect of pH When an ethanolic solution of DPBH was added to an iron(I1) solution a green complex was formed immediately. This complex was soluble in Triton X-100 solution and had two absorption maxima at 427 and 615 nm. The absorption spectrum of the iron(I1) - DPBH complex is shown in Fig. 1 together with its second-derivative spectrum. A study of the effect of pH on the complexation of DPBH with iron(I1) gave a constant absorbance in the pH range 4.5-8.4 at 427 nm and 3.0-9.6 at 615 nm respectively (see Fig. 2). Effect of DPBH Concentration As can be seen from Fig. 3 more than 6-fold the iron concentration (molar ratio) of DPBH was necessary for complete complexation to occur and the excess of DPBH did not interfere. 0.3 I I+ $ 0.2 8 Q 0.1 C n 0 B U t 1-I I ' 4-0 5 10 15 20 40 I 5 10 15 20 40 [DPBH]/[ Fe] Fig.3. Effect of DPBH concentration. Wavelength, (A) 427 nm and (B) 615 nm; iron(II) 410 p.p.b.; pH 4.9. Other conditions as in Fig. 2. Effect of Triton X-100 Concentration Triton X-100 concentration. complete solubilisation and for obtaining a constant absorbance. 4% (maximum tested) did not interfere. The effect of Triton X-100 concentration on the absorbance was examined by varying At least 0.3% Triton X-100 solution was necessary for A concentration of up to TABLE I1 TOLERANCE LIMITS FOR OTHER IONS ON THE DETERMINATION OF 10.3 pg OF IRON Tolerable error f 3 yo. Ions C1- Br- I- SCN- ClO,- NO,- S,0a2- PO4,- tartrate citrate, thiourea . . .. . . . . Ca(II) Mg(I1) . . . . Mn(I1) . . . . Pb(I1) . . . . . . . . . . Ti(1V) . . . . . . . . . . Hg(I1) . . . . ,. . . . . Co(I1) . . . . . . . . . . Pd(I1) . . Al(III) Cd(II) *Cr(IIIj,* Cr(VI) Zn(II)*$t . . Ag(I),* Cu(II),*v$ Ni(II),* V(V) . . . . . . . . * Gave a positive error at 427 nm but did not interfere at 615 nm. t A l-ml volume of 1 M sodium citrate solution was added. $ A 0.5-11-11 volume of 1 M thiourea solution was added. Tolerance limi 1124 Composition of the Complex metal to ligand ratio of the complex was 1 3. SINGH et al. SPECTROPHOTOMETRY AND ANALOGUE Analyst VoZ. 108 A molar ratio plot (see Fig. 3) and Job’s method of continuous variation showed that the Stability of the Complex absorbance remained constant even after 15 h.The complex formed under the recommended conditions was very stable so that the Calibration Graph Sensitivity and Precision mended procedure. Straight line calibration graphs passing through the origin were obtained using the recom-The equations of the lines obtained by least-squares treatment were Fe (p.p.m.) = 1.64A (at 427 nm) . . . . - * (1) Fe (p.p.m.) = 4.548 (at 615nm) . . * * (2) where A is the absorbance. The optimum ranges for the iron determination were 0.1-1.6 and 0.2-4.5 p.p.m. at 427 and 615 nm respectively. The sensitivities for an absorbance of 0.001 and the molar absorptivities calculated from equations (1) and (2) were 1.64 ng cm-2 and 3.41 x 1041mol-lcm-l at 427 nm and 4.5411gcm-~ and 1.23 x lO41mo1-lcm-l at 615 nm respectively.Ten standard solutions containing 10.3 pg of iron were analysed by the recommended procedure. The results gave relative standard deviations of 0.7 and 0.5% at 427 and 615 nm, respectively. Effect of Diverse Ions Solutions containing 10.3 pg of iron(I1) and various amounts of other ions were prepared and the recommended procedure for iron determination was followed. The results are summarised in Table 11 from which it can be seen that chromium(III) zinc(II) silver(I), copper(II) nickel(II) cobalt(I1) and palladium(I1) interfere with the determination but the interferences due to these ions (except the last two ions) can be removed by measuring the absorbance at 615 nm or by using masking agents. Therefore the procedure seems to be selective although the tolerance limits for cobalt (11) and palladium(I1) are low.Application to Actual Samples tion of iron in both well and tap waters. with results obtained by atomic-absorption spectrophotometry carried out for comparison. In order to confirm the usefulness of the proposed method it was applied to the determina-The results are summarised in Table I11 together TABLE I11 DETERMINATION OF IRON IN WATER SAMPLES Iron found p.p.m. Sample Well water . . Tap water . . r- 1 Proposed method (-*- Atomic-absorption 427nm 615nm method 0.57 0.55 0.66 0.56 0.55 0.56 0.55 0.33 0.30 0.31 0.33 0.31 0.33 0.31 Sensitisation by Employing Analogue Derivative Spectrophotometry Ishii and Kohl1 and Ishii and Satoh12 have previously reported that derivative spectro-photometry using an analogue differentiation circuit is extremely effective for the sensitisa-tion of ordinary spectrophotometry.As an example of sensitisation the second-derivative spectrophotometric determination of iron is described here September 1983 DERIVATIVE SPECTROPHOTOMETRY OF IRON WITH DPBH 1125 Selection of conditions for measurement of the second-derivative s$ectrurn In second-derivative spectrophotometry the second-derivative spectrum of the deter-minant is recorded and the second-derivative value (the vertical distance from a peak to a trough or that from the base line to a trough of the spectrum) is measured. Because the second-derivative value depends on both the time constant of the analogue differentiation circuit and the scan speed of the spectrophotometer these need to be selected to give a well resolved large peak (to give good selectivity and higher sensitivity in the determination).This is done on the basis of the breadth of the band in the ordinary absorption spectrum. Our apparatus has six circuits with different time constants. They are represented with circuit numbers from 1 to 6 and an increase in the circuit number means an increase in the time constant. In general a large time constant and/or a fast scan speed should be used for a broad band in the absorption spectrum. In Fig. 4 the second-derivative spectra of the iron(I1) - DPBH complex solution measured with varying circuit number or scan speed are shown; circuit No. 6 and a scan speed of 300 nm min-I are seen to be preferable for the iron determination.1 0 2 3 a) I I I 400 500 600 240 300 i - 600 400 500 600 Wavelengthhm Fig. 4. Influence of (a) circuit number (with scan speed 300 nm min-I) and (b) scan speed (with circuits all No. 6) on second-derivative spectra of iron(I1) - DPBH complex solution. Iron(II), 410 p.p.b.; DPBH 2 x ~ O - * M ; pH 4.9; Triton X-100 0.4%; reference reagent blank. Numerical values indicate first- and second-differentiation circuits in (a) and scan speed in ( b ) 1126 SINGH et al. SPECTROPHOTOMETRY AND ANALOGUE Analyst V d . 108 Calibration graphs The calibration graph prepared by plotting the second-derivative value versus the iron concentration gave a straight line graph passing through the origin when the peak to trough value or the base line to trough value was plotted.The equations for each graph measured by employing a combination of circuit No. 6 and a scan speed of 300 nm min-l at 427 and 615 nm are as follows for peak to trough measurements: Fe (p.p.b.) = 1270 (at 427 nm) . . * * (3) Fe (p.p.b.) = 3020 (at 615 nm). . . . * - (4) and for base line to trough measurements: . . * - (5) Fe (p.p.b.) = 1450 (at 427 nm) . . Fe (p.p.b.) = 5340 (at 615 nm). . * . - * (6) where D is the second-derivative value represented by the conversion of the value into absorbance. An example of a calibration graph is shown in Fig. 5 from which it can be seen that iron down to the 40 p.p.b. level can be easily determined by the proposed method. 0 100 200 300 400 Iron concentration p.p.b.Fig. 5. An example of the calibration graph in second-derivative spectrophotometry. Circuits all No. 6; scan speed 300 nm min-1; recorder sensitivity, x 1/4; reference reagent blank; I 11 I11 and IV correspond to equations (3) ( 5 ) (4) and 6 respectively. Ironconcentrates A 41; B 83; C,220; andD 414p.p.b. Comparison with Other Methods Sensitive chromogenic reagents for the determination of iron are 3-(4-phenyl-2-pyridyl)-5-phenyl-l,2,4-triazine (PPPT) ,14 phenyl-a- (4-ethylpyridyl) ketoxime,15 biacet ylmon oxime-4-phenyl-3-thiosemicarbazonel6 and di-2-pyridyl ketone azine.17 Among the hydrazones use September 1983 DERIVATIVE SPECTROPHOTOMETRY OF IRON WITH DPBH 1127 are diacetyldihydrazone,18 2-benzoylpyridine hydrazone,l9 di-2-pyridylglyoxaldihydrazone20 and 2,2’-dipyridyl-2-pyridylhydra~one.~l None of these reagents are as sensitive as DPBH.They either involve an extraction procedure or are associated with interferences from many metal ions. On the other hand the synthesis of DPBH is easy and the proposed method is simple sensitive and selective. The determination of iron can be carried out over a wide pH range. The reaction is instantaneous there is no need for extraction and the complex formed is very stable. DPBH is not only one of the most sensitive hydrazones used so far for the determination of iron but compares in sensitivity with PPPT and can be used in aqueous phases containing Triton X-100. Further the method can be applied to more dilute samples by employing analogue derivative spectrophotometry.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. References Katyal M. and Dutt Y. Talanta 1975 22 151. Singh R. B. Jain P. and Singh R. P. Talanta 1982 29 77. Odashima T. and Ishii H. Nippon Kagaku Kaishi 1973 729. Odashima T. and Ishii H. Nippon Kagaku Kaishi 1975 83. Odashima T. and Ishii H. Anal. Chim. Acta 1975 74 61. Ishii H. and Odashima T. Nippon Kagaku Kaishi 1975 1332. Odashima T. and Ishii H. Anal. Chim. Acta 1976 86 231. Odashima T. and Ishii H. Bunseki Kagaku 1977 26 678. Odashima T. Anzori F. and Ishii H. Anal. Chim. Acta 1976 83 431. Odashima T. Satoh S. and Ishii H. Nippon Kagaku Kaishi 1982 1322. Ishii H. and Koh H. Nippon Kagaku Kaishi 1980 203. Ishii H. and Satoh K. Fresenius 2. A n d . Chem. 1982 312 114. Silverstein R. M. and Bassler G. C. “Spectrometric Identification of Organic Compounds,” Second Edition John Wiley New York 1967 (translated by Araki S. and Mashiko Y. Tokyo Kagaku Dohjin Tokyo 1969). Schilt A. A. Yang T. A. Wu J . F. and Nitzki D. M. Talanta 1977 24 685. Schilt A. A. and Taylor P. J. Talanta 1969 16 448. Can0 Pavon J . M. Sanchez J . C. J. and Pino F. Anal. Chim. Acta 1975 75 335. Valcarcel M. Martinez M. P. and Pino F. Analyst 1975 100 33. Graciani Constante E. and Olias Jimenez J . M. An. Quim. 1971 67 615. Graciani Constante E. An. Quim. 1974 70 695. Graciani Constante E. An. Quim. 1971 67 607. Alexaki-Tzivanidou H. Anal. Chim. Acta 1975 75 231. Received March 14th 1983 Accepted March 30th 198

ISSN:0003-2654    

DOI:10.1039/AN9830801120

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

1128 Analyst September 1983 Vo2. 108 pp. 1128-1134 Spectrophotometric Study and Analytical Applications of the Complexes of Copper( II) and Zinc(ll) with some Sulphonated Azo Dyes Maria Pesavento and Teresa Soldi Dipartkento d i Chimica Generale Universitit d i Pavia Viale Taranaelli 12 27100 Pavia Italy The complexing ability of some water-soluble azo dyes towards copper (11) and zinc(I1) were studied in 0.1 M sodium perchlorate and perchloric acid solutions a t 25.0 "C. Solutions with C;LZ > CL where CM is the concentra-tion of metal and CL the concentration of ligand and CL > CM (CM up to M CL up to 2 x M and pH up to the precipitation of metal hydrox-ides) were investigated spectrophotometrically. The ligands T-azo-C and 3,6-disulyho-TAN are able to complex both copper(I1) and zinc(II) but amaranth only copper(I1).Besides tlie usual complexes having 1 1 metal to ligand molar ratios a 2 1 complex has been detected in the system copper(I1) - T-azo-C. The results are compared with others obtained previously. The copper(I1) complexes are much more stable than those of zinc(I1) and those of many other metal ions. Therefore a method is proposed for the spectrophotonietric determination of copper(II) which is almost free from interferences. Keywords Copper determination ; copper and zinc complex formation ; sulphonated azo dyes ; spectroplzotometry Nickel(II) copper(I1) and zinc(I1) form complexes with the sulphonated azo dye l-tetrazol-azonaphth-2-ol-3,6-disulpl~onic acid (T-azo-R) their stability constants decreasing (according to the Irving - William series) in the order copper>nickel>zinc a t a rate of approximately 2 log units.ls2 Further the nickel(I1) ion forms complexes with some other azo dyes structur-ally related to T-azo-R which provides reliable methods for the analytical determination of nickel(I1) .3 To give a more complete view of the observed trend in the stability constants for T-azo-R complexes and to provide new selective spectrophotometric methods for the determination of copper and zinc a study of their complex formation with T-azo-R analogues is reported.The following systems were investigated Zn(I1) - T-azo-C; Zn(I1) - 3,B-disulpho-TAN ; Zn(I1) -TAAC; Zn(I1) - amaranth; Zn(I1) - Cr-2B; Cu(I1) - T-azo-C; Cu(I1) - 3,6-disulpho-TAN; and Cu(I1) - amaranth.Investigations on Cu(I1) - TAAC and Cu(I1) - Cr-2B equilibria have been described previously4y5 and molecular structures and the protonation constants of all the mentioned azo dyes are reported in Fig. 1. The acidic dissociation of the second phenolic group of the three 1,8-dihydroxynaphthalene-3,6-disulplionic acid (chromotropic acid) derivatives has not been considered the dissociation constants being as low as about 10-14. Experimental The copper(I1) perchlorate stock solution was prepared according to Bottari et nL6 by dis-solving a suitable amount of the pure metal; the zinc(I1) perchlorate solution was obtained by dissolving the metal in a slight excess of percliloric acid. Amaranth and Cr-2B were of analytical-reagent grade (Fluka) ; T-azo-C TAAC and 3,B-disulpho-TAN were prepared as described previo~sly.~s~ss The final solutions containing fixed concentrations of cation and ligand and 0.1 M sodium perchlorate solution or percliloric acid were prepared by dilution.Solutions with C,>C or C,>C, where C is the concentration of metal and C, the concentration of ligand and acidity varying from 0.1 M perchloric acid to the precipitation of the metal ion hydroxide have been examined. M of the metal ion can be studied. Owing to the high absorbance of tlie free ligand only solutions with The ligand stock solutions were prepared by weighing the solid compounds. Solutions containing up to 2 PESAVENTO AND SOLD1 1129 H \ N-N Fig. 1. Structures and protonation constants of the azo dyes showing (A) T-azo-R log fll = 9.10 and log p2 = 12.01’; (B) T-azo-C log p1 = 8.23 and log p2 = 11.73’; (C) 3,6-disulpho-TAN log fll = 7.40 and log /I2 = 8.301l; (D) TAAC log/I = 5.12 and log /I2 = 5.72*; (E) Cr-2B log fl = 8.705; and (F) amaranth, log fll = 9.80.3 a ligand concentration up to 2 x More concentrated ligand solutions were not considered in order to avoid association reactions between the azo-dye molecules.Absorption spectra for different pH values at given constant concentrations of ligand and metal were obtained by varying the acidity of the solutions by the addition of small amounts of concentrated perchloric acid or sodium hydroxide solution. The absorbance was measured with a Spectracomp 601 C spectrophotometer equipped with l-cm quartz cells (Carlo Erba). The pH values were determined potentiometrically in the cell Glass electrode 1 test solution I 0.1 M NaClO I saturated calomel electrode standardised for the hydrogen ion concentration according to the previously outlined pro~edure.~ At an acidity as high as pH 1-2 the hydrogen ion concentration was calculatedfrom the poten-tial of the cell ( E ) by means of the equation [H+] = [10W,’-E)/59.16- d/I]-l where E,’ and d are two constants simultaneously determined as described previouslyg and I is the ionic strength.E,’ includes the standard potential of the glass electrode and the activity coefficient of the hydrogen ion; d depends on the molar conductances of perchloric acid and sodium perchlorate. The equation given allows the determination of the hydrogen ion concentration when there is a large and varying junction potential.In order to eluci-date the nature and the stability of the species existing at the equilibrium in all the considered conditions the experimental data were treated according to a graphical method,1° by which the determination of the molar absorptivity of the different species at each wavelength can also be determined. Owing to the acidity ranges investigated the hydrolysis reactions of the metal ions were always neglected. M can be investigated spectrophotometrically 1130 Zinc( 11) Complexation hydroxide (pH S) only by T-azo-C and 3,6-disulpho-TAN. C at different pH values are shown in Figs. 2 and 3. 1 1 (metal to ligand) is formed in solutions with both Cz,> C and C,> Cz,. plexation reactions and the formation constants are reported in Table I.PESAVENTO AND SOLDI SPECTROPHOTOMETRY OF COMPLEXES Analyst Vol. 108 Results Undcr these experimental conditions zinc can be chelated before the precipitation of the Spectra of solutions with C,,> Only a complex with a molar ratio of The com-0.4 0) t e s 2 0.2 300 380 460 540 620 700 Wavelengthhm Fig. 2. Absorption spectrum of Zn - T-azo-C. p H (1) 1.940; (2) 2.650; (3) CFn = 4.12 x low4 M 3.462; (4) 5.050; ( 5 ) 5.865; (6) 6.715; (7) 7.700; and (8) 8.802. and CL = 2.01 x 10-5 M. Copper( 11) Complexation Copper(I1) forms stable complexes with all the azo dyes considered here. The spectra of solutions containing C,,>C at different acidities are reported in Figs. 4 5 and 6 respectively, for the ligands T-azo-C 3,6-disulpho-TAN and amaranth.With T-azo-C two complexes are successively formed when C,,>C, and the solution is made more basic. The former having a metal to ligand molar ratio of 1 1 can also be detected in solutions with C,>C,,. The second with a metal to ligand ratio of 2 1 is formed only in solutions containing a large excess 0.3 0.2 2 -P s 0.1 0 300 380 460 540 620 700 Wavelengthhm Fig. 3. Absorption spectrum of Zn - 3,6-disulpho-TAN. p H (1) 2.142; (2) 2.550; (3) 3.036; (4) 3.292; (5) 3.542; (6) 3.733; (7) 4.060; (8) 4.307; (9) 4.567; (10) 4.826; (11) 5.0S2; (12) 5.340; (13) 5.600; (14) 5.921; (15) 6.360; and (16) 6.215. Czn = 4.12 x lo-* M and CL = 2.65 x M September 1983 of metal ion and is not very soluble.complex with a metal to ligand molar ratio of 1 1. reported in Table I. OF CLJ(II) AND ZN(II) WITH SULPHONATED AZO DYES 1131 With the other two ligands copper(I1) forms only one The complexation reactions and the formation constants of the copper(I1) complexes are TABLE I COMPLEXATION REACTIONS OF SOME AZO DYES WITH COPPER, NICKEL AND ZINC Log of equilibrium Ligand Metal ion Complexation reaction constant Log Kf Reference T-azo-R . . . . T-azo-C . . . . 3,6-Disulpho-TAN. . TAAC Amaranth Cr-2B Copper Nickel Zinc Copper Nickel Zinc Copper Nickel Zinc Copper Nickel Zinc Zinc Copper Nickel Zinc z:; Cu + HL f CuL + H Ni + HL + NiL + H NIL + HL $ NIL + H Zn + HL .$ ZnL + H Cu + H,L +$HL + H CuHL + Cu - Cu,L + H Ni + H,L + NiHL + H NiHL + H,L f.Ni(HL) + Zn + H,Lf ZnHL + H Cu + HL - CuL + H Ni + HL $ NiL + H NiL + HL $ NiL + H Zn + HL =? ZnL + H Cu + HLL.+pC"HL + H CuHL + Cu + Cu,L + H Ni + H,L + NiHL + H NiHL + Ni + Ni,L + H - . No reaction Cu + HL + CuL + H Ni + HL 2 NiL + H No reaction Cu + H,L + CuHL + H No reaction No reaction H 3.70 f 0.03 1.27 f 0.03 -2.80 f. 0.05 -0.97 0.02 1.63 f 0.02 -1.80 f 0.05 -0.88 f 0.02 -3.52 f 0.04 -2.72 f 0.03 3.44 f 0.03 1.16 f 0.02 -1.62 f 0.05 -0.94 f 0.02 -0.35 f 0.02 -2.26 f 0.05 -2.64 0.03 -4.84 f 0.05 -1.30 f 0.02 -4.35 f. 0.03 -1.90 f 0.03 13.20 10.77 17.47 8.53 9.93 7.42 12.12 5.58 10.85 8.57 14.36 6.47 4.79 2.50 ---8.50 5.45 6.75 1 1 1 2 This paper This paper 3 3 This paper This paper 3 3 This paper 4 4 3 3 This paper This paper 3 This paper 5 3 This paper Discussion The final results are reported in Table I together with some others previously obtained on the complexation of copper(I1) and zinc(I1) with T-azo-R.The complexation reactions of nickel(II) the constants of the complexation reactions and the stability constants (Kf) of the complexes are also reported in Table I. These values have been calculated from the equilibrium constants of the proton displace-ment reactions given in Table I. In order to calculate these values it is necessary to know which proton is displaced. For instance copper(I1) forms a 1 1 monoprotonated complex with Cr-2B.The following complex structures are possible: I II By observing that the derivatives of the naphth-2-01-3,B-disulphonic acid (R-salt) form the metal complexes in more acidic solutions and bind the proton more strongly with respect to the corresponding chromotropic acid derivatives I1 seems more likely. As the same holds for nickel and zinc all the stability constants reported in Table I have been calculated assuming that chelation occurs through the azo-nitrogen and the ortho-phenolic oxygen. The presence of a donor atom in the diazo moiety enhances the complexing ability of the ligand possibly because it behaves as a tridentate chelate. Only one exception is observed, the complex of copper(I1) with TAAC is less stable than the corresponding one with Cr-2B.The gain in the stability of the complex due to the formation of a tridentate chelate seems to be more than counter balanced by the electron-attracting capability of the thiazolyl heterocycle, which makes the free anion of TAAC more stable than that of Cr-2B. This is supported by the fact that the protonation constant of Cr-2B is much higher than that of TAAC. The anio PESAVENTO AND SOLD1 SPECTROPHOTOMETRY OF COMPLEXES 1132 0.8 0.6 W c (II e a 0.4 0.2 0.0 4 Analyst Vol. 108 400 480 560 640 720 800 Wavelengthhm Fig. 4. Absorption spectrum of Cu - T-azo-C. pH (1) 1.071; (2) 1.798; (3) 2.208; (4) 3.486; (5) 4.572; (6) 4.978; and (7) 5.164. CcU = 9.96 x produced by the dissociation of the phenolic group of 3,6-disulpho-TAN is much more basic than that of TAAC as is evident from a comparison of the protonation constants.Consequently the 3,6-disulpho-TAN being tridentate and sufficiently basic is able to form more stable complexes than the corresponding ligand amaranth while the opposite is true for TAAC and Cr-2B. Compared with the nickel(I1) behaviour neither copper(I1) nor zinc(I1) form 1 2 (metal to ligand) complexes. Copper forms binuclear complexes not only with TAAC which has a great tendency to form 2 1 metal to ligand c~mplexes,~ but also with T-azo-C. Possible structures of the binuclear complexes of copper(I1) are M and CL = 4.07 x M. The former seems more likely after considering that copper(I1) is not able to form binuclear complexes with the corresponding R-salt derivatives T-azo-R and 3,6-disulpho-TAN where the peri-hydroxy structure is not present.The previous observation3y7 that azo dyes derived from the R-salt bind cations more strongly than those derived from chromotropic acid (with the only exceptions of magnesium and aluminium) has also been confirmed for copper and zinc. The reason for this is not yet understood. The order of the stabilities of the complexes formed by all the ligands agrees with the Irving - William series The differences for logKf are of about 2 log units. Analytical Applications As already discussed for nickel(II) only amaranth shows a blue shift on chelation. In this instance the spectra of copper(I1) and nickel(I1) complexes are very similar with an absorbance maximum at wavelength 463 nm (molar absorptivity cC == 1.1 x lo4 1 mol-1 cm-l) and an isosbestic point at 473 nm.For the other ligands the red shift of the absorption bands com-pared with the corresponding ones of the free ligands increases in the order Zn(II) Ni(II), Cu(I1) complexes. In all the examples the spectra of the complexes formed with the thre September 1983 OF CU(II) AND ZN(II) WITH SULPHONATED AZO DYES 1133 I I I I 300 380 460 540 620 Wavelengthhm 700 Fig. 5. Absorption spectrum of Cu - 3,Bdisulpho-TAN. pH (1) 1.090; (2) 1.340; M and CL = 8.21 x (3) 1.637; (4) 1.931; (5) 2.230; and (6) 5.043. CcU = 1.04 x lom6 M. 0.8 -0.6 -al S m 9 $ 2 0.4 -0.2 0.0 300 380 460 540 Wavelengthhm 620 700 Fig. 6. Absorption spectrum of Cu - amaranth. pH (1) 1.062; (2) 2.186; (3) 3.002; (4) 3.280; (5) 3.456; (6) 3.'696; (7) 3.946; (8) 4.241; (9) 4.515; (10) 4.777; and (11) 5.020.CcU = 1.96 x 10-3 M and CL = 4.08 x 10-5 M. cations overlap considerably. Consequently copper( I I) even at very low concentration does interfere in the spectrophotometric determination of both nickel and zinc with all the sulphon-ated azo dyes considered. In contrast the determination of copper can be performed select-ively by choosing the correct acidity and wavelength. 3,6-Disulpho-TAN T-azo-C and T-azo-R appeared to be the best spectrophotometri 1134 PESAVENTO AND SOLD1 reagents for copper(I1) determination because they are able to form undissociated complexes in acidic solutions (pHG4) with C = 1-2 x A method is described for the determination of copper in some steel samples which also contain nickel(I1) using 3,6-disulpho-TAN.Table I1 shows the results of the analyses ob-tained under the following conditions ca. 0.2 g of the sample was dissolved in concentrated nitric acid and the solution was made up to 50 ml with water and sufficient nitric acid to avoid iron(II1) precipitation. A l-ml volume of this solution was added with 1 ml of a 3,6-disulpho-TAN M solution and 0.1 ml of 85% phosphoric acid and diluted to 10 ml with water. The acidity of the final solution was adjusted to a pH of about 1 with concentrated sodium hydroxide solution. The absorbance was measured at 580 nm against a blank containing the ligand at the same acidity and the concentration of copper determined by the standard additions method. The photometric sensitivity of the outlined procedure is 0.004 pg cm-2 slightly worse than that obtainable at lower acidity (0.002 pg cm-2).On the other hand at pH 1 only few inter-ferences exist. The values of C,/C, that produce an error of 5% in the determination of copper(I1) are as follows Ni(I1) 6.07 ; Pb(I1) 200; Mn(I1) Co(I1) Cd(I1) and Zn(I1) no inter-ference up to 5 x M ; Ag(I) 7.23; Bi(III) 0.06; Pd(II) 0.20; Hg(II) no interference up to CHg = C,; and Cr(III) 28 (the colour is completely developed after 20 min). It is also possible to determine the nickel content of the same samples if its concentration is at least 10% of that of copper. The following procedure was used 1 ml of the original solution was added with 1 ml of M 3,6-disulpho-TAN solution 1 ml of acetate buffer (1 M pH = 4.5) and diluted to 10 ml with 0.15 M tartrate solution.The absorbance of the final solution was recorded at 580 nm against a blank consisting of all the components except nickel. In the blank copper(I1) was also included whose concentration was known from the above described procedure at pH 1 (the molar absorptivity of copper using the conditions of the nickel determination was 2.8 x lo4 1 mol-l cm-1). The concentration of nickel was obtained by the standard additions method. Many heavy metal ions interfered in this determination of nickel3 and of course, also in the determination of copper when performed at the same acidity. It is therefore, more preferable to perform the analysis at a pH of 1. M. An excess of nitric acid does not react with 3,6-disulpho-TAN.Final results of this analysis are also shown in Table 11. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. TABLE I1 SPECTROPHOTOMETRIC DETERMINATION OF COPPER(II) AND NICKEL(II) IN STEEL Sample %m/m Copper found,* yo %mlm Nickel found,* yo 1 0.14 0.14 f 0.005 0.07 0.08 f 0.008 2 0.10 0.10 f 0.003 0.06 0.06 f 0.008 3 0.17 0.17 f 0.04 0.15 0.15 f 0.012 4 0.18 0.18 f 0.004 0.14 0.14 f 0.012 Copper content Nickel content, * Mean value of four independent determinations. References Fulle Soldi T. Bertoglio Riolo C. Gallotti G. and Pesavento M. Gazz. Chim. Ital. 1977 107 347. Riolo Bertoglio C. Pesavento Frau M. and Soldi Fulle T. Ann. Chim. (Rome) 1978 68 651. Pesavento M. and Soldi T. Ann. Chim. (Rome) submitted for publication. Bertoglio Riolo C. Fulle Soldi T. Gallotti G. Pesavento M. and Spini G. Gazz. Chim. Ital., Bertoglio Riolo C. Fulle Soldi T. Gallotti G. and Pesavento M. Gazz. Chim. Ital. 1974 104 193. Bottari E. Liberti A. and Ruffolo A. J . Inovg. Nucl. Chem. 1968 30 2173. Pesavento M. Riolo Bertoglio C. Soldi Fulle T. and Cervio G. Ann. Chim. (Rome) 1979 69 649. Busev A. I. Krysina L. S. Zholondkovskaya T. N. Pribilova G. A. and Krysin E. P. Zh. Anal. Pesavento M. Riolo C. Soldi T. and Garzia R. Ann. Chim. (Rome) 1982 72 217. Langova M. Klabenesova I. Kasiura K. and Sommer L. Collect. Czech. Chem. Commun. 1976, Pesavento M. Ann. Chim. (Rome) 1983 73 173. 1975 105 221. Khim. 1970 25 1575. 41 2386. Received February 21st 1983 Accepted March 23rd 198

ISSN:0003-2654    

DOI:10.1039/AN9830801128

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Analyst September 1983 Vol. 108 99. 1135-1140 1135 Substituted N-Hydroxy-NN‘-diary1 benzamidines as Selective Extractants for the Spectrophotometric Determination of Vanadium(V) in the Presence of Acetic Acid or an Azide Miss Abha Rani Jha and Rajendra Kumar Mishra Department of Chemistry Ravishankar University Raipur 492 010 India The reactions of 1 1 newly synthesised N-hydroxy-NN’-diarylbenzamidines (HOA) with vanadium(V) in the presence of acetic acid or an azide have been studied spectrophotometrically to determine the effect of substituents on the complexing properties of the ligand. The investigations showed the formation of 1 2 1 [vanadium(V) HOA acetic acid] and 1 2 2 [vanadium-(V) HOA azide] complexes in chloroform. Based on the mixed complex formations simple rapid sensitive and selective methods for the spectro-photometric determination of microgram amounts of vanadium(V) have been developed.The methods have been applied to the determination of the vanadium content of BCS steels. Keywords ; Vanadium determination ; spectrophotovnetry ; N-hydroxy-NN’-diarylbenzavtzidines ; steel analysis ; solvent extraction Numerous methodsl-l1 have been recommended for the spectrophotometric determination of vanadium(V). However these methods are not very satisfactory as they suffer from serious interference from one or more elements such as iron copper chromium nickel manganese, cobalt and tungsten which are commonly associated with vanadium in alloys and complex materials. In these methods either the interference cannot be removed or the modified pro-cedures are lengthy.Svehla and Tolg2 proposed disubstituted aromatic hydroxylamines of hydroxamic acids as being the most sensitive spectrophotometric reagents €or vanadium. Although N-benzoyl-N-phenylhydroxylamine and its analogues3 are useful reagents molyb-denum titanium zirconium manganese chromium and tungsten interfere with the determina-tion of vanadium.12 Also low and erratic results were obtained by this method.13-16 Pal et a1.l’ have now found that substituted hydroxamic acids are unsuitable for determining vanadium in various alloy steels containing chromium molybdenum arid tungsten. The proposed methods based on the chloroform extraction of mixed-ligand complexes of vanadium (V) with I\r-hydroxy-l\l-~-chlorophenyl-N’-(2,3-dimethyl)phenylbenzamidine hydro-chloride (HCPDMPBH) and acetic acid or an azide compare favourably with the above mentioned methods.The reagent does not absorb at the A,,,. of the complexes hence chloroform can be used as a reference. Vanadium(V) is not reduced and the absorptivity and the position of A,,,. of the complexes are not affected by the presenceof ethanol addedasa stabiliser to the chloroform. The methods are sensitive (Sandell sensitivities of 0,0127 and 0.0095) and are applicable in 0.5-10.0 M of acetic acid in the acetic acid system and in a pH range of 0.6 to 5.2 in the azide systems respectively. The methods are free from interferences by most of the common ions and tolerate considerable amounts of Zr(IV) Ti(IV) Nb(V), Ta(V) and W(V1). Iron(II1) does not interfere in the acetic acid system and it can be masked with trisodium phosphate in the azide system.Large amounts of copper can also be masked with thiourea. Hence these methods have the added advantages of not requiring an initial separation of iron and of one single extraction being sufficient for the complete extraction of vanadium (V) , Experimental Apparatus was used for absorptiometric measurements. Systronic pH meter Type 322. An ECIL ultraviolet - visible spectrophotometer Model GS-865 with matched silica cells The pH measurements were made with 1136 Chemicals and Reagents JHA AND MISHRA SUBSTITUTED HOA AS EXTRACTANTS FOR Analyst Vol. I08 All of the chemicals used were of analytical-reagent grade. Standard vanadium(V) solution.A stock solution of vanadium(V) was prepared by dis-solving ammonium metavanadate in doubly distilled water and standardising titrimetrically.18 Dilute solutions were prepared from this solution. Hydroxyamidines were prepared by the condensation of equimolar amounts of N-arylbenzimidoyl chloride with the appropriate N-arylhydroxylamine in diethyl ether.l9 The resulting hydrochloride was recrystallised from absolute ethanol containing a few drops of concentrated hydrochloric acid. The free bases were obtained by treatment of the hydrochlorides with dilute ammonia solution and crystallised from benzene - light petroleum (2 + 1). Satisfactory results were obtained for the elemental analysis of these compounds. Hydroxyamidines. Caution-Benzene is highly toxic and appropriate precautions should be taken.Extraction solutions. A 2% m/V solution of sodium azide in water and a 0.1% m/V solution Caution-Chloroform is hazardous and appropriate precautions should be taken carrying out the of the hydroxyamidine in chloroform were used for all extraction work. extraction procedure under a fume hood. Procedure Place an aliquot of solution containing 100 pg of vanadium(V) in a separating funnel. To this add 9 ml of glacial acetic acid and dilute to 25 ml or add 5 ml sodium azide solution and maintain the pH at 2.5 in the final dilution of 25 ml using hydrochloric acid - potassium chloride buffer for the acetic acid and the azide system respectively. Equilibrate the aqueous phase with 15 ml of the chloroform solution of the hydroxyamidine for 2 min.Transfer the organic phase into a 50-ml beaker containing anhydrous sodium sulphate (2 g). Wash the aqueous phase with 2 x 3-ml portions of chloroform. Decant the combined extract into a 25-ml calibrated flask and dilute to volume with chloroform. Read the absorbance at the wavelength of maximum absorption of the complexes using chloroform as a blank. 0.7 0.6 0.5 0 0 2 0.4 n 5: a 0.3 0.2 0.1 0 400 500 600 700 Wavelengthtnm Fig. 1. Absorption spectra of mixed complexes of vanadium(V) with HCPDMPBH and acetic acid or azide CV = 7.85 x M and CHCPDMPBH = 0.003 M. (A) Vanadium(V) - HCPDMPBH -azide CN3- = 0.10 M ; (B) vanadium-(V) - HCPDMPBH - acetic acid, CCH~COOH = 6.0 M ; (C) vanadium-(V) - HCPDMPBH; and (D) HCPDMPBH in chloroform September 1983 SPECTROPHOTOMETRY OF V(V) WITH ACETIC ACID OR AZIDE Absorption Spectra Results and Discussion 1137 The absorption spectra of the complexes and the reagent in chloroform are illustrated in Fig.1. The reagent reacts with vanadium(V) in the absence of adduct forming substances giving a blue - violet complex in chloroform ( E = 1900 1 mol-l cm-l at Amax. 570 nm). In the presence of acetic acid or azide intensely coloured mixed complexes are produced ( E = 4000 and 5350 1 mol-l cm-l at Amax. 580 and 590 nm in the acetic acid and azide systems respectively). The reagents show negligible absorption in the 450-700 nm region. Choice of Solvent Of the various solvents (benzene chloroform and carbon tetrachloride) tried chloroform is used as a diluent because in this the distribution coefficient of the reagents and adducts are high.Effect of Acidity acid. HCPDMPBH was found to be 0.5-10.0 M acetic acid. In the acetic acid system acidity of the aqueous phase was maintained with glacial acetic The optimum acidity range for the quantitative extraction of the complex with An acidity above 10 M cannot be TABLE I SPECTRAL CHARACTERISTICS OF VANADIUM(V) MIXED COMPLEXES WITH VARIOUS HYDROXYAMIDINES AND ACETIC ACID OR AZIDE I N CHLOROFORM Sample No.* 1 2 3 4 5 6 7 8 9 10 11 X m-C1 m-C1 m-C1 m-C1 m-C1 m-C1 m-C1 m-C1 H P-CH3 p-Cl Acetic acid system Azide system A f 'P Sandell's ~ / 1 mol-1 cm-1 sensitivity/ ~ / 1 mol-l cm-l sensitivity/ Y (Amax./nm) pg V (Amax./nm) pg V H 3 800 0.0134 4 400 0.01 1 6 0.009 9 O-CH 2 900 0.0175 5 150 0.011 6 m-CH 3 750 0.0136 4 400 0.0104 0.012 7 p-OCH 3 200 0.0159 4 000 0.009 7 295 (CH3) 2 3 300 0.0154 5 250 0.0424 c 1200 2,6(CH3) No colour 0.009 5 3(CH3) 2 3 700 0.0137 5 350 0.010 1 283 ((3%) 2 3 600 0.014 1 5 050 0.009 9 2,3(CH3) 2 3 950 0.0130 5 150 2,3(CH3)2 4 000 0.012 7 5 (595) 350 0.009 5 (585) (595) (585) P-CH 3 600 0.014 1 4 900 (585) (590) (590) (600) (590) (590) (570) (570) (570) (565) (580) (575) (575) (580) (585) (680) (590) * Compounds 1 4 and 5 are free bases 1138 JHA AND MISHRA SUBSTITUTED HOA AS EXTRACTANTS FOR A?dySt V d .108 attained due to the miscibility of acetic acid with chloroform. The acidity of the aqueous phase in the acetic acid system was therefore maintained at 6 M throughout the investigation, for consistency of operation.The pH of the aqueous phase was adjusted with 2 M hydrochloric acid and dilute ammonia solution in the azide system. The optimum pH range for complete extraction of the metal was found to be 0.6-5.2. Effect of Reagents At least 12- and 10-fold molar excesses of HCPDMPBH were necessary for the complete extraction of the metal in the acetic acid and azide systems respectively. Addition of more reagent up to a 100-fold molar excess caused no adverse effect on the values of Amax. or on the absorbance of the complexes. An acetic acid concentration of 0.5 M and a 140-fold excess of azide were found to be necessary for the complete extraction of vanadium(V).Effect of Substituents N-Hydroxy-NN’-diarylbenzamidine20s21 has three possible sites for substitution and hence has wider scope for improvement in its complexing properties compared with other re-a g e n t ~ . ~ ~ $ ~ ~ In this investigation the effect of substituents on the complexing properties of the -N=C-N(0H)-group has been evaluated on the basis of potentialities of 11 analogues of hydroxyamidines towards the extractive spectrophotometric determination of vanadium(V) in chloroform. The absorption spectra of the ternary complexes of these reagents were scanned and the molar absorptivities of the complexes were calculated on the basis of metal content a t their respective values of Amax. (Table I). It is observed that the substitution in the N - or N’-phenyl ring has little effect on the position of Amax.however it affected the molar absorptivity of the complex. Firstly the introduction of methyl groups in both of the ortko positions in the N’-phenyl ring decreased the absorptivity markedly prob-ably owing to steric hinderance; the 0-CH and 2,5(CH3) groups in this phenyl ring also showed a hypochromic shift in the acetic acid system. Substitution in this phenyl ring with a p-OCH group decreased the absorptivities in both the systems. However substitution in the N’-phenyl ring in the azide system showed a hyperchromic shift in the order 2,3(CH,),, 2,5(CH,),>O-CH3>~-CH (Table I 1-8). Secondly substitution in the N-phenyl ring with p-Cl m-Cl and 9-CH groups caused a hyperchromic shift (Table I 8-11).The steric hinder-ance observed due to the substitution of the N-phenyl ring of hydroxyamidine apparent in the acetic acid system may be explained by the following probable structures of the vanadium mixed-ligand complexes I V(V) - HOA - CH3COOH complex (1 2 1) ; and 11 V(V) - HOA -Na- complex (1 2 2). The apparent trends observed are as follows. CH&OOH . Ar” 8 1 - 1 Ar-N-O\ jJN=C- Ar’ Ar-C-N’l I ‘0-N-Ar I I OH Ar” I HOA 0 v \ II /N3- Ar-N-0 Ar-C=N/* ‘N3-I In the acetic acid mixed complex the acetic acid molecule is an adduct while both the mole-cules of hydroxyamidines are directly co-ordinated with the metal hence there is more possi-bility of steric hinderance which is shown by the hypochromic shift in the complexes of ortko-substituted hydroxyamidines.However in the azide mixed complex one molecule of hydroxyamidine is directly co-ordinated with vanadium(V) and the second molecule is an adduct similar to the vanadium(V) - oxine - SCN- complexes.24 Effect of Other Variables 15 to 40 “C and the volume of the aqueous phase from 15 to 60 ml are not critical. The order of addition of reagents concentration of reagents variation of temperature from A time o September 1983 1139 2 min was sufficient for complete extraction of vanadium(V) and the complexes are stable for at least 40 h at room temperature. SPECTROPHOTOMETRY OF V(V) WITH ACETIC ACID OR AZIDE Beer’s Law Sensitivity and Precision of the Method The acetic acid and azide systems (with HCPDMPBH) followed Beer’s law in the ranges 0.8-1 1.2 and 0.8-8.8 p.p.m.of vanadium respectively. The optimum concentration ranges, on the basis of a Ringbom plot25 were found to be 2.4-10.4 and 1.8-7.6 p.p.m. of vanadium for the acetic acid and azide systems respectively. The Sandell sensitivities of t h e colour reac-tions for vanadium are 0.012 7 and 0.0095 pg cm-2 (acetic acid and azide systems respectively). The mean absorbance values of the acetic acid and azide systems were found to be 0.315 and 0.420 and relative standard deviations & 1.047 and & 1.047% respectively (10 measurements were made on solutions each containing 100 pg of vanadium per 25 ml). Composition The ratio of hydroxyamidine and acetic acid or azide to the metal was determined by Job’s method of continuous variation26 and curve-fitting methods.27 The results obtained showed the formation of 1 2 1 (V HOA CH,COOH) and 1 2 2 (V HOA N3-) mixed c o m p l e x e ~ .~ ~ ~ ~ Influence of Diverse Ions To evaluate the effect of different ions solutions containing 4 p.p.m. of vanadium(V) and varying amounts of other ions were analysed with HCPDMPBH as described in the recom-mended procedure. Up to 2 500 p.p.m. of chloride bromide nitrate sulphate urea thiourea, alkali and alkaline earth metal ions and lanthanides do not interfere in the determination of vanadium as the acetic acid and azide mixed complexes. Thiocyanate interferes seriously as it changes the colour of the complex to green and shifts A,,, to 600 nm with an increase in E . The amounts tolerated (in parts per million) of other ions in the acetic acid and azide systems are shown respectively in parentheses fluoride iodide phosphate thiosulphate (800 1 000) ; citrate tartrate (400 400) ; triethanolamine selenate (800 900) ; Bi3+ (1 000 800) ; Sb3+ TlPF (1 000 1000) ; Pb2+ (1 200 1000) ; Ni2+ Co2+ Cd2+ (800 2000) ; Mn2+ Cu2f (800 800) ; A13+, Cr3+ Fe3+ (800 1000); Ti4+ (60 200); Zr4+ (20,40); Mo6+ (200 600); W6+ (20 40); and U6+ (400 1000).Application of the Method The validity of the proposed methods have been tested by determining the vanadium content of three BCS steel samples two vanadium - tungsten steels BCS 64 and BCS 241/1 and one tungsten-free steel BCS 252. A weighed amount of the sample containing approximately 2 mg of vanadium was dissolved in 40% nitric acid in a 400-ml beaker.The solution was evaporated to near dryness and 5 ml of concentrated hydrochloric acid were added to this and again evaporated to near dryness; 25 ml of water were added and the solution was boiled. The yellow precipitate of tungstic acid, along with other insoluble materials was filtered off. The residue was washed with hot water The results are summarised in Table 11. TABLE I1 DETERMINATION OF VANADIUM IN VARIOUS BCS STEELS Sample* BCS 64a alloy steel . . BCS 241/1 high-speed steel BCS 252 low-alloy steel . . . Certified Vanadium value found,t % . 1.570 A 1.556 B 1.552 . 1.570 A 1.552 B 1.550 . 0.460 A 0.452 B 0.450 Standard deviation yo f0.0102 f 0.006 0 f0.0064 f 0.005 9 f0.005 6 f 0.006 0 * BCS British Chemical Standards Bureau of Analysed Samples Ltd.Newham t Average of six determinations. A Acetic acid system and B azide system. Hall Middlesbrough 1140 JHA AND MISHRA and 1% hydrochloric acid. The filtrate was again evaporated to a small volume. The pro-cess was repeated 2-3 times and then finally diluted to volume in a 100-ml calibrated flask with water containing one drop of potassium permanganate solution to ensure that all the va-nadium is in the vanadium(V) state. The vanadium content of the sample was determined as described under Procedure. Conclusion Simple rapid sensitive and selective methods for the extractive spectrophotometric deter-mination of vanadium with N-h ydrox y-N-p-chloropheny1-N’- (2,3-dimet h yl) phen ylbenzamidine hydrochloride (HCPDMPBH) and acetic acid or an azide have been developed.The methods are free from the rigid control of various analytical variables and follow Beer’s law. The methods are applicable in the acidity range 0.5-10.0 M of acetic acid for the acetic acid system and in a pH range of 0.6-5.2 for the azide system. The effect of substituents has been discussed the steric effect is more obvious in the acetic acid system. The substitution of the N’-phenyl ring in both the ortho positions i.e. 2 and 6 resulted in the disappearance of the colour in the acetic acid system however this decreased the absorptivity of the complex in the azide system. Most of the common metal ions associated with vanadium in its ore and alloys do not inter-fere. The methods have been applied satisfactorily to determine the vanadium content of standard steel samples.Thanks are due to Professor S. G. Tandon Head Department of Chemistry Ravisliankar University Raipur for providing research facilities and to the Council of Scientific and Industrial Research New Delhi for awarding a Senior Research Fellowship to one of us (A.R. J.) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 26. 26. 27. References Sandell E. B. “Colorimetric Determination of Traces of Metals,” Third Edition Interscience New Svehla G. and Tolg G. Talanta 1976 23 755. Majumdar A. K. “N-Benzoyl-N-phenylhydroxylamine and its Analogues,” Pergamon Press, Yatirajam Y. and Arya S. P. Anal. Chim. Acta 1976 86 209. Anjaneyulu Y.Sarma B. S. R. and Rao V. P. R. Anal. Chim. Acta 1976 86 313. Ushida F. Yamada S. and Tanaka M. Anal. Chim. Acta 1976 83 427. Raje R. B. and Sane R. T. J . Indian Chem. Soc. 1977 54 416. Sanke Gowda H. and Shakunthala R. Analyst 1978 103 1215. Akama Y. Nakai T. and Kawamura F. Analyst 1981 106 250. Sharma Y. Anal. Chim. Acta 1981 126 233. Yoshimura K. Kaji H. Yamaguchi E. and Toretani T. Anal. Chim. Acta 1981 130 345. Shendrikar A. D. Talanta 1969 16 51. Goto H. and Kakaita Y. Bunseki Ir‘agaku 1961 10 904. Vita 0. A. Levier W. A. and Litteral E. Anal. Chim. Acta 1968 42 87. Baughman W. J. and Waterbury G. R. U.S. A t . E. C. 1968 LA 3843. Donaldson E. M. Talanta 1970 17 583. Pal B. K. Mitra B. K. and Chattopadhyay S. Talanta 1976 23 554. Charlot G. and Bezier D. “Quantitative Inorganic Analysis,” John Wiley New York 1957. Deb K. K. and Mishra R. K. J . Indian Chem Soc. 1976 53 178. Satyanarayana K. and Mishra R. I<. Anal. Chem. 1974 46 1605. Satyanarayana K. and Mishra R. K. J Indian Chem. Soc. 1976 53 63 469 and 928. Vogel A. I. “A Text Book of Quantitative Inorganic Analysis,” Longmans Green London 1964. Sogani N. C. and Bhattacharya S. C. Anal. Chem. 1956 28 81 and 1616. Rao V. P. R. and Anjaneyulu Y. J . Inovg. Nucl. Chem. 1971 33 3567. Iiingbom A. 2. Anal. Chem. 1939 115 332. Job P. Ann. Chim. (Paris) 1928 9 113. Sillen L. G. Acta. Chem. Scand. 1956 10 185. York 1959 p. 926. Oxford 1972. Received November 17th 1982 Accepted Febmayy 2Ist 198

ISSN:0003-2654    

DOI:10.1039/AN9830801135

出版商:RSC    

年代:1983

数据来源: RSC

摘要:

Analyst September ,1983 SHORT PAPERS 1141 Spectrophotometric Determination of Phosphate in Polluted Waters by Solvent Extraction of Molybdenum Miss Abha Chaube and V. K. Gupta" Department of Chemistry Ravishankar University Raipztr 492 010 India Keywords Spectrophotometry ; phosplzate ; polluted waters ; solvent extraction ; molybdenum blue Phosphorus which is an important nutrient occurs widely in the environment. In the aqueous environment phosphorus is a prerequisite for microbial growth. Heavy algal growth occurs when phosphorus is present in water and as such is undesirable. The determination of phosphorus is therefore of importance to water analysts and limnologists. Fertilisers, detergents and sewage are major sources of phosphorus as phosphate. A level of 0.03-0.40 mg 1-1 of total inorganic phosphate is a common criterion for the maximum acceptable levell in waters.'The molybdenum blue procedure for assaying inorganic phosphate is a well established method. I t involves the formation of 12-molybdopliosphoric acid from orthophosphate and an excess of molybdate in acid solution followed by reduction to give molybdenum blue. Various reducing agents2-8 have been reported in the literature but most of them suffer from some drawbacks such as stability of the colour inteferences from arsenic and copper the length of time required for full colour development sensitivity and absorption by the blank. Solvent extraction has been proposed by some workers to enhance the sensitivity as well as to reduce interference of some expected interferent~.~ In this investigation a new reducing agent malonyldihydrazide is used for the reduction of molybdophosphoric acid to give molybdenum blue.Extraction of molybdenum blue has been carried out for the determination of phosphorus at sub-microgram levels in tap water and in polluted water. Extraction in butanol (Amax. 780 nm Emax. = 3.40 x lo4 1 mol-l cm-l) gives satisfactory results. The proposed method is simple and sensitive and the colour is stable for 2 d. The method is free from the interference of molybdenum arsenic and copper (after removal by cupferron). The blank gives no colour which is a major advantage of the method. The influence of various parameters such as acidity and reagent concentration on the colour system have been investigated.Experimental Apparatus The spectral measurements were made using a Carl Zeiss recording ultraviolet - visible spectrophotometer Model GS-865 with 10-mm matched silica cells and a Systronics digital spectrophotometer Model 105. A short-stemmed pear-shaped separating funnel of 125-ml capacity was used for extraction purposes. Reagents All chemicals used were of analytical-reagent grade. Maloizyldihydrazide. Malonyldihydrazide was prepared by adding hydrazine hydrate (0.1 mol in ethanol) to an ethanolic solution of diethyl malonate (0.05 mol in ethanol) in a molar ratio of 2 1. The white solid obtained was recrystallised from water and had a melting-point of 150 "C. A 1% aqueous solution of the reagent was prepared. * To whom correspondence should be addressed 1142 SHORT PAPERS Analyst Vol.108 A stock solution of potassium dihydrogen phosphate containing 1000 pg ml-l of phosphorus was prepared in de-aerated doubly distilled water. Working standards were prepared by appropriate dilution of the stock solution. Phosphate stock solution. Sulphuyic acid. Sulphuric acid (10 N) was prepared in doubly distilled de-aerated water. Ammonium molybdate solution. A 2.5% ammonium molybdate solution was prepared. Procedure An aliquot (about 50 ml) of water containing 1-7 pg (0.02-0.14 p.p.m.) of phosphorus was taken in a 150-ml boiling-tube and 2 ml of ammonium molybdate solution were added. The acidity was adjusted to 2 N with sulphuric acid. Then 1 ml of malonyldihydrazide solution was added and the solution was kept on a boiling water-bath for 2 min.Development of a blue colour indicated the formation of molybdenum blue owing to the presence of phosphorus in the sample. The solution was allowed to cool for 10 min for full colour development. This solution was then transferred into a 125-ml separating funnel and extracted with two 5-ml portions of butanol. The blue extract was dried over anhydrous sodium sulphate and the absorbance was measured at 780 nm. A reagent blank prepared by following a similar pro-cedure showed negligible absorbance hence all measurements were made with butanol as the reference. The concentration of phosphorus was calculated from a calibration graph prepared by treating standard solutions in a similar manner. Results and Discussion All spectral studies were made with 50 ml of distilled water containing the required amount The absorption spectra of molybdenum blue has absorption maxima at The reagent blank showed negligible absorption in The colour of the molybdenum blue is stable for 6 h in water and for 2 d of phosphorus.780 nm in butanol and 820 nm in water. both these instances. in butanol. Effect of Varying Reaction Conditions Maximum colour was developed in the acidity range of 1.6 to 2.4 N sulphuric acid. An acid concentration of greater than this caused a decrease in the absorbance whereas below 1.0 N the blank also produced a colour. It was observed that maximum colour intensity was obtained when the solution was kept for 2 min on a boiling water-bath and later cooled for 10 min. It was found that 2 ml of 2.5% ammonium molybdate solution and 0.8-1.2 ml of a 1% malonyldihydrazide solution were needed for full colour development.Beer’s Law Molar Absorptivity Sensitivity and Reproducibility Beer’s law was obeyed for 0.02-0.14 p.p.m. of phosphorus extracted from 50 ml of de-ionised water and measured at 780 nm. The regression analysis of absorbance on concentration gave a regression coefficient of 0.11 with a correlation coefficient of 0.99 and an intercept of 0.00. The molar absorptivity and sensitivity as defined by Sandell of the colour reaction were found to be 3.40 x lo4 1 mol-l cm-l and 0.0009 pg cm-2 (as calculated using the butanol extract) respectively. The reproducibility of the method was studied by replicate analysis of a standard phosphorus solution over a period of 7 d.The data were employed to calculate the standard deviation and relative standard deviation which were found to be h0.004 and &0.56% respectively. Effect of Foreign Ions As this method was developed mainly for the analysis of water samples the effects of foreign ions commonly present in tap water river water and polluted water were studied. The effect was studied by taking samples of tap water and polluted water as well as samples obtained by adding known amounts of ions normally present in effluents to 50-ml aliquots containing 0.08 p.p.m. of phosphorus and proceeding as recommended in the Procedure. Arsenic(III), the interferent most expected did not interfere when up to a 750-fold excess was present as shown in Table I. Copper(I1) did not interfere up to a 250-fold excess after cupferron had been used to separate it as its cupferronate.1° Silicon interferes with the method September 1983 SHORT PAPERS 1143 TABLE I EFFECT OF FOREIGN IONS Concentration of phosphorus 4 pg per 50 ml.Diverse ion Tolerance limit*/mg Ca2+ Sr2+ Cd2+ Pb2+ C1- C0,2- Sod2- . 6.0 Fe3+ Fe2+ Sb3+ As3+. . . . . . 3.0 Mg2+ . . . . 1.6 cu2+ . . . . 1.ot NO,- phenol . . . . . . . . . . 1.0 * Tolerance limit was the amount of diverse ion that caused &2% error. t When cupferron was used. Effect of Solvent Extraction Solvent extraction using various higher alcohols was employed to increase the sensitivity of the reaction. Butanol was found to be the most suitable of the various alcohols tried; the colour was found to be less stable in isoamyl alcohol and complete extraction did not take place in hexanol or octanol.Application of the Method to the Analysis of Tap Water and a Polluted Water factory and tap water. Before analysis water samples were filtered through a Whatman No. 41 filter-paper. were analysed by the proposed procedure. Samples were collected from a stream receiving effluent from a superphosphate fertiliser No pre-treatment was required for the phosphorus determination. Samples Comparison with the conventional ascorbic acid methodll Values obtained for the above samples by the conventional ascorbic acid method were com-pared with those obtained by the proposed method. The coefficient of correlation of the two methods was found to be 0.98 for the determination of phosphorus in 12 water samples.The results of the proposed method are almost identical with those of the conventional method so the accuracy of the method for the analysis of water samples is satisfactory. Results are shown in Table 11. TABLE I1 ANALYSIS OF TAP WATER AND A POLLUTED WATER Phosphorus found/pg ml-I Sample Tap water . . Polluted water . . Proposed method 2.00 2.02 2.05 2.01 2.00 2.03 8.00 8.10 7.99 8.10 8.02 8.09 Ascorbic acid methodli 2.03 2.00 2.00 1.99 1.99 2.03 7.97 8.00 8.09 8.10 8.05 8.10 Conclusion The proposed method is satisfactory for the analysis of polluted waters. Stability excellent reproducibility and lack of interference from a large number of foreign ions are the major advantages of the method.The extraction method makes the method versatile and useful for the determination of phosphorus in a large number of samples 1144 SHORT PAPERS Analyst Vol. 108 The authors thank the Head of the Department of Chemistry and Department of Bio-Sciences for providing laboratory facilities and the University Grants Commission New Delhi, for the award of a Junior Fellowship to one of them (A.C.). 1 . 2. 3. 4 . 5. 6 . 7. 8. 9 . 10. 1 1 . References Black J. A. “Water Pollution Technology,” Reston Publishing Company Inc. Reston VA 1977. Schaffer F. Fong J. and Kirk P. Anal. Chem. 1953 25 343. Chalmers R. A. and Thomson A. D. Anal. Chim. Acta 1958 18 575. Fogg D. N. and Wilkinson N. T. Analyst 1958 83 406. Ging N. S. Anal. Chem. 1956 28 1330. Telep G. and Ehrlich R. Anal. Chem. 1958 30 1146. Djurkin V. Kirkbright G. F. and West T. S. Analyst 1966 91 89. Bauminger B. B. and Walters G. Analyst 1966 91 205. Babko A. K. and Pilipenko A. T. i n Rosinkin A. Translator “Photometric Analysis-Methods of Furman M. F. Mason W. B. and Pekola J. S. Anal. Chem. 1949 21 1324. American Public Health Association American Water VC‘orks Association and Water Pollution Control Federation “Standard Methods for the Examination of Water and Wastewater,” Fourteenth Edition American Public Health Association Washington D.C. 1976. Determining Non-metals,” Mir Moscow 1976. Received November 16th 1982 Accepted March 3rd 198

ISSN:0003-2654    

DOI:10.1039/AN9830801141

出版商:RSC    

年代:1983

数据来源: RSC