ISSN:0003-2654    

DOI:10.1039/AN98611FX033

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALAO 111(9) 1001-1112 (1986) September 1986100110131017102310291033103910451051105910651069107310771085108910951099110311071109TheThe Analytical Journal ofAnalystThe Royal Society of ChemistryCONTENTSAnalytical Applications of the Catalysed Iodine - Azide Reaction. A Review-G. Ramis Ramos, M. C. GarciaUse of Solid Boric Acid as an Ammonia Absorbent in the Determination of Nitrogen-Darryl D. SiemerAutomatic Nitrogen-15 Analyser for Use in Biological Research-Joha J. Therion, Hendrik G. C. Human, CorneliusClaase, Roderick I. Mackie, Al brecht KistnerApplication of Electrothermal Atomic Absorption Spectrometry t o the Determination of Trace Amounts of Indium inMetallic Zinc and Lead-Krystyna Brajter, Ewa Olbrych-SleszyhskaDetermination of Arsenic(V) in Aqueous Solutions by D.c.Argon Plasma Emission Spectrometry. InterferenceStudies-Kimmo Srnolander, Matti KauppinenStudy of Organic Interferences in the Spectrophotometric Determination of Nitrite Using Composite Diazotisation -Coupling Reagents-George Norwitz, Peter N. KeliherSpectrophotometric Determination of Some Pharmaceutical Carbonyl Compounds Through Oximation and Sub-sequent Charge-transfer Complexation Reaction-Saied Belal, Afaf A. El Kheir, Magda M. Ayad, Sobhi A. Al Ad1Spectrophotometric Study of the Iron(lll) - Morin Complex in a Micellar Medium-F. Hernandez Hernandez, J. MedinaEscriche, R. Marin Saez, M. C. Roig BarredaSpectrophotometric and High-performance Liquid Chromatographic Determination of the Kinetics and Mechanisms ofHydrolysis, lsomerisation and Cyclisation of Both E and Z isomers of 2-{ [(2-Amino-5-Chlorophenyl)phenyl-methylene]amino}]acetamide-Maurice Bernard Fleury, Sabine Letellier, Jean-Pierre Porziemsky, Bernard MomponSampling and Gas Chromatographic Analysis of Volatile Sulphur Compounds and Gases at Sub-v.p.m.Levels in thePresence of Ozone-Philip G. Slater, Leigh Harling-BowenEffects of Slow Heating Rates on Products of Polyethylene Pyrolysis-Thomas P. Wampler, Eugene J. LevyDetermination in Urine of Diisocyanate-derived Amines from Occupational Exposure by Gas Chromatography - MassDetermination of Organic Sulphides by Enthalpimetry Using Chromyl Chloride-Mieczybaw Wronski, Awn S. AbbasDescription of Air Pollution by Means of Pattern Recognition.Part 2-Geert Jan H. Roelofs, Frans W. Pijpers, Gfred A. P. E.Flow Cell Studies with lmmobilised Reagents for the Development of an Optical Fibre Sulphide Sensor-RamaierAlvarez-Coque, R. M. Villanueva CamariasFragmentography-Christi na Rosen berg, Heikki Savolai nenJakobsNarayanaswamy, Fortunato Sevilla IllREPORT OF THE ANALYTICAL METHODS COMMllTEECollaborative Studies of Methods for the Detection of Residues of Monensin in Chicken TissuesSHORT PAPERSSimple Fibre Optic pH Sensor for Use in Liquid Titrations-Nira Benaim, Kenneth T. V. Grattan, Andrew W. PalmerTitrations in Non-aqueous Media. Part II. Basicity Order of Aliphatic Amines in Nitrobenzene Solvent-Turgut Gunduz,Titrations in Non-aqueous Media. Part 111. Basicity Order of Aniline, N-AIkyl and N-Aryl-substituted Anilines andProblems in the Dissolution of Silicates by Acid Mixtures-CeIia Maqueda, Jose Luis Perez Rodriguez, Angel JustoNed2 Gunduz, Esma Kili~, Adnan Kenar, Gulay CetinalPyridine in Nitrobenzene Solvent-Turgut Gunduz, Ned2 Gunduz, Esma KIIIG, Adnan KenarBOOK REVIEWSTypeset and printed by Heffers Printers Ltd, Cambridge, Engtan

ISSN:0003-2654    

DOI:10.1039/AN98611BX035

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ii

ISSN:0003-2654    

DOI:10.1039/AN98611BP037

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST SEPTEMBER 1986 VOL. 111 1001 Analytical Applications of the Catalysed Iodine - Azide Reaction A Review G. Ramis Ramos M. C. Garcia Alvarez-Coque and R. M. Villanueva Camaiias Departamento de Quimica Analitica Facultad de Quimica Universidad de Valencia Valencia Spain Summary of Contents Introduction Mechanism of reaction and catalytic activity Induction coefficient and sensitivity Qua I it at ive a na I ysis Quantitative analysis Fixed time methods Fixed signal methods Initial slope methods Open system methods Other catalytic methods Differential kinetic methods Masking and derivatisation Separation methods Selectivity Determination of metal ions Other applications References Keywords Review; iodine - azide reaction; kinetic methods Introduction In acidic or neutral media a solution of sodium azide and iodine remains practically unchanged for a long time but in the presence of divalent sulphur compounds the irreversible redox reaction 2N3- + I2 -+ 21- + 3N2 proceeds rapidly giving rise to bleaching of the solution and the evolution of nitrogen bubbles.This reaction was described by Raschig in 1904,’ and has since then been the object of numerous studies. In addition, many analytical procedures have been developed that allow the sensitive and selective identification and determination of the compounds that catalyse the reaction. In this paper the analytical applications of the iodine -azide reaction are reviewed and the diverse significant aspects involved are critically examined together with the methods that have been used to follow the reaction and the analytical characteristics of the procedures proposed.Mechanism of Reaction and Catalytic Activity The iodine - azide reaction is catalysed by free sulphide and diverse metal sulphides by thiocyanate thiosulphate and carbon disulphide and by thiols disulphides and thioketones, among other organic compounds that contain sulphide sulphur. In contrast the reaction is not catalysed by sulphite, sulphate and organic sulphoxy compounds.2 Crystalline or coagulated elemental sulphur does not produce the catalysis, but it does catalyse the reaction as a finely dispersed su~pension.3~4 Hydrogen selenide does not show any catalytic activity ,5 but some selenium compounds such as N-benzoylsele-noureas Ph-CO-NH-CSe-R where R = piperidino or morpholino catalyse the reaction their catalytic activity being lower than that exhibited by the analogous sulphur compounds.6 However the catalytic activity of compounds such as Na2Se(S203)2.3H20 and Na2Te(S203)2.2H20 seems to be due almost exclusively to the presence of thiosulphate ions.7 Certain charcoals and carbon blacks also show catalytic activity due to the presence of active C02 complexes in their composition .g The mechanism of the reaction has been studied by several authors.9-17 Dahl and Pardue,l6 studying diverse disulphides and mercaptans established a reaction pathway that explains the behaviour of many catalysts. In the mechanism suggested, the reaction begins with the attack of the 12N3- complex by a disulphide and consists of six reaction steps: 12N3- + RSSR F== RSI + RSN3 + I- .. . . (a) . . RSI + N3- e RSN3 + I- . . . . . . . . ( b ) RSN3 + N3- e 3N2 + RS- . . . . . . . . (c) RS- + I2N3- e RSI + I- + N3- . . . . . . (d) RSN3 + RS- e RSSR + N3- . . . . . . . . U, RSI + RS- e RSSR + I- . . . . . . ( e ) Steps (a) (e) and u> represent themselves a catalytic cycle, whereas steps (b) ( c ) and (d) are an inner cycle. The reaction pathway for sulphydryl compounds is the same but step (d) would be the starting point. In this instance an initial rapid period is observed due to the high concentration of the RS-species. The reaction slows down when this species is completely oxidised to disulphide as step (a) where the reactive intermediates are regenerated is the principal rate-determining step.16 Many other catalysts such as inor-ganic sulphides thioureas and dithiocarbamates exhibit a similar behaviour.lg20 After some time the catalysed reac-tion can stop completely owing to further oxidation of the sulphur to inactive species.The following order of decreasing catalytic activity has been o bserved:21 RSH > RR’C=S > R-S-S-R‘ > H which agrees with the proposed mechanism. The differences in activity are related to the difficulty of cleaving the differen 1002 ANALYST SEPTEMBER 1986 VOL. 111 bonds to form the RS- species. Thus the unexpected high catalytic activity exhibited by lipoic acid: may be explained by the strain existing on the five-membered ring. 16 A study performed with different dithio acids of the form RR'PS2H shows that the catalytic activity decreases when steric hindrance exists on the sulphur atom.22 The same effect is observed when the active sulphur can form intramolecular hydrogen bonds.16923 Thus for example cystamine which differs from cystine only by its lack of a carboxyl group on each extreme of the molecule exhibits a catalytic activity 2.05 times larger. 16 On the other hand electrophilic groups enhance the catalytic activity which may be attributed to the greater susceptibility of the disulphide bond to nucleophilic attack. l6 Thus dithiodiglycollic acid which differs from cystine in its lack of an amino group on each side of the disulphide bond and in the fact that carboxyl groups are on the a-carbons instead of on the @-carbons shows a catalytic activity 3.2 times larger than cystine.The increase in activity may be explained by the greater difficulty in forming intramolecular hydrogen bonds and by the closer proximity of the electrophilic carboxyl groups to the disulphide bond.16 Miiller et al.23 established correlations between the struc-ture of several N-benzoylthiourea derivatives (RR'N-CS-NH-CO-Ph) and their catalytic activity. Disubstituted com-pounds are generally ten times as active as monosubstituted compounds and among the latter compounds with electron-withdrawing groups are more active than those with electron-donating groups. Induction Coefficient and Sensitivity Because of the oxidation side reactions of the catalysts it is convenient to distinguish between catalytic activity and the induction coefficient.The catalytic activity is a value propor-tional to the rate constant of the catalysed reaction and to its initial rate whereas the induction coefficient24 or reactivity number25 has been defined as moles of iodine consumed per mole of active sulphur initially present. Therefore it depends not only on the activity of the catalyst but also on its resistance to oxidation and on reaction time. Relative catalytic activities are obtained from the measure-ments of initial reaction rates whereas the induction coeffi-cient is calculated from the extent of the catalysed reaction after a certain time period.26 Not always enough attention has been paid to this distinction ambiguous expressions such as effectiveness catalytic effect or reactivity being used.21927Jg The induction coefficient of a substance is a measure of the sensitivity that can be reached in its determination.24.29 Its value depends on the relative rates of the different competitive reactions involved and therefore it changes considerably with the experimental conditions.The induction coefficient usually increases with increasing azide and iodine concentra-tions19.30; however it can decrease when the concentration of iodine exceeds a certain value as observed for sulphide,31,32 and dithiocarbarnates.19727 An increasing iodide concentration usually causes an increase of the induction coefficient as observed for cysteine30 and free sulphide ,31-33 although for metal sulphides20 and thiocyanate ,34 a diminuition of the coefficient is produced above a certain iodide concentration.Optimum conditions are not always utilised. Thus for example moderate concentrations of azide and iodide much lower than the optimum are often used in order to achieve cheaper determinations.19,2OV*4 The induction coefficient usually decreases rapidly above pH 8 owing to the dismutation of iodine. It is inconvenient to work below pH 5 owing to the volatility and toxicity of hydrazoic acid (log KH = 4.7 at 25 "C boiling-point 37 "C).35 In the pH range 5-8 the reaction may be independent of pH as occurs with ZnS20 or it may show a more or less marked dependence as observed for substituted thioureas.28J6 Some compounds show a maximum and a minimum near pH 5-6 and 7-8 respe~tively.16~3~ Most procedures recommend a pH in the range 5.5-6.5 where the azide - hydrazoic system shows some buffer capacity.The induction coefficients may change considerably with the order of addition and even with the rate of mixing of the reagents.l9Jg This could be due to different local iodine concentrations during the process. Thus for cysteine a maximum value is obtained when the order of mixing is I-, N3- HC1 cysteine and 12 being only half as much if cysteine is added last.39 Dependences on the dielectric constant, visc0sity2~ and total saline content40 have also been reported. The induction coefficients of a variety of substances in different experimental conditions have been established.41.42 Most of the catalysts have values below 600 but in some instances they are much higher.Thus thioammeline (4,6-diamino-l,3,5-triazine-2-thiol) shows a value of 4400 in a 2% azide solution.40 Qualitative Analysis The identification of diverse organic and inorganic sulphur compounds making use of the catalysis of the iodine - azide reaction was suggested by Feigl and Anger.43.44 The test may be performed either on filter-paper impregnated with the iodine - azide reagent or on a watch-glass where the formation of nitrogen bubbles is also observed together with the loss of iodine colour. The limits of detection are in the range 0.02-0.5 pg for sulphide thiosulphate and carbon disulphide being about 2 pg for thiocyanate.2.35 Extremely low values have been obtained for some organic compounds such as thioacetic acid (0.3 ng) rhodanine (3 ng) and thiourea (5 ng).M Some qualitative uses are the detection of mercaptans in bacterial cultures45 and sulphur compounds in urine which is used for the diagnosis of intoxication by dithiocarbamate fungicides46 and by accelerators of rubber polymeri~ation.~7 A method has been proposed to differentiate between animal and vegetable fibres ( e .g . wool and cotton). After melting with sodium or potassium only animal fibres give positive results.48 Finally an iodine - azide solution is employed to develop sulphur compounds in column adsorption49 and paper chromatography.50 Quantitative Analysis The reaction has been widely used to determine small amounts of sulphur compounds and has also been applied to the determination of metal ions that form stable complexes with sulphur-containing ligands.The diverse procedures proposed for the determination of non-metallic inorganic and organic compounds are examined in this section and their detection limits and other characteristics are shown in Table 1. The analytical applications of the reaction are determined by the oxidation side reactions of the catalysts. In only a few instances under restricted conditions and during short periods may it be considered that the concentration of the catalyst is constant and that the reaction proceeds according to definite simple kinetics. For this reason most of the pro-cedures described have been developed on a purely empirical basis without establishing the corresponding kinetic -equa-tions.In most of these procedures an excess of iodine is rapidly added to the stirred mixture of azide and catalyst, some control over addition and stirring rates24 and tempera-ture being necessary. Many analytical techniques based on the measurement of unreacted iodine evolved nitrogen or heat produced have been used to monitor or establish the advance of the reaction ANALYST SEPTEMBER 1986 VOL. 111 1003 Table 1. Determination of non-metallic inorganic and organic compounds Substance determined Method Characteristics of the procedures samples and references s2- . . . . . . . . . . Titration with arsenite after a After 15 min of reaction determination of 0.03-3 p.p.m. and 3-300 fixed time. Visual end-point detection in the presence of starch p.p.b.using 0.02 ~10.01 M and 5 m ~ / 2 . 5 mM iodine - azide solutions respectively; CV* < 6%. Previous separation as H,S by Titration with arsenite after a fixed time with amperometric end-point detection Gasometry Spectrophotometric determina-tion of iodine in the presence of starch after a fixed time Turbidimetry of nitrogen and enthalpimetry Appearance of fluorescence up to Direct injection enthalpimetry a fixed value Potentiometric sensor with two platinum microelectrodes Coulometric generation of iodine Addition of iodine in the presence of ascorbic acid. Biampero-metric end-point detection FIA method with biamperometric determination of iodine Elemental sulphur . . . . Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch As above As above with amperometric end-point detection Effects derived from nitrogen bubble formation (see text) Turbidimetry of nitrogen and enthalpimetry Appearance of fluorescence up to a fixed value Coulometric generation of iodine FIA method with biamperometric determination of iodine s4062- .. . . . . . Gasometry SCN- . . . . . . . . . . Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch point detection As above with amperometric end-Gasometry Spectrophotometric determina-tion of iodine at 350 nm after fixed time Appearance of fluorescence up to a fixed value Coulometric generation of iodine. Biamperometric end-point detection Amperometric determination of iodine Turbidimetry of nitrogen and en t halpime t ry sweeping with nitrogen; volatile mercaptans interfere.51 Similar -procedures applied to industrial waste waters sewage52 and steel31 with LODt 0.5 p.p.b.; air (LOD 1.5 pg)53-54 and bile55 Determination of 0.2-200 p.p.m.56 Determination of 1-200 pg.Water and water - organic solvent After 10 min of reaction. Applied to Ti and Zr.58 Similar procedure mi~tures38.5~ for steel32 Determination of 300-3OOO p.p.m. in 100 p1 and 0.1-100 p.p.m. if H2S is recovered from a nitrogen stream. Applied to the measurement of bacterial activity of E. coli on cysteine59 indicator60 CV < 3%. Previous separation as H2S by sweeping with nitrogen; applied to commercial copper with a CV < 5%33 Determination of 8-1000 ng CV 10-2.7% LOD 5 ng.For H2S in a gas stream; considerable amounts of SO2 HCHO CS2 and methylmercaptan can be tolerated61 Determination of 0.5-1 p.p.m. in 1 ml. Rhodamine B recommended as After 2 rnin of reaction determination of 1.2-32 p.p.m. in 0.5 ml, Determination of 4-40 p.p.b. in 25 ml CV2%62 Determination of 0.3-1.6 p.p.m. in 50 p1 CV 2.5% for 1.3 p.p.m., LOD 10 ng63 LOD 0.2 p.p.m. in 10 p164 Determination of 0.1-3.5 p.p.m. in 50 ml CV < 0.24%. Previous extraction in a miscible organic solvent. Applied to metallic suIphides,65,66 metals and alloys67 with CV < 4% ointments,4 rubber and polymers with CV < 6%6+70 Determination of 5-80 pg in 1-10 ml of extract CV < 4.9% in the range 6.6-31.8 pg. Previous extraction in DMF and oxidation to S2O32- with nitrite.Applied to carbonates71 Determination of 2 pg-0.3 g CV <2.5%. Thiocyanate is masked with Determination of 2-800 p.p.m.56 KI34,72 Determination of 4-40 p.p.m. in 12 m173 Determination of 320-3240 p.p.m. in 100 pl (batch method) and Determination of 0.5-3.5 p.p.m. in 1 ml. Rhodamine B recommended Determination of 20-280 p.p.b. in 25 ml CV < 2% ,62 112-1120p.p.m. (flow method)59 as indicator" LOD 0.1 p.p.m. in 10 p164 LOD 0.01 pg CV 2-3% .74 Similar procedure for mixtures with thiosulphate previous separation by paper chromatography75 After 30 s of reaction determination of 2.5 p.p.b.-70 p.p.m. CV 2.5 Yo 76 Determination of 2-3500 p. p. m .56 - 57,77,78 After 3 rnin of reaction determination of 5-80 p.p.b.79 Determination of 0.2-6 p.p.m.Rhodamine B recommended as Determination of 20-120 p.p.b. CV < 3%. In mixtures with thiouream indicator" Determination of 10-100 ng81 Determination of 0.7-7 g 1- in 100 pI5 1004 Table l-continued Substance determined (CH2SCN)* and PhCHzSCN . I 3-Butenyl isothiocyanate cs2 . . I . . . . . Cysteine . . . . . . Glutathione . . . . . . Ergothioneine . . . . 2,6-Thio-4-pyrimidine . . 2-Mercaptobenzimidazole 6Mercaptopurine, 2-thiouracil and 2-mercaptopyrimidine 2- 6- and 8-mercaptopurine Thioammeline . . . . BismuthiolI . . . . . . . . . , . . . . . . . . . . . . . . . . Method Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch As above As above As above Fixed signal method.Bleaching time is measured Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch As above Gasometric measurement after a fixed time Spectrophotometric determi-nation of iodine in the presence of starch after a fixed time Turbidimetry of nitrogen and enthalpimetry FIA method with biamperometric determination of iodine Coulometric generation of iodine Addition of iodine in the presence of ascorbic acid. Potentiometric end-point detection Titration with arsenite after a fixed time. Visual end-point detectiun in the presence of starch FIA method with biamperometric Coulometric generation of iodine determination of iodine Titration with arsenite after a fixed time.Visual end-point detection in the presence of starch Addition of an iodine - azide solution at a constant rate Spectrophotometric determina-tion of iodine after a fixed time FIA method with biamperometric determination of iodine Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch As above . Asabove ANALYST SEPTEMBER 1986 VOL. 111 Characteristics of the procedures samples and references Determination of 2-20 p.p.m. and 4-40p.p.m. in 1 ml respectively. Applied to contaminated air53.54 After 30 s of reaction. Microdetermination in a half rape seed, After 30 min of reaction determination of 20-120 p.p.m. in 5 ml. Determination of mixtures with ethyl xanthate in sewage83384 After 30 rnin of reaction determination of 0.05-1 pg CV 5 % .Industrial wastes.85 Similar procedure for air53,54 Bleaching times are in the range 6-100 min. Determination of previous synthesis of the corresponding thiourea with methylamine82 0.005-1.5% CS2 in 1 ml of organic solvent. Mercaptans and H2S interfere.86 Similar procedures for 7.5 X 10-4-4 X in 1 ml of benzene and for 0.1-2.25% in 1 ml of benzene or chl0roform5~ After 4 h of reaction determination of 0.4-2.4 p.p.m. (with 1 g of NaN at pH 6) and 0.1-0.8 p.p.m. (with 2.5 g of NaN at pH 5.2) in 50 ml CV < 5%. In mixtures with glutathione29 and erg0thioneine.2’3~~ In erythrocytes previous separation by gel chromatography5 After 20-30 s of reaction determination of 13 p.p.b.-2.4 p.p.m. in 75 ml CV 4%.Cystine and methionine up to 0.8 and 0.2 mg, respectively can be tolerated. Applied to milk beer and wheat.88 Similar procedure in ref. 30. After 10-60 rnin of reaction determination of 24-120p.p.m. in 2 ml, LOD 8.6 p.p.b. in 7 mP9 CV 8%15 Determination of 350-3500 p.p.m. in 100 $59 LOD 0.2 p.p.m. in 10 p1@ Determination of 2-16 p.p.b. in 25 ml CV <2%. In albuminw CV 4.6% for 28 ng of S as cysteine2, After 5 min of reaction determination of 2-12 p.p.m. (with 1 g of NaN at pH 6) and 1-5 p.p.m. (with 2.5 g of NaN3 at pH 5.2) in 50 ml CV < 5%. In mixtures with erg0thioneine,8~ and cysteine and ergothioneine.29 In erythrocytes previous separation by gel chromatography.5 Similar procedure after 15-20 s of reaction in 0.5-3 ml of lemon and orange juices91 LOD 0.2 p.p.m.in 10 ~1~ Determination of 40-240 p.p.m. in 25 ml CV < 2%w After 5 rnin of reaction determination of 0.02-8 p.p.m. in 50 ml, CV 3%. Cysteine and glutathione are blocked in 5 min with N-ethylmaleimide.29.87 In erythrocytes previous separation by gel chromatography5 Determination of 10-100 p.p.m. in 10 ml LOD 2 p.p.m.92 Determination of 50-200 p.p.b. CV 1.9-5.6%. In zinc electroplating baths. Zn2+ is masked with EDTAg3 LOD 0.1 p.p.m. in 10 ~ 1 6 4 Determination of 3-12,l-12 and 5-80 pg of the three compounds, with CV 2.5,2.0 and 2.0% respectively” Determination of 1CL150 p.p.b. in 50 ml CV2%. Other mercaptans are blocked with N-ethylmaleimide and Cuz+ is masked with oxalic acid95 Determination of 0.06-1 p.p.m. in 50 m ANALYST SEPTEMBER 1986 VOL.111 Table l-continued 1005 Substance determined Methionine . . . . . . . . As above Ethionine . . . . . . . . As above Sulphathiazole . . . . . . Asabove Penicillin G . . . . . . . . As above Thiamine . . . . . . . Promazine chloropromazine andpromethazine . . , Cystine Enerbol Lipoic acid Thiourea Method As above with biamperometric end-point detection As above visual end-point detection in the presence of starch As above Gasome try Spectrophotometric determi-nation of iodine in chloroform extracts at 525 nm after a fixed time. Also gasometry Fixed signal with potentiometric monitoring of iodine Initial slope method with potentiometric monitoring of iodine enthalpimetry Turbidimetry of nitrogen and Titration with arsenite after a fixed time.Visual end-point detection in the presence of starch As above As above Gasometry Addition of an iodine - azide solution at a constant rate Spectrophotometric monitoring of iodine in the presence of starch FIA method with biamperornetric determination of iodine Coulometric generation of iodine. Biamperometric end-point detection Turbidimetry of nitrogen and enthalpimetry Effects derived from nitrogen bubbles formation (see text) Characteristics of the procedures samples and references After 30 min of reaction determination of 7-230 p.p.m. in 50 ml, CV 3%. Cystine up to 20 p.p.m. can be tolerated97398 Determination of 10-700 p.p.rn. in 50 rnl CV 1-3%oY9 After 2 h of reaction determination of 0.3-10 p.p.m.in 50 ml, CV ~ 2 % . Drugslm Determination of 0.2-6 p.p.m. of S in 5 ml equivalent to 10-350 pg of potassium salt or 15-500 pg of procaine salt CV 5% for 1 p.p.m. S. The compound is previously hydrolysed in basic medium. Streptomycin and aureomycin do not interfere1o1 After 10 min of reaction determination of 0.15-2.2 g 1-1 in 10 ml, CV 2% for 1 g 1-1. In drugs ascorbic acid is titrated separately in absence of a ~ i d e 3 ~ After 1 h of reaction. Drugs102 After 30 min of reaction determination of 1-20 p.p.m. in 50 ml, CV 3%. Insulin hair wool and beer.103 Similar procedure applied to beer; cysteine up to 1.6 p.p.m- can be determined in the same sample; methionine up to 2.7 p.p.rn. is tolerated.88 Similarly after 2 h of reaction 0.05-5 p.p.m.(CV 2-3%0) can be determined in 100 ml The procedure is applied to insulin wool and hair104 LOD 1 ng c v 2 . 5 % 0 ~ ~ After 30-60 min determination of 36-144 p.p.m. in 0.5 ml confidence limits f 6 p.p.m. ; determination of 0.2 k 0.04 g cystine in samples with 1 g of protein nitrogen (99% confidence level). Applied to human serum protein hydrolysates14 Measured times are in the range 20-200 s; determination of 0.25-2 p.p.m. and5-25 p.p.m. in 2 ml CV 1 and 2% respectively. Alanine, leucine hystidine and glycine up to 500 p.p.m. and Ca2+ Zn*+, Co2+ and Cd2+ up to Measurements taken in less than 30 s determination of 0.25-2 p.p.m. and 2-25 p.p.m. in 1 ml CV 1 and 2%0 respectively106 M do not interfere105 Determination of 1.5-15 g I-' in 100 p159 CV 1 % in drugs107 Determination of 10-700 pg CV 1.5%.lo8 Similar procedure applied to urine4 Determination of 10-30 p.p.m. in 2 ml. Acid copper electroplating Determination of 2-200 pg in water and water - organic solvent Determination of 7-80 p.p.m. in 10 ml LOD 2 p.p.m. ,92CV5%110 baths.109 Similar procedure for 5-150 pg in citric fruits4 mixtures38 Determination of 10-100 p.p.b.111 LOD 0.2 p.p.m. in 10 p P CV < 3% for 16 p.p.b. In mixtures with thiocyanateso Determination of 0.38-3.8 g 1-1 in 100 pP9 Determination of 1-10 p.p.m. in 12 m17 1006 Table l-continued Substance determined Thiourea and substituted thioureas . . . . . . ANALYST SEPTEMBER 1986 VOL. 111 Method Characteristics of the procedures samples and references Titration with arsenite after a fixed time.Visual end-point detection in the presence of starch Spec tropho tometric de termina-tion of iodine in the presence of starch after a fixed time Direct injection enthalpimetry Thioureas substituted thioureas and tetrarnethylthiuram sulphide Biamperostatic method 2-Thiobarbituric acid . . 2-Thiobarbituric acid and derivatives . . . . Rubeanic acid and derivatives . . . . Dithiocarbamates . . . . Ethyl xanthate Merthiolate . , h i d e . . . . . . . . . . . . . . . . . . . . * Coefficient of variation. t Limit of detection. After 30 s of reaction determination of 2-300 pg of S . 1 l 2 Mixture of thioureas previous separation by paper chromatography5O LOD 3 p.p.b. thiourea and 7 p.p.b. phenylthiourea in 7 m P After 30 s of reaction determination of 0.05-1 pmol in 100 p1, CV < 3% .18 Similar procedure for 50-500 pmol in 5 pl LOD 5 pmol CV < 2%; applied to the determination of mixtures of thioureas in the presence of mercaptans isothiocyanates, thiosulphate and sulphide after separation by TLC CV 6% referred to the sample36 Determination of 2.6-26 nmol thiourea and phenylthiourea 69-624 nmol benzoylthiourea and 0.6-6.1 nmol tetramethylthiuram sulphide in 5 ml CV 2.5% for 13 nmol thiourea25 FIA method with biamperometric LOD 0.1 p.p.m.in 10 1.1164 determination of iodine Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch Determination of 0.02-7 p.p.m. in 100 ml previous separation by paper chromatography113 As above LODO.l-0.2p.p.m.in50m1 CV3-6Y0114 As above After 2 min of reaction determination of 0.1-3.2 p.p.m. sodium diethyldithiocarbamate in 50 ml CV < 3% ; Fe3+ and AP+ can be masked with fluoride.24 Similar procedures for other derivatives,1*5J16 and mixtures of N-monoalkyl derivatives, previous separation by high-voltage electrophoresis1~7 in water and water - organic solvent mixtures38 CV 2% at a 30 p~ levelly Gasome try Direct injection enthalpimetry Determination of 0.5-100 pg sodium tetramethylenedithiocarbamate After 10 s of reaction determination in the 5-65 p~ range in 5 ml, Titration with arsenite after a fixed time. Visual end-point detection in the presence of starch After 15 s of reaction determination of 20-55 pg of S.In mixtures with CS2 in sewage83 Spectrophotometric determina- LOD 1 p.p.m. in 7 mlgy tion of iodine in the presence of starch after a fixed time Back titration with thiosulphate Determination of 0.1-700 mg of N3-.118A 15-20 min reaction time after total consumption of azide in the presence of thiocyanate is recommended119 Fixed Time Methods The selection of an optimum reaction time implies a compro-mise between the current rate and extent of the reaction. When the rate slows down excessively a longer time of reaction does not yield any valuable gain in sensitivity. The most suitable reaction time can be easily established when a signal - time graph is recorded. The extent of the reaction is obtained either by titrimetry of the remaining iodine with sodium arsenite3.56,65,71,98 o r hydrazine sulphate,51 or by a spectrophotornetri~,79~93J~~J~~ enthalpimetri~19933~36 or gasometric method.57.77.A flow injection analysis method (FIA) has been described, in which 10-pl samples are injected into an iodine and azide solution stream. Measurement of the concentration of iodine is carried out biamperometrically .64 In enthalpimetric methods the temperature rise produced by the reaction is proportional t o the amount of iodine and azide consumed and therefore t o the amount of catalyst. These methods are very sensitive as the iodine - azide reaction is highly exothermic. Methods based on the measurement of released nitrogen are less sensitive although together with enthalpimetric methods they offer the advantage over other instrumental methods of not being affected by turbidity or precipitates in the sample ANALYST SEPTEMBER 1986 VOL.111 1007 Fixed Signal Methods Pardue and Shepherd105 proposed a fixed signal method for the determination of cystine. A potentiometric concentration cell sensitive to iodine is used and the time required for the cell voltage to reach a given value is measured. The concentrations of azide and iodide are large compared to that of iodine and remain essentially constant. Under these conditions and at cystine concentrations below 2 p.p.m. the rate of the reaction is proportional to the instantaneous concentrations of cystine C and iodine: where kl is a rate constant. When these pseudo-first-order kinetics are obeyed the time interval At required for the voltage interval AE to be overcome is inversely proportional to cystine concentration which may be expressed by AE 1 kkl At .. . . . . ' * (2) c=- X-where k is a temperature-dependent constant from the Nernst equation. However above 2 p.p.m. of cystine the reaction order n with respect to cystine increases gradually leading to positive errors. Equation (1) should be rewritten in the form -- d[121r - kl C" [I2It . . . . . . . . dt where n b 1. Equation (2) then becomes (3) (4) which states that C is proportional to the nth root of the reciprocal of the time interval. Under the conditions used and for the range 5-25 p.p.m. when the average value used is n = 1.24 the errors are within 2%. A method has been proposed for the determination of CS2 based on the measurement of the time needed for iodine to disappear which may be carried out by visual observation.86 Rhodamine B has been used as a fluorescent indicator in the determination of sulphide thiosulphate and thiocyanate.The fluorescence of the indicator quenched by iodine appears gradually as the reaction proceeds. The reciprocal of the time required for the fluorescence to reach a certain intensity is proportional to the catalyst concentration .60 Initial Slope Methods Pardue106 proposed an initial slope method to determine cystine. The initial rate of the reaction is measured by the decrease in iodine concentration which is monitored poten-tiometrically. A linear response with a slope proportional to the cystine concentration is obtained.The relationship is given by The method suffers from the same drawbacks of the corre-sponding fixed signal method,lOs leading to high values above 2 p.p.m. of cystine. Open System Methods Some procedures have been described in which an iodine -azide solution is added at a constant rate to the sample solution. When the catalyst has been completely destroyed an increase in the iodine concentration is observed. The volume of reagent added is proportional to the initial concentration of the catalyst.92.110,122.123 The iodine concentration must be kept low and constant in order to achieve reproducible results. Therefore the addition and stirring rates must be carefully controlled. A precise and convenient control of the concentration of iodine may be attained by using competitive reactions, coulometric iodine generation or stat methods.Muller et al.23 proposed the addition of iodine in the presence of a given amount of ascorbic acid which competes with the catalysed reaction giving rise to a larger consumption of iodine. The time required for iodine to appear in the solution is propor-tional to the initial concentration of the catalyst. A similar method for microsamples has been described.63 Jedrzejewski and Ciesielski62~80~90 used the anodic gener-ation of iodine. The end-point was detected biamperometric-ally. In the biamperostatic method proposed by Pantel,25 the iodine concentration is measured continuously and a biamper-ometric signal is used to control the rate of addition from an automatic burette.In this way the iodine concentration is kept constant at a value below 0.2 mM. The determinations are accomplished in a short time. Other Catalytic Met hods Weisz and Meiners73 described an unconventional method, where drops of a chloroformic iodine solution are introduced with a capillary into a long and narrow vertical glass tube containing the mixture of azide and catalyst. The nitrogen produced in the catalytic reaction is retained on the falling drops and their downward movement gradually stops. The depth of fall and the time needed for the drops to return to the upper end of the tube are non-linearly related to the concentration of the catalyst. In the turbidimetric method described by Weisz et al. ,59 the measurements are carried out in a 3 + 1 glycerol - water medium which regularises the formation of nitrogen bubbles.During a given time and owing to light scattering the absorbance increases and then decreases giving rise to a maximum. The height of the maximum and the time required to reach it are both a measure of the concentration of the catalyst although the relationships are not linear. The ratio of the two parameters gives more precise results than either used separately. The authors employed this method and direct injection enthalpimetry simultaneously to determine several catalysts in batch and flow-through systems (double indi-cation). Enthalpimetric detection provides a very precise indication of the zero time point. A sensor for determining hydrogen sulphide in a gas stream has been designed.61 The end of the sensor is a sintered-glass ball which exudes an iodine - azide solution at a constant flow-rate.Hydrogen sulphide transported by the carrier gas catalyses the reaction giving rise to a potential difference measured by two platinum electrodes one inside and the other outside the ball. Selectivity The selectivity of most of the procedures is given by the selectivity of the iodine - azide reaction itself. Iodine and iodide oxidants and iodine reductants interfere. However, these interferences may usually be overcome by means of blank determinations in the absence of azide.37.52 Inter-ferences due to cations that precipitate or complex iodide or the catalyst may be avoided in some instances with masking agents.24352.95 The amount of iodine consumed by the oxi-dation of the catalyst is always negligible.Procedures allowing the selective determination of different catalysts found together in the sample make use of differen-tial kinetic methods of analysis other selective reactions to block or destroy certain catalysts and previous separation 1008 ANALYST SEPTEMBER 1986 VOL. 111 Differential Kinetic Methods Differential kinetic methods can only be applied in some favourable instances when the behaviour of the catalysts is sufficiently different. A procedure for the simultaneous determination of xanthates and CS2 in sewage based on the difference in their induction coefficients has been described. The extent of the reaction 15 s and 3 min after the introduc-tion of iodine is measured in parallel experiments.The amount of reacted iodine is proportional to the initial concentration of xanthates and to the sum of xanthates and CS2 respectively.83 Mixtures of cysteine and cystine can be resolved by a similar procedure. Cysteine is determined 20-30 s after the beginning of the reaction whereas the sum of both compounds is obtained after 30 min. The procedure was applied to their determination in milk beer and wheat.88 Jedrzejewski and Ciesielskiso applied their coulometric titration method to the selective determination of mixtures of thiourea and thiocyanate. The method differs from those cited above because the concentration of the generated iodine is controlled instead of the induction time. Thiourea catalyses the reaction at a lower iodine concentration and at a greater rate than thiocyanate.Finally the biamperometric sensor of Kiba and Furusawa61 allows the determination of hydrogen sulphide in the presence of substantial amounts of carbon disulphide and methylmer-captan owing to their different catalytic activities and the transient nature of the signal. Carbon disulphide and methyl-mercaptan react more slowly so that a 100-fold amount with respect to hydrogen sulphide merely broadens the peaks, leaving their heights unaffected. Masking and Derivatisation The resolution of mixtures of diverse sulphydryl compounds has been performed making use of their reactions with a$-unsaturated carbonyl compounds such as N-ethylmale-imide and with aldehydes. Mercaptans are added to the N-ethylmaleimide double bond giving thioethers which have very low catalytic activities : O G O + R’SH - 0 xo I R N-Ethylmaleimide blocks cysteine and glutathione in less than 5 min whereas no reaction is observed with ergo-thioneine even after a few hours.Analytical procedures based on these differences have been described for determin-ing mixtures of the two former compounds with the latter.g7 Similarly thioammeline can be determined in the presence of other sulphydryl compounds.95 Aldehydes also block sulphydryl groups giving thioacetals: Mixtures of cysteine glutathione and ergothioneine have been resolved making use of the different rates of reaction with formaldehyde.29 The percentages of blocked compounds in a 0.3% formaldehyde solution at pH 8-9.5 and after 30 s reaction time are 92,5 and 27% respectively the reproducibi-lity of the blocking percentages being within 2%.To determine mixtures of cysteine (x) and glutathione (y) a first experiment is carried out after 4 h of reaction time giving rise to a result A proportional to the sum of both substances ( A = C + Cy). In a second experiment the sample reacts with formaldehyde during the 30 s before the introduction of azide. RCHO + 2 R ’ S H j RCH(SR’)2 + H20 The second result obtained B is related to the concentration of the compounds according to B = 0.O8Cx + O.95Cy The concentrations are found by solving the equation system. Similarly a series of three parallel experiments using both reagents formaldehyde and N-ethylmaleimide permits the resolution of mixtures of cysteine glutathione and ergo-thioneine .29 Sulphite although it does not catalyse the reaction reduces iodine and therefore interferes with the determination of sulphide.This interference may be avoided in a similar way by blocking sulphite with formaldehyde. 121 The selective determination of isothiocyanates in the presence of other catalysts of the iodine - azide reaction may be performed by treating the sample with an amine. The corresponding thiourea which is a much more active catalyst, is formed. The micro-determination of 3-butenyl isothio-cyanate in a half rape seed has been carried out in this way, after prior extraction of the compound with ethanol. An aliquot of the extract is treated with methylamine to give N-methyl-N’-(3-butenyl)thiourea the induction coefficient of which is much greater than the coefficients of the other sulphur-containing compounds present in the seed.A blank determination is also made with an aliquot of the extract not treated with me thylamine .82 Mixtures of thiocyanate with sulphide or thiosulphate have been resolved making use of the inhibitory action of iodide on the activity of thiocyanate.34 Separation Methods Among the compounds that exert a catalytic action only H2S, CS2 and lower alkyl mercaptans volatilise at room tempera-ture. The determination of sulphide can be performed by sweeping H2S with a nitrogen stream and absorbing it in dilute NaOH33>5* or in a zinc(I1) or cadmium(I1) solution.51 It is, however faster and simpler to absorb it directly in the iodine -azide solution.54 Interference from thiosulphate and thiocyanate in the identification of the sulphide ion is overcome by precipitation of sulphide with zinc or cadmium carbonate.2 Similarly, sulphide and thiosulphate are eliminated by precipitation with HgC12 for the identification of thiocyanate.35 Small amounts of elemental sulphur have been determined in metals and alloys ,67 metal sulphides,65,66 vulcanised rub-ber,68@ sulphur-containing polymers68370 and ointments,4 after extraction with an organic solvent.Elemental sulphur must be finely dispersed to act as a catalyst which is achieved by using a miscible organic solvent such as dimethylformam-ide and adding an aliquot of the extract into the iodine - azide aqueous solution .3,65770 Thin-layer chromatography has been used to resolve mixtures of substituted thioureas.36 The separation is per-formed on silica gel layers with dioxane - benzene - acetic acid (90 + 80 + 1) as the eluent.The spots are located by spraying chromatograms of standard solutions with an iodine - azide solution. The corresponding areas on the chromatogram of the sample solution are scraped off and suspended in an azide solution and the components are determined by direct injection enthalpimetry . Substituted thioureas50 and mixtures of thiosulphate and tetrathionate75 have been determined after separation by paper chromatography and high-voltage electrophoresis in a pH 9.2 borax buffer has been used for substituted dithiocarba-mates.117 The determination of cysteine glutathione and ergo-thioneine in haemolysate of erythrocytes has been performed after separation in a Sephadex G-10 column.By elution with a pH 6.8 phosphate buffer two fractions are obtained the first corresponding to the mixture of cysteine and glutathione ANALYST SEPTEMBER 1986 VOL. 111 1009 Table 2. Determination of metal ions Element Hg(1) and Hg(I1) . . Co(I1) * Ni(I1) . Cu(I1) . Zn(I1) . . Pd(I1) . . Fe(II1) . . Rh(II1) . . Au(II1) . . Bi(II1) . . Ir(1V) . . Pt(1V) . . . . . . . . * . . . . . . . . . . . . . . I Ru(VII1) and Os(VII1) . I . . . . . . . . . . . . . . . . . . . . . . . . Ligand* Characteristics of the procedures samples and references Ethylenediamine- Determination of 0.5-8 pg115 dithiocarbamate 6-Mercaptopurine Determination of 1-30 pgl24 Pyrrolidinedithio- After 10 min of reaction determination of 0.5-15 pg, car bamate CV < 2.570 previous separation from Ni Mn Cu Fe and Zn by ion exchangel25 5-100 ml CV 5% (9'/0 at 10 p.p.b.level) Fe(II1) and AI(II1) can be masked with fluoride. 126 Applied to nickel chloride and drugs; Co(I1) is previously separated from Ni(I1) and Cu(I1) by ion exchange4" Diethyldithiocarbamate After 2 min of reaction determination of 10-240 p.p.b. in As above After 2 min of reaction determination of 28-280 p.p.b. in 5 ml CV < 8%. In margarine and drugs Ni(I1) is previously separated from Cu(I1) and Co(I1) by ion exchange40 After 10 s of reaction determination of 60-840 p.p.b.in 5 ml CV < 3%. In Zamack alloys previous separation by Determination of 1-30 pg124 As above? extraction with dimethylglyoxime19 6-Mercap topurine Diethyldithio- Determination of 6 X Cu in 1 g NaC1120J27 carbarnates Thioammeline Determination of 6-250 p.p.b. in 50 ml CV < 10%. In zinc salts without separation or pre-concentration of copperl28 Determination of 3-24 p,p.b. in 10 ml and 2-240 p.p.b. in 50 ml. In tap and distilled water commercial azide acetic and oxalic acids. 129 In drugs previous separation of Co(I1) and Ni(I1) by ion exchange4" As above 8-Mercaptopurine Thiopental Determination of 0.04-1.6 p.p.m. in 50 m1130 Determination of 20-200 p.p.b. in 50 ml previous separation of Zn(I1) by ion exchange131 As above Determination of 40-400 p.p.b.in 50 ml previous separation of Cu(I1) by ion exchange131 6-Mercaptopurine After 10 min of reaction determination of 0.01-3 p.p.m., CV 4% but 10% at ng level. Other Pt-group metals can be tolerated in the presence of masking agents132 As above Determination of 0.1-30 pg1Z4 2-Mercaptopurine Determination of 0.2-lOp.p.m. CV 3°/~133 As above 2- and 6-mercapto-purine Determination of 0.1-0.8 p. p.m. CV 5.1 YO 134 LOD 20 p.p. b. in 5 m1135 Bismuthiol Determination of 0.1-20 p.p.m. in 50 ml CV 2-10%; Sn(II), Pd(II) Cu(II) Fe(III) Pb(I1) and Zn(I1) interfere96 2-Mercaptopurine Determinationof 10-400p.p.b. CV8.2%133 6-Mercaptopurine Determinationof0.2-5 p.p.m. CV 1lYO136 As above Determination of 0.1-1 p.p.m. Ru(VII1) and0.02-1 p.p.m.Os(VII1) in 5 ml CV 6.1 and 6.4% respectively. Previous separation by volatilisation as Ru04 and 0s04137 2- and 6-mercapto- LOD 2 and 5 p.p.b. in 5 ml respective1yl35 purine * Except in the instances indicated the procedure involves the determination of the free ligand by titration of the unconsumed iodine with 1- Determination of the free ligand by direct injection enthalpimetry. $ Extraction of the complex displacement of the ligand and spectrophotometric determination of the unconsumed iodine after a fixed reaction arsenite after a fixed reaction time. time 1010 ANALYST SEPTEMBER 1986 VOL. 111 which in the column are oxidised to disulphides and the second corresponding to ergothioneine .5 The selective deter-mination of cysteine and glutathione is carried out as described above ,29 after reduction of the disulphides with H2Se.Determination of Metal Ions Metal complexes with sulphur-containing compounds usually exhibit a lower catalytic activity than do free ligands owing to the blocking of active sulphur atoms by coordination. More-over the catalytic activity of the complexes is often almost cancelled. This inhibition effect may be used in the indirect determination of metal ions following the extent of the iodine - azide reaction by either of the techniques described above. The characteristics of the proposed procedures are sum-marised in Table 2. The inhibition effect may also be used to establish the stoicheiometries and conditional stability constants of the As corresponds to methods based on an inhibitory effect, calibration graphs show negative slopes and the upper limit of the application range depends on the analytical concentration of the ligand and on the conditional stability constant of the complex.The smaller the constant is the shorter the calibration linear range will be.139 In several instances side-reactions of complex formation with azide are important and must be ~onsidered.~5J~~ Although most metal - azide complexes are weak the use of a large excess of azide can lead to a decrease of the conditional constant .I39 Metal ion - catalyst - azide equilibria are often attained slowly. Therefore it may be necessary first to mix these components and to wait some time before starting the catalysed reaction. In the determination of Co2+ 126 and Ni*+ 19140 with DDTC a 10-min waiting time has proved to be adequate.In addition to the methods based on the inhibitory effect a direct method for Rh(III) based on the catalytic activity of the Rh(II1) - 2-mercaptopurine complex has been described.133 A procedure for determining Cu(II) Pb(I1) and Cd(I1) on a different basis has also been proposed. The complexes of these metals with DDTC are extracted with chloroform the excess of ligand is eliminated by washing with NaOH and the extract is added to a methanolic iodine - azide solution. Metal -DDTC complexes are partially decomposed in the presence of an excess of azide there being free DDTC proportional to the amount of metal. The determination is performed by measur-ing spectrophotometrically the amount of unreacted iodine 5 min after the beginning of the catalysis.The sensitivity of the method is 50 times better than that of the usual spectropho-tometric method where a direct measurement of the Cu(I1) -DDTC absorbance is carried out. 120,127 The selectivity of the methods for determining metal ions depends on the selectivity of the sulphur compound as a ligand. When this is not enough interferences may be avoided with masking agents132 or by previous separation by volati-l i ~ a t i o n ~ ~ ~ e x t r a c t i ~ n l ~ J ~ ~ or ion-exchange chroma-tography. 40,131 complexes. 126,134,136,138-141 Other Applications The determination of azide has been performed by titration of the unconsumed iodine with thiosulphate after the total consumption of azide in the presence of an excess of some stable catalysts.118 Thiocyanate has been recommended for this purpose. 119 The iodine - azide indicator reaction has been proposed for end-point detection in titrations with sulphide in acidic media. The first drop of excess titrant causes the evolution of H2S which is transferred by a nitrogen stream into a vessel containing the iodine - azide reagent.142 The reaction has also been employed for the semi-quantitative measurement of the properties of diverse materials. Thus the induction coefficients of cyclic tetra-sulphides of primary amines have been used to establish cross-linking activity in elastomer vulcanisation An increase of the molecular weight of the substituents related to cross-linking activity leads to higher induction coefficients.143 The photographic activity of some sulphur compounds, which are used as additives in photography is related to their catalytic activity.233144 However methods based on the measurement of the extent of the reaction after a relatively large time period are not suitable as the oxidation side reactions of the catalysts lead to wrong conclusions. Muller et al.23 proposed the use of ascorbic acid to avoid the influence of these side-reactions. A corrosivity test has been described based on the observation of nitrogen evolution from surfaces impregnated with the iodine - azide reagent.145 Finally the reaction is useful for studying the stability of sulphur compound solutions of analytical interest in diverse media.24,146,147 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. References Raschig F. Chem. Ztg. 1908 32 1203. Burns D. T. Townshend A, and Carter A. H. in Chalmers, R. A. and Masson M. Editors “Inorganic reaction Chemi-stry,” Volume 2 Part B Ellis Horwood Chichester 1981. Kurzawa Z. andPuacz W. Chem. Anal. (Warsaw) 1977,22, 833. Puacz W. and Puacz J.,in Braetter P. and Schramel P., Editors “Trace Element Analysis in Chemistry Medicine and Biology Proceedings of the 1st International Workshop, 1980,” Walter de Gruyter Berlin p. 469. Kurzawa J. and Kurzawa Z. Chem. Anal. (Warsaw) 1974, 19 931. Muller H.; and Beyer L. Z. Chem. 1979 19,295. Hofman-Bang N.and Holten B. Acta Chem. Scand. 1950, 4 1390. Puri B. R. Sud V. K. and Kalra K. C. Indian J . Chem., 1972 10,76. Friedmann E. J. Prakt. Chem. 1936 146 179. Lovtrup S. C. R . Trav. Lab. Carlsberg Ser. Chim. 1949 27, 63. Hofman-Bang N. Acta Chem. Scand. 1950,4 456. Hofman-Bang N. Acta Chem. Scand. 1950,4 856. Hofman-Bang N. Actu Chem. Scand. 1950,4 1005. Strickland R. D. Mack P. A. and Childs W. A. Anal. Chem. 1960 32,430. Whitman D. W. and Whitney R. Mc. Anal. Chem. 1953, 25 1523. Dahl W. E. and Pardue H. L. Anal. Chem. 1965,37,1382. Beck M. T. and Bazsa G. M q y . Kem. Foly. 1971 77 76. Kiba N. Suto T. Furusawa M. and Takenchi T. in Chihara H. Editor “Proceedings of the 5th International Conference on Thermal Analysis Kyoto Japan 1977,” Heyden London 1977 p.38. Medina Hernandez M. J. Garcia Alvarez-Coque M. C., Ramis Ramos G. and Mongay Fernhndez C. Thermochim. Acta 1985 90 277. GozAlvez Rodriguez J. C. Licenciature Thesis Valencia, 1986. Awe W. and Naujoks E. Osterr. Apoth.-Ztg. 1952 6 534. Gabdullin M. G. Garifzyanov A. R. and Toropova V. F., Izv. Vyssh. Uchebn. Zaved. Khim. Khim. Tekhnol. 1984,27, 181. Muller H. Beyer L. Muller C. and Schroter C. Z. Anorg. Allg. Chem. 1978,446 216. Kurzawa Z. and Kubaszewski E. Chem. Anal. (Warsaw), 1974 19 263. Pantel S. Anal. Chim. Acta 1983 152 215. Bartz J . Dobak J. and Kurzawa Z. Zesz. Nuuk. Uniw. Imienia Adama Mickiewicza Poznaniu Mat. Fiz. Chem., 1967 151. Babko. A. K. Markova L. V. and Prikhod’ko M. U. Ukr. Khim. Zh. 1966 32 72 ANALYST SEPTEMBER 1986 VOL.111 101 1 28. 29. 30. 31. 32. 33. Suzuki S. Bunseki Kagaku 1962 11 299. Kurzawa Z. and Kurzawa J. Chem. Anal. (Warsaw) 1974, 19 755. Kurzawa Z. Chem. Anal. (Warsaw) 1961 6 813. Babko A. K. and Segeda A. S . Zh. Anal. Khim. 1968,23, 727. Babko A . K. and Markova L. V. Zavod. Lab. 1959 25, 1283. Kiba N. Nishijima M. and Furusawa M. Talanta 1980,27, 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. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73 * 74. 1090. Kurzawa Z. Chem. Anal. (Warsaw) 1960 5 741. Burns D. T. Townshend A . and Carter A. H. in Chalmers, R. A. and Masson R.Editors “Inorganic Reaction Chemistry,” Volume 2 Part A Ellis Horwood Chichester, 1981. Kiba N. Suto T . and Furusawa M. Talanta 1981,28 115. Perez Ruiz T. Martinez Lozano C. and Hernhndez Lozano, M. An. Quim. 1982,78B 241. Puacz W. Mikrochim. Acta 1981 2 155. Segeda A. S . Ukr. Khim. Zh. 1970 36 213. Kurzawa Z . and Kubaszewski E. Chem. Anal. (Warsaw), 1976 21 565. Bartz J. Makowski B. and Kurzawa Z. Zesz. Nauk. Uniw. Imienia Adama Mickiewicza Poznaniu Mat. Fiz. Chem., 1967 157. Zuk A. and Wejchan M. Poznan. Tow. Przyj. Nauk Pr. Kom. Nauk Podstawowych Stosow 1972,3,45. Feigl F. and Anger V. “Spot Tests in Inorganic Analysis,” Sixth Edition Elsevier Amsterdam 1972. Feigl F. and Anger V. “Spot Tests in Organic Analysis,” Elsevier Amsterdam 1966. Iverson W.P. J . Bacteriol. 1964 87 478. Villa P. Sciarra M. and Dell‘Uomo R. Zg. Mod. 1978 71, 623. Fournier E. Petit L. and Lecorsier A J . Eur. Toxicol., 1971 3 337. Vydra F. and Soukupova M. Chemie (Prague) 1952,8,219. LeRosen A. L. Monaghan P. H. Rivet C. A. Smith E. D., and Suter H. A. Anal. Chem. 1950 22 809. Karska B. and Balcerkiewicz L. Chem. Anal. (Warsaw), 1973 19 421. Kiba N. and Furusawa M. Talanta 1981,28 601. Kurzawa Z. Chem. Anal. (Warsaw) 1960 5 555. Gershkovich E. E. Novoe V . Oblasti Sanit. Khim. Anal., 1962 53. Gershkovich E . E. Tr. Nauchn. Sessii Leningr. Nauchn.-Issled. Inst. Gig. Tr. Profzabolev 1958 153. Hankiewicz J. and Suckel B. Polski Tyg. Lek. 1965 20, 959. Kurzawa Z. Chem. Anal. (Warsaw) 1959 4 649. Shiokawa T.and Suzuki S . J . Chem. SOC. Jpn. Pure Chem. Sect. 1950 71 629. Markova L. V. Zavod. Lab. 1961 27 379. Weisz H. Meiners W. and Fritz G. Anal. Chim. Acta, 1979 107 301. Goto H. and Shishiokawa T. J . Chem. SOC. Jpn. 1944 65, 673. Kiba N. and Furusawa M. Talanta 1976 23 637. Jedrzejewski W. and Ciesielski W. Chem. Anal. (Warsaw), 1981 26 743. Klockow D. Auffarth J. and Graf G . F. Fresenius Z. Anal. Chem. 1982,311,244. Kurzawa J. Anal. Chim. Acta 1985 173 343. Puacz W. and Kurzawa Z. Mikrochim. Acta 1977 2 623. Puacz W. and Kurzawa Z. Mikrochim. Acta 1978 2 263. Puacz W. and Kurzawa Z. Zh. Anal. Khim. 1979,34,734. Puacz W. and Wejchan-Judek M. in “Materialy Kon-ferencyjne Ogolnopolskiego Sympozjum Siarkowynch lst, 1977,” Politech Poznan Inst. Technol.Chem. Poznan 1978, p. 56. Kurzawa Z. and Puacz W. Chem. Anal. (Warsaw) 1978,23, 389. Puacz W. and Wejchan-Judek M. Polimery (Warsaw) 1979, 24 355. Puacz W. and Puacz J. Chem. Inz. Chem. 1979 13 69. Kurzawa Z . Chem. Anal. (Warsaw) 1960 5 567. Weisz H. and Meiners W. Talanta 1979 26 769. Holter H. and Lmtrup S . C.R. T. Lab. Carlsberg Ser. Chim. 1949 27 72. 75. 76. 77. 78. 79 # 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. Krzyminska A. and Kot B. Chem. Anal. (Warsaw) 1979, 24 373. Kurzawa Z. Chem. Anal. (Warsaw) 1960 5 731. Senise P. Mikrochem.Ver. Mikrochim. Acta 1951 35136, 210. Karasik V. M. and Nemchinskaya V. L. Zh. Obshch. Khim. 1948 18 1228. Utsumi S. Okutani T. and Yamada T. Bunseki Kagaku, 1975 24 799. Jedrzejewski W. and Ciesielski W. Chem. Anal. (Warsaw), 1979 24 861. Michalski E. and Wtorkowska A. Chem. Anal. (Warsaw), 1961 6 365. Kurzawa Z . and Krzymien M. Chem. Anal. (Warsaw) 1970, 15 915. Kurzawa Z. Wojciak W. and Solecki R. Chem. Anal. (Warsaw) 1967 12 1007. Wojciak W. Solecki R. and Kurzawa Z . Chem. Anal. (Warsaw) 1967 12 849. Kurzawa Z. and Meybaum Z . Chem. Anal. (Warsaw) 1960, 5 333. Gershkovich E. E. Zavod. Lab. 1962,28 1437. Kurzawa Z. and Suszka A. Chem. Anal. (Warsaw) 1968, 13 743. Kurzawa Z. and Suszka A Chem. Anal. (Warsaw) 1960,5, 327. Markova L.V. and Glasivtsova N. Z. Anal. Khim. Ekstr. Protsessy 1970 96. Jedrzejewski W. and Ciesielski W. Chem. Anal. (Warsaw), 1984 29 85. Kurzawa Z. and Suszka A. Chem. Anal. (Warsaw) 1962,7, 645. Suzuki S. and Kawagoe S. Bunseki Kagaku 1952 1 87. Gabdullin M. G. Vanyashima M. N. andToropova V. F., Izv. Vyssh. Uchebn. Zaved. Khim. Khim. Tekhnol. 1983,26, 1277. Kurzawa Z . Matusiewicz H. and Matusiewicz K. Chem. Anal. (Warsaw) 1974 19 1175. Kurzawa Z. and Zietkiewicz M. Chem. Anal. (Warsaw), 1975 20 707. Kurzawa Z. Kurzawa J. and Swit Z. Chem. Anal. (Warsaw) 1978 23 409. Kurzawa Z. Chem. Anal. (Warsaw) 1960 5 331. Kurzawa Z. Chem. Anal. (Warsaw) 1961 6 399. Kurzawa Z. and Lemieszek Z . Chem. Anal. (Warsaw), 1975 20 147. Kurzawa Z . and Krzyminska A.Chem. Anal. (Warsaw), 1973 18 1103. Kurzawa Z . and Suszka A. Chem. Anal. (Warsaw) 1960,5, 897. Kurzawa Z. Balcerkiewicz L. and Krzyminska A. Chem. Anal. (Warsaw) 1974 19 333. Kurzawa Z . Chem. Anal. (Warsaw) 1960 5 325. Kurzawa Z . Chem. Anal. (Warsaw) 1961 6 1013. Pardue H. L. and Shepherd S . Anal. Chem. 1963 35 21. Pardue H. L. Anal. Chem. 1964 36 633. Kurzawa Z. and Krzyminska A. Chem. Anal. (Warsaw), 1974 19 1263. Kurzawa Z. Kurzawa J. and Swit Z . Chem. Anal. (Warsaw) 1977 22 961. Kurzawa Z . and Zietkiewicz M. Chem. Anal. (Warsaw), 1974 19 119. Suzuki S . and Ishida T. Bunseki Kagaku 1963 12 395. Richmond J . Rainey C. and Meloan C. E. Anal. Lett., 1976 9 105. Kurzawa Z. and Krzymien M. Chem. Anal. (Warsaw) 1968, 13 1047. Kurzawa Z.and Dobrzanska-Jajszczyk A. Chem. Anal. (Warsaw) 1974 19 1071. Kurzawa Z . and Szukalska A. Chem. Anal. (Warsaw) 1976, 21 297. Byr’ko V. M. Tikhonova T. I . and Pavlova G. I. Zh. Anal. Khim. 1976,31 1086. Kurzawa Z. and Karska B. Chem. Anal. (Warsaw) 1980, 25 465. Kurzawa Z. and Krzyminska A. Chem. Anal. (Warsaw), 1977 22 671. Wtorkowska-Zaremba A. Chem. Anal. (Warsaw) 1969 14, 847 1012 ANALYST SEPTEMBER 1986 VOL. 111 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. Polito W. L. and Neves E. A. An. Simp. Bras. Eletroquirn. Eletroanal. 3rd 1982 2 637. Babko A. K. and Markova L. V. Prikhod’ko M. U. Zh. Anal. Khim. 1966 21 935. Sakuragawa A. Harada T. Okutani T. and Utsumi S., Bunseki Kagaku 1980 29 264.Suzuki S. Bunseki Kagaku 1962 11 299. Suzuki S. Bunseki Kagaku. 1962 11 384. Matusiewicz H. Chem. Inz. Chem. 1980 15 103. Kurzawa Z. Puacz J. and Puacz W. Fresenius Z. Anal. Chem. 1979,296 160. Kurzawa Z. and Kubaszewski E. Chem. Anal. (Warsaw), 1974 19 483. Babko A. K. and Markova L. V. Metody Anal. Khim. Reakt. Prep. 1966 13 117. Kurzawa J. Kurzawa Z . and Swit Z. Chem. Anal. (Warsaw) 1976 21 791. Kurzawa Z. and Zietkiewicz M. Chem. Anal. (Warsaw), 1976 21 13. Kurzawa Z. and Matusiewicz H. Chem. Anal. (Warsaw), 1975 20 687. Kurzawa Z. and Dobrzanska-Jajszczyk A. Chem. Anal. (Warsaw) 1978 23 905. Kurzawa Z. and Matusiewicz H. Chem. Anal. (Warsaw), 1975 20 465. Matusiewicz H. Chem. Anal. (Warsaw) 1981,26 11. Kurzawa Z. Matusiewicz H. and Matusiewicz K. Chem. Anal. (Warsaw) 1976 21 797. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144 145. 146. 147. Matusiewicz H. and Kurzawa Z . Chem. Anal. (Warsaw), 1978 23 363. Matusiewicz H. Chem. Anal. (Warsaw) 1978 23 63. Matusiewicz H. and Kurzawa Z . Chem. Anal. (Warsaw), 1976 21 1035. Kurzawa Z. and Zietkiewicz M. Chem. Anal. (Warsaw), 1976 21 3. Garcia Alvarez-Coque M. C. Villanueva Camanas R. M., Ramis Ramos G. Medina Hernandez M. J. and Mongay Ferniindez C. Thermochim. Acta in the press. Kurzawa Z . and Matusiewicz H. Chem. Anal. (Warsaw), 1975 20 257. Kurzawa Z. and Dobrzanska-Jajszczyk A. Chem. Anal. (Warsaw) 1978 23 897. Weisz H. and Schlipf J. Anal. Chim. Acta 1980 121 257. Zuk A. and Lecka J. Poznan. Tow. Przyj. Nauk Pr. Kom. Nauk Podstawowych Stosow. 1972 3 51. Suzuki S. Bunseki Kagaku 1962 11,306. Tuovinen 0. H. Lapple W. J. and Mair D. M. J. Am. Water Works Assoc. 1981 73 126. Kurzawa Z. and Krzyminska A. Chem’. Anal. (Warsaw), 1978 23 177. Kurzawa Z. and Puacz J. Chem. Anal. (Warsaw) 1978,23, 417. Paper A6/67 Received February 28th 1986 Accepted March 20th 198

ISSN:0003-2654    

DOI:10.1039/AN9861101001

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST SEPTEMBER 1986 VOL. 111 Use of Solid Boric Acid as an Ammonia Determination of Nitrogen Darryl 0. Siemer WINCO PO Box 4000 ILF 0-25 Idaho Falls ID 83403 USA 1013 Absorbent in the An inexpensive trap - de-mister assembly utilising solid crystalline boric acid as an ammonia absorbent was developed to replace the specialised trap - condenser - impinger apparatus normally used in Kjeldahl-type distillations. The boric acid is subsequently dissolved in water and a final determination of the ammonium ion is made either by acidimetric titration a conductance measurement or spectrophotometric measurement after the addition of Nessler‘s reagent. The apparatus fits into the neck of a calibrated flask during the distillation and is itself constructed from disposable plastic pipette tips glass-wool and rubber tubing.Both the high surface area and the tortuosity of the paths through the randomly packed bed of crystals makes solid boric acid an efficient ammonia absorbent. The system is useful for small-scale separations only because, when the volume of solution from which the ammonia is distilled exceeds 25 ml the acid crystals become saturated with water and begin to dissolve before all of the ammonia is trapped. Keywords Solid-state ammonia trap; boric acid; Kjeldahl distillation; nitrogen determination Taras’s review of nitrogen determination procedures empha-sises the importance of what will hereafter be referred to as the “Kjeldahl approach.”l The two main steps in the Kjeldahl approach are firstly a chemical conversion of the nitrogen in the sample to ammonia or ammonium ion and secondly the distillation of the ammonia into an acid liquid trapping agent.The ammonia in the trapping agent is then determined by any of a number of suitable finish techniques. Whereas the original Kjeldahl technique was developed for “organic nitrogen” determinations the same basic approach can be used for determining practically all of the chemical forms of nitrogen if the distillation step is preceded by a suitable chemical pre-treatment of the sample. The bottleneck in many of these analytical procedures is the distillation - separation process not the initial chemical conversion step. Traditionally this distillation step is per-formed with specialised all-glass stills that have a water-cooled condenser the free end of which is immersed in the trapping solution (e.g.the Pregl - Parnas - Wagner apparatus). These distillation systems typically possess high internal surface areas and considerable dead volumes. Consequently a relatively large volume of water must be boiled over to flush the ammonia released from the sample digest quantitatively through to the trapping agent. This paper discusses a simple and inexpensive solid-state ammonia trap which significantly reduces the time necessary for these separations in many situations. An outline of some practical applications is included in order to illustrate the utility of the system. Experimental Apparatus Fig. 1 shows an ammonia trap assembly inserted into the neck of the distillation flask. It consists of a standard 1-ml disposable plastic micropipette tip filled approximately one third with boric acid crystals (about 0.5 g).Several batches of boric acid from different manufacturers were used with essentially equivalent results. Crystal sizes were such that more than 95% of the boric acid was retained between sieves passing 850 and 150 pm particles (Tyler 20-100-mesh screens). A plug of glass-wool supports the crystals and a plastic “chimney” prevents steam from pushing the plug of moist crystals out of the top of the trap. The chimney is made by cutting both the top and bottom off another pipette tip. A plastic retainer cut from yet another pipette tip is forced into Fig. 1. Trap - de-mister assembly. A Chimney; B pipette tip used for body of trap; C boric acid crystals; D retainer; E glass-wool; F, tip used for de-mister; G gum - rubber gasket; H neck of distillation flask the trap body over the glass-wool before the boric acid is added.This prevents the glass-wool from being washed out along with the boric acid crystals after the distillation step has been completed. The ammonia trap is inserted into another pipette tip loosely filled with glass-wool that serves as a “de-mister” to intercept the particles of spray released during the boiling of the sample digestate. The de-mister serves the same purpose as the trap usually incorporated in commercial Kjeldahl systems but it provides the necessary surface area with far less dead volume and solution “hang up.” A short length of gum-rubber tubing is forced over the bottom of the de-mister to serve as a gas-tight gasket between it and the neck of the flask.Both the de-mister and the trap assembly can be re-used an indefinite number of times. After the boric acid used in a prio 1014 ANALYST SEPTEMBER 1986 VOL. 111 determination has been removed the system is prepared for the next by connecting it to a sink-aspirator vacuum source and then pulling water acetone and finally air through it. Distillation Procedure Standard 25- or 50-ml calibrated flasks and a magnetic stirrer -hot-plate were used for both the sample preparation and distillation steps. To perform an ammonia distillation the contents of the digestion flask are first cooled if necessary, and then diluted to approximately 5 ml with water.If a magnetic stir-bar was not used during the sample preparation step one is now added to the flask. Alternatively a few boiling chips can be added. This is necessary because “bumping” can cause expulsive loss of the trap - de-mister assembly and/or its contents. Enough strong base to provide an excess of at least 2-3 mmol of free hydroxyl ion is then added and the filter - de-mister assembly is immediately inserted into the neck of the flask. The flask is placed on to a pre-heated stirrer - hot-plate and its contents are rapidly brought to boiling-point. The trap is removed after an approximately 1.5 min distillation time. The chimney is pulled out of the trap and any boric acid crystals adhering to it are rinsed into a small beaker containing hot water.The contents of the trap are then back-washed into the same beaker using a plastic wash-bottle with a spout modified to fit snugly over the bottom tip of the trap. Finish Techniques For sample digestates containing relatively large amounts of ammonium ion a titration of the dissolved boric solution proved satisfactory. Standardised hydrochloric acid was added with a 2-ml micrometer burette (Cole Parmer) until the methyl orange visual end-point (ca. pH 4) was reached. For low concentrations a spectrophotometric finish tech-nique was investigated. This involves the neutralisation of the boric acid with a slight excess of strong base (about 1.1 equivalents of NaOH per mole of boric acid in the trap) followed by the addition of Nessler’s reagent and dilution to a suitable volume.The absorbance of the fully developed yellow chromogen was then determined in a spectrophotometer at 425 nm in a l-crn path-length cuvette. An alternative and usually superior finish technique consists of the measurement of the conductivity of the boric acid solution. Boric acid is an extremely weak acid (pK = 10.1) and consequently its aqueous solutions are very poor electrical conductors. On the other hand ammonium borate is a typical ionic salt and gives rise to a relatively large electrical response. For this work the detection module of a Dionex Model 2000i ion chromatograph was re-plumbed so that the columns were by-passed and the solution could be sucked directly through the conductivity cell with a plastic syringe. After the cell is filled with the solution the measurement is made under “zero-flow” conditions.Results and Discussion The initial testing of the solid-trapping agent concept was performed by simply drawing headspace fumes from a reagent “ammonia” bottle through one of the traps with a syringe and then blowing the contents of the syringe past the investigator’s nose. The fact that no ammoniacal odour could be detected indicated that boric acid is an effective ammonia absorber. Table 1 outlines a study performed to determine the effect on ammonia recoveries of using varying distillation times. Aliquots of a standard ammonium sulphate solution were used so that no prior sample preparation step was needed. The amount of ammonium ion taken was chosen to be relatively large (0.1392 mmol) so that the precise titrimetric finish could be employed.The “primary standard” on which the recovery Table 1. Ammonia recovery as a function of distillation time. In all instances the sample consisted of 0.100 ml of 1.392 M ammonium sulphate solution and a total solution volume of approximately 5 ml containing 10 mequiv. of free hydroxyl ion was boiled Mass of 20 0.35 85.0 30 0.70 99.85 45 0.82 100.01 90 1.40 99.36 150 2.03 100.20 240 3.30 99.71 * The interval from when the condensate first appeared in the distillate/g Recovery YO Time/s* de-mister to when the filter was removed. values in both this and the following tables are based is the sodium carbonate used to standardise the hydrochloric acid titrant. In this series of experiments the total solution volume in the distillation flask prior to placing it on to the hot-plate was approximately 5 ml.The flask was weighed both before and after the distillation to determine the amount of solution boiled away and passing through the de-mister - trap assembly. The total time required for each distillation was approximately 30 s greater than the figure listed in the first column because that interval was required to bring the solutions to the boil. The data indicate that when solution volumes are relatively small essentially complete ammonia recovery is obtained within 30 s of the onset of rapid boiling. In this instance this corresponds to 10-15% of the solution in the flask. As might be expected a series of further experiments performed with different volumes and compositions of the solutions in the distillation flask revealed that the distillation times required for quantitative recoveries were a function of both the volume of the solution being boiled and of its total salt concentration.As a rule of thumb quantitative ammonia recovery is achieved by the time that 15% of the aqueous phase has been boiled away. Higher total salt contents raise the boiling-point of the solution which in turn enhances the volatilisation of the ammonia relative to that of the water. The upper limit to the solution volume from which the ammonia can be effectively trapped is determined by the solubility of the boric acid in the water absorbed/condensed in the trap during the distillation process.After complete water saturation of the boric acid occurs a portion of the resulting boric acid solution formed subsequently might either drip back into the de-mister or be carried out of the top of the chimney as a spray. This effectively limits the amount of steam that can be passed through the trap to no more than that generated by boiling away approximately 3-4 ml of the solution in the flask. With high salt concentrations (the equivalent of approximately 4 M NaOH) present ammonia can be quantitatively recovered from a maximum digestate volume of approximately 25 ml. However to provide a good margin of safety in this respect solution volumes of no more than 15 ml are recommended. Tables 2 and 3 give the results of some applications of these traps to nitrogen determinations in a number of different types of compounds.In the experiments described in Table 2, enough of the compound was weighed into a 25-ml calibrated flask to provide 0.3-0.5 mmol of “amino - amido” nitrogen. A 0.5-ml volume of 9 M sulphuric acid was added and the sample was then digested for 5 min on a pre-heated hot-plate with a surface temperature of approximately 400 “C. Under these conditions the water immediately boils off and the sample digests in quietly refluxing concentrated sulphuric acid at a temperature of approximately 300 “C. No digestion catalyst was used or needed with these readily hydrolysed compounds ANALYST SEPTEMBER 1986 VOL. 111 1015 Table 2. Recoveries of nitrogen from various compounds. Samples were digested with 0.25 ml of sulphuric acid for 5 min Compound Recovery Yo Iron(I1) ammonium sulphate .. 100.85 Thiourea . . . . . . . . . . 98.87 Ammonium sulphate . . . . . . 99.97 Sulphamic acid . . . . . . . . 99.97 Table 3. Recoveries of nitrate-nitrogen from various compounds. Samples containing approximately 0.4 mequiv. of nitrogen were reduced with aluminium powder in strong base prior to the distillation step. The relative standard deviation of the nitrate determinations performed at this concentration level was approximately 1 YO Compound Recovery YO Nitric acid . . . . . . . . . . 99.89 Cadmium nitrate . . . . . . 99.85 Calcium nitrate . . . . . . . . 99.67 Sodium nitrate . . . . . . . . 100.10 The distillation - separation procedure outlined above was then applied.The excellent nitrogen recoveries obtained for the hydro-lysed sulphamic acid samples were of special interest to this laboratory because the original focus of this project was to develop a rapid procedure to determine sulphamate that did not rely on the often troublesome titrimetric determination of nitrite. Indeed in many instances the accuracy and precision of sulphamic acid determinations performed with the “Kjel-dahl” approach proved to be much better than could be routinely obtained by a more conventional redox titrimetric method involving nitrite.2 However as a general-purpose analytical method for sulphamate at this facility the technique proved to have a major weakness. This is that any concomitant nitrate severely interferes during the hydrolysis step.Nitrate (or one of its decomposition products in the hot sulphuric acid) apparently oxidises some or all of the sulphamate’s “amido nitrogen” to elemental nitrogen. Table 3 shows the results of nitrate determinations per-formed using the same final separation and finish steps after prior reduction of the sample with metallic aluminium.3 The procedure involved placing a suitable amount of sample (containing 0.3-0.5 mmol of nitrogen) into a 50-ml flask and then dissolving it in about 2-3 ml of water. Approximately 150 mg of powdered aluminium followed by 2 ml of 10 M NaOH were added and the ammonia trap - de-mister assembly was immediately seated into the neck of the flask. After the initial vigorous reaction had subsided the flask was placed on the hot-plate and the ammonia remaining in the flask distilled as described above.Recoveries were again excellent, The spectrophotometric finish approach can be recommen-ded only for those situations in which its excellent analytical sensitivity is actually required by the problem at hand. Nessler’s reagent is notoriously difficult to use in high-salt sample solutions and in this work also tended to give erratic results unless considerable care was taken to reproduce exactly all the relevant conditions in preparing the solutions for the spectrophotometer. The absolute and relative amounts of boric acid and base used solution temperatures reagent addition technique reagent volume etc. all need to be reproduced carefully for good over-all analytical precision.Preliminary experiments also revealed that the “phenate” procedure often recommended for water analyses was even less satisfactory.4 Both boric acid and borates apparently inhibit the development of the indophenol chromogen. The most convenient finish technique proved to be the conductance measurement. It was developed to serve as a substitute for the more troublesome Nesslerisation technique for low-nitrogen samples. The mean conductance of a series of six blanks run through the distillation procedure was 7.15 pS and the standard deviation was 0.53 CIS. In each instance 0.50 k 0.02 g of boric acid was used in the traps and the final solution volume was 50 ml. Essentially identical conductance values (a mean of 7.21 pS with a standard deviation of 0.48 pS) were obtained with 1% boric acid - water solutions prepared directly without the intermediate distillation and transfer steps.This indicates both that the de-mister does a very effective job of preventing spray carry-over and that, with reasonable care avoidance of contamination of the boric acid with miscellaneous salts throughout the handling process is possible. The analytical response generated by ammonia added to these solutions is approximately 5.6 pS per p.p.m. of ammonia and the response is linear to at least the 1000 pS maximum conductivity range of the detector used for this work. This translates to an analytical range from approximately 0.25 p.p.m. (the detection limit based on the 30 criterion) to about 180 p.p.m. a range encompassing trace to macro levels of nitrogen in the original sample.Conclusions The limited solution volume from which ammonia can be quantitatively recovered probably constitutes the most serious limitation of the ammonia traps. This factor effectively limits their application to relatively small-scale work only. However, for many applications the fact that their use permits good analytical results to be obtained quickly with otherwise non-specialised glassware and equipment makes them an attractive alternative to classical Kjeldahl still set-ups. The most serious limitation of the conductance finish technique is its lack of specificity. Any ionic species present either in the original boric acid or inadvertantly picked up during transfer steps can give serious blank problems. However with reasonable care it is not too difficult to avoid contamination. References Taras M. J. in Boltz D. Editor “Colorimetric Determi-nation of Nonmetals,” Interscience New York 1958 Chapter Whitman C. L. Anal. Chem. 1957 29 1684. Bartow E. and Rogers J. S . Waf. Sew. Ser Ill. Univ. 1909, 7 14. American Public Health Association American Water Works Association and Water Pollution Control Federation “Stan-dard Methods for the Examination of Water and Wastewater,” Fourteenth Edition American Public Health Association, Washington DC 1975 pp. 416-417. IV pp. 87-91. Paper A6143 Received February 12th 1986 Accepted April 2nd 198

ISSN:0003-2654    

DOI:10.1039/AN9861101013

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST SEPTEMBER 1986 VOL. 111 1017 Automatic Nitrogen4 5 Analyser for Use in Biological Research Joha J. Therion Animal and Dairy Science Research Institute Private Bag X2 Irene 1675 Republic of South Africa Hendrik G. C. Human and Cornelius Claase National Institute for Materials Research South African Council for Scientific Research Pretoria 000 1, Republic of South Africa Roderick I. Mackie Animal and Dairy Science Research Institute Private Bag X2 Irene 1675 Republic of South Africa and Albrecht Kistner laboratory for Molecular and Cell Biolog y South African Council for Scientific Research Pretoria 000 I , Republic of South Africa An automatic nitrogen-I5 analyser capable of analysing 12-20 samples per hour has been developed. The generation of pure nitrogen gas from the sample is carried out automatically the nitrogen isotope ratio is determined by emission spectrometry and the results are calculated as atom-% nitrogen-I 5.The total nitrogen content of the sample is also determined. Sample masses can vary between 0.5 and 100 mg and only 2-10 pg amounts of N are necessary for accurate determinations. Keywords Automatic nitrogen analysis; emission spectroscopy; nitrogen- 15 isotope ratio; automated data handling Nitrogen is one of the most important and familiar elements in the biological sciences as it is one of the main constituents of amino acids peptides proteins and nucleic acids. It is conspicuous however that little work has been carried out on the kinetics of the assimilation of nitrogenous compounds into cells their quantitative partitioning between pools and their flow to different metabolites.An important reason for this apparent lack of interest is that all the radioactive isotopes of nitrogen have short half-lives; the longest is only 10 min for nitrogen-13. The use of radioactive isotopes is therefore both difficult and expensive although nitrogen-13 has been utilised to study ammonia assimilation pathways by different bac-teria 1-5 The stable isotope nitrogen-15 has been used for studying the kinetics of ammonia assimilation by Derxia qumrnosa6 and for the dynamics of nitrogen metabolism in ruminants.7 Nitrogen-15 has also been used extensively in plant and soil studies .8,9 In most of these studies isotope-ratio mass spectrometers have been used to measure nitrogen-15; this is a relatively expensive and slow procedure although a high accuracy can be obtained.Broida and Chapmanlo introduced a photoelec-tric method to measure nitrogen-15 abundance and it became evident from this that emission spectrometry could be very useful for the isotopic analysis of microgram amounts of nitrogen. It has several advantages over mass spectrometry in that an emission spectrometer is less expensive than a mass spectrometer less demanding of the operator and the minimum amount of nitrogen required for determination is much smaller.11.12 Since 1958 emission spectrometry has been developed and applied to agricultural studies and many other research areas.12-16 However the methods for the pre-paration of samples to isolate nitrogen for isotopic analysis have so far been time consuming and have proved to be a limiting factor in extensive surveys such as nutritional studies.17 Goulden and Salter16 devised an automatic emission spectrometer capable of analysing 60 samples per hour. However biological samples had to be subjected to Kjeldahl digestion and ammonium chloride ultimately recovered, which was then injected into a reactor tube where nitrogen was generated and passed into the discharge tube of the emission spectrometer. This is a tedious process that has to be carried out manually and on average one operator can only process and measure 70 samples per week.12 This paper describes a fully automated system that measures total nitrogen and nitrogen-15 in biological materials without prior sample preparation.Experimental The system consists of two major components an automatic nitrogen analyser for determining total nitrogen and an emission spectrometer for measuring nitrogen-15. Automatic Nitrogen Analyser The automatic nitrogen analyser (Model NA1500 Carlo Erba Strumentazione Milan Italy) was used with a minor modifi-cation to the outlet tube for coupling to the emission spectrometer. A schematic diagram is shown in Fig. 1. This instrument is designed for micro and macro determinations of total nitrogen present in a wide range of organic and inorganic substances in both solid and liquid form. The analyser can operate completely automatically and is capable of perform-ing 50 analyses sequentially using the automatic sampler.It has been used with great success over a wide range of nitrogen concentrations from 10 pg to 2 mg. The instrument operates on the principle of the Dumas combustion procedure. The sample is placed in a tin container and introduced into the automatic sampler where it is purged with helium for 20 s before it drops into the combustion column which is maintained at 1000 "C. The container melts and the heat of the reaction of the tin with a small amount of pure oxygen, introduced simultaneously primes the flash combustion of the sample at ca. 1700 "C. Under these conditions even thermally resistant substances are completely oxidised. The combustion products are carried by a constant flow of helium through chromium oxide granules maintained at 1000 "C.The oxi 1018 ANALYST SEPTEMBER 1986 VOL. 111 m source 1/4m m o n oc h ro m at o r Printer - Computer Fig. 1. Schematic diagram of the automatic nitrogen-15 analyser dation is completed in a 5-cm layer of silver-coated cobalt oxide at the bottom of the combustion column. This also retains interfering substances produced during the combustion of halogenated compounds. The combustion products a mixture of C02 various oxides of nitrogen and H20 pass through a reduction reactor filled with reduced copper maintained at 650°C. At this temperat-ure the nitrogen oxides are reduced to elemental nitrogen, which together with the C 0 2 and H20 pass through two absorbent filters the first containing Mg( C104)2 to absorb water and the second containing NaOH absorbed on asbestos particles to retain CO2.The elemental nitrogen enters the chromatographic column and together with the carrier gas, flows through the thermal conductivity detector which generates an electrical signal proportional to the concen-tration of nitrogen present. The signal is amplified the area integrated and the result printed out by the data processor (DP110 Carlo Erba). By calibrating the instrument with a standard it is possible to calculate the percentage of nitrogen in the samples. Emission Spectrometer The analysis of nitrogen-15 by emission spectrometry depends OR the property of nitrogen at reduced pressure (2-10 Torr), to emit light of characteristic wavelengths when energised by radio- or microwaves. The presence of an extra neutron in the 15N nucleus causes shifts resulting in a readily measurable displacement of the bandhead that usually appears in the ultraviolet region from 297.7 nm for the l4N'4N molecule to 298.3 nm for 14N15N and 298.8 nm for 15N15N.These bands can be readily separated by a small monochromator with ca. 1 A resolution. The intensities of the bands are proportional to their nitrogen content so that in an equilibrium mixture of the three molecular forms of nitrogen the proportion of 15N can be calculated from the ratio of the bands of mass 28 and 29 in a manner analogous to that used in mass spectrometry.12 The spectrometer features a microwave source for the excitation of the nitrogen gas in a stream of He carrier gas a monochromator that scans rapidly and repeatedly over the relevant spectral region and a computer for the rapid acquisition of spectral information and data processing.Scanning Monochromator A Jarrell-Ash 0.25 m Ebert monochromator (Jarrell-Ash Division Fischer Scientific Waltham MA USA) with a 2360 lines mm-1 grating blazed for 300 nm and a dispersion of 1.65 nm mm-1 was used. Wavelength scans obtained with the monochromator using slits of 25,50 and 150 ym respectively, with a helium and natural nitrogen mixture excited by a microwave discharge at 20 W showed that the 50 pm slits were the best set to use as with these there was only a marginal loss in the definition of the spectrum compared with the 25 pm slits. With the 150 pm slits the resolution was definitely impaired. The rapid scanning over the spectral region was accom-plished with a rotating quartz plate QP (1.9 mm thick) behind the entrance slit (Fig.1). The refraction of the light beam (at approximately 300 mm) is such that a 40" rotation of the plate shifts the image at the exit slit by 0.6 nm the difference in wavelength between the two peaks to be measured. A synchronous motor is used for driving the refractor plate in order to ensure the constant speed of rotation that is necessary as the computer measures intensity at regular intervals (700 ps between measurements) so that the two peaks are separated on a time basis rather than on a wavelength basis. Constant rotational speed also eliminates the effect of a 50 Hz ripple on the emission from the microwave source and on the output of the amplifier.The speed of rotation is 1 rev s-1 resulting in a time lapse of 0.11 s between the measurement of the two peaks. The dispersed signal by photomultiplier tube 1 (PMT1) is constantly related to a signal measured by photomultiplier tube 2 (PMT2) as shown in Fig. 1. The latter measures the intensity of the 337.1 nm bandhead of nitrogen belonging to the same band system through a narrow-band interference filter that excludes any extraneous light. This serves as an efficient reference for the analytical channel to eliminate the effect of the evolutionary nature of the nitrogen content of the gas mixture given off by the source. Excitation Source A Microtron 200 Mark 111 (Electro-Medical Suppliers Green-ham UK) microwave generator (2450 MHz) was used for the excitation of the gas mixture.The instrument is equipped with a power meter a reflected power meter and a magnetron protection cut-out device that becomes operative when the reflected power exceeds 75 W. A 214L type resonant cavity accommodating discharge tubes of up to 13 mm diameter (RC) was used as the termination on the coaxial cable (Fig. 1). The source can supply power to a maximum of 200 W but the minimum of 20 W is normally used ANALYST SEPTEMBER 1986 VOL. 111 1019 Pumping System The Carlo Erba nitrogen analyser supplies the nitrogen -helium gas mixture at a flow-rate of approximately 80 ml min-l at atmospheric pressure. Inside the discharge tube the pressure must be ca. 5 Torr and the reduction in pressure is accomplished using a vacuum pump (Fig.1). The pumping speed of the rotatory vacuum pump (Alcatel 2004 A CIT Alcatel Paris France) is reduced by a fixed diaphragm D (ca. 1 mm diameter) in the line. The discharge cavity is located between this diaphragm and the needle valve V which is adjusted so that the combined effect of the pump D and V is such that the pressure on the supply side is near to atmospheric pressure in order not to interfere with the operation of the nitrogen analyser. A Pirani gauge type pressure meter (Alcatel API 101 T CIT Alcatel Paris France) is available for measuring the pressure in the discharge section whereas a diaphragm type meter measures the pressure on the inlet side. A second needle valve N is available for bleeding air into the system. As this is a constant source of nitrogen for the discharge rather than one changing with time as supplied by the nitrogen analyser it can be conveniently used for preliminary measurements and for checking the operation of the system.Electronics A single stabilised high-voltage supply (Tennelec C 952 with <3 mV regulation and less than 0.001% drift per hour 0-10 mA current) is used for the Hamamatsu 1P28 photomultiplier on the monochromator and the Hamamatsu R166 solar blind photomultiplier tube on the reference channel. The light flux is sufficiently high so that the photomultiplier tubes can be used with a high voltage supply of approximately 360 V. Each photomultiplier signal is amplified and converted by a 12-bit analog to digital converter. The parallel digital infor-mation is loaded into a Zenith microcomputer (Zenith Data Systems Illinois USA) via two parallel interface ports.An optical switch on the rotating quartz plate drive supplies a reference pulse to the computer that initiates a measurement cycle. In this way the measurements are synchronised with the quartz plate position and therefore with wavelength. The computer triggers the A/D converters for each conversion. Measurement by Computer Analysis of a sample begins when the NA1500 cycle is started at time to. After the time interval necessary for the gases to pass through the instrument the first detectable nitrogen appears at the microwave excitation source at time tl and is present until time t2. Programmed measurements begin at ta and end at tb [Fig.2(a)]. Ideally this interval should include the region of maximum intensity. The programme runs through 25 cycles of measurements in the interval ta-tb. Each cycle duration is 1 s (the period of revolution of the rotating quartz plate) but measurements are made only during a fraction of this 1 s period viz. over 80” or an 80/360 s period. During this period the spectral region of interest i. e. 297.4-298.6 nm is scanned. Only three values of intensity are required from the spectrum viz. the intensity at the strong I4N14N peak at 297.7 nm the intensity at the weak 14N15N peak at 298.3 nm and a background value. The first value is easily selected by the computer from the 320 elements stored in the memory of channel 2 by looking for the maximum of the l4NI4N peak [Ip, Fig.2(b)]. Next the intensity at the 14N15N peak is located 56 measurements after the first peak ( I p +56) and this separation remains fixed. Similarly the 93rd measurement after the first peak (Ip +93) represents a reliable position (wavelength) at which to measure the background intensity [Fig. 2(b)]. These three values from channel 2 are normalised with respect to Measure and display I I 1 1 1 I I + 56 I + 93 Fig. 2. spectrum showing the positions at which measurements are made (a) Sequence of events during an analytical cycle; and (b) N channel 1 the reference channel by dividing by the corre-sponding values of the averaged set of 25 reference intensities. Results and Discussion Evaluation of Data Although the background on the long wavelength side of the 14N15N peak appears flat and reproducible it does not represent the real background at the wavelength of this peak, as using this value for calculating the I4N15N to 14N14N peak ratio yields a value of ca.1 50 instead of the value of 1 137 for natural (0.365 atom-%) 15N abundance. This is due to the fact that the emission peak of the 14N2 molecule at 297.7 nm produces a “skirt” of spectral emission towards the long wavelength side that interferes at 298.3 nm the wavelength of the 14N15N molecule maximum. This is a real feature of the spectrum and not due to inadequate resolution by the monochromator. A sample of known 15N abundance is therefore used to establish the background value. For this purpose a sample at natural abundance has proved to be adequate.Measurements are made as follows the intensity of radiation is measured at the 14N14N peak the 14N15N peak and at a minimum in the spectrum on the long wavelength side. These values are called P1 P2 and B respectively. The concentration of 15N C is given by c= - loo (inatom-%) . . . . . . 2R + 1 where P i - A P2 - A R E - . . . . . . . . For a sample of natural abundance R = 137 and the value of the true background A is given by . . . . . . . RP2 - Pi R - 1 A = - (3 1020 ANALYST SEPTEMBER 1986 VOL. 111 Comparing this value with B the background measured away from the peak it was found that the ratio k = A/B was both positive and constant regardless of the signal sizes. The signal size was varied in a 1 8 range by varying the nitrogen concentration and also by varying the high voltage on the photomultiplier tube but no correlation of this ratio with the signal size was found.After establishing the value of k accurately by running three or more samples of natural abundance samples of unknown 15N content can be run and the concentration calculated from equation (1) by substituting kB for A the true background in equation (2) so that R can be defined in terms of the background measured away from the peak ( B ) and the concentration of 15N (C) calculated without needing to know the true absorbance. Hence, 100 2R + 1 c= -where Pi - kB P2 - kB R=-Following this procedure and running a number of natural abundance samples as unknowns a standard deviation of better than 0.02 atom-% of 15N is usually found.Repeated analyses were carried out on defined inorganic and organic substances to test the accuracy and reproducibility of the results obtained with the nitrogen analyser (ANA). These are shown in Table 1. Excellent reproducibility was obtained with all three substances with coefficients of variation between 0.3 and 0.85%. The agreement between the known nitrogen content of the compounds and the values obtained was very good. The difference was 0.05 for (NH4)2S04 whereas it was 0.08 for haemin an organic compound from which it is difficult to release nitrogen. The difference for albumin an organic macromolecule was only 0.01. Table 2 shows a comparison of nitrogen analyses of biological samples performed using the Kjeldahl method and the ANA respectively.Coefficients of variation varied between 1.5 and 2.7% which is good considering the heterogeneity of the samples and with the maize cell walls, the extremely low nitrogen content. It is important to note that the accuracy with which the sample is weighed into the tin cup will influence the performance of the ANA. The use of large samples would therefore increase the accuracy of the method although the size of the tin cup dictates the sample size to a great extent especially when using biological samples of low density. The sample masses shown in Table 2 were the maximum that could be fitted into the sample cups except for the duodenal contents and sheep faeces where smaller samples were also used.From the results of the latter two substances it seems that the problem of heterogeneity of the samples can be overcome by carrying out repeat analyses on the same batch of material. The linearity of the measurements performed on the emission spectrometer was determined using 15N-labelled (NH4)2S04 standards (obtained from Isotope Services Los Alamos National Laboratory Los Alamos NM USA). Table 3 shows the excellent agreement obtained between the nominal values of the standards and those measured with the automatic emission spectrometer. A correlation co-efficient of 0.9996 was found with a y-intercept of -0.061. The slight deviation from a straight line with a y-intercept = 0 could be attributed to the fact that a higher degree of accuracy can be obtained with a mass spectrometer.The memory effect of the automatic 15N analyser is not Table 1. Accuracy and reproducibility of nitrogen analyses performed using the automatic nitrogen analyser Sample (NH4)2S04 . . . . Haemin . . . . Mass/ mg 5.07 5.04 5.05 5.05 5.00 5.00 10.07 10.05 10.05 10.00 10.05 10.06 Atom-% of '5N Measured Mean Expected SD cv 21.29 21.15 21.10 0.094 0.44 21.20 21.15 21.17 21.08 21.02 8.49 8.51 8.59 0.025 0.29 8.53 8.53 8.48 8.53 8.54 Albumin . . . . 14.20 14.96 15.09 15.10 0.121 14.19 14.96 14.23 15.19 14.22 15.20 14.19 15.04 14.24 15.21 0.79 Table 2. Determination of atom-% 15N in biological samples using Kjeldahl procedures and the automatic nitrogen analyser Mean mass/ Sample mg Bacteria .. . . 2.04 Maizecellwalls . . 6.51 Duodenal contents 2.04 10.02 Sheepfaeces . . . . 2.02 10.04 Atom-% N in sample Kjeldahl ANA SD cv 6.66 6.66 0.136 2.0 0.14 0.18 0.004 2.27 3.30 3.85 0.105 2.74 3.30 4.09 0.064 1.57 2.51 2.41 0.038 1.59 2.51 2.51 0.056 2.2 ANALYST SEPTEMBER 1986 VOL. 111 0.357 0.373 0.391 0.342 0.356 , 1021 ' b 0.364 5 0.019 Table 3. Measured 15N atom-% in a range of 'SN standards. Standards obtained from Isotope Services Los Alamos National Laboratory, Los Alamos NM USA Measured atom-% 1sN Nominal atom-% 15N x (n = 6) SD cv 0.37 0.358 0.006 1.68 0.5 0.481 0.018 3.74 1.0 0.898 0.022 2.45 2.0 1.764 0.025 1.42 5.0 4.528 0.081 1.79 10.0 9.521 0.118 1.24 Table 4. Test of the memory effect in the discharge tube of the ADSRI nitrogen-15 analyser Atom-% 1SN Sample No.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Nominal 0.365 0.365 0.365 0.365 0.365 16.603 16.603 16.603 16.603 16.603 0.365 0.365 0.365 0.365 0.365 significant and can be ignored. The results in Table 4 show that even under extreme conditions where the highly enriched samples were immediately followed by a sample at natural abundance no significant increase could be detected. Goulden and Salterl6 also found no memory effect when they subjected the NIRD automatic analyser to a highly enriched sample followed by one at natural abundance. The most important feature of the automatic nitrogen-15 analyser is the fast rate of analysis possible (12-20 samples per hour compared with 6-8 samples per week with manually operated emission spectrometers1*).It is also more advanced than the automatic system described by Goulden and Salter,16 as biological samples can be analysed directly without prior Kjeldahl digestion and subsequent recovery of ammonium chloride. Further the system is fully automated with a direct printout of atom-% 15N for each sample. Although mass spectrometry remains the most accurate means of analysing for nitrogen-15 with a reproducibility of 0.001 atom-% possible in modern apparatus automatic emission spectrometry offers a number of advantages. The most important are the much faster rate of analysis possible and the very small amount of nitrogen (10 pg) required for accurate determinations compared with the 200-2000 pg required for determination by mass spectrometer.The auto-matic nitrogen-15 analyser described in this paper is therefore ideally suited for biological research where metabolites are often only available in small amounts. The authors thank the Protein Research Fund Protein Advisory Committee Republic of South Africa for providing funds for the construction of the instrument. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. References Kenealy W. R. Thompson T. E. Schubert K. R. and Zeikus J. G. J. Bacteriol. 1982 150 1357. Kim H-C. and Hollocher C. H. J. Bacteriol. 1982,151,358. Meeks J . C. Wolk C. P. Thomas J. Lockau W. Shaffer, P. W. Austin S. M. and Galonsky A. J. Biol. Chem. 1977, 252 7894. Meeks J. C. Wolk C. P. Lockau W. Schilling N. Shaffer, P. W. and Chien W.-S. J. Bacteriol. 1978 134 125. Wolk C . P. Thomas J. Shaffer P. W. Austin S. M. and Galonsky A. J . Biol. Chem. 1976 251 5027. Wang R. and Nicholas D. J. D. Phytochemistry 1985 24, 1133. Leng R. A. and Nolan J. V. J. Dairy Sci. 1984 67 1072. Bremner J. M. J . Assoc. Off. Anal. Chem. 1985 68 155. Yamamuro S. Soil. Sci. Plant Nutr. 1981 27 405. Broida H. P. and Chapman N. W. Anal. Chem. 1958,30, 2049. Ohmori M. Iizumi H. and Hattori A. Anal. Biochem., 1981 111 83. Salter D. N. Proc. Nutr. SOC. 1981 40 355. Ito O. Yoneyama T. Akiyarna Y. and Kumazawa K., Radioistope (Jpn.) 1976 25 448. Fiedler R. and Proksch G . Plant Soil 1972,36,371. Lloyd-Jones C. P. Adam J. Judd G. A. and Hill-Cottingham D. G. Analyst 1977 102 473. Goulden J . D. S . and Salter D. N. Analyst 1979 104 756. Salter D. N. and Smith R. H. Br. J . Nutr. 1977 38 207. Paper A6157 Received February 20th 1986 Accepted April 7th 198

ISSN:0003-2654    

DOI:10.1039/AN9861101017

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST SEPTEMBER 1986 VOL. 111 1023 Application of Electrothermal Atomic Absorption Spectrometry to the Determination of Trace Amounts of Indium in Metallic Zinc and Krystyna Brajter and Ewa Olbrych-Sleszynska Department of Chemistry University of Warsaw Pasteura 1 02-093 Warsaw Poland Trace amounts of indium in metallic zinc and lead have been determined by graphite furnace atomic absorption spectrometry. The detection limits for three operated furnace systems were evaluated for SP9-01 (1 ) 6.0 x 10-13 g; for HGA-500 (2) 7.0 x 10-13 g; and for GRM-1268 (3) 1.1 x 10-11 g each using 20 pl of sample. Linear calibration graphs were obtained between 0.01 and 0.1 p.p.m. (I) 0.05 and 0.6 p.p.m. (2) and 0.05 and 0.2 p.p.m. (3). Ion-exchange separation employing Xylenol Orange modified Amberlyst A-26 anion exchanger was used as the preliminary step for the determination of indium in Zn - In and Pb - In alloys as interferences from other metals were observed in the indium absorbance.In order to compare the results two other separation methods for the lead matrix and several different graphite furnace atomisers were used. Molar excesses of Ni Co Fe Cu and Zn of less than 1000 and between a 100 and 1000 molar excess of Al cause a decrease in the indium absorbance. With molar ratios greater than 1000 the suppression caused by Al Co and Zn disappears and that of Fe becomes less pronounced. Ga causes an enhancement of the indium atomic absorption signal. Equivalent amounts of As Sb Bi do not interfere. Keywords Indium determination; electrothermal atomisation; atomic absorption spectrometry; ion-exchange separation; Xylenol Orange modified Amberlyst A-26 Indium is frequently found in trace amounts in metallic zinc and lead.Its determination in metals and ores (in nature indium occurs in some zinc and lead ores) is important from an analytical point of view. The determination of indium, especially in the presence of the other metal matrices presents many problems for the analytical chemist. The aim of the work reported in this paper was to determine trace amounts of indium in metallic zinc and lead using atomic absorption spectrometry (AAS) with electrothermal atomisation (ETA). AAS was chosen for indium determination as the most convenient means of comparing the results with other methods of separation.In preliminary experiments we found that there were interferences from matrix metals. Preliminary separation of indium from matrix metals was adopted as the most practical procedure to avoid matrix interferences. Three independent methods of separation were investigated co-precipitation of hydroxides precipitation of lead sulphate and ion-exchange separation. We found ion-exchange separation on Xylenol Orange modified anion exchanger to be the most convenient method. Experimental Instrumentation A Pye Unicam SP9-01 atomic absorption spectrometer a Perkin-Elmer Model 2380 atomic absorption spectrometer equipped with an HGA-500 graphite furnace atomiser and a Beckman Model 1272D spectrophotometer equipped with an Unilam 1288 burner and a Pye Unicam GRM-1268 graphite furnace atomiser were used.The three instrument systems were equipped with deuterium background correction. The deuterium lamp background compensator could not eliminate all the matrix interferences observed. The operating con-ditions are given in Table 1. The SP9-01 atomic absorption spectrometer was used for the determination of indium in all the metallic lead and zinc samples. For the examination of the indium content in the weighed samples and for comparison of results indium was also determined using the Beckman 1272D spectrophotometer and the GRM-1268 graphite furnace atomiser. Table 1. Optimum operating conditions for the determination of indium by ETA atomic absorption spectrometry Parameter Wavelengthhm . . . . . . Dryingtemperature/"C .. . . Drying time/s . . . . . . . . Charring temperaturePC . . . . Charringtimels . . . . . . Atomisation temperature/'C . . Atomisationtimek . . . . . . CleaningtemperaturePC . . . . Cleaningtimek . . . . . . Band-passhm . . . . . . . . Injectionvolumelpl . . . . . . Argon flowratell rnin- I . . . . HGA-500 Pye Unicam Pye Unicam (Perkin-Elmer GRM- 1268 SP9-01 2380) (Beckman 1272A) 303.9 110 15 500* 15 2400 15 3000 5 0.7 20 7 L. 303.9 100 20 1200 20 2500 10 2500 3 0.7 20 Standard (0.3) 303.9 100 20 1100-1350 30 2700-2900 10 3000 3 0.7 20 Standard ( 1 3) * Explanation in text 1024 Reagents A standard indium solution was prepared by dissolving indium (spectral grade 1.000 g) in 2 M HCl and diluting to 500 ml.This solution was further diluted as required. Metal ion solutions were generally prepared by dissolving the appropriate masses of the sulphates and nitrates of the metals in doubly distilled water and were standardised by EDTA titration. The stock solutions were diluted as required. ANALYST SEPTEMBER 1986 VOL. 111 Procedure B Ion-exchange Columns Columns 25 cm long and 5 mm i.d. with a stopcock at the end, were used. They were packed with the macroporous strong base anion-exchange resin Amberlyst A-26 with a bead size of 0.1-0.2 mm (Rohm and Haas) modified by the use of Xylenol Orange (XO) tetrasodium salt.1 In procedures A and B in order to obtain XO-loaded resin the Amberlyst A-26 chloride form with a bead size of 0.1-0.2 mm was shaken with a 4 x 10-5 M aqueous solution of XO until the supernatant became colourless.The resin was then filtered off washed with water and ethanol dried and stored in a refrigerator.' The modified resin had a capacity of 0.4 mmol of XO per gram of Amberlyst A-26 in the primary chloride form. A 4 cm bed height of 200 mg of modified resin was used in both procedures. Flow-rates of 40 ml h-1 were used. Ion-exchange Separation of Indium from Zinc and Lead Preliminary experiments confirmed the usefulness of XO-loaded resin for the separation of indium from zinc and lead. To optimise the conditions for the separation of trace amounts of indium from the great excess of lead and zinc some investigations were performed with the use of synthetic solutions simulating the composition of the real metallic zinc and lead samples being analysed.The effect of bed height and the size of the resin beads was investigated. The following procedures for the rapid separation of indium from zinc (A) and from lead (B) were deduced. Procedure A A 13-ml volume of solution containing 0.005 mg of In and 669 mg of Zn adjusted to pH 2.0 was introduced into the column. Zn was eluted with 25 ml of water acidified to pH 4.0 with H2S04. Indium was eluted with 20 ml of 0 . 2 ~ HN03 into a 25-ml Calibrated flask. After dilution to volume indium was determined according to the conditions given in Table 1. The indium concentration was obtained from a calibration graph. The mean obtained from six separate determinations was 0.0049 L- 1.3 X 10-4 mg (95% confidence limit).Each result was the mean of four AAS measurements by the GRM-1268. The mean obtained using the Pye Unicam SP9-01 spectropho-tomer was 0.0048 k 1.2 x mg (95% confidence limit). A 13-ml volume of solution containing 0.005 mg of In 669 mg of Pb and 2 mg of XO adjusted to pH 3.0-3.8 was introduced into the column. Pb was not retained on the column and passed into the eluent. To wash the Pb from the void volume, 25 ml of water acidified to pH 4.0 with HN03 were used. Indium was eluted with 20 ml of 0 . 2 ~ HN03 into a 25-ml calibrated flask. The volume was adjusted to the mark and indium was determined by AAS. The indium concentration was obtained from the calibration graph. The mean obtained from six separate determinations was 0.0049 _+ 1.3 x 10-4 mg (95% confidence limit).Each result was the mean of four AAS measurements by the GRM-1268. The mean obtained using the Pye Unicam SP9-01 spectrometer was 0.0049 k 1.2 x 10-4 mg (95% confidence limit). Analysis of Metallic Zinc Procedure 1 and Metallic Lead, Procedures 2 3 and 4 The sample of metallic zinc was dissolved by heating it with 5 ml of HN03 (1 + 1). The solution was evaporated twice and the residue was dissolved in 0 . 2 ~ HN03 transferred into a 25-ml calibrated flask and diluted to volume. Five different samples of metallic lead were dissolved in the same way and transferred into five 25-ml calibrated flasks. Procedure 1 A 5-ml aliquot was diluted to 10 ml adjusted to pH 2.0 and introduced into the ion-exchange column.Zn was not retained on the resin. Zinc remaining in the void volume was washed out with 25 ml of water acidified to pH 4.0 with H2S04. Indium was then eluted with 20 ml of 0.2 M HN03 into a 25-ml calibrated flask and was diluted to volume. The indium was determined by AAS according to the conditions given in Table 1. The concentration was obtained from the calibration graph, and the results of the analysis are presented in Table 2. Procedure 2 A 5-ml aliquot was diluted to 10 ml and adjusted to pH 3.0-3.8; 2 mg of XO were then added and the aliquot was introduced into the column. Pb was not retained on the resin bed and passed into the eluent. Lead was washed out with 25 ml of water acidified to pH 4.0 with HN03. Indium was eluted with 20 ml of 0 .2 ~ HN03 into a 25-ml calibrated flask. The solution was diluted to volume and indium determined by AAS; the concentration was obtained from the calibration graph. The results are presented in Table 3. Procedure 3 Five samples of metallic lead containing different amounts of indium were analysed. They were dissolved by heating with 5 ml of nitric acid (1 + 1) and then 20 mg of metallic iron, dissolved in 1 + 1 nitric acid were added to all the samples and metal hydroxides were coprecipitated with ammonia solution. Table 2. Determination o f indium in metallic zinc by ETA - AAS De t t' rm i na t ion after ion-exchange separation+ (Procedure 1 ) D i rec t de t t' r ni i n ;i t ion Instrument Sam ple/mg 111. '%I Sa ni plt'img In '%, GRM-1268 .. 1020 0.0009 I125 0.0005 * 0.0002 SP9-01 . . . I005 0.0009 1015 0.0000 5 0.0003 * Obtained from 10 AAS measurements without zinc separation after dissolving the sample as described before Procedure 1. t Mean and range (95% confidence limit) of four separate determinations ANALYST SEPTEMBER 1986 VOL. 111 Q 0.030 0.020 0.010 1025 (a) - /x-x-x ,x-x -4 X -I I I Table 3. Determination of indium by ETA - AAS in five different samples of metallic lead after the separation step obtained for the GRM-1268 ( a ) and SP9-01 (b). All samples analysed contained a mixture of metal ions (Cu. Ag. Bi. As Sh. Sn. TI) in the range 0.045-0.0006% for each metal ion Separation by coprecipitation of hydroxides (Procedure 3) Saniple/mg 544.7 512.0 1126 1 108 933.8 921.4 1010 1013 1103 1105 In," %" 0.028 0.025 0.0 12 0.01 1 0.002 0.003 0.0007 0.0006 0.008 0.008 Separation by precipitation of lead sulphate (Procedure 4) Sample/mg In.* O/" 512.1 0.020 543.2 0.020 1140 0.013 1114 0.01 1 1002 0.00 1 101 1 0.002 1115 0.0006 1112 0.0005 1232 (1.007 1114 0.007 Ion-exchange separation (Procedure 2) Sample/mg In,? YO 532.0 0.025 k 0.001 508.1 0.024 k 0.002 1012 0.015 k 0.004 1007 0.012 2 0.003 1 104 0.0018 k 0.0003 0.002 ? 0.0004 1103 120s 0.0007 ? 0.0003 120 1 0.0006 ? 0.0003 1125 0.0070 ? 0.00 1 8 1112 0.0070 k 0.0021 * Obtained from four AAS measurements for one sample of metallic lead.after one separation step. i- Mean and range (95% confidence limits) for four AAS measurements in each of four separate determinations.This was repeated twice. The hydroxide precipitate was dissolved in 5 ml of HN03 (1 + l ) transferred into a 25-ml calibrated flask and diluted to volume. Indium was deter-mined by AAS according to the conditions given in Table 1. The results were corrected for blanks determined under the same conditions as the lead samples. Procedure 4 For the comparison of results the lead matrix in all alloy samples was separated as lead sulphate after the dissolution step described above and indium was determined in solution after filtering off the precipitate. For this 5 ml of concentrated H2S04 were added to the aliquot and the sample was heated until white fumes were observed.The sample was then cooled and after adding 50 ml of water heated to boiling. In this instance the indium standards were mixed with lead nitrate and then analysed in exactly the same way. The results obtained are presented in Table 3. Results and Discussion Results were obtained by three different ETA - AAS systems. The Beckman 1272D system belongs to an earlier generation of spectrophotometers but is still used by many laboratories. The results obtained using it are comparable with those from a newer type of spectrophotometer the Pye Unicam SP9-01. This implies that even those laboratories that possess older types of spectrophotometers may use our method with good results. To optimise the charring temperature the dependence of absorbance on charring temperature was investigated (Fig.1). A charring temperature of 1200°C applied for 20 s was the maximum temperature at which no loss of analyte occurred when the HGA-500 was used. For the SP9-01 spectrometer a charring temperature of 500 "C applied for 15 s was advised by Pye Unicam for indium determination. There was no differ-ence in the absorption peak at charring temperatures in the range 500-1000°C in the determination of indium. For the GRM-1268 an optimum charring temperature of 1100 "C applied for 20 s was chosen. Atomisation temperature versus absorbance is presented in Fig. 2. The maximum temperatures available using the HGA-500 2400°C for the SP9-01 and 2700-2900°C for the GRM-1268 were chosen. These maximum temperatures were also recommended by Dittrich.2~3 As expected the best analytical results were obtained using the HGA-500 and SP9-01 systems.For the GRM-1268 L'vov's pyrolytic graph-ite platform was used but no improvement was obtained. The 3 c I 500 1000 1500 Tc ha rl0c Fig. 1. Height of indium absorption peak as a function of charring temperature. GRM-1268 Tat,, = 2800 "C cln = 0.1 p.p.m tchar. = 30 s 1500 2000 2500 3000 T,t"lll/"C Fig. 2. (a) Absorbance of indium; and (b) peak height as a function of atomisation temperature. ( a ) HGA-500 Tchar. = 1200 "c cIn = 0.4 p.p.m. ( b ) GRM-1268 Tchar. = llOO°C cIn = 0.1 p.p.m. detection limit with the GRM-1268 is similar to2,4 or better than3 those in the literature. The presence of NaOH HCl or HN03 has a strong influence on the determination of indium, as has also been observed by others.2,4,5 According to Dittrich,zJ the suppression of the indium absorbance in the presence of HC1 is due to InCl formation.We think that a more probable explanation for the decrease in the indium signal is the diffusion of volatile InC13 from the furnace in the period of time before the atomisation temperature has been reached. The presence of HCl and NaOH causes a propor-tional suppression of absorbance whereas the addition of HN03 (>O. 18 M) stabilises the signal. Hence 0.2 M HN03 was chosen as the best medium for the determination of indium. Linear calibration graphs were observed between 0 and 0.6 p.p.m. for the HGA-500 between 0 and 0.2 p.p.m. for the GRM-1268 and between 0 and 0.1 p.p.m. for the SP9-01.In all instances a 20+1 aliquot was used 1026 ANALYST SEPTEMBER 1986 VOL. 111 Table 4. Detection limit and sensitivity (for 0.0044 absorbance) for the determination of indium with different spectrophotometers. In all instances indium was determined under the optimum conditions given in Table 1; 20-yl volumes were used Perkin-Elmer Beckman 1272D Pye Unicam 2380 (Pye Unicam Parameter SP9-0 1 (HGA-500) GRM- 1268) Detection limithg ml-l . . 0.030 0.035 0.55 Detection limit/pg . . . . 0.60 0.70 11 Sensitivityhgml-1 . . . . 0.050 0.050 2.2 Sensitivitylpg . . . . . . 1 .0 1 .0 44 Reproduci bilityhg ml- 1 . . k 0.0052 k 0.0052 +0. 10 Fig. 3. Effect of Al Fe Co Ni Cu Zn Ga and Pb on the AAS signal of In expressed as relative change in signal for the HGA-500 and GRM-1268 systems.A Ga; B Co; C Fe; D Al; E Zn; F Ni; G Pb; and H Cu Zinc lead and other matrix metals have a marked effect on the indium absorbance; metals were taken as nitrates through-out to avoid the influence of chlorides. Molar excesses of Ni, Co Fe Cu and Zn of less than 1000 and of A1 between 100 and 1000 cause a decrease in the indium absorbance. With molar ratios greater than 1000 the suppression caused by Al Co and Zn disappears and that of Fe becomes less pronounced. Gallium causes an enhacement of the indium atomic absorp-tion signal. These results were obtained by the HGA-500 system and by the GRM-1268 system for comparison and are presented in Fig. 3. Equivalent amounts of As Sb and Bi do not interfere with the indium atomic absorption signal.The complex nature of the chemical process taking place in the graphite furnace makes the correct explanation of the interference effects observed extremely difficult. Campbell and Ottaway6 and Aggett and Sprott7 have postulated that atoms are produced by direct reduction of the metal oxides by graphite. If this is true the suppression of the indium absorbance in the presence of an excess of the other metals, especially of those forming more stable carbides or refractory oxides may be explained simply by the occlusion of the indium or the graphite surface. Under such conditions the absorption peak of indium was observed to be delayed and was of a different shape. We observed the suppression of the indium absorption signal in the presence of Fe Co and Ni, which form carbides with high melting-points and in the presence of A1 (refractory oxide) Zn (high sublimation point) Pb and Cu.A strong influence of the presence of zinc and lead on the absorption signal of indium was observed; this was the reason for developing Procedures 1 2 3 and 4. Some modification of the charring step by an extension of time could not remove the interference of lead and zinc on the indium signal. Magnesium nitrate was investigated as a matrix modifier but it did not remove the interferences. Mg(N03)2 causes an enhacement of the indium atomic absorption signal related to its concentration in the sample. If the excess of interfering metal is sufficiently large a new effect is observed an increase of the indium signal. This also suggests that the mechanism of atomisation may have been changed.This effect is very similar to that observed if a modified graphite furnace (modified by means of some metal ions forming stable carbides or other compounds) is used for the atomisation process.sl0 No presence of indium in interfering reagents even at high interfering metal concen-trations was reported. Owing to the complicated effects on the indium signal resulting from the presence of other metal ions we decided to eliminate matrix interferences by a preliminary separation of indium from the sample of metallic zinc and lead by ion exchange. The methods were applied to the determination of indium in metallic zinc and lead. The results are presented in Tables 2 and 3. In order to compare the determination in metallic lead, we also used other separation methods namely coprecipi-tation and precipitation of the matrix by the use of H2S04.In the coprecipitation method indium as In( OH)3 coprecipi-tates on Fe(OH)3. This method was also adopted for the separation of indium. The results are presented in Table 3. The XO-loaded resin permits the simple fast separation of indium from an excess of lead and zinc using a very short resin bed. No influence from other metal ions present in up to a 10-fold molar excess was observed ANALYST SEPTEMBER 1986 VOL. 111 References 1. 2. 3. 4. 5. Brajter K. and Olbrych-Sleszynska E. Talanta 1983 30, 355. Dittrich K. Talanta 1977 24 735. Dittrich K. Talanta 1977 24 725. Yudelevich I. G. Burynova L. M. Bakhturova N. F. and Korda T. M. Zh. Anal. Khim. 1977 32 28. Martinsen I . and Langmyhr F. J. Anal. Chim. Acta 1982, 135 137. 1027 6. 7. 8. 9. 10. Campbell W. C . and Ottaway J. M. Talanta 1974 21 837. Aggett J . and Sprott A. J. Anal. Chim. Acta 1974 72 49. Brajter K. and Kleyny K. Talanta 1985 7 521. Thomson K. C . Godden R. G . and Thomerson D. K., Anal. Chim. Acta 1975 74 289. Lagas P. Anal. Chim. Acta 1978,98 261. Paper A51349 Received September 30th) 1985 Accepted April 14th) 198

ISSN:0003-2654    

DOI:10.1039/AN9861101023

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1986, VOL. 111 1029 Determination of Arsenic(V) in Aqueous Solutions by D.c. Argon Plasma Emission Spectrometry. Interference Studies Kirnrno Srnolander” and Matti Kauppinen Department of Chemistry, University of Joensuu, P.O. Box 1 1 I , SF-80101 Joensuu 10, Finland The linear dynamic range, detection limits, accuracy and precision of the determination of As(V) in aqueous solutions by d.c. argon plasma emission spectrometry were studied at four emission lines of arsenic(\/). The interference and matrix effects of eight common acids and 12 cations [(Na(l), K(I), Mg(ll), Ca(ll), AI(III), Cr(lll), Cr(VI), Fe(lll), Co(ll), Ni(ll), Cu(ll) and Zn(ll)] on the determination of arsenic(\/) in aqueous solutions were investigated at the arsenic emission line at 193.696 nm.The linear range covered concentrations from 0.05 to 100 pg ml-1 of As(V). The RSD increased from 0.5 to 5% when the concentration of As(V) decreased from 10 to 0.75 pg ml-1 and from 15 to 50% in the range 0.5-0.05 pg ml-1 of As(V). The minimum detection limit was 0.063 pg ml-I, calculated to 30 of ten measurements of blank, as recommended by IUPAC. Keywords: Arsenic(V) determination; plasma emission spectrometry; interference studies; matrix effects Arsenic and its compounds are extremely toxic and concern about their levels in the environment has stimulated great interest in the development of analytical methods for the determination of arsenic in a variety of sample matrices. Several different plasma emission methods for the determi- nation of arsenic have been investigated by various workers.These include the direct determination of arsenic by direct current plasma emission spectrometry (DCP-AES)l.2.3 and inductively coupled plasma emission spectrometry (ICP- AES) ,4-7 and determination by ICP-AESG10 with hydride generation systems (HY-ICP-AES), microwave-induced plasma emission spectrometry and DCP-AES with hydride generation systems (HY-MIP-AESI’ 3 1 2 and HY-DCP-AES3). Extensive investigation of hydride-forming elements have been carried out by Thompson et al.879 using HY-ICP-AES. Simultaneous multi-element determinations including arsenic have been made by ICP-AES476.7 and by HY-ICP-AES.839J3 In comparison with the ICP-AES technique and the methods relying on hydride generation, the direct determination of arsenic by DCP-AES has been relatively little studied.Experimental Instrumentation A Spectrametric Spectraspan 111 single-channel plasma emission spectrometer with a d.c. argon plasma source and an Cchelle monochromator with an average resolution of 0.003 nm was used for the emission measurements. Data output was via a Hewlett-Packard 85A processor. The slit openings, as recommended by the manufxturer, had an entrance of 50 x 300 pm and an exit of 100 x 300 pm. The photomultiplier voltage used was 750 V. The argon gas flowed through the electrode sleeves and nebuliser at 50 and 30 lb in-2, respectively. The solution uptake rate was main- tained at 1.6 ml min-1 by a peristaltic pump. An integration period of 3 s was used, with three integrations per sample and standard.The spatial profile of the plasma was peaked horizontally and vertically. The optimum position in the plasma source, where the maximum net signal to background ratio is obtained, was at approximately 0.2 mm below the intersection of the three plasma legs when the plasma was at about the “-1” position specified in the manual.14 Reagents All reagents were of analytical-reagent grade. A standard stock solution of arsenic (2000 pg ml-1) was prepared by diluting an ampoule of As205 with water (Merck, Titrisol). Except for HBr, the strong acids (Merck) were diluted to 10 M solutions before being used to prepare the solutions used in the investigation. The water was distilled and de-ionised immediately before use (Milli Q system). All glassware was acid-washed before use.Results and Discussion Instrumental Parameters The most common emission line for the determination of arsenic by plasma emission spectrometry is 193.696 nm.473 This is also the strongest line in ICP-AES,16 although many other lines can and have been used.16J7 For example, Thompson et aZ.9 used the arsenic lines at 228.812 and 234.984 nm for their investigation of hydride-forming elements by HY-ICP-AES. Barnett et a1.12 selected the line at 234.984 nm for their measurement with a miniature HY-MIP-AES. Degners investigated seven emission lines of arsenic, but he reports only the detection limits, not the intensities. Urasa2 determined arsenite and arsenate by DCP-AES using the line at 197.197 nm, and Panaro and KrulP used the line at 228.812 nm with HY-DCP-AES.Important emission wavelengths and their relative intensities are tabulated in Table 1. We selected four emission lines for the investigation of arsenic(V), namely, 193.696, 197.197, 199.048 and 228.812 nm (Fig. 1). All of these were expected to be relatively free from direct spectral overlap by foreign ele- ments. Table 1. Relative intensities of common analytical wavelengths for arsenic with different plasma emission methods Relative intensity Wavelength/ nm ICP HY-ICP HY-MIP DCP 193.696 I 197.197 I 198.970 I 199.048 I 200.334 I 228.812 I 234.984 I 245.653 I Reference 83 145 108 - - 111 43 67 77 100 100 100 100 100 51 186 - - 23 - - - - - - - - - - - - - 16 9 12 Thispaper * To whom correspondence should be addressed.1030 ANALYST, SEPTEMBER 1986, VOL.111 I 0.2 nm - H Fig. 1. Spectral profiles of selected arsenic emission lines. (-), As(V) in water (2.5 pg ml-1); (---), water. (a) h = 193.696 nm; ( b ) h = 197.197 nm; (c) h = 199.048 nm; and ( d ) h = 228.812 nm Table 2. Detection limits (DL), calculated to 3a of ten measurements of blank as recommended by IUPAC, l9 background equivalent concentrations (BEC) and relative sensitivity for arsenic emission lines measured by DCP-AES. Wavelength/ DL/ BEC/ Relative 193.696 0.063 3.9 83 197.197 0.295 12.1 111 199.048 0.441 21.9 67 228.812 0.117 10.1 100 nm pg ml-1 pg ml-1 sensi tivity Detection Limit The detection limits (DL) of different wavelengths and methods differ appreciably, depending on the analytical line selected, the instrument used, the method of analysis and the form of arsenic determined. Urasazobtained DL of 20 ng ml-1 for As(II1) and 40 ng ml-1 for As(V) using DCP-AES, whereas Panaro and Krull3 found about 300 ng ml-l for both As(II1) and As(V) using the same method.The hydride generation system is much more sensitive than conventional plasma spectrometry. Oliveira et al. 13 have reported DL as low as 1 ng ml-1 for As(II1) and As(V) by HY-ICP-AES. Very low DL have been obtained by hydride generation system AAS (HY-AAS), e.g., 0.35 ng ml-l.l8 We determined the detection limit for aqueous solutions of As(V) at the line 193.696 nm using three different methods. The first method was to calculate three times the standard deviation of ten measurements of blanks, as recommended by IUPAC.19 The second method involved the measurement of the method detection limit for the 99% confidence level1 and the third was based on the equation DL = 3(RSD) (I,,/Za)Ca, where Ib and I, are the mean background analyte intensities and RSD the relative standard deviation of n measurements of the analyte at a concentration Ca.20 The DL values for the different methods were 0.063, 0.088 and 0.079 yg ml-l, respectively.The reproducibility of the detection limits varied from day to day by k30%, as calculated by the first method. Calibration Graphs and Precision The linearity and the analyte concentration range of the As(V) determination was investigated at the four emission lines with standards of 0, 1, 5 , 10, 50 and 100 pg ml-1. Not surprisingly, the 197.197 nm line gave the best intensity and sensitivity.16,17 According to the “Tables of Spectral Line Intensities, Part 1,”17 the line at 199.048 nm should have had better intensity I I Fig. 2. Spectral profiles of arsenic emission lines at 193.696 nm. (-), As(V) in water (2.5 pg ml-l); (. * .), HOAc (a) or HF (b), both 5 M; and (---), water and sensitivity than those recorded here. However, the values in the Tables are based on arc emission spectrometry, and in the corresponding table of Boumans,l6 where the sensitivities are corrected for ICP-AES, the line is one of the weakest. The calibration graphs of the three most sensitive lines (197.197, 228.812 and 193.696 nm, respectively) are linear from 0 to 100 pg ml-1. The weakest line (199.048 nm) consistently curves upwards. The background equivalent concentrations (BEC) calculated for different concentrations are given in Table 2 together with the sensitivities of the four lines.The emission line at 193.696 nm was selected for more detailed study in this investigation because it had the lowest detection limit and good sensitivity. The linearity, precision and accuracy of the selected line were studied with three standard series (0.05-2.5, 5-75 and 100-400 pg ml-1). The linear range covered 0.05 to 100 pg ml-1. Above 100 pg ml-1, the deviations from linearity began to increase, although the relative standard deviation values (RSD) were only about 0.5%. The deviations were less than 0.02 and 0.08 pg ml-1 in the first and second standard series, respectively. The precision, however, was significantly decreased at the lower concentration levels: the RSD values increased from 0.5 to 5% when the concentration decreased from 10.0 to 0.75 pg ml-1 and from 15 to 50% in the range 0.5-0.05 pg ml-1.Effects of Acid Concentration In order to avoid errors in the determination of As(V), it is necessary to match the total acid content of the samples and standards as closely as possible. Viscosity changes produced by variations in the total acid content affect sample transport properties and therefore the analytical sensitivities? Other factors, such as aerosol transport losses and changes in the excitation conditions in the plasma, also play a role.15 The effects of eight acids (HF, HC1, HBr, HN03, HC104, H2SO4, H3P04 and HOAc) on determinations of 2.5 yg ml-1 As(V) solutions were studied with pure acids at concen- trations of up to 5 M.Only two of the acids has a significant background effect. HF showed an emission line at the same wavelength as As(V) [Fig. 2(b)], and HOAc gave many lines, as can be seen in Fig. 2(a), and the background level was higher than with other acids. These two acids caused the largest signal enhancements of As(V); they increased the intensity of the As line in an almost linear manner in up to 5 M acid solutions, in which the intensities were +60 and +110% higher, respectively, than in an aqueous solution. A linear correlation between intensity and acid concentration was also observed for HBr, the increase being +30% in 5 M acid solutions (Fig. 3). In 0.1 M H2SO4 and 0.5 M HC104 solutions the intensity was decreased by about -4 and -lo%, respectively, from where it decreased linearly with increasing acid concentration to -30% for H2SO4 and remained approximately constant for HC104.The H3P04 solutions initially increased the intensity slightly, with a maximum at about 1.5 M, and then there was a linear decrease to - 18%. The deviations in HCl and HN03 solutions remained approximately constant at about - 2 to -5% lower than in water solutions.ANALYST, SEPTEMBER 1986, VOL. 111 1031 80 - - 40 0 1 2 3 4 5 Acid concentration/mol I-’ Fig. 3. Effect of different acids on 2.5 vg ml-1 of As(V) in water at the emission line at 193.696 nm. (a) A, HOAc; B, HF; C, HBr; and D, HC1. (b) E, H3P04; F, HNO,; G, H,SO,; and H, HC104 Our results are in agreement with the slight suppressing effect of HN03 and HC1 and the very strong suppressing effect of H2SO4 reported for the determination of As by ICP- AES,15,21 and also with the acid effect of HC104 reported in a multi-element determination by ICP-AES.7 A h I I I I 1 0 1 2 3 4 Log(Cr concentration/vg ml-1) Fig.4. at the emission line at 193.696 nm. A, Cr(IIfi; and B, Cr(V1) Effect of Cr(II1) and Cr(V1) on 2.5 p ml-1 of As(V) in water Effect of Foreign Elements The 2.5 pg ml-l As(V) solution was spiked with 10-5000 pg ml-1 of different nitrate salts [except Cr, which was added as Cr(II1) chloride and Cr(V1) oxide]. The cations Mg(II), Ca(II), Al(III), Fe(III), Co(II), Ni(II), Cu(I1) and Zn(I1) increased the intensity of the As(V) line oniy slightly: deviations in 1000 pg ml-1 solutions were less than 4%, whereas in solutions of 5000 pg ml-1, Na and K caused deviations of less than 5%, Mg, Ca, Ni, Cu and Zn deviations of about 10% and Al, Fe and Co deviations of about 24,19 and 14y0, respectively.The greatest effect was caused by Cr(II1) chloride and Cr(V1) oxide, both of which enhanced the intensity linearly up to 500 pg ml-1, where the deviations were about 14 and 8%, respectively. In 1000 and 5000 pg ml-1 solutions the intensities were 23, 81 and 15, 61% stronger, respectively (Figs. 4 and 5). Literature values for the spectral interference effects of 1000 gg ml-1 of A1 as determined by ICP-AES are larger than the values in this study.6.7 according to Tao et LzZ.,~ 1000 pg ml-l of A1 are equivalent to 23.9 pg ml-1 of arsenic at the emission line 193.696 nm, however, we observed only a slight increase in deviation with 1000 yg ml-1 and a +24% deviation with 5000 yg ml-1 of A1 at the same wavelength.Other elements reported to cause an interference effect are Fe, Mg and Ti. Solutions of 1000 pg ml-1 of these elements were found to be equivalent to 2.92,0.206 and 0.400 pg ml-1 of As, respectively.6 Degners investigated the interference effects of Fe, Cr and Cu at seven different emission lines of arsenic by ICP-AES. The maximum concentrations of these elements that caused no interference effects were 10, 100 and 1000 pg ml-1 for Fe, Cr and Cu, respectively. In our study, the deviations of arsenic with 1000 pg ml-1 of Fe, Cr(II1) and Cr(V1) added were about +4, +23 and +15 pg ml-1, whereas no effect was found with Cu addition.Elements such as Ca, Mg, Na and K often cause a stray light effect in ICP-AES6 and DCP-AES.1.22.23 We observed a Fig. 5. Spectral profiles at 193.696 nm of 2.5 p ml-1 As(V) in water (---) and of 5000 pg ml-* Cr(V1) in water (* . .! after subtracting the signal due to water deviation of only about 3% for 1000 pg ml-1 solutions, and of about +10 and +5% deviation for 5000 pg ml-1 solutions of Ca, Mg and Na, K, respectively. The foreign elements have less interference effects in our study by DCP-AES than in the reported studies by ICP-AES. This is due to the very good resolution of the Cchelle monochromator that provides a two-dimensional spectral pattern with an average resolution of 0.003 nm.14 Conclusions The As(V) emission line of 193.696 nm gave the lowest background and detection limit and good sensitivity.The acids HOAc and HF caused strong spectral interference and increased intensities the most; HCl and HN03 had the least effect. Among the cations, Cr(II1) and Cr(V1) had the greatest enhancing effects. The stray light effects of Ca, Mg, Na and K, and the matrix effects of Al, Fe, Co, Ni, Cu and Zn, were almost non-existent at 1000 pg ml-1 concentrations. The precision decreased significantly below As( V) concen- trations of 0.75 pg ml-1, which limits the usefulness of the direct method. Sensitivities are considerably better for envi- ronmental samples using the hydride generation system in AAS and AES. At high concentrations of arsenic the DCP-AES method without hydride generation has the advantage of allowing the simultaneous determination of other elements.We express our thanks to Professor Heikki Hyvarinen of the Department of Biology, University of Joensuu, for placing the Spectraspan I11 instrument at our disposal.1032 ANALYST, SEPTEMBER 1986, VOL. 111 1. 2. 3. 4. 5. 6 . 7. 8. 9. 10. 11. 12. 13. 14. References Dellefield, R. J., and Martin, T. D., At. Spectosc., 1982,3, 165. Urasa, I. T., Anal. Chem. 1984, 56, 904. Panaro, K. W., and Krull, I. S . , Anal. Lett., 1984, 17, 157. Que Hee, S. S . , Macdonald, T. J., and Boyle, J. R., Anal. Chem., 1985,57, 1242. Degner, R., Fresensius 2. Anal, Chem., 1982, 311, 94. Tao, H., Iwata, Y., Hasegawa, T., Nojiri, Y., Haraguchi, H., and Fuwa, K., Bull. Chem. SOC. Jpn., 1983, 56, 1074. McQuaker, N.R., Kluckner, P. D., and Chang, G. K., Anal. Chem., 1979,51,888. Thompson, M., Pahlavanpour, B., Walton, S. J., and Kirk- bright, G. F., Analyst, 1978, 103, 705. Thompson, M., Pahlavanpour, B., Walton, S. J., and Kirk- bright, G . F., Analyst, 1978, 103, 568. Nahakara, T., Anal. Chim. Acta, 1981, 131,73. Lichte, F. E., and Skogerboe, R. K., Anal. Chem., 1973,45, 399. Barnett, N. W., Chen, L. S . , and Kirkbright, G. F., Spectro- chim. Acta, Part B, 1984, 39, 1141. Oliveira, E., McLaren, J. W., and Berman, S. S . , Anal. Chem. 1983,55,2047. “Instruction Manual and Handbook,” Spectrametrics, Andover, MA, 1982. 15. 16. 17. 18. 19. 20. 21. 22. 23. Botto, R. I., Spectrochim. Acta, Part B , 1985, 40, 397. Boumans, P. W. J. M., Spectrochim. Acta, Part B , 1981, 36, 169. Meggers, W. F., Corliss, C. H., and Scribner, B. F., “Tables of Spectral Line Intensities, Part I-Arranged by Elements,” Second Edition, National Bureau of Standards, Washington, DC, 1975. Florino, J. A., Jones, J. W., and Capar, S. G., Anal. Chem., 1976, 48, 120. Nakahara, T., and Kikui, N., Spectrochim. Acta, Part B, 1985, 40, 21. Hulmston, P., Jefferies, A. C., and Davies, J. A., Talanta, 1984, 109,519. Schramel, P., and Ovcar-Pavlu, J., Fresensius 2. Anal. Chem., 1979, 298, 28. Johnson, G. W., Taylor, H. E., and Skogerboe, R. K., Spectrochim. Acta, Part B, 1979, 34, 212. Johnson, G . W., Taylor, H. E., and Skogerboe, R. K., Appl. Spectrosc., 1979, 33, 451. Paper A6133 Received February 4th, 1986 Accepted April lst, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101029

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1986, VOL. 111 1033 Study of Organic Interferences in the Spectrophotometric Determination of Nitrite Using Composite Diazotisation - Coupling Reagents George Norwitz and Peter N. Keliher" Chemistry Department, Villanova University, Villanova, PA 19085, USA A study was made of organic interferences in the spectrophotometric determination of nitrite by the diazotisation - coupling technique using three composite reagents [sulphanilamide and N41- naphthy1)ethylenediamine (NED); sulphanilic acid and NED; and 4-nitroaniline and NED]. Many organic substances interfere, usually causing low results. The interference is usually less with the 4-nitroaniline - NED and sulphanilamide - NED methods than with the sulphanilic acid - NED method. The interferents tested included aliphatic amines (primary, secondary and tertiary), aromatic amines (primary, secondary and tertiary), various phenolic compounds and miscellaneous organic compounds (sucrose, dextrose, lactic acid, succinic acid, acetamide, acetanilide, ethylenediamine tetraacetate, cholesterol, rennin, dodecyl sodium sulphate, acetophenone, urea, citric acid, caffeine, saccharin, morpholine, L-asparagine, gelatin, benzoic acid, formaldehyde, cinchonine, nicotinic acid, trypsin, creatine, starch, albumin, gum tragacanth, casein, formic acid, sorbic acid, ascorbic acid and acetaldehyde).The effect of detergents and soap was also examined. Large amounts of water-miscible solvents (methanol, ethanol, acetone and glycerin) can be tolerated. Water-immiscible solvents do not affect the colour.Keywords: Nitrite determination; organic interferences; composite diazotisation - coupling reagents; spectrophotometry The authors have previously formulated conditions for the spectrophotometric determination of nitrite using three com- posite diazotisation - coupling reagents [sulphanilamide and N-( 1-naphthy1)ethylenediamine (NED); sulphanilic acid and NED; and 4-nitroaniline and NEDl.1 Recently, they have studied inorganic interferences in the methods.2 The purpose of the work reported here was to study organic interferences. The problem of organic interferences is important because nitrite is frequently determined in the presence of organic materials, as in the characterisation of waters, wastes, food and chemical processes. No comprehensive study of organic interferences in the spectrophotometric determination of nitrite by the diazotisation - coupling technique has been made and even qualitative data are ~ c a n t .3 . ~ Experimental Reagents The organic compounds used in this investigation were purchased from Eastman Kodak or Aldrich Chemical. The detergents and soap were ordinary commercial products. Standard nitrite solution A , 1 ml = 100 pg of N02-N. Sodium nitrite (0.4926 g) was dissolved in water and diluted to 1 1 in a calibrated flask. Sodium nitrite solution B, 1 ml = 1.00 pg of N02-N. This was prepared fresh daily from standard nitrite solution A. Sulphanilamide - NED reagent. Sulphanilamide (2.5 g) was dissolved in 650 ml of 1 M hydrochloric acid, 30 ml of 0.20% NED solution were added and the solution was diluted to 1 1.Sulphanilic acid - NED reagent. Sulphanilic acid (5.00 g) was dissolved in a mixture of 750 ml of water and 35 ml of 1 M hydrochloric acid by heating, the solution was cooled to room temperature, 25 ml of 0.20% NED solution were added and the solution was diluted to 1 1. 4-Nitroaniline - NED reagent. 4-Nitroaniline (2.50 g) was dissolved in 300 ml of sulphuric acid (P + 1) by heating, the solution was cooled to room temperature, 50 ml of 0.20% NED solution were added and the solution was diluted to 1 1. * To whom correspondence should be addressed. The NED solution, sulphanilamide - NED and sulphanilic acid - NED reagents were stored in brown bottles in a refrigerator. The 4-nitroaniline - NED reagent was stored in a brown bottle at room temperature. Preparation of Calibration Graphs Appropriate aliquotsl of standard nitrite solution B were transferred into 50-ml calibrated flasks, 10 ml of composite reagent were added from a graduated cylinder, the volume was diluted to the mark and the solution was mixed.The absorbance was measured against the reagent blank at the following wavelengths after the following times: sulphanil- amide - NED, 542 nm and 15 min; sulphanilic acid - NED, 541 nm and 30 min; and 4-nitroaniline - NED, 542 nm and 10 min. The absorbance was plotted against micrograms of N02-N per 50 ml. Study of Organic Interferences Standard nitrite solution B (5 ml) and water were added to 50-ml calibrated flasks. Various amounts of interferent solution were then added and the solution was allowed to stand for about 10 min.Composite reagent (10 ml) was added, the solution diluted to the mark and the absorbance measured as described under Preparation of Calibration Graphs. After consideration of the many analyses for each interferent for each of the three composite reagents, the tolerance limit for the interferent (the limit beyond which the interferent produced an error greater than 0.015 absorbance unit) and the effect of the interferent beyond the tolerance limit were established. Results and Discussion Aliphatic Amines The results for the interference of aliphatic amines, aromatic amines, phenolic compounds, miscellaneous organic com- pounds and typical detergents and soap are shown in Table 1. Aliphatic amines can cause low results with all three reagents.The tolerance limit for the aliphatic amines with the1034 ANALYST, SEPTEMBER 1986, VOL. 111 Table 1. Tolerance limits for organic interferences. ide - NED = sulphanilamide - NED; ic - NED = sulphanilic acid - NED; 4N - NED = 4-nitroaniline - NED Tolerance limit/mg per 50 ml Effect beyond tolerance limit? Interferent Added as* Aliphatic amines: Methylamine . . . . . . 4% W soln. Ethylamine . . . . . . . . 5% W soh. Dimethylamine . . . . . . 5% W soln. Diethylamine . . . . . . 5% W soln. Trimethylamine . . . . . . 2.4% W soh. Triethylamine . . . . . . 1% W soln. Aromatic amines: An i 1 in e . . . . . . . . 4-Chloroaniline . . . . . . 2,4-Dichloroaniline . . . . Anthranilic acid (2-aminobenzoic acid) . . Anthranilamide (2-aminobenzamide) .. . . Naphthionic acid (Camino-l-naph t hal- enesulphonicacid) . . . . N,N-Dimethylaniline . . . . N-Methylaniline . . . . Phenolic compounds: Phenol . . . . . . . . Salicylicacid . . . . . . Resorcinol . . . . . . . . 3,4-Xylenol . . . . . . . . Pyrogallol . . . . . , . . Carvacrol . . . . . . . . Vanillin . . . . . . . . 1 -Naph tho1 . . . . . . . . 1 Yo M soln. 1% M soln. 1 YO M soh. 0.25% M soh. 0.25% M soln. 0.25% M soh. 5% M soh. 5% M soh. 2% W soln. 0.5% M soln. 0.2% W soh. 0.2% M soln. 0.2% W soln. 0.2% M soln. 0.2% M soln. 0.2% M soln. Miscellaneous organic compounds: Sucrose . . . . . . . . 10% W soh. Dextrose [(+)-glucose] . . . . 10% W soh. Lactic acid . . . . . . . . 10% W soh. Succinic acid . . . . . . 5% W s o h Acetamide . . . . .. . , 10% W soh. Acetanilide . . . . . . . . 0.5% W soln. Ethylenediaminetetra- acetate (Na,salt) . . . . 1% W soh. Cholesterol . . . . . . . . 2% W soln. Rennin . . . . . . . . 1% W soh. Dodecyl sodium sulphate . . 2% W soh. Acetophenone . . . . . . Acetophenone Urea . . . . . . . . . . 5% Wsoln. Citric acid . . . . . . . . 5% W soh. Caffeine . . . . . . , . 2% W soln. Saccharin . . , . . . . . 0.4% W soln. Morpholine . . . . . . . . 1 YO W soln. L-Asparagine . , . . , , 0.5% W soln. Gelatin . . . . . . . . 1% W soh. Benzoic acid . . . . . . 0.1 YO W soh. Formaldehyde . . . . . . 0.5% W soh. Cinchonine . . . . . . . . 0.05% W soln. Nicotinic acid . . . . . . 0.1 YO W soh. Trypsin . , . . . . . . 0.02% W soh. Creatine hydrate . . . . . . 0.1% W soln. Starch (potato) .. . . . . 0.5% W soh. Albumin (egg) . . . . . . 0.1% W soh. Gum tragacanth . . . . . . 0.1 YO W soh. Casein . . . . . . . . . . 0.1 YO in 0.01 M KOH Formic acid . . . . . . . . 0.05% W soln. Sorbic acid . . . . . . . . 0.05% W soh. Ascorbic acid . . . . . . 0.05% W soln. Acetaldehyde . . . . . . 0.05% W soln. ide - NED ic - NED 100 150 30 150 25 100 10 10 10 1 .o 7.5 0.2 50 100 400 50 1.5 2 0.05 3 10 0.2 3000 3000 1500 1500 1000 150 300 600 300 400 300 500 50 60 75 50 20 15 2.0 3 15 10 6 3 12.5 12.5 5 5 0.02 0.1 0.2 0.02 3 10 2 5 10 15 5 5 5 0.3 2 0.15 10 10 100 50 1.5 0.5 0.05 2 5 0.2 2000 2000 500 1500 1000 150 200 600 300 300 300 300 30 30 75 50 10 15 1 .o 3 5 10 6 3 10 10 5 5 0.02 0.1 0.1 0.02 4N - NED 800 900 900 1500 150 300 100 50 100 10 50 0.4 200 150 100 75 5 0.2 0.05 3 15 0.2 3000 3000 2000 1500 1500 150 300 600 300 400 300 50 400 200 75 75 50 75 2.5 40 2.5 5 6 5 25 12.5 5 5 0.05 0.05 0.2 0.1 ide - NED Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low --j: Low Low - - - - - - Low 9 Low Low Low Low Low Low Low Low Low Low Low High (colloid) High (colloid) High (colloid) High (colloid) 7 Low Low Low Low - - ic - NED Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low - - Low - - Low 5 Low Low Low Low Low Low Low Low Low High Low Low High (colloid) High (colloid) High (colloid) High (colloid)7 Low Low Low Low - 4N - NED Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low Low - - Low Low - - - - Low 0 Low Low Low Low Low Low Low Low Low High Low Low High (colloid) High (colloid) High (colloid) High (colloid) 7 Low Low Low Low -ANALYST, SEPTEMBER 1986, VOL.111 Table l-continued 1035 Interferent Detergents and soap: ‘‘All’’ detergent (Lever Brothers) . “Sparkleen” detergent “Ivory” soap (Fisher Scientific) . . . . (Proctor and Gamble) . . Tolerance limithg per 50 ml Effect beyond tolerance limitt Added as* ide - NED ic - NED 4N - NED ide - NED ic - NED 4N - NED 0.1% W soln. 10 10 10 High, High, High, low low low (colloid) (colloid) (colloid) 0.1% W soln. 10 111 111 Low Low Low 0.1% W soln. 0.03)) 0.0311 0.0311 High, High, High, low (ppt.) low (ppt.) low (ppt.) * W = Water solution and M = methanolic solution; all solutions mlV. t “Low” refers to a negative absorbance error greater than 0.015 in the presence of 5.00 pg of N02-N and “high” refers to a positive absorbance error greater than 0.015 in the presence of 5.00 pg of N02-N.$ -, The effect beyond the indicated tolerance limit could not be established because of the limited solubility of the interferent in the additive solution. 0 Larger amounts did not affect the colour but caused difficulty because of floating droplets. 7 Precipitate also formed. 11 See text. three reagents increases markedly in the order sulphanilic acid - NED, sulphanilamide - NED and 4-nitroaniline - NED and is surprisingly high for the 4-nitroaniline - NED reagent. In the sulphanilamide - NED and 4-nitroaniline - NED methods (but not the sulphanilic acid - NED method), more primary and secondary aliphatic amine can be tolerated than tertiary aliphatic amine.In all the methods, larger amounts of the ethylamines (primary, secondary and tertiary) can be toler- ated than the corresponding methylamines. The interference from aliphatic amines is probably mainly caused by the reaction of the aliphatic amine with the nitrite after the addition of the acidic composite reagent. In neutral solution, primary, secondary and tertiary aliphatic amines would merely form amine nitrite salts with the nitrite.5 However, in the presence of acid, primary aliphatic amines would react with nitrite to produce nitrogen, secondary aliphatic amines would react to produce nitroso compounds and tertiary aliphatic amines would still only form amine nitrite salts.5 It would seem from these reactions that tertiary aliphatic amines would interfere less than primary and secondary aliphatic amines but, as indicated above, this does not happen in most instances.It is understandable why the ethylamines (primary, secondary and tertiary) would interfere less than the corre- sponding methylamines, as the ethylamines contain a lower percentage of nitrogen. The 4-nitroaniline - NED method is clearly the method of choice for the determination of nitrite in the presence of aliphatic amines. Aromatic Amines Aromatic amines can cause low results in all three methods and the interference is moderate or strong depending on the particular aromatic amine and the method. The tolerance limit for aromatic amines with the three reagents increases in the order sulphanilic acid - NED, sulphanilamide - NED and 4-nitroaniline - NED, Primary aromatic amines interfere more than secondary and tertiary aromatic amines in all the methods.There is no easy explanation for this. In neutral solution, primary, secondary and tertiary aromatic amines would be expected to form amine nitrite salts with the nitrite (although for the secondary and tertiary aromatic amines this salt formation would be limited by the low solubility of these amines in water). In acidic solution, primary aromatic amines would react with nitrite to form diazonium salts, secondary aromatic amines would react to form N-nitroso compounds and tertiary aromatic amines would react to form p-nitroso compounds.5 In the methods, it might be expected that the interfering primary aromatic amines would be diazotised and then coupled with the NED, but this is by no means certain.Aromatic amines could possibly interfere by acting as coupling agents in place of the NED. However, in the 4-nitroaniline - NED and sulphanilamide - NED methods such coupling seems unlikely (it is known that aromatic amines can act as coupling agents only in neutral or very weakly acidic soh- tions6). The 4-nitroaniline - NED method is the method of choice for the determination of nitrite in the presence of primary, secondary and tertiary aromatic amines. The prob- lem of the determination of nitrite in the presence of primary aromatic amines is encountered in the analysis of wastes from industrial diazotisation processes. Phenolic Compounds Phenolic compounds can cause low results with all three methods.Phenol shows only moderate interference whereas substituted phenols usually produce strong interference. Generally, the interference from phenolic compounds is less with the 4-nitroaniline - NED and sulphanilamide - NED methods than with the sulphanilic acid - NED method. Probably the most important cause of the interference of phenolic compounds is the nitrosation or oxidation of the phenolic compounds by the nitrite after the addition of the composite reagent. It is known that such reactions readily take place in acidic solutions.5 As would be expected, the interference is especially strong with a compound such as 1-naphthol that is easily nitrosated or a compound like pyrogallol that is easily oxidised.It is not believed that phenolic compounds interfere by acting as coupling agents in place of the NED (phenolic compounds can act as coupling agents only in neutral or alkaline solution.6) Either the 4-nitroaniline - NED or sulphanilamide - NED method is the method of choice for the determination of nitrite in the presence of phenolic com- pounds. Miscellaneous Organic Compounds The extent of the interference from miscellaneous organic compounds is varied. Low results are usually produced by an excess of the interferent. Large amounts (greater than 150 mg per 50 ml) of sucrose, dextrose, lactic acid, succinic acid, acetamide, acetanilide, ethylenediaminetetraacetate (diso- dium salt), cholesterol, rennin, dodecyl sodium sulphate and acetophenone can be tolerated in all the methods.Large amounts of urea can be tolerated in the sulphanilamide - NED and sulphanilic acid - NED methods and large amounts of1036 ANALYST, SEPTEMBER 1986, VOL. 111 Table 2. Interference of water-miscible solvents. Abbreviations etc. as in Table 1. Tolerance lirnithl per 50 ml Effect beyond tolerance limit Solvent ide - NED ic - NED 4N - NED ide - NED ic - NED 4N - NED Methanol . . . . 5 5 20 High Low Low Ethanol . . . . 10 10 20 High Low Low Acetone . . . . 10 10 25 High Low Low Glycerin . . . . 10 10 20 Low Low Low citric acid and caffeine can be tolerated in the 4-nitroaniline - NED method. Moderate amounts (15-75 mg per 50 ml) of saccharin, morpholine, L-asparagine and gelatin can be tolerated in all the methods. Small amounts (usually several milligrams per 50 ml) of benzoic acid, formaldehyde, cincho- nine, nicotinic acid, trypsin, creatine hydrate, starch, albu- min, gum tragacanth and casein can be tolerated in all the methods. Trace amounts (in some instances less than 0.1 mg per 50 ml) of formic acid, sorbic acid, ascorbic acid and acetaldehyde can cause low results in all the methods. Formic acid, sorbic acid, ascorbic acid and acetaldehyde are all used as food additives,’ so the tolerance limits should be of practical interest.In general, it is believed that the cause of the low results from most of the miscellaneous organic compounds is nitro- sation or oxidation of the compounds by the nitrite, particu- larly after the addition of the composite reagent. However, a reaction between the organic compound and the diazotised aromatic amine of the composite reagent or even a reaction between the organic compound and the dye cannot be ruled out, The conditions and mechanism involved in the oxidation of organic compounds are different from those involved in the oxidation of inorganic compounds.The high results caused by starch, albumin and gum tragacanth are due to the colloidal nature of the solutions of these substances. Reasonably accurate results can be obtained in the presence of several times the recommended limits for these substances by making the absorbance measurements against a blank solution not containing the composite reagent. Casein interferes because of the colloidal nature of its solution and because a precipitate is formed on adding the composite reagent.Satisfactory results cannot be obtained by filtering solutions containing starch, albumin, gum tragacanth and casein. Generally, the 4-nitroaniline - NED and sulphanilamide - NED methods will tolerate greater amounts of the miscellaneous organic com- pounds than the sulphanilic acid - NED method. However, the interference from urea is much greater with the 4-nitroaniline - NED method than the other two methods (probably because of the greater acidity of the 4-nitroaniline - NED reagent). Detergents and Soap The interference from detergents and soap is diverse. As indicated in Table 1, the maximum limit for “All” detergent is 10 mg per 50 ml for all the methods and the maximum limit for “Sparkleen” detergent is 1 mg per 50 ml for all the methods.When 10 mg of “Sparkleen” per 50 ml were present, the recoveries in the sulphanilamide - NED, sulphanilic acid - NED and 4-nitroaniline - NED methods were 91,22 and 8l%, respectively. The maximum limit for “Ivory” soap is 0.03 mg per 50 ml for all the methods. When more than this amount is present, a curdy precipitate (stearic acid) is produced on adding the composite reagent. However, if the precipitate is filtered off after adding the composite reagent, up to 3 mg of the soap can be tolerated. More than 3 mg of the soap causes incomplete colour development in all the methods. Either the 4-nitroaniline - NED or sulphanilamide - NED method is recommended for the determination of nitrite in the presence of detergents and soap. Solvents The interference from water-miscible solvents is shown in Table 2.It can be seen that large amounts of water-miscible solvents can be tolerated and the tolerance limit for the solvents tested increased in the order glycerin, methanol, ethanol and acetone. The tolerance limits for water-miscible solvents is greater for the 4-nitroaniline - NED method than the other two methods. The effect of water-immiscible solvents (chloroform, hexane, toluene, ethyl acetate and diethyl ether) was also tested. These solvents were found not to affect the colour or extract it and do not interfere if they are separated from the sample solution by use of a separating funnel before or after the development of the colour. For all the organic substances tested in this work the recovery of nitrite obtained at the recommended limits for the organic interferents was approximately 95 YO and the reprodu- cibility at these limits was good.On adding an increasing amount of interferent beyond the recommended limit, the error for the recovery of nitrite tended to increase gradually but the results became somewhat erratic (this would be expected by the nature of the interference). The rate of increase of the error for the recovery of nitrite on adding the increasing amount of interferent varied for the 61 organic substances for the three methods. Organic Compounds in Distilled Water We conducted almost all of our work using reagent-grade (Fisher Scientific) distilled water. However, we also per- formed some work using water that is distilled for the Villanova University general chemistry laboratories by means of a high-capacity still.We found that the absorbance obtained for 5 pg of N02-N was consistently lower (by about 0.02 absorbance unit) when using the Villanova distilled water. This necessitated the preparation of a different calibration graph. We believe that the cause of the low results is organic matter, which is not removed in a large-scale distillation. The blank obtained using this water was insignificant and the pH was about the same as that of reagent water, ca. 6.5. The water distilled on a large scale did not contain detectable amounts of inorganic materials and boiling it (to remove gases) made no difference in the results for nitrite. The standard permanga- nate test for organic matter, which consists of acidification with sulphuric acid, addition of potassium permanganate and noting any decrease in colour after a few minutes standing, is not sufficiently sensitive to detect small amounts of organic matter. References 1. 2. Norwitz, G., and Keliher, P. N., Analyst, 1984, 109, 1281 Norwitz, G., and Keliher, P. N., Analyst, 1985, 110, 689.ANALYST, SEPTEMBER 1986, VOL. 111 1037 3. 4. Williams, W. J., “Handbook of Anion Determination,” 1972, pp. 10, 66, 330 and 783. 5. Boltz, D. F., and Howell, J . A., “Colorimetric Determination of Nonmetals,” Wiley, New York, 1978, pp. 216-220. Butterworths, London, 1979, pp. 147-151. Noller, C . R., “Chemistry of Organic Compounds,” Third Edition, W. B. Saunders, Philadelphia and London, 1965, pp. 261-262,529-530 and 554. Fierz-David, H. E., and Blangey, L., “Fundamental Processes of Dye Chemistry,” Interscience, New York, 1949, pp. 7. National Research Council, “Food Chemicals Index,” Second Edition, National Academy of Sciences, Washington, DC, 6. 249-257. Paper A 61 77 Received March 7th, I986 Accepted April 29th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101033

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, SEPTEMBER 1986, VOL. 111 1039 Spectrophotometric Determination of Some Pharmaceutical Carbonyl Compounds Through Oximation and Subsequent Charge-transfer Complexation Reactions Saied Belal* College of Pharmacy, Department of Pharmaceutical Analytical Chemistry, Alexandria University, Alexandria, Egypt Afaf A. El Kheir, Magda M. Ayad and Sobhi A. Al Ad1 College of Pharmacy, Department of Pharmaceutical Chemistry, Zagazig University, Zagazig, Egypt Two spectrophotometric procedures are presented for the determination of six pharmaceutical carbonyl compounds through charge-transfer complexation of their oximes with either the a-acceptor iodine or the n-acceptor choranil. The optimum assay conditions and their applicability to the determination of the test compounds in pharmaceutical products are described.The results obtained, which compared favourably with those given by a pharmacopoeial method, illustrate the accuracy, sensitivity and simplicity of the developed procedures. Keywords: Pharmaceutical carbonyl compound determination; charge-transfer complexation; oximation; spectrophotometry Progesterone, testosterone propionate and nandrolone phenylpropionate are A4-3-ketosteroid drugs in common use in therapeutics as progestational, androgenic and anabolic agents, respectively. A number of methods have been described for their determination including titrimetric,l col- orimetric,2-7 spectrophotometric,8 polarographic9J0 and chromatographicll-13 procedures. The pharmacopoeial method of analysis for their injectable solutions uses the isoniazide colorimetric14 method.Griseofulvin is a potent antifungal antibiotic; carvone and menthone (constituents of oil of caraway and peppermint, respectively) are used as carminatives and flavouring agents in pharmaceutical preparations. The latter is also used as an antipruritic agent. A review of the assay methods for griseofulvin, carvone and menthone, including titrimetric, spectrophotometric, electrochemical and chromatographic procedures, has been described by the authors,15 and methods of analysis for carbonyl compounds have been surveyed.16 Official compendia describe several different procedures for the determination of carbonyl compounds in volatile oils, including the acid - base titrimetric - hydroxylamine methods of the BPI7 and EP18 and the USPI9 method using the neutral sulphate.Charge-transfer complexation reactions have been exten- sively used for the determination of electron-donating basic nitrogenous compounds using as reagents the a-acceptor iodine or polyhalo or polycyano quinone n-acceptors in organic solvents.2G29 The introduction of a Schiff's base moiety from the reaction of carbonyl compounds with substituted ammonia or hydrazines was first used by the authors30 in the determination of corticosteroid drugs through solvent extrac- tion and charge-transfer complexation of their phenylhydraz- ones. The authors have used the oximation reaction for the indirect titrimetric and/or nitrite - diazo coupling spectropho- tometric determination of carbonyl drugs.7J5 Until now, the analytical use of charge-transfer complexation reactions of oximes or carbonyl compounds amenable to oximation has not been reported.This paper describes the development of a spectrophotometric analysis of ketonic drugs through their oxime formation, followed by solvent extraction and reaction with electron acceptor reagents. * Present address: College of Medicine and Allied Sciences, King Abdulaziz University, Jeddah, Saudi Arabia. Experimental Reagents and Materials All chemicals used were of analytical-reagent grade; solvents were of spectroscopic grade. Stock standard solutions of the drugs were prepared by accurately weighing 100 mg of the cited drugs into 100-ml calibrated flasks, dissolving in absolute ethanol and diluting to volume with the same solvent. Working standard solutions of the drugs were prepared by diluting the stock standard solutions with ethanol to give solutions of 0.1 mg ml-1 concentration.Hy droxylammonium chloride solution, 5% mlV in 60% ethanol. a-Acceptor, 2 x 10-4 M iodine solution in chloroform. n-Acceptor, 2 X 10-4 M chloranil solution in chloroform. Drugs. The drugs determined were progesterone, testo- sterone propionate and nandrolone phenylpropionate powders, griseofulvin powder, carvone and menthone (Aldrich), oil of caraway and oil of peppermint. Instrument A double-beam spectrophotometer (Uvidec-320) with 1-cm quartz or glass cells was used. Procedures Oximation step To an ethanolic solution equivalent to 10 mg of the drug were added 4 ml of hydroxylammonium chloride reagent and a drop of glacial acetic acid, and the mixture was refluxed on a water-bath for 1 h (A4-3-ketosteroids), or 20 min (other ketones).The reaction mixture was then evaporated to dryness at 70 "C and the residue was transferred into a 100-ml separating funnel with the aid of 10 ml of each of chloroform and water. The mixture was then shaken and the chloroform layer separated and transferred into a 100-ml calibrated flask. Extraction was continued with six successive 10-ml portions of chloroform, which were collected in the same calibrated flask. The extract was diluted to volume with chloroform and then subjected to either Procedure A or Procedure B. Colour development Procedure A. A 0.5-1-ml aliquot of the ketoxime extract was transferred into a 10-ml calibrated flask, treated with 2 ml1040 ANALYST, SEPTEMBER 1986, VOL.111 of iodine solution and allowed to stand for 30 min, before diluting to volume with chloroform. The absorbance of the solution was measured at 300 nm (ketosteroids) or 295 nm (griseofulvin, carvone and menthone) in a I-cm cell against a reagent blank. Procedure B. A 1-ml aliquot of the extract was transferred into a 10-ml calibrated flask, treated with 5 ml of chloranil solution and allowed to stand for 45 min at room temperature, or placed in a water-bath at 45 “C for 15 min, cooled and then diluted to volume with chloroform. The absorbance of the solution was measured at 430 nm (ketosteroids) or 440 nm (other ketones) in a 1-cm cell against a reagent blank. The concentration of the drug was calculated from the calibration graph obtained by applying either procedure A or B to standard solutions equivalent to 1-5 mg of drug per 100 ml, or from the corresponding linear equation describing the calibration graph.Application to A4-3-ketosteroid drugs in ampoules An accurately measured volume of the injection equivalent to 10 mg of the drugs was dissolved in 40 ml of light petroleum (40-60 “C, saturated with 90% ethanol) and extracted with four 20-ml portions of 90% ethanol (saturated with light petroleum). The alcoholic extracts were collected in a 100-ml calibrated flask and diluted to volume with 90% ethanol. A portion was subjected to the oximation and colour develop- ment procedures described above. Application to griseofulvin tablets Twenty tablets were finely powdered and an accurately weighed amount equivalent to 40 mg of drug was refluxed with 75 ml of absolute ethanol for 15 min.Sufficient ethanol was added to dilute to 100 ml, the mixture was centrifuged and a 25-ml aliquot of the supernatant liquid was subjected to the oximation and colour development procedures. Application to caraway or peppermint oil A 1-ml volume of the volatile oil was diluted to 100 ml with ethanol and then 1 ml of the resulting solution was subjected to the oximation procedure, followed by procedure A or B. For the standard additions method, a suitable aliquot of the standard solution of carvone or menthone was added to a 1-ml aliquot of the previously analysed diluted oil, then the assay was carried out and the recovery of the added amount was calculated.and griseofulvin contain two keto groups per molecule, they also gave a monoxime, a fact which was noted in the literature15J1 and in this work from molar absorptivity calculations applying Job’s33 method and other experimental observations. Iodine Acceptor Procedure Oximation results in the introduction of a basic centre to the carbonyl drug molecule. Mixing the oxime and iodine in chloroform media resulted in a change of the violet colour of iodine to yellow, owing to a charge-transfer complexation reaction between the ketoxime n-donor and the a-electron acceptor iodine followed by the formation of a triiodide ion pair, as shown by the following suggested routes, which agree with reports on similar reactions in the literature25.26: OH OH OH OH Outer complex Inner complex Triiodide ion pair oLmax.= 295-300 n n The absorption graphs of the reaction products (Figs. 1 and 2) show a high absorption band in the region of 300 nm and a lower band with a maximum at 365 nm, characteristic of the n-donor - iodine charge-transfer complexes. In chloroform, iodine itself has a maximum absorption at about 512 nm and the free oxime has an absorption peak at 34&350 nm. However, the iodine - oxime product has an absorption maximum at 270-300 nm. This variation is due to the modification of the absorption maximum of the triiodide ion formed from the release of iodide ions in the reaction by the accompanying ketoxime cation.25.26.32 o-6 I Results and Discussion Oximation The addition - elimination oximation reaction of the studied compounds is assumed to give an oxime according to the following route: 260 300 340 380 Wavelengthlnm Fig.1. Absorption spectra of oxime - iodine Complexes in chloro- form for: I, griseofulvin (0.019 mg ml-1); 11, carvone (0.016 mg ml-1); and 111, menthone (0.02 mg ml-1) ‘C=O + HzN-OH -‘C=N- OH + H20 / / The authors’ earlier studies7J5 of the optimum conditions for carrying out this reaction have led to the described conditions of oxime preparation and manipulation of the reaction mixture prior to its solvent extraction. Chloroform, the chosen extraction solvent, was also an appropriate medium for the charge-transfer complexation reaction, as it has the desired solubilising and lipotropic characteristics, in addition to a reasonable degree of polarity. Stoicheiometry As carvone, menthone, testosterone propionate and nandrol- one phenylpropionate contain only one carbonyl group, they should give a monoxime on reaction.Although progesterone I I I O ’ 2;o 300 340 380 Wavelengttilnm Fig. 2. Absorption spectra of oxime - iodine complexes in chloro- form for: I, progesterone (0.025 mg ml-I); 11, testosterone propionate (0.028 mg ml-I); and 111, nandrolone phenylpropionate (0.03 mg ml-1)ANALYST, SEPTEMBER 1986, VOL. 111 1041 C hloranil Acceptor Procedure Mixing the chloroform oxime extract (Amax. 340-350 nm) with chloroformic chloranil solution (Amax, 290 nm) resulted in the development of a red chromogen. The colour reached its maximum intensity and stability on standing for 45 min at room temperature or on heating in a water-bath at 45 "C for 15 min.This interaction is due to charge-transfer complexation between the ketoxime, n-donor and the chloranil n-acceptor , which, because of the polarity of the medium, may lead to a radical ion pair28 as shown in Scheme 1. The chloranil radical ion pair had a A,,,, of 430-440 nm (Figs. 3 and 4). The A,,,. of the chloranil radical ion would hence be slightly modified by the accompanying ketoxime cation, otherwise different ketox- ime complexes would have given an identical A,,,, 0 0 Oxime Ch lora n iI n - JC complex / in chloroform / 400 440 480 520 Wavelengthlnm Fig. 3. Absorption spectra of oxime - chloranil complexes in chloroform for: I, progesterone (0.045 mg ml-I); 11, testosterone propionate (0.045 mg ml-1); and 111, nandrolone phenylpropionate (0.044 mg ml-1) I n I I I ' 400 440 480 520 Wavelengthlnm 0- Radical ion pair Scheme 1 Fig.4. Absorption spectra of oxirne - chloranil complexes in chloroform for: I, griseofulvin (0.033 mg ml-l); 11, carvone (0.038 mg ml-l); and 111, menthone (0.035 mg ml-I) Table 1. Beer's plot data for ketoxime charge-transfer complexes in chloroform Iodine acceptor (procedure A) Chloranil acceptor (procedure B) Slope Progesterone . . . . . . . . 0.174 Testosteronepropionate . . . . 0.145 Nandrolone phenylpropionate . . 0.122 Griseofulvin . . . . . . . . 0.242 Carvone . . , . . . . . . . 0.215 Menthone . . . . . . . . 0.198 Drug (b) Intercept 0.04 0 -0.01 0.02 0.02 -0.01 (a) Slope 0.200 0.181 0.154 0.262 0.200 0.162 (b) Intercept -0.02 -0.01 -0.03 -0.03 0 0.02 (a) Table 2.Results of assay of pharmaceutical carbonyl compounds using iodine acceptor charge-transfer complexation method (procedure A) and the BP 1980 method Recovery* k S.D., YO Preparation Proposed method BP method Pure progesterone . . . . . . . . 100.15 f 0.38 100.19 k 0.73 Progesterone (Lutone ampoules) . . 99.99 k 0.41 99.63 k 0.44 Pure testosterone propionate . . . . 100.31 5 0.20 100.47 k 0.51 Testosterone propionate (Testone E. ampoules) . . , . . . 99.97 k 0.39 99.71 2 0.77 Pure nandrolone phenylpropionate . . 100.14 k 0.53 99.96 k 0.49 Nandrolone phenylpropionate (Durabolin ampoules) . . . . . . 99.84 2 0.35 100.02 f 0.63 Pure griseofulvin . . . . . . . . 99.87 f 0.55 100.11 k 0.7 Griseofulvin (Griseofulvin tablets) . . 99.98 f 0.74 100.26 f 0.66 Pure carvone .. . . . . . . . . 100.19 f 0.65 100.52 f 0.88 Carvone (in caraway oil) . . . . . . 99.75 k 0.45 98.98 k 0.55 Pure menthone . . . . . . . . . . 100.21 k 0.62 100.41 k 0.68 Menthone (in peppermint oil) . . . . 99.75 k 0.45 99.55 k 0.52 * Mean of five determinations k standard deviation. t Figures in parentheses are tabulated values of t and F a t the 95% confidence limit. t (2.3)t 0.11 1.38 0.66 0.66 0.54 0.54 0.6 0.62 0.69 2.28 0.49 0.66 F (6.39) 1- 3.65 1.12 6.30 4.0 1.12 3.25 1.63 1.25 1.83 1.5 1.2 1.351042 ANALYST, SEPTEMBER 1986, VOL. 111 Table 3. Results of assay of pharmaceutical carbonyl compounds using chloranil acceptor charge-transfer complexation method (procedure B) and the BP 1980 method Recovery* f S.D., Yo Preparation Proposed method BP method (2.3)t (6.39) t t F Pure progesterone .. . . . . . . 100.20 k 0.66 100.19 k 0.73 0.02 1.20 Progesterone (Lutone ampoules) . . 100.27 f 0.47 99.63 f 0.44 2.28 1.16 Pure testosterone propionate . . . . 100.42 f 0.49 100.47 k 0.51 0.16 1.08 Testosterone propiona te (Testone E. ampoules) . . . . . . 99.77 k 0.50 99.71 k 0.77 0.15 2.40 Pure nandrolone phenylpropionate . . 100.01 f 0.41 99.96 f 0.49 0.18 1.47 Nandrolone phenylpropionate (Durabolin ampoules) . . . . . . 99.76 f 0.44 100.02 f 0.63 0.76 1.95 Pure griseofulvin . . . . . . . . 100.10 k 0.77 100.11 f 0.70 0.02 1.22 Griseofulvin (Griseofulvin tablets) . . 100.34 * 0.60 100.26 f 0.66 0.20 1.22 Pure carvone . . . . , . . . . . 99.95 k 0.52 100.52 k 0.88 1.26 2.85 Carvone (in caraway oil) .. . . . . 99.49 f 0.40 98.98 f 0.55 1.70 1.87 Pure menthone . . . . . . . . . . 100.19 _t 0.49 100.41 k 0.68 0.59 1.96 Menthone (in peppermint oil) . . . . 99.99 k 0.59 99.55 k 0.52 1.29 1.29 * Mean of five determinations ? standard deviation. t Figures in parentheses are the tabulated values o f t and F a t the 95% confidence limit. Assay Parameters Effect of temperature After its full development, the complex may dissociate with a corresponding decrease in intensity on heating above room temperature. Effect of complex formation Leaving the reaction mixture to stand, as described under Procedures, was essential for the complete development of the complexes by transformation of the outer complexes to inner complexes.32 Effect of varying acceptor concentration The concentration of the acceptor should not be excessive in order to avoid the formation of ter-molecules or higher complexes. 32 Linearity of Calibration Graphs A linear relationship was obtained for the absorbance of the ketoxime acceptor reaction products when the concentration of the parent drugs was in the concentration range 0.01-0.05 mg ml-1 in the final measured solutions.The graphs show negligible or zero intercepts and are described by the regression equations A = a + bC (A, absorbance of a 1-cm layer; b, slope; a, intercept; and C, concentration of the measured solution in mg per 100 ml) obtained by the least-squares method.34 The data for the compounds studied (Table 1) that were used for the calibration graphs indicate the sensitivity of the proposed procedures.Quantification, Accuracy and Precision of Procedures A and B The validity of the proposed procedures for the determination of the studied compounds in their pure state and in phar- maceutical forms was tested by analysing these products with the proposed procedures and an official method.17 A standard additions technique was used for the determination of carvone and menthone in volatile oils of caraway and peppermint, respectively, to give the percentage recovery of an added amount. The results obtained (Tables 2 and 3) were compar- able to those obtained by the official method as both the t and Fvalues did not exceed the theoretical values.34 This indicated that the proposed procedures were as accurate and precise as the applied official methods and that no interference from excipients and vehicles was encountered.This was expected because the procedure involves solvent extraction of the oxime formed and a preliminary clean-up step with 90% ethanol was used for the injections. The procedure proved to be directly applicable to volatile oils with no preliminary separation step. The proposed procedures offer the advan- tages of accuracy and simplicity of reagents and apparatus. Among the spectrophotometric methods used to determine ketones in volatile oils, the proposed procedure is one of the most sensitive, simple and accurate and is of potential use as a general method. No preliminary clean-up step before oxima- tion is needed if the chloranil acceptor method is used to determine A4-3-ketosteroid drugs in injections.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. References Roushdi, I. M., El Sebai, A. I., and Belal, S . , Pharmazie, 1973, 28, H.1, 41. Emil, F., Dawoud, A. Y., and Nagi, W., Analyst, 1976, 101, 616. Matsui, M., and Takashi, M., Anal. Biochem., 1976, 75, 441. Wu John, Y. P., J. Assoc. Ofl. Anal. Chem., 1971, 54, 617. Myrick, J. W., Puge, D. P., and Pfabe, Y. H., J. Assoc. Off. Anal. Chem., 1972, 55, 1175. Julia, D., Diss. Pharm. Pharmacol., 1970, 22, 337. Magda, M. A., Belal, S., A1 Adl, S., and Abou El Kheir, A., Anal. Left., 1985, 18, B10. Georoge, S . , Fresenius 2. Anal. Chem., 1981, 309, 97. Chatten, L. G., Yadaw, R. N., and Madan, D. K., Pharm. Acfa Helv., 1976, 51, 381. Cantin, D., Alery, J., and Cocur, A., J. Pharm. Belg., 1977, 32, 255.Cochran, R. C., Darrey, K. J., and Ewing, L. L., J. Chrornatogr., 1979, 173,349. Cook, S. J., Rawlings, N. C., and Kennedy, R. I., Steroids, 1982, 40, 369. Archambault, A., Begue, R., Faure, Z., and Grandin, B., J. Chromatogr., 1984, 284, 261. “The United States Pharmacopeia, 18th Revision,” Mack, Easton, PA, 1970, p. 939. Ayad, M. M., Belal, S . , A1 Adel, S. A., and El Kheir, A. A., Analyst, 1985, 110, 823. Johnson, D. R., in Meites, L., Editor, “Handbook of Analy- tical Chemistry,” Volume 12, McGraw-Hill, New York, 1965, “British Pharmacopoeia 1980,” HM Stationery Office, Lon- don, 1980, p. 104. p. 99.ANALYST, SEPTEMBER 1986, VOL. 111 1043 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. “Egyptian Pharmacopoeia,” Cairo University Press, Cairo, 1972, pp. 266 and 720. “The United States Pharmacopeia, 20th Revision,” US Pharmacopeial Convention, Rockville, MD, 1980, pp. 1216- 1222. Taha, A. M., Ahmed, A. K. S . , Gomaa, C. S . , and El Fatatary, H., J . Pharm. Sci., 1974, 63, 1853. Gornaa, C., and Taha, A. M., J . Pharm. Sci., 1975, 64, 1398. Tan, H. S . I., Gerlach, E. D., and Dimartio, A, S., J. Pharm. Sci., 1977, 66, 767. Belal, S., Abdel Hady, E. S . M., Abdel-Hamid, M. E., and Abdine, H., Analyst, 1980, 105, 774. Rizk, M. S., Walash, M. I., and Ibrahim, F. A., Analyst, 1981, 106, 1163. Taha, A. M., El Rabbat, N. A., and Abdel-Fattah, F. A., Analyst, 1980, 105, 568. Taha, A. M., El Rabbat, N. A., and Abdel Fattah, F. A., J. Pharm. Belg., 1980,35, 437. Rao, C . N. R., Bhat, S. N., and Dwivedi, P. O., Appl. Spectrosc. Rev., 1972, 5, 1. 28. 29. 30. 31. 32. 33. 34. Taha, A. M., and Rucker, G., Arch. Pharm. (Weinheim, Ger.), 1977, 310, 485. Belal, S., Abdel Hady, M. A., Abdel-Hamid, M., and Abdine, H., J. Pharm. Sci., 1981, 70, 1927. Ayad, M. M., Belal, S . , El Adl, S . M., and A1 Kheir, A. A , , Analyst, 1984, 109, 1419. Oxford, A. E., Raistrick, H., and Simanart, I., Biochern. J., 1939, 33, 240. Foster, R., “Organic Charge-transfer Complexes,” Academic Press, London, 1969, pp. 61 and 191. Job, P., Ann. Chim. (Paris), 1936, 16, 97. Bauer, E. L., “Statistical Manual €or Chemists,” Academic Press, London, 1971, p. 61. Paper A6122 Received January 23rd, 1986 Accepted February 24th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101039

出版商:RSC    

年代:1986

数据来源: RSC