ISSN:0003-2654    

DOI:10.1039/AN98207FX013

出版商:RSC    

年代:1982

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98207BX015

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

1v SUMMARIES OF PAPERS I N THIS ISSUE April, 1982Summaries of Papers in this IssueCritical Evaluation of a Multi-element Scheme Using PlasmaEmission and Hydride Evolution Atomic-absorptionSpectrometry for the Analysis of Plant and Animal TissuesAn analytical scheme that uses inductively coupled argon plasma emissionspectroscopy (ICAP) and hydride evolution atomic-absorption spectrometry(HEAA) for the determination of trace elements in plant and animal tissueshas been evaluated. The scheme incorporates the ion-exchange procedure ofKingston et al., which uses Chelex 100 resin to concentrate trace elements andremove potentially interfering alkali and alkaline earth metals. The separa-tion procedure is included in a scheme designed to maximise the number ofanalyte metals that can be determined from a single digestion of a biologicalmatrix. Acid-digested samples are divided into two fractions.One fraction(5% of the total) is measured directly by ICAP for alkali and alkaline earthmetals and phosphorus and for transition metals such as iron and manganese,which are not well behaved on the resin. The other fraction (95% of thetotal) is subjected to the separation procedure whereby a number of biologicallyimportant trace elements, including cadmium, copper, molybdenum, nickel,vanadium and zinc, are initially sequestered by the resin, and then stripped intoa small volume of dilute nitric acid for ICAP measurement of the “matrix-free”analytes. Arsenic, selenium and antimony, which are not retained by theresin, are coll cted with the initial column effluent, acidified and determinedby HEAA.#he reliability of the scheme is influenced by the nature of theacid digestion procedure used to oxidise the organic matrix. The scheme wastested by analysis of ten National Bureau of Standards .biological referencematerials.Keywords : Plasma emission ; hydride evolution atomic-absorption spectrometry ;Chelex 100 separation scheme ; NBS reference materials ; plant and animaltissuesJOHN W. JONES and STEPHEN G. CAPARDivision of Chemical Technology, US Food and Drug Administration, 200 C Street,S.W., Washington, D.C. 20204, USA.and T. C. O’HAVERDepartment of Chemistry, University of Maryland, College Park, MD 20742, USA.Analyst, 1982, 107, 353-377.Determination of Mercury in Pharmaceutical Products byAtomic-absorption Spectrophotometry Using a Carbon Rod AtomiserAn electrothermal atomisation procedure using a carbon rod atomiser isdescribed for the determination of mercury after its extraction with dithizone(diphenylthiocarbazone) into chloroform.The increased stability of mercuryafter extraction allows drying and ashing to be carried out adequately withoutloss prior to atomisation. The sensitivity to mercury in the carbon rodatomiser is 1.1 x 10-lo g to give 1% absorption a t 253.7 nm. Calibrationgraphs are linear over the range 0.2-2.0 pg ml-l of mercury. The method canbe applied directly, without a preliminary digestion procedure, to solutionscontaining organic mercurial preservatives or bactericides [phenylmercury (11)acetate or nitrate or thiomersal (sodium ethylmercurithiosalicylate)], and t otrace determination in basic and some neutral and acidic compounds, solublein water or in dilute acids.A standard additions procedure is recommendedto overcome possible matrix effects, and to allow the detection of contamin-ation errors. The results compare favourably with those obtained by theconventional cold vapour technique.Keywords Mercury determination ; dithizonate ; pharmaceuticals ; atomic-absorption spectrophotometry ; electrothermal atomisationPAMELA GIRGIS TAKLA and VICTOR VALIJANIANWelsh School of Pharmacy, University of Wales Institute of Science and Technology,King Edward VII Avenue, Cardiff, CF1 3NU.Analyst, 1982, 107, 378-384April, 1982 SUMMARIES OF PAPERS IN THIS ISSUEGravimetric Determination of Nickel with ThiosemicarbazonesA systematic study of the use of thiosemicarbazones in the gravimetricdetermination of nickel is reported.The compounds tested were furfuralthiosemicarbazone (FAT), thiophen-2-aldehyde thiosemicarbazone (TAT)and furfural 4-phenyl-3-thiosemicarbazone (FAPT) . FAPT is the mostappropriate reagent owing to its sensitivity and selectivity. FAPT has beenapplied to the gravimetric determination of nickel in diverse standardsamples. The reagents tested in this work were compared with classicalvic-dioximes used in the gravimetric determination of nickel.Keywords: Furfural thiosemicarbazone; thiophen-2-aldehyde thiosemi-carbazone ; furfural 4-phenyl-3-thiosemicarbazone ; nickel determination ;gravimetryD.ROSALES and J. M. CANO-PAVONDepartment of Analytical Chemistry, Faculty of Chemistry, University of Seville,Seville-4, Spain.Analyst, 1982, 107, 385-39 1.VPrecipitation of Selenium and Tellurium from HomogeneousSolutions by Dimethyl SulphiteElemental selenium and tellurium are precipitated from homogeneous solutionsby employing dimethyl sulphite for the in situ production of sulphur dioxide.Results obtained show that both selenites and selenates as well as telluritesand tellurates are quantitatively reduced by this reagent. The advantagesof this precipitation over the conventional precipitation procedure are out-lined. Based on the different conditions under which the two elements areprecipitated, a procedure for the separation of selenium and tellurium fromone another is reported.Keywords : Selenium determination ; tellurium determination ; homogeneousprecipitation ; dimethyl sulphite ; separationB.V. NARAYANA and N. APPALA RAJUDepartment of Metallurgy, Indian Institute of Science, Bangalore-560 012, India.Analyst, 1982, 107, 392-397.Automatic Potentiometric Titration of Thiocyanate - CyanideMixtures in Hydrometallurgical EffluentsMixtures of cyanide and thiocyanate in hydrometallurgical effluents heavilyclouded with particulates are titrated quickly and successfully with silvernitrate solution by using a potentiometric automatic titrator fitted with asilver working electrode and a glass reference electrode.When thiocyanateis t o be determined, cyanide is masked with formalin. Titrations over awide range of concentration and ratio of the two species require minimumpre-treatment of the samples and give sharp end-points and good replication.Keywords: Thiocyanate - cyanide wixtures; silver - glass electrode pair;hydrometallurgical ejYEuentsG. F. ATKINSONDepartment of Chemistry, University of Waterloo, Waterloo, Ontario, Canada,N2L 3GI.J. J. BYERLEYDepartment of Chemical Engineering, University of Waterloo, Waterloo, Ontario,Canada, N2L 3GI.and B. J. MITCHELLDepartment of Chemistry, University of Waterloo, Waterloo, Ontario, Canada,N2L 3GI.Analyst, 1982, 107, 398-402vi SUMMARIES OF PAPERS IN THIS ISSUEDead- stop Determination of EDTA and NTA in CommerciallyAvailable DetergentsApril, 1982A rapid and selective method for the determination of ethylenediaminetetra-acetic acid (EDTA) and/or nitrilotriacetic acid (NTA) in commercially avail-able detergents has been developed.It is based on a titration with a standardsolution of copper(I1) in acetate buffer, where the end-point is revealed bymeans of a “dead-stop” system with two polarised copper electrodes.Furthermore, it is possible to determine in the course of the same titrationthe relative amounts of both chelating agents by using 4-(2-pyridylazo)-resorcinol (PAR), which changes colour at the end-point of the first reaction(copper - EDTA) . Most detergent constituents, including polyphosphates,have been observed to have no effect on the determination; interference fromsome constituents (perborates and zeolites) can easily be removed. Themethod has been shown to give good results in the analyses of differentcommercially available products.Keywords : E D T A and N T A determination ; copper(II) sulphate titrant;dead-stop indicator ; PAR indicator ; detergentsR. CALAPAJ and L.CIRAOLOIstituto di Merceologia, Universitk di Messina, 981 00 Messina, Italy.F. CORIGLIANOIstituto di Chimica Industriale, Universitii di Messina, 98100 Messina, Italy.and S. DI PASQUALEIstituto di Chimica Analitica, Universitk di Messina, 98 100 Messina, Italy.Analyst, 1982, 107, 403-407.Comparison of Laser-excited Fluorescence and PhotoacousticLimits of Detection of Polynuclear Aromatic HydrocarbonsAn extremely simple and sensitive system that employs both fluorescenceand photoacoustic detection modes simultaneously has been used to obtainlimits of detection rapidly for 30 polynuclear aromatic hydrocarbons. Thesensitivities of the two techniques are compared with both fixed and tunablelaser excitation and a “levelling” effect is demonstrated by the photoacousticresults. High-energy pulse excitation demonstrates that the photoacousticsignal does not increase indefinitely with incident pulse energy. Applicationof the simultaneous detection scheme to the measurement of fluorophorquantum efficiencies in solutions and to antifluorochrome stains is suggested.Keywords : Photoacoustic ; fluorescence ; polynuclear aromatic hydrocarbons ;laser detectionEDWARD VOIGTMAN, ARTHUR JURGENSEN and JAMES D. WINE-FORDNERDepartment of Chemistry, University of Florida, Gainesville, FL 3261 1, USA.Analyst, 1982, 107, 408-413

ISSN:0003-2654    

DOI:10.1039/AN98207FP033

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

April, 1982 SUMMARIES OF PAPERS IN THIS ISSUEDetermination of Saccharin by Desorption ofFerroin from Silica GelFerroin is strongly adsorbed on silica gel but its ion-association complexesare easily desorbed. A simple, rapid and accurate procedure for the deter-mination of saccharin has been devised utilising these properties. Theprocedure is based on the quantitative formation of the ion-association com-plex [(Fe phen.J2+(saccharin),-] when saccharin dissolved in water, 60%aqueous ethanol, 70% aqueous acetone or 80% aqueous methanol is shakenwith ferroin-impregnated silica gel. The absorbance of the characteristicorange colour of the ferroin ion-association complex is measured at 510 nm.There are no interferences from glucose, sucrose and dulcin even whenthese constituents are present a t 1600 times the concentration of saccharin.Sodium cyclamate, sorbic acid, benzoic acid, citric acid and sodium chloridegive positive errors.Although the presence of sodium hydrogen carbonateresults in a low recovery of saccharin, this compound may be eliminated easilyby the addition of dilute sulphuric acid.A procedure for the routine determination of saccharin in saccharin tabletsis proposed.Keywords : Saccharin determination ; silica gel adsorption ; &is( 1,lO-phenan-throline) iron (11)E. ROY CLARK and EL-SAYED A. K. YACOUBDepartment of Chemistry, University of Aston in Birmingham, Gosta Green,Birmingham, B4 7ET.Analyst, 1982, 107, 414-421.Assay of Hydrazine in Isoniazid and its Formulations byDifference SpectrophotometryA rapid procedure is described for the determination of hydrazine at the verylow levels that may occur as a result of hydrolysis in certain isoniazid formula-tions.The method is based upon the measurement of the absorbancedifference between a solution of the 4-dimethylaminobenzalazine derivative ofhydrazine and a similarly prepared solution containing 3% V/V acetone.The absorbance difference, which is proportional to the concentration ofhydrazine, is unaffected by the presence of a large excess of isoniazid. Thechoice of experimental conditions which provide the maximum accuracy,sensitivity and specificity is discussed.Keywords Hydrazine determination ; ultraviolet spectrophotometry ; differencespectrophotometry ; isoniazid ; isoniazid formulationsA.G. DAVIDSONDepartment of Pharmaceutical Chemistry, School of Pharmaceutical Sciences,University of Strathclyde, Glasgow, G1 1XWAnalyst, 1982, 107, 422-427.vi...vl11 SUMMARIES OF PAPERS IN THIS ISSUEHighly Sensitive Spectrophotometric Determination of TraceAmounts of Aluminium with Chromazol KS andCetylp yridinium BromideApril, 1982A highly sensitive spectrophotometric method has been developed for thedetermination of aluminium, based on the formation of a ternary complexwith chromazol KS (CALKS) and cetylpyridinium (CP) bromide in aqueoussolution in the presence of 25-40% of ethanol. The pH range for the forma-tion of the ternary complex is 5.8-6.7, and the wavelength of maximumabsorbance is 625nm.The ternary system obeys Beer’s law for between0.02 and 0.32 pgml-l of aluminium. A high molar absorptivity of =1.02 x lo5 1 mol-1 cm-l and a Sandell’s sensitivity of 0.00026 pg cm-2 havebeen obtained. The complex has the composition Al(0H) (CALKS),(CP),,as established by Job’s method of continuous variation and the equilibrationshift method. The method has good selectivity and so can be applied to thedirect spectrophotometric determination of trace amounts of acid-solublealuminium in steel.Keywords : Aluminium determination ; spectrophotometry ; chromazol KS ;cetylpyridinium bromideLIU SHAO-PUCentre Laboratory, Diesel Engine Plant of Chengdu, Chengdu, 610081, People’sRepublic of China.Analyst, 1982, 107, 428-432.Rapid Determination of Residual Chlorine by Flow Injection AnalysisFlow injection determination of residual chlorine in solution has been carriedout spectrophotometrically by methyl orange decolorisation and by formationof a yellow complex with o-tolidine.The carrier streams are 0.180 mM methylorange in pH 2 buffer solution or 0.01 M hydrochloric acid, and 3.0 mM o-toli-dinium dichloride (in 2 M hydrochloric acid) , respectively. A sampling rate of288 samples per hour has been obtained with the latter reagent and thedetection limit is 0.08 p.p.m. of chlorine. Influence of flow parameters suchas flow-rate, tube length and diameter and the effect of interferents on thedetermination have been investigated. A study of the hyprochlorite - ammon-ium - o-tolidine reaction system has been performed and a method for thesimultaneous determination of NH,+ and OC1- is described.Keywords : Chlorine determination ; flow injection analysis ; methyl orange ;o-toladineD.J. LEGGETT, N. H. CHEN and D. S . MAHADEVAPPAUniversity of Houston, Department of Chemistry, Houston, TX 77004, USA.Analyst, 1982, 107, 433-441.Improving the Yield, Purity and Molecular Integrity of SkeletalMuscle RNA Isolated by Phenol ExtractionShort PaperKeywords : R N A extraction ; R N A degradation ; ribonuclease inhibitorsFRANCIS N. ONYEZILIDepartment of Clinical Pathology, College of Medicine, University of Lagos, P.M. B.12003, Lagos, Nigeria.Analyst, 1982, 107, 442-445April, 1982 SUMMARIES OF PAPERS IN THIS ISSUESpectrophotometric Determination of Beryllium(I1) Using aTrisazosalicylic Derivative of TriphenylamheShort PaperKeywords : Beryllium( I I ) determination ; spectrophotometry ; triphenylaminetrisazosalicylic acidM.E. M. KHALIFADepartment of Chemistry, Faculty of Science, Mansoura University, Mansoura,Egypt.Analyst, 1982, 107, 446-449.Titrimetric Determination of the Yield of Sulphide Formed byAlkaline Degradation of CephalosporinsShort PaperKeywords : Cephalosporins ; alkaline degradation ; sulphide ; titrimetryA. G. FOGG, M. A. ABDALLA and H. P. HENRIQUESChemistry Department, Loughborough University of Technology, Loughborough,Leicestershire, LEll 3TU.Analyst, 1982, 107, 449-452.Investigations on the Determination of Germanium inOrganogermanium Compounds Using Carbon Furnace AtomisationShort PaperKeywords : Germanium determination ; organogermanium ; carbon furnaceatomisation ; atomic-absorption spectrophotometryD.THORBURN BURNS and D. DADGARDepartment of Analytical Chemistry, The Queen’s University of Belfast, Belfast,BT9 5AG, Northern Ireland.Analyst, 1982, 107, 452-455.Differential-pulse Polarographic Determination of Red 10B FormedFrom the Permitted Food Colour Red 2GShort PaperKeywords : Red 10B ; Red 2G ; food colours ; diflerential-pulse polarographyA. G. FOGG and M. R. WHETSTONEChemistry Department, Loughborough University of Technology, Loughborough,Leicestershire, L E l l 3TU.Analyst, 1982, 107, 455-459.Determination of Morphine and Morphine-6-nicotinate by anIn Situ Reaction on Chromatographic Plates with Dansyl ChlorideShort PaperKeywords : Morphine and morphine-6-nicotinate determination ; morphine-3,6-dinicotinate ; dansylation ; derivatisation by overspottingixR. WINTERSTEIGERInstitut fur Pharmazeutische Chemie, Universitat Graz, A-8010 Graz, Austria.Analyst, 1982, 107, 459-46 1

ISSN:0003-2654    

DOI:10.1039/AN98207BP037

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

APRIL 1982 The Analyst Vol. 107 No. 1273 Critical Evaluation of a Multi-element Scheme Using Plasma Emission and Hydride Evolution Atomic-absorption Spectrometry for the Analysis of Plant and Animal Tissues John W. Jones and Stephen G. Capar Division of Chemical Technology US Food and Drug Administration 200 C Street S . W. Washington, D.C. 20204 USA and T. C. O’Haver Department of Chemistry University of Maryland College Park MD 20742 USA An analytical scheme that uses inductively coupled argon plasma emission spectroscopy (ICAP) and hydride evolution atomic-absorption spectrometry (HEAA) for the determination of trace elements in plant and animal tissues has been evaluated. The scheme incorporates the ion-exchange procedure of Kingston et al. which uses Chelex 100 resin to concentrate trace elements and remove potentially interfering alkali and alkaline earth metals.The separa-tion procedure is included in a scheme designed to maximise the number of analyte metals that can be determined from a single digestion of a biological matrix. Acid-digested samples are divided into two fractions. One fraction (5% of the total) is measured directly by ICAP for alkali and alkaline earth metals and phosphorus and for transition metals such as iron and manganese, which are not well behaved on the resin. The other fraction (95% of the total) is subjected to the separation procedure whereby a number of biologically important trace elements including cadmium copper molybdenum nickel, vanadium and zinc are initially sequestered by the resin and then stripped into a small volume of dilute nitric acid for ICAP measurement of the “matrix-free” analytes.Arsenic selenium and antimony which are not retained by the resin are collected with the initial column effluent acidified and determined by HEAA. The reliability of the scheme is influenced by the nature of the acid digestion procedure used to oxidise the organic matrix. The scheme was tested by analysis of ten National Bureau of Standards biological reference materials. Keywords Plasma emission ; hydride evolution atomic-absorption spectrometry ; Chelex 100 separation scheme ; NBS reference materials ; plant and animal tissues Acquisition of data on the trace element content of plant and animal tissues has become a major analytical effort for many laboratories.Despite recent instrumental advances such as electrothermal atomic-absorption spectrometry (EAAS) and argon plasma emission spectros-copy [inductively coupled (ICAP) or d.c. (DCP) plasmas] formidable problems remain for the determination of many trace elements. We have encountered both chemical and instru-mental limitations in attempting to apply ICAP spectroscopy to the determination of low concentrations of toxic and nutritive elements in a wide variety of foods and related biological materials. Similarly, many biologically important elements are present in organic materials at microgram per kilogram levels. I t is seldom possible to mineralise a tissue sample by conventional wet or dry ashing procedures and to dissolve the resulting ash in an appropriate solvent without diluting the original analyte concentration to some extent.Attempts to minimise this dilution factor by 353 The detection powers of ICAP are usually in the microgram per litre range 354 JONES et al. MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst VoZ. 107 ashing a relatively large sample and diluting it to a small volume (e.g. 1 g dry mass in 10 ml of solution) are usually frustrated by the limited solubility of the inorganic matrix elements,lS2 e.g. alkali or alkaline earth metal pyrophosphates for dry ashed samples or potassium per-chlorate for perchloric acid digestions. Further for samples that contain high concentrations of alkali and alkaline earth metals dissolving the ash from large samples and concentrating the solution to small volumes results in high solution concentrations of these elements.This can be particularly troublesome for ICAP measurements because of the adverse effect on nebulisa-tion efficiency3s4 and the contribution to subtle but significant changes in spectral background due either to true spectral structure or to the.introduction of stray light into the spectro-meter system.6s6 Recently chelating ion-exchange resins have been shown to be an effective means of con-centrating and separating trace elements from complex matrices prior to AAS or ICAP measurement .7-9 Kingston et al.1° demonstrated the applicability of Chelex 100 cation-exchange resin to the isolation of eight transition metals from a seawater matrix. They were able to concentrate the trace elements from the seawater matrix 100-fold and eliminate the high concentrations of alkali and alkaline earth metals before EAAS measurement of the transition elements.Sturgeon et used a similar approach to determine trace metals in seawater. The qualitative metal composition of seawater is similar to that of mineralised biological materials ; sodium potassium calcium and magnesium predominate in both matrices. Because of this similarity we have examined the possible utility of the Chelex 100 method of Kingston et al. to separate and concentrate several toxic and nutritive elements from acid digests of plant and animal tissues. We have incorporated the separation procedure into a scheme that uses ICAP spectroscopy and hydride evolution atomic-absorption spectrometry (HEAA) to determine up to 18 ele-ments in samples of plant and animal tissue.The behaviour of ten US National Bureau of Standards (NBS) Biological Standard Reference Materials (SRMs) was used to judge the effec-tiveness of the analytical scheme for a variety of complex matrices and to identify those trace elements which could be determined successfully from a single digestion. Experimental Apparatus Inductively coupled plasma ICAP measurements were performed with a Jarrell-Ash Model 975 Plasma Atom Comp 0.75-m direct-reading spectrometer operated at 1.1 kW source power at 27.12 MHz. Potass-ium measurements were performed on an attached Jarrell-Ash 0.5-m scanning monochro-mator (“N + 1”). A limited number of cobalt measurements were obtained on a second 0.75-m Jarrell-Ash plasma spectrometer.Analytical lines and standardisation concentrations used for all measurements are listed in Table I. Emission measurements were performed at TABLE I ICAP AND HEAA ANALYTICAL LINES AND MEASUREMENT CONDITIONS Line/ Element nm* Cd . . 226.5 c o t . . . . 228.6 Cr . . 357.9 Mo . . . . 203.8 Ni . . 231.6 v . . 292.4 Standardisation concentration/ pg ml-l 1 .oo 1 .oo 1.00 1.00 1.00 1.00 Standardisation Background Line/ concentration/ Background correction Element nm* pg ml-l correction Yes A1 . . . . 308.2 10.0 No Yes Fe . . . . 259.9 10.0 No Yes Y e s Yes P . . . . 214.9 100 No Yes Mg . . . . 279.5 35.0 NO Ca . . 393.3 5.00 No Ca . . . . 370.6 200 No c u . . 324.7 5.00 Yes KI . . . . 766.5 200 No Mn .. . . 257.6 5.00 No Pb . . 220.3 5.00 Yes As$. . . . 193.7 - Yes Yes Zn . . . . 206.2 5.00 Yes Se§ . . . . 196.0 -Sb§ . . . . 217.6 - Yes * ICAP measurements were made a t 16 mm above the coil. t Co measurements were performed a t 228.6 nm using an independent ICAP polychromator system. 1 K measurements were obtained on an “N + 1” 0.5-m scanning monochromator. Fj Determined by HEAA using calibration graphs prepared from 25-500 ng of each elernent.l A+ril 1982 AND HYDRIDE EVOLUTION AAS FOR PLANT AND ANIMAL TISSUE 355 16mm above the induction coil. The Jarrell-Ash crossflow nebuliser system was slightly modified by bubbling the nebuliser argon (0.5 1 min-l) through a fritted 20 x 5 cm cylindrical column of de-ionised water located at the gas inlet of the nebuliser.The resulting water-saturated nebuliser argon reduced crusting of salts at the crossflow nebuliser needle orifices. Solutions were delivered to the nebuliser by a peristaltic pump (Gilson Minipuls 2) at 1.1 ml min-l. Wavelength modulation (“spectrum-shifter”) background correction was used for some elements as indicated in Table I to com-pensate for minor base-line drift during the analyses. The argon coolant -plasma flow was 18lmin-l. Hydride evolution atomic-absorption spectrometer spectrometer using a semi-automatic hydride generator as previously described.12 HEAA measurements were performed on a Perkin-Elmer Model 403 atomic-absorption Glassware Borosilicate volumetric ware was first hand-washed with hot tap water and rinsed with distilled water then soaked in nitric acid (20% V / V ) for 48 h and rinsed with distilled de-ionised water (DIW) before use.Borosilicate 100-ml Kjeldahl digestion flasks were cleaned by boiling a mixture of concentrated nitric perchloric and sulphuric acids in them and rinsing with copious amounts of DIW before use. Resin columns Resin columns consisted of 200 x 8 mm polypropylene columns with a polyethylene frit resin support (Kontes Co. No. K-420160). Sample and reagent reservoirs were 125-ml FEP Teflon separating funnels (Nalgene 4301-0125). The drain stems of the separating funnels were removed and replaced with polypropylene micropipette tips cut to fit securely into the Teflon drain stopcock of the funnel. Reservoirs and columns were cleaned before the resin was loaded by soaking in warm 20% nitric acid for 2-3 h followed by rinsing with copious volumes of DIW.Reagents Concentrated nitric and sulphuric acids. Concentrated Perchloric acid redistilled from Vycor. Concentrated hydrofluoric acid. Analytical-reagent grade. Concentrated ammonia solution. Analytical-reagent grade. Chelex 100 iwinodiacetate chelating ion-exchange resin 200400 mesh N a f form. Metal standard solzdions 1 000 p.p.m. Water distilled and de-ionised (DIW). L4mmonium acetate solution 1 M. Distilled by sub-boiling procedure from quartz. G. F. Smith Chemical Co. Columbus, Ohio. Bio-Rad Commercial or prepared from high-purity metals or Milli-Q (Millipore Corp.). Laboratories Richmond CA. metal salts. Prepared from analytical-reagent grade crystals.As prepared the solution was contaminated with unacceptable levels of copper zinc and lead; it was cleaned before use by adjusting the pH to 5.3 with nitric acid or ammonia solution and passing the solution through columns of Chelex 100 resin (NH,+ form). Although the resulting reagent was clean enough for this study a better preparation is to mix high-purity acetic acid and ammonia solution as described by Kingston et al.l0 Sodium tetrahydroborate(Il1). Sodium iodide crystals. Analytical-reagent grade. Concentrated hydrochloric acid. Analytical-reagent grade. High purity pellets (Ventron Corp. Beverley MA). Digestion Samples were digested by either of two procedures using a heating rack and perchloric acid fume trap. Binary nitric - perchloric acid digestions were performed by adding 25 ml of con-centrated nitric acid and 5 ml of concentrated perchloric acid to 1-3 g of dry sample in a 100-nil Kjeldahl flask.The digestion was continued through the vigorous perchloric aci 356 JONES et al. MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst VoZ. I07 reaction and halted after an additional 5 min of boiling. Ternary nitric - perchloric - sul-phuric acid digestions were conducted in a similar manner except that 2 n l of concentrated sulphuric acid were added at the beginning of the digestion. These digestions were continued until the perchloric acid reaction subsided. At this point the heater temperature was in-creased and the digestion was allowed to proceed until all of the residual perchloric acid was distilled from the flask and copious fumes of sulphur trioxide were evolved.The digestion was discontinued at this point. For all digestions sample charring was avoided in order to mini-mise losses of volatile elements (e.g. selenium). Digests were diluted to 100 ml with DIW to dissolve the acid-soluble portion of the samples. Caution-As with all uses of perchloric acid these procedures should be followed only by experienced Post-digestion treatment of insoluble silicates with hydrofluoric acid was performed for some samples and is described later. Separation For this study we followed as closely as possible the separation conditions used by Kingston et aZ.,1° except for the column reservoir system which was inconvenient for our application. Instead we used the simpler gravity flow system described under Apparatus.A solution of pH 5.0-5.5 containing trace transition elements was passed through a bed of Chelex 100 resin (NH,+ form).1° A number of metals were strongly retained by the resin whereas alkali and alkaline earth elements at pH 5 were minimally retained. Ammonium acetate was then used to elute any residual alkali and alkaline earth elements from the resin. Trace elements were removed from the resin as a group with 7-10 ml of dilute nitric acid. This approach is similar to that reported by Baetz and Kenner,' who used elution with ammonium sulphate to remove residual sodium (and probably other alkali and alkaline earth metals) from Chelex 100 before sulphuric acid elution of cadmium sequestered from acid digestions of foods.Resin columns were prepared by pipetting 10.0 ml of a magnetically stirred slurry of 30 g of resin in the Na+ form + 150 ml of DIW into the column. The approximately 2 g of water-saturated resin was then treated successively with two 15-ml portions of DIW two 15-ml portions of 15% V/V nitric acid two 15-ml volumes of DIW to rinse excess of acid from the resin 10 ml of 15% V/V ammonia solution to convert the resin into the functional NH,+ form, and finally two 15-ml volumes of DIW to remove excess of ammonia. A 2-cm layer of water was left above the resin bed. For this study the resin was discarded after a single use although it can be regenerated by the above procedure. The 100 d of diluted digest solution was divided into two fractions. A 5-ml volume of this solution was transferred into a 25-ml calibrated flask and diluted to volume with 20% V/V perchloric acid (the non-Chelexed fraction).The remaining 95 ml were transferred into the Teflon separating funnel. The pH was adjusted to 5.3 & 0.2 with concentrated ammonia solution followed by dilute ammonia solution and dilute nitric acid as required. A 1-ml volume of 1 M ammonium acetate solution (pH 5.3) was added to buffer the solution as in the reference method. At this pH some NBS reference materials developed a slight flocculent white pre-cipitate which did not interfere with the separation other than to slow the flow-rate of sample through the resin. Sample solution was passed through the gravity-flow column by repetitive manual filling from the separating funnel.The effluent was collected in a Teflon bottle, capped and stored for the determination of arsenic selenium and antimony (the HEAA fraction) . The column was washed with 40 ml of 1 M ammonium acetate solution (pH 5.3) to remove sequestered alkali and alkaline earth metals. Excess of ammonium acetate was rinsed from the resin with 10 ml of DIW. Next 10 ml of 15% V/V nitric acid were added to the column to strip the sequestered trace elements from the resin. The nitric acid eluent was collected in calibrated 25-ml stoppered graduated cylinders. Residual nitric acid was rinsed from the resin with a single 4-ml volume of DIW which was collected in the cylinders. The eluents were diluted to 15 ml with water. This 10% V/V nitric acid solution of the trace elements was called the Chelexed fraction.The shrinking and swelling of the Chelex 100 resin with changes in sorbed ions13 did not drastic-ally affect the separation time (usually 4-8 h) although different eluent flow-rates were observed for different cationic forms of the resin. We investigated only the 200400-mesh resin. Larger resin particles may reduce the separation time. personnel familiar with their potentially hazardous properties April 1982 AND HYDRIDE EVOLUTION AAS FOR PLANT AND ANIMAL TISSUE 357 ICAP Measurement Two-point calibrations using a standard blank and the standard concentrations shown in Table I were performed. Calibrations were based on the average of two sequential 21-s exposures of multi-element standards ie. 14 s on line 7 s on background. All sample intensities were measured by averaging two back-to-back 21-s integration periods.These intensity integrations were corrected for inter-element interferences and/or background fluctuations as required and out-put by the ICAP minicomputer as micrograms per millilitre. Raw concentration data were transferred from the spectrometer via magnetic tape to a larger computer for manipulation.’* Each non-Chelexed fraction and the corresponding Chelexed fraction were measured three times first against one calibration using multi-element standards in 20% V/V perchloric acid or 10% V/V nitric acid for non-Chelexed and Chelexed fractions respectively and then by two subsequent calibrations and measurements for a total of six ICAP measurements of each digested sample. These data provided sufficient information to establish an estimate of the ICAP contribution to the over-all analytical variability.The non-Chelexed fractions were measured directly by ICAP against multi-element standards in 20y0 V/V perchloric acid for alkali and alkaline earth elements and for trace elements such as iron manganese and zinc, which were usually present at readily measurable concentrations. The purpose of the relatively concentrated perchloric acid was to provide a sufficiently viscous solution matrix such that minor differences in residual acid concentrations between sample digests would be “swamped” by the perchloric acid and therefore would not contribute to variability in aerosol transport to the plasma.2 This diluent was used for all non-Chelexed fractions from all digestion procedures and for ICAP calibration standards as the viscosity of this diluent was sufficient to overcome any aerosol transport variability associated with residual perchloric sulphuric or hydrofluoric acid from the original digestions.Concentrations of all solutions were determined under the conditions in Table I. HEAA Measurement HEAA measurements for arsenic selenium and antimony were conducted as previously described.12 The initial column effluents of the neutralised sample solution were collected in Teflon bottles treated with 48 ml of concentrated hydrochloric acid and diluted to 160 ml with DIW. Aliquots of this solution ranging from 1 to 20 ml (depending on the analyte concentra-tion) were transferred into test-tube reaction cells and diluted to 20 ml as necessary with 30% V/V hydrochloric acid.Calibration standards covering the range 25-500 ng of each analyte were prepared in 30% V/V hydrochloric acid. Samples for arsenic and antimony measure-ment were treated with sodium iodide pre-reductant to convert the pentavalant species into trivalent states. Hydrides of arsenic selenium and antimony were generated by adding a basic 4% nz/V solution of sodium tetrahydroborate(II1) from a semi-automatic hydride gener-ator. Analyte gases were passed directly into a hydrogen (nitrogen-diluted) flame for measure-ment of the transient absorption signals. Results and Discussion Preliminary Experiments Before evaluation of the Chelex 100 procedure using NBS SRMs preliminary separations were conducted on biological matrix-simulating standard solutions and on trace element-fortified food samples digested with the binary acids.In these experiments nanogram to microgram amounts of Al(III) As(III) Be(II) Cd(II) Co(II) Cr(III) Cu(II) Fe(III) Mn(II), Mo(VI) Ni(II) Pb(II) Se(IV) Sb(III) Sn(II) Te(IV) Ti(III) Tl(I) V(II1) and Zn(I1) were added to the acidified standard solutions of alkali and alkaline earth elements and phosphorus (milligram amounts of calcium magnesium sodium phosphorus and potassium) before pH adjustment and separation. Similar trace element spikes were added to food samples before binary acid digestion and subsequent separation. These trial separations indicated that cadmium copper molybdenum nickel and zinc could be consistently recovered quantitatively in the Chelexed fraction from most sample types examined.Cobalt chromium iron manganese lead and vanadium were occasionally recov-ered at about 100% for several (dissimilar) matrices but frequently exhibited low (less than 80%) recovery. Aluminium and titanium were recovered quantitatively from undigeste 358 JONES et ai. MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst VOJ. I07 standards but yielded variable recoveries (60-120~0) at low spike levels for acid-digested foods presumably owing to leaching of aluminium and titanium from the borosilicate digestion flasks. Recoveries of these elements in the Chelexed fraction at spikes greater than about 10 pg were quantitative. Beryllium and tin recoveries in the Chelexed fractions were generally less than 60%.No arsenic selenium antimony thallium or tellurium was detec-table in the Chelexed fraction of any of the test samples. However HEAA measurement of the initial column effluent indicated that arsenic and selenium added as spikes passed through the column quantitatively in most instances. Matrix elements (calcium magnesium and potassium) were present at very low concentra-tions in the Chelexed fraction usually less than 1 pg ml-l. Sodium was not measured by ICAP because no direct-reading channel was available but no visible 589-nm emission was observed. The amount of phosphorus in the Chelexed fraction was usually less than 20 pg ml-l and appeared to be directly associated with the amount of aluminium and iron in the final 10% V/V nitric acid eluent. All of the matrix elements (except sodium) were readily measurable in the non-Chelexed fraction.Evaluation Using NBS Standard Reference Materials The analytical scheme employing ICAP measurement of the Chelexed and non-Chelexed fractions and HEAA measurement of the initial column effluents was evaluated for several of the above elements using variations of the binary and ternary acid digestions. NBS SRMs were used for this evaluation in order to avoid differences in chemical behaviour of “synthetic spikes” versus the behaviour of “naturally occurring” trace elements. Sample solubility These SRMs may be grouped into two general categories totally acid-soluble and partially acid-soluble. The term “soluble” in this context refers to the absence of visible insoluble materials after digestion with the binary or ternary acids and dilution to volume.SRM bovine liver wheat flour rice flour and tuna are totally soluble whereas orchard leaves, spinach tomato leaves pine needles and brewers’ yeast (and to a much lesser extent oyster tissue) contain siliceous materials that are not readily dissolved by the acid digestion. Brewers’ yeast contains the greatest relative amount of the insoluble materials ; pine needles and oyster tissue contain the least. NBS certification for trace elements in the reference samples is based on total solubilisation of the matrix. We therefore subjected the Chelex 100 procedure to four commonly used digestion procedures binary binary plus hydrofluoric acid ternary and ternary plus hydrofluoric acid. The digestion procedures are similar to those used by NBS to characterise biological reference materials.15 To dissolve high-silicate SRMs acid digests were initially diluted to 90 ml with DIW mixed thoroughly and filtered through nitric acid-cleaned polyethylene frits to collect the residues.The silicate residues collected on the frits were then dissolved by adding 0.5-0.9 ml of hot 47% m/V hydrofluoric acid directly to the polyethylene filter. This hydrofluoric acid solution was then returned to the original 90 ml of digest solution. The solutions were diluted to 100 ml in polypropylene calibrated flasks sampled for the non-Chelexed fraction and carried through the Chelex procedure. For this experiment solutions were initially made basic to pH paper with ammonia solution in order to avoid damage to the glass pH electrodes by hydrofluoric acid.No hydrofluoric acid treatments were performed for bovine liver tuna wheat flour, rice flour or oyster tissue. The high-silicate SRMs required hydrofluoric acid treatment of the insoluble residues to obtain non-Chelexed results that agreed with NBS certified or informational values. This was particularly true for aluminium iron and manganese. The presence of hydrofluoric acid during the separation step however caused precipitation and/or competitive complexation, which inhibited the reaction of Chelex 100 with aluminium iron manganese chromium and lead. No useful Chelexed results were obtained for these elements when hydrofluoric acid was present. Other elements such as cadmium copper nickel molybdenum and zinc were ade-quately recovered from the resin in the presence of hydrofluoric acid.E f e c t s of digestion procedure on removal of matrix the reference materials. Table I1 gives separation efficiencies for calcium magnesium phosphorus and potassium in These results were obtained by comparing masses of matrix element April 1982 AND HYDRIDE EVOLUTION AAS FOR PLANT AND ANIMAL TISSUE TABLE I1 EFFICIENCY OF REMOVAL OF MATRIX ELEMENTS FROM SRMs 359 Proportion of original mass removed yo* No. of A > Digestion samples Ca Mg K P HN03 - HCIO . . . . 40 99.3 f 0.4 99.4 f 0.5 97 f 2 91 f. 6 HNO - HC10 - H F . . 15 94 f 2 9 l f 3 9 5 f 3 9 8 f 1 HNO - HC104 - H,SO 24 98 f 2 9 7 f 3 9 8 f 1 9 o f . 5 HNO3 - HC104 - HzSO4 - HF . . . . 15 92.7 f 0.6 94 f 3 96 f 2 99.0 f 0.4 * Variability expressed as f 1 standard deviation.in the Chelexed and non-Chelexed fractions. All four elements were removed with at least 90% efficiency from the resin before the final nitric acid eluate containing the transition metals. Both the binary and the ternary acid digests yielded essentially quantitative removal of calcium magnesium and potassium from all samples. Phosphorus was removed less efficiently because of the aforementioned association of phosphorus with aluminium and iron. For the hydrofluoric acid-treated samples removal of phosphorus was quantitative. Conversely, h ydrofluoric acid-treated samples gave poorer separation for calcium and magnesium appar-ently because of precipitates (probably Ca2+ Mg2+ Si0,- and F- systems) that formed at pH 5.3 and partially redissolved during the nitric acid elution.Procedural versus measurement precision A minimum of three replicates were digested for each SRM for each digestion procedure, except for the ternary acid digestion of the high-silicate SRMs (orchard leaves tomato leaves, spinach pine needles and brewers’ yeast) for which a single sample was digested and for the binary acid digestion of tomato leaves and pine needles for which duplicate samples were examined. For the purpose of method evaluation each ICAP measurement was treated as a separate “result .” Therefore nine “results” per element (triplicate ICAP results from tripli-cate samples) were available from which to estimate the accuracy and precision of the chemical method and of the ICAP measurement.For the “soluble” SRMs 4-6 samples per SRM were digested by each of the binary and ternary acid procedures representing up to 18 individual “results” per element. Table I11 summarises the elemental results for the non-Chelexed and Chelexed fractions of each SRM for each digestion procedure. Values are given as apparent concentration in micrograms per gram or mass per cent. dry mass as calculated from the measured solution concentrations. The over-all mean and accompanying uncertainties for each element in each SRM are shown in Table 111. The non-parenthetical error estimates expressed as &l stand-ard deviation represent the over-all method error including the variability incurred during sample preparation and during the ICAP measurement step.In addition the uncertainty of the ICAP measurement step (ICAP error) is included in parentheses. These values are the “within” standard deviations which were obtained by an analysis of variance on the triplicate ICAP measurements of each set of replicate sample solutions. They represent the uncertainty associated with the ICAP calibration - measurement step a t a given solution concentration. The data in Table I11 can be used to differentiate between chemical method failure and ICAP error. For example when the ICAP error approaches or exceeds the over-all method error, instrumental uncertainty limits the precision of the analysis. Conversely when the over-all method error is much larger than the ICAP error method imprecision arising from chemical or manipulative variability controls the precision.Poor over-all method precision in the presence of good measurement precision is indicative of method failure. This is illustrated in Table I11 for the determination of manganese in binary acid digests of orchard leaves. For the Chelexed fraction the over-all method mean and standard deviation for three independent digests were 51 & 11 pg g-l. The ICAP measurement precision obtained for triplicate measurements was 1 pg g-l indicating that method failure was responsible for the over-all poor precision and consequently poor accuracy. On the other hand when ICAP measure-ment precision is poor ie. near the detection limit the chemical method performance cannot be satisfactorily evaluated. Using these criteria the data in Table 111 were evaluated to judge the effectiveness of the proposed analytical scheme 360 JONES et al.MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst Vol. I07 Individual elements In Table I11 the symbol $ indicates the experimental result that appears to be closest to the “true” analyte concentration. This symbol is used primarily to distinguish those samples which yield variable results due to silicate occlusions and only one of them is used for all samples although two or more procedures may have yielded mean values with over-lapping errors. For poorly behaved elements such as chromium or lead the symbol is omitted as none of the digestions gave Chelexed values that agreed consistently with NBS or literature values. Comparisons between Chelexed and non-Chelexed values for aluminium determined in digests that were not treated with hydrofluoric acid show that soluble aluminium can be recovered semi-quantitatively from solution for either the binary acid (90 & 6% recovery) or the ternary acid (87 5 12% recovery) digestion by the Chelex procedure.This compares well with recoveries obtained for the diet samples fortified with aluminium at con-centrations above contamination levels leached from the borosilicate glass. Precipitation at pH 5.3 and resin saturation may have contributed to aluminium losses. The accuracy of Chelex determination of aluminium appears to be limited primarily by sample insolubility (silicates). Total dissolution using hydrofluoric acid was required in order to obtain non-Chelexed results that agreed with NBS or literature values.Blank contamination leached by the non-hydrofluoric acid digestions from the digestion flasks prevented evaluation for alu-minium in bovine liver tuna wheat flour rice flour and oyster tissue. Inadvertent contact of hydrofluoric acid with glass volumetric ware prevented aluminium determination for the ternary acid digestions of high-silicate SRMs. Results for cadmium and nickel obtained for several of the SRMs showed generally good agreement with NBS certified values; agreement with NBS informa-tional values for the high-silicate SRMs were not as good. However the Chelexed results agreed well with literature values particularly for nickel in spinach and pine needles. Simi-larly good agreement with literature values was obtained for cadmium in tomato leaves.For cadmium in spinach eight of a total of ten samples gave values very near the 1.5 pg g-l NBS informational value but the other two results were drastically higher (one each in binary and ternary acids) causing apparently poor results for the non-hydrofluoric acid digestions. The source of the two spurious values was unknown. For the other samples the cadmium method performance was primarily controlled by ICAP precision particularly at the lower concentra-tions. Molybdenum behaved well through the Chelex separation for most of the digestion procedures. Some procedural imprecision occurred for the low-molybdenum con-centrations in the high-silicate SRMs but ICAP measurement imprecision appeared to be the more limiting factor.Silicate occlusion of molybdenum did not appear to be a major problem, although the brewers’ yeast in particular did release more molybdenum after hydrofluoric acid treatment. Lead. Results for lead for the high-silicate SRMs were consistently lower than NBS values. Only for the orchard leaves at 45 pg g-l did the Chelex procedure agree well with the NBS value. Lead was lost from the resin during ammonium acetate elution of the samples as seen in our initial attempts to remove lead contamination from the 1 M ammonium acetate solution by passing the reagent through Chelex 100 (NH,+). Lead elution (breakthrough) was observed with continued passage of ammonium acetate solution through the resin. In addition a sample of NBS pine needles (binary acid) subjected to the Chelex procedure with-out the ammonium acetate elution gave a lead value of 10.5 * 0.3 pg g-l which is in good agreement with the NBS value of 10.8 & 0.5pgg;l.Therefore it appears that the ammonium acetate elution step may have to be modified in order to ensure quantitative retention of lead on the resin before nitric acid elution. Additionally measurement precision was poor for lead determinations below about 1 pg g1 for the SRMs (less than 0.06 pg rn-1 of lead for a I-g sample) preventing evaluation of the separation procedure for lead below this level. Chromium and manganese exhibited erratic behaviour on the resin in the SRM analyses like that observed in preliminary standard and food digest separa-tions. I t is unclear from this work whether losses of chromium and manganese are associated with loss from the resin before the nitric acid elutionlOJ1 or are due to retention by the resin during the acid elution.27 However substantial loss of manganese from the resin during the Aluminium.Cadmium and nickel. In addition nickel appeared to be associated with the silicate materials. Molybdenum. Chromium and manganese A@%? 1982 AND HYDRIDE EVOLUTION AAS FOR PLANT AND ANIMAL TISSUE 361 ammonium acetate elution is suspected because as with lead a sample of pine needles digested with binary acid and subjected to the separation procedure without ammonium acetate elution gave a manganese value of 620 pg g-l. This is lower than the NBS value of 675 & 15 pg g1 but significantly higher than the 475 43 pg g-l obtained for the binary acid digestions with ammonium acetate elution.The ternary acid digestions gave a consistently poorer Chelexed recovery for manganese than did the binary acid digestion. This suggests that in addition to losses associated with am-monium acetate elution the digestion procedure significantly affects manganese (and possibly chromium) behaviour on the resin. Ternary acid digestion also resulted in apparently poorer results for chromium in pine needles and tomato leaves relative to the binary acid digestion of the same samples. For spinach and orchard leaves however ternary acid yielded values in good agreement with the NBS values. Such inconsistent behaviour suggests the possibility that mixed oxidation states of chromium and manganese may be responsible for the variable results.The characteristic pink colour of MnO was observed in the 2 ml of residual sulphuric acid from ternary acid digests of pine needles but was absent from the residual perchloric acid from the binary acid digests. Chromium(II1) - chromium(V1) mixtures have been reported to exhibit different behaviour on the resin.28 Other possible causes of variable chromium recovery include volatilisation or silicate occl~sions.~~ Except for tomato leaves the recovery of soluble iron by Chelex 100 from the binary or ternary acid digestion (comparison of Chelexed veuus non-Chelexed values) was generally greater than 90%. This was particularly true for ternary acid digestion of the low-silicate SRMs where good agreement with NBS values occurred. Resin saturation may have contri-buted to loss of iron from the tomato leaves which have the highest soluble iron concentration of the ten SRMs examined.Kingston et aZ.1° recovered an average 90-93% of added iron from sea-water. Our results for much higher concentrations in digested biological matrices generally agree well with their findings. However because iron may be measured directly in the non-Chelexed fraction the latter procedure is preferred. Vanadium exhibited a distinct difference in behaviour between binary and ternary acid digestion. Much lower vanadium results were obtained for samples prepared by the binary digestion than for those digested with ternary acid. This effect was not attribut-able solely to a more complete solubilisation of the silicate matrix by the ternary acid mixture, as the hydrofluoric acid-solubilised samples exhibited similar behaviour i.e.nitric - perchloric -hydrofluoric acid yielded lower vanadium results than nitric - perchloric - sulphuric - hydro-fluoric acid. With the exception of the brewers’ yeast all of the SRMs that contained ICAP-measurable vanadium gave higher results for the ternary acid digestion than for binary acid digestion. Vanadium may be present in the neutralised binary acid digest solution as a mix-ture of two or more species that exhibit different affinities for the resin under the separation conditions employed. Addition of sulphuric acid to the digestion apparently shifts any equilibrium towards the species readily amenable to the separation procedure. No NBS certified or informational values except an uncertified value of 2.8 pg g1 in oyster tissue are available for vanadium in these SRhls.Our results using ternary acid or ternary plus hydro-fluoric acid agree well with other values reported in the literature. With the exception of the high-silicate brewers’ yeast most vanadium in the SRMs appears to have been released by the ternary acid digestion without the addition of hydrofluoric acid. Chelexed and non-Chelexed zinc results were in general agreement with NBS values for all digestion procedures. There appeared to be silicate occlusion of zinc in some sample types e.g. pine needles and brewers’ yeast but it was a small fraction of the total zinc present. Chelexed and non-Chelexed results agreed very well indicating that except for zinc levels below about 5 pg g-l dry mass the Chelex separation offers no advantage over direct ICAP measurement.A few results for SRMs were obtained on Chelexed samples of hydrofluoric acid-treated SRNls. Values for triplicate digestions of tomato leaves were 0.57 0.07 and 0.57 & 0.04 pg 8-l of cobalt for the binary plus hydro-fluoric acid and ternary plus hydrofluoric acid digestions respectively. The NBS uncertified value for cobalt is 0.6 p g g-l. For spinach the results were 1.25 &- 0.06 and 1.15 & 0.03 pg g1 of cobalt for binary plus hydrofluoric acid and ternary acid digestions respectively; the NBS Iron. Vaizadium. Zinc. Cobalt. No cobalt values are given in Table 111 TABLE I11 EFFECT OF DIGESTION PROCEDURE ON THE BEHAVIOUR OF THE ANALYTICAL SCHEME SRM Orchard leaves SRM 1571 .. Tomato leaves SRM 1573 . . Spinach SRM 1570 . . . Pine needles SRM 1575 . . Brewers' yeast SRM 1569 . . Oyster tissue SRM 1566 . . Bovine liver SRM 1577 . . Albacore tuna RM 50 . . Wheat flour SRM 1567 . . Rice flour SRM 1568 . . HNOS - HC104 - H F -0.19 2.03 f 0.02 (f0.02) HNO, -0.006 1.97 ( -0.004 1.98 f 0.04 (f0.03) 2.09 f 0.03 3.04 f 0.05 (f0.03) 3.00 f 0.03 1.24 f 0.08 (+0.02) 1.35 k 0.03 0.40 &- 0.01 (f0.004) 0.41 & 0.02 0.227 f 0.007 (f0.005) -0.008 -0.02 -0.04 -0.005 --0.002 0.153 & 0.003 (fO.001) 0.15 & 0.02 0.0121 -+ 0.0003 (&0.0002) 0.0124 * 0.002 0.042 & 0.004 (iO.0009) 0.045 f 0.0061s 0.0194 5 0.0006 (f0.0006) 0.019 * 0.001 0.0148 f 0.0005 (&0.0004) 0.014 f 0.002 t0.0001 <0.0001 <0.0001 <0.0001 .. Chelexed No n - C h e 1 ex e d NBS or 1it.t Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. C h e 1 ex e d Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. . . Chelexed Non-Chelexed NBS or lit. . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. -0.19 3.08 f 0.05 (fO.01) -0.01 3.10 ( -0.2 1.36 f 0.04 (f0.005) -0.02 1.25 (* -0.17 0.410 f 0.003 (f0.003) -0.05 0.409 -0.05 0.242 f 0.004 (+0.003) -0.007 0.229 -0.0005 0.151 < 0.000 1 0.0120 < 0.000 1 0.041 <0.0001 0.0199 < 0.000 1 0.0148 HNO - HClO HNO - HClO - H F HNO, Orchard leaves SRM 1571 .. Tomato leaves SRM 1573 . . Spinach SRM 1570 . . . . . . Chelexed Non-Chelexed NBS or lit. . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. -0.07 -0.000 2 0.60 & 0.02 (*0.01) 0.62 & 0.02 0.62 * 0.01 -0.000 3 -0.07 0.67 0.02 (&0.02) 0.70 f 0.02 (0.7) 0.60 & 0.06,'' 0.7318 -0.003 -0.1 -0.0005 *0.01) 0.598 -0.006 f0.02) 0.69 (& -0.002 k0.03) 0.88 ( 0.86 * 0.04 (60.02) 0.73 0.05,17 0.90,18 0.89 i 0.0119 0.89 5 0.0 TABLE 111-continued Mg % mlm r-HNO - HClO HNO - HClO - H F HNO, -0.003 -0.02 -0.009 0.118 f 0.003 (fO.001) 0.119 j 0.002 (j0.002) 0.120 0.12 f 0.02,'' 0.1518 SRM Pine needles SRM 1575 .. . . . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. C h e 1 ex e d Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Brewers' yeast SRM 1569 . . . . . . -0.0008 0.173 f 0.007 (+0.004) 0.178 + O.OIOzo -0.03 0.190 f 0.006 (-+0.006) -0.001 0.187 ( Oyster tissue SRM 1566 . . . . . . -0.000 2 -0.0008 -0.141 - 0.143 f 0.004 (f0.004) 0.128 f 0,009 Bovine liver SRM 1577 . . . . . . <0.0001 0.066 f 0.002 (fO.OO1) 0.0604 f 0.0005 <0.0001 0.0657 Albacore tuna RM 50 . . . . . . Wheat flour SRM 1567 . . . . . . <0.0001 0.119 <0.0001 0.042 f 0.001 (fO.001) 0.0427 f O.OO1lle <0.0001 0.0429 Rice flour SRM 1568 .. . . . . . . <0.0001 0.051 f 0.002 (fO.001) 0.0517 f 0.00071@ <0.0001 0.051 p %:Im C HNO - HClO HNO - HC10 - H F HNO, -0.02 -0.001 -0.07 0.206 f 0.004 (f0.005) 0.198 ( 0.197 f 0.007 (f0.006) 0.21 f 0.01 Orchard leaves SRM 1571 . . . . . . Tomato leaves SRM 1573 . . . . . . Spinach SRM 1570 . . . . . . . . Pine needles SRM 1575 . . . . . . Brewers' yeast SRM 1569 . . . . . . Oyster tissue SRM 1566 . . . . . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. -0.05 0.34 f 0.01 (fO.01) 0.34 + 0.02 -0.001 0.35 f 0.01 (fO.01) -0.06 0.35 (fO.-0.05 -0.002 -0.07 0.52 f 0.02 (fO.O1) 0.55 f 0.02 (f0.02) 0.53 (kO. 0.55 f 0.02 0.119** 0.002 (fO.001) 0.117 f 0.004 (f0.004) 0.119 ( 0.12 f 0.02 1.00 f 0.04 (f0.04) 1.02 f 0.03 (f0.03) 1.04 ( -0.04 -0.004 -0.04 -0.04 -0.000 7 -0.07 --0.02 - -0.02 0.78 f 0.02 (f0.02) - 0.79 f (0.81 TABLE I I I-continued SRM Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 . . HNOs - HClO4 HNOS - HCIO --0.01 -1.18 1 0.03 (k0.02) -0.82 1 0.03 (10.03) -(l.l) 1.21 f 0.041D -0.002 ---0.000 8 -0.137 f 0.005 (+0.005) -0.1239 + 0.003310 -0.000 3 -0.160 5 0.006 (k0.006) -0.1443 f 0.003510 H F HNO, -0.01 1.21 &- Chelexed Non-Chelexed NBS or lit.Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Chelexed -0.003 0.85 &--0.000 9 0.142 -0.0004 0.163 HNO - HClO, -0.004 1.47 & 0.07 (50.05) 1.47 f 0.03 -0.009 4.3 & 0.2 (10.2) 4.46 1 0.03 -0.01 3.6 f 0.2 (&0.1) 3.56 f 0.03 0.353 f 0.008 (f0.004) 0.37 f 0.02 -0.009 -0.005 HNOS - HCIO - H F -0.06 1.43 f 0.07 (&0.08) HNO, -0.003 1.47 (f0.02) Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 . . Pine needles SRM 1575 Brewers’ yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit.Chelexed Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Xon-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. -0.13 4.4 f 0.1 (f0.1) -0.009 4.4 (k0.2) -0.1 3.7 f 0.1 (kO.1) -0.02 3.7 (f0.04) -0.05 0.34 f 0.02 (k0.02) -0.01 0.362 -0.07 1.45 f 0.05 (k0.04) -0.009 1.450 1.4 f 0.1 (f0.06) 1.55 f O.O5*O -0.006 0.98 f 0.04 (f0.04) 0.969 f 0.005 -0.002 -0.002 0.98 f -0.003 0.99 f 0.97-f 0.05 (k0.02) 0.97 f 0.06 -0.006 1.23 f (1.22) 0.09 -0.006 1.20 f <0.0001 0.130 f 0.136 f t0.0001 0.005 0.004 (f0.006) <0.0001 0.131 -= 0.000 1 0.114 f 0.115 f 0.008 (+0.008) 0.112 f 0.00 TABLE 111-continued A11w g-l SRM Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 .. Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 . . HNO - HClO HNO - HClO4 - H F HNOs ._ Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. . . . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. . . . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. .. . . Chelexed . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. . . Chelexed 158 f 23 (f3) 178 f 28 (+23) 347 f 8,21 330a' 78 f 18 (f1) 322 f 18 (f20)$ 446 f 10 ( f 6 ) 473 f 20 ( f 2 1 ) (1 200) 443 f 36 ( f 6 ) 463 f 39 (f24) 870 f 50 202 + 43 (f15) 1170 f 60 (540)z 178 & 35 ( f 2 ) 782 f 31 (f33)$ 414 f 19 ( f 6 ) 472 & 12 (&lo) 545 f 30 307 f 24 (f13) 526 & 17 (f19)$ 476 f 67 (f10) 585 f 150 (h35) 2300 4 lozo -7 --465 & 34 ( f 6 ) 2000 f 56 (f61)$ -7 -7 -a -7 -7 -7 ----7 -7 -Cd/I% g-' I HNO - HClO HNO - HClO - H F HNO, 0.17 f 0.07 (*0.07) 0.12 & 0.08 (&0.09) 0.2 0.11 f 0.02 - - Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1670 .. _ _ Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 2.56 + 0.06 (k0.05) (3) 2.7 f 0.41' 2.6 f 0.1 ( f O . l ) - - 2.40 2.2 * 1 (*0.1) 1.40 f 0.08 (fO.09) 2.8 - -(1.5 TABLE III-continued Cd/w g-' t HN0,- HCIO HNO - HClO - H F HNO, 0.14 + 0.07 (f0.07) 0.16 f 0.09 (kO.09) 0.3 SRM Pine needles SRM 1575 . . Brewers' yeast SRM 1569 . . Oyster tissue SRM 1566 . . Bovine liver SRM 1577 . . Albacore tuna RM 50 . . Wheat flour SRM 1567 . . Rice flour SRM 1568 . . . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Non-Chelexed NBS or lit.Chelexed Chelexed Chelexed Non-Chelexed NBS or lit. - -(<0.5) 0.18 5 0.03l' 0.08 f 0.04 - (k0.04) 0.18 f 0.07 --0.12 k0.08) 3.61 & 0.03 (fO.03) 3.5 f 0.4 0.35 5 0.05 (&0.06)$ 0.27 f 0.04 0.06 & 0.03 (&0.04)$ ----0.04 5 0.01 (&O.Ol)$ 0.032 & 0.007 0.04 f 0.02 (fO.Ol)t 0.029 f 0.004 --3.54 0.39 0.08 0.05 0.06 Crlw g-' I HNO8 - HClO HNO HClOp - H F HNO, 2.4 f 0.1 ( f O . l ) 2.6 f 0.3 - <0.05 - 2.6 Orchard leaves SRM 1571 . . Tomato leaves SRM 1573 . . Spinach SRM 1570 . . . . Pine needles SRM 1575 . . Brewers' yeast SRM 1569 . . Oyster tissue SRM 1566 . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit.Non-Chelexed NBS or lit. Chelexed . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 3.8 f 0.2 (50.2) 4.5 f 0.5 -2.28 <0.05 -3.6 f 0.5 (50.3) 4.6 -f 0.3 - t0.05 - 4.6 2.9 f 0.2 (f0.2) 2.6 f 0.2 - t0.05 - 1.3 1.2 f 0.6 (f0.1) 2.12 5 0.05 - 0.7 <0.05 -0.34 f 0.09 (h0.07) 0.69 & 0.27 -0. TABLE III-continued SRM Bovine liver SRM 1577 . . . . . . Chelexed Non-Chelexed NBS or lit. Non-C helexed NBS or lit. Albacore tuna RM 50 . . . . . . Chelexed Wheat flour SRM 1567 . . . . . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Rice flour SBM 1568 . . . . . . . Chelexed Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 . . Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 .. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. . . Chelexed . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. . . Chelexed Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. CrII.% g-' HNO - HCIO HNO - HClO - H F HNO, 0.4 f 0.5 (f0.5) 0.088 f 0.009 -1.4 f 0.3 (fO.1) 1.75 f 0.12" -0.4 f 0.2 (f0.07) -0.2 + 0.2 (f0.05) -0.4 1.5 0.3 0.08 Culw g-' r HNO - HClO HNOs - HClO - H F HNO, 12.0 12.0 f 0.4 - (+0.3) 12 f 1 10.4 f 0.5 (f0.6)$ 9.8 f 0.4 (A0.4) 8.2 11 f 1 11.1 f 0.5 (f0.5) 11.2 5 0.4 (f0.3) 10.9 12 & 2 12.0 f 0.8 - (f0.4) - -- -3.0 f 0.5 (f0.6)$ 3.0 f 0.3 2.7 f 0.2 (f0.2) 2.9 - -13 f 1 (h0.6) 18.1 f 0.7 (A0.3) 17.7 11 f 280 62.9 f 0.5 (&0.6)$ - 61.8 - -- -63.0 f 3.6 185 f 9 (*8) 193 A 10 - 189 - -3.2 f 0.3 (f0.3) - 3.1 - -2.38 0.9," 3.27" 1.8 - 1.9 f 0.2 (f0.2) 1.9 f 0.2 (50.2) - 1.9 - -2.0 f 0.3 - -2.2 f 0.TABLE 111-continued F e l 2 g-l r HNO - HClO HNO - HClO - H F HNOI 238 * 20 ( f 5 ) 99 f 9 ( f 3 ) 271 494 f 10 (fll) 256 f 11 (f8) 290 It 6 ( f 4 ) 283 300 f 20 631 f 14 (f14) 642 f 17 (f 13) 625 690 f 25 193 f 34 (f7) 538 SRM Orchard leaves SRM 1571 Chelexed Non-Chelexed NBS or lit.Chelexed ,. Tomato leaves SRM 1573 Non-Chelexed NBS or lit. Spinach SRM 1570 . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. C h e 1 ex e d Non-Chelexed NBS or lit. 462 f 33 ( f 9 ) 491 f 20 (fll) 550 f 20 154 f 2 ( f 3 ) 175 f 7 (+4) 200 f 10 136 f 22 ( f 2 ) 541 f 15 (f14) Pine needles SRM 1575 79 13 ( 5 3 ) 194 f 4 ( f 4 ) Brewers’ yeast SRM 1569 217 f 5 ( f 4 ) 257 f 34 ( f 9 ) 707 16*O 344 f 23 (f8) 660 f 15 (f17) 499 440 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 50 f 2 (k0.8) 52 f 2 (*I) 57 f 2,1° 58.1aa 17.1 f 0.8 (f0.2) 17 f 1 (*0.8) 18.3 f 1.0 Wheat flour SRM 1567 18 17.9 Rice flour SRM 1568 .. 1.1 f 0.4 ( f O . l ) r . 3 5 0.4 (10.3) 8.7 f 0.6 HNO - HClO - H F 23 f 2 (f0.6) 37 89 f 1 (*I) 84 HNO, Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 26.8 22 TABLE III-continued SRM Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 .. HNO - HCIO, 475 f 43 (*lo) 652 f 15 (f6) 675 f 15 6.4 & 0.2 ( f O . 1 ) 9.1 f 0.6 (f0.6) 7.0 f 0.89O Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed IVBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed . . Chelexed Non-Chelexed NBS or lit. 2.70 f 0.08 (fO.09) 10.4 f 0.8 (fO.9) 2.35 9.6 17.2 f 0.2 (f0.2) 17.8 f 0.9 ( f l ) 17.5 f 1.2 9.7 f 0.8 (f0.3) 10.5 f 0.3 (f0.4) 10.3 f 1.0 6.2 17.4 4 f 10.4 0.49 f 0.04 (h0.07) 0.7 f 0.4 (f0.3) (1.3) 0.54 f 0.02,16 0.61'' 8.0 f 0.4 ( f O . l ) 8.2 & 0.3 (f0.3) 8.5 f 0.5 19.1 f 0.9 (fO.2) 20.2 f 0.5 (h0.5) 20.1 f 0.4 0.27 0.6 3 8.3 8 20.1 A -HNO - HClO - H F HNO HNO - HClO, 0.2 f 0.1 (*0.1)$ -0.3 f 0.1 Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 .. Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit. 0.2 f 0.1 (fO.1) - 0.2 0.5 f 0.1 ( i O . 1 ) 0.65 & 0.10," 0.62 f 0.0418 0.2 f 0.1 (f0.08) --0.3 i 0.117 0.2 f 0.1 ( f O . 1 ) 0.1017 -3.3 + 0.3 ( i O . 1 ) --0.1 -+ 0.1 (hO.1) (<0.2) -3.1 f 0.5 ( f O .l ) 3.4 f 0.1 -0.5 f 0.1 (fO.l)$ - 0.4 0.4 f 0.2 (fO.O9)$ - 0.3 0.13 f 0.06 (f0.04)$ - 0.2 3.9 f 0.2 (h0.1) - 3.4 t0.07 3. TABLE III-continued Moll&3 g-' r HNO - HClO4 - H F HN03 HNO - HClOd SRM Albacore tuna RM 50 0.1 f 0.1 (fO.1) --0.39 f 0.09 (kO.09) -(0.4) 1.59 f 0.07 (f0.08) -(1.6) Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 0.42 Wheat flour SRM 1667 1.59 Rice flour SRM 1568 . . Nilwg g-' I HNO - HClOd HNOI - HClOd - H F HNOI 1.27 1.15 f 0.07 (f0.08) - 1.15 f 0.09 (f0.08) 1.3 f 0.2 1.12 f 0.06 (f0.05) 1.2 f 0.318 -1.10 1.12 f 0.08 (kO.09) - -Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 .. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 4.9 f 0.2 (fO.l) - 5.1 f 0.1 (kO.1) - 5.4 Chelexed Non-Chelexed NBS or lit. (6) 4.1 f 0.5,'8 5.4 f LO,*, 5.68 f 0.2818 2.07 -f 0.07 (f0.06) -(3.5) 2.3 f 0.2," 2.218 4.6 f 0.3 (10.2) --0.97 f 0.09 (f0.05)$ 1.03 f 0.19 -t0.06 --0.7 f 0.2 (f0.04) 1.02 f 0.121' 0.16 f 0.04 (40.04) --(0.18) 0.15 f 0.02 (f0.02) (0.16) -2.2 f 0.1 (f0.1) - 2.24 Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 Wheat flour SRM 1567 Rice flour SRM 1568 . . Chelexed Non-Chelexed NBS or lit. Chelaxed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit.Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 4.8 5.9 i 0.2 (f0.2) -0.92 0.9 0.20 0.1 TABLE 111-continued Pblwg g-1 r HNO8 - HClO HNOS - HCIO - H F HNO, 44 f 2 (f0.08) 12 f 2 (f0.4) 46 45 f 3 - -SRM Orchard leaves SRM 1571 Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Tomato leaves SRM 1573 5.0 f 0.2 (*0.2) 6.3 f 0.3 0.8 f 0.3 (*0.2) 1.2 f 0.2 9.6 f 0.4 (f0.4) ---10.8 f 0.5 0.6 f 0.5 (f0.5) --0.5 f 0.2 (f0.3) 0.48 f 0.04 0.3 f 0.3 (f0.2) 0.34 f 0.08 0.5 f 0.3 (f0.3) ----1.1 * 0.2 (f0.2) 4.3 -Spinach SRM 1570 .. 0.2 f 0.1 (kO.1) - 0.8 Pine needles SRM 1575 8.4 -f 0.4 - (f0.3) 9.8 Brewers' yeast SRM 1569 0.5 f 0.1 (fO.1) - 0.2 Oyster tissue SRM 1566 0.5 Bovine liver SRM 1577 0.4 --Albacore tuna RM 50 0.4 Wheat flour SRM 1567 Rice flour SRM 1568 . . <0.1 - -VIW g-l r HNO - HClO HNO - HCIO - HF HNO, 0.3 f 0.1 - (f0.05) 0.60 f 0.2" 0.61'' 0.25 f 0.08 - (f0.06) 0.54 Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 .. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. 0.79 f 0.07 (f0.03) 0.69 f 0.09 (fO.08) 1.19 - -1.3 f 0.218 0.6 f 0.1 (k0.06) 0.74 f 0.06 (f0.07) 1.34 1.06 f 0.1718 - - . SRM Pine needles SRM 1575 . . Brewers' yeast SRM 1569 . . Oyster tissue SRM 1566 . . Bovine liver SRM 1577 . . Albacore tuna RM 50 . . Wheat flour SKM 1567 . . Rice flour SRM 1568 . . . . . . Orchard leaves SRM 1571 . . Tomato leaves SRM 1573 . . Spinach SRM 1570 . . . . Pine needles SRM 1575 . . Brewers' yeast SRM 1569 . . Oyster tissue SRM 1566 . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed Non-Chelexed NBS or lit.Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed Chelexed . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed . . Chelexed Non-Chelexed NBS or lit. . . Chelexed Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Non-Chelexed NBS or lit. Chelexed . . Chelexed TABLE 111-continued VIW g-l I HNO - HC10 HNO - HClOp - H F HNO, 0.37 0.33 f 0.05 (f0.06) 0.47 f 0.081B 0.28 f 0.08 (f0.09) - -1.46 1.31 f 0.09 (f0.06) 4.1 f O.l*O 4.27 f 0.09 (f0.06) - -0.68 f 0.06 (f0.06) -(2.8) <0.05 -0.055 f 0.0022e 2.44 0.09 HNO - HCIO, 25 f 1 (f0.9) 25 f 3 56 f 2 ( f 2 ) 59 f 3 (f1) 62 f 6 46 f 1 (f0.8) 46 f 2 ( f 2 1 50 f 2 60 f 3 ( f 3 ) 63 f 3 ( f 3 ) 53.5 f 2.0,'' 52,ln 61 f 10'" 59 f 6 ( f 3 ) 63 f 2 ( f 2 ) 70 f 2eo 859 f 9 ( f W $ 878 f 15 (f12) 852 f 14 25 f 1 Znl!-G g-' HNO - HCIO - H F HNO, 23.5 f 0.9 (fl) 26 f 2 (fl) 46.2 f 0.6 (f0.5) 48 f 3 ( f 2 SRM Bovine liver SRM 1577 .. . . . , Albacore tuna RM 50 . . . . . . Wheat flour SRM 1567 . . . . . . Rice flour SRM 1568 . . . . . . Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. Chelexed Non-Chelexed NBS or lit. TABLE 111-continued HNO - HClO, 134 f 7 (f3) 135 f 4 ( f 5 ) 130 f 10 14.1 f 0.3 (f0.4) (13.6 * l) 14.7** 14 + 1 (rti) 10.6 f 0.5 (f0.2) 10.6 f 0.7 (f0.8) 10.6 f 1.0 19.3 f 0.7 (f0.6) 19.4 & 1 20 f 1 (k0.9) HNO - HClO - HF HNO, - 134 135 -14.2 14 10.6 10.5 19.8 20.2 * HNO - HClO - H,SO digestion of NBS orchard leaves tomato leaves spinach pine needles and brewers’ yeast represented by single samples; i NBS certified values given as mean f95% confidence interval.NBS informational values given in parentheses. $ These values represent our “best estimate” of the concentration of the analyte in a given SRM. § A1 contamination occurred as a result of contact of hydrofluoric acid with volumetric glassware subsequent to sample digestion. 7 A1 contamination occurred from glassware during sample digestion. Literature references given under the HNO - HCIO column for convenience no digestion description is intended. No hydrofluoric acid was involved 374 JONES et al.MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst VoZ. I07 uncertified value is 1.5 pg g-l. Chelex 100 separation although additional verification is required. These results indicate that cobalt may be amenable to the HEAA SRM results Table IV summarises the results for arsenic and selenium obtained by the Chelex 100 procedure. No hydrofluoric acid data are given for arsenic because of arsenic contamination in samples treated with hydrofluoric acid. In general the data indicate that the ternary acid digestion must be used if both arsenic and selenium are to be determined in the same sample solution using rapid HEAA with sodium tetrahydroborate( 111) reduction. TABLE IV SUMMARY OF RESULTS OBTAINED BY HEAA FOLLOWING THE CHELEX PROCEDURE AND COMPARISON WITH NBS AND LITERATURE VALUES Found/pg g-l SRM Orchard leaves SRM 1571 Tomato leaves SRM 1573 Spinach SRM 1570 .. Pine needles SRM 1575 Brewers' yeast SRM 1569 Oyster tissue SRM 1566 Bovine liver SRM 1577 Albacore tuna RM 50 . . Wheat flour SRM 1567 Rice flour SRM 1568 . . Element ._ As Se As Se As Se As Se As Se As Se As Se As Se As Se As Se r Binary acid**? 12.0 f 0.6 (f0.4) <0.1 <0.1 <O.l 0.29 f 0.02 (f0.02) 0.15 f 0.01 (fO.01) 0.19 f 0.01 (fO.01) co.1 0.56 f 0.03 (f0.03) 0.98 f 0.05 (f0.08) 2.32 f 0.09 (fO.l) 2.22 f 0.03 (fO.1) 1.0 f 0.1 (f0.2) 3.7 f 0.2 (+0.2) <0.1 < 0.1 ~0.05 <0.05 0.76 f 0.08 (f0.05) 0.28 f 0.03 (f0.05) Ternary acid**? 12.9 (f0.4) <0.1 <0.1 t0.1 0.29 (fO.01) 0.16 (fO.01) 0.19 (f0.03) 0.53 (f0.08) 1.01 f 0.06 (f0.07) 2.42 f 0.08 (*0.1) 1.03 f 0.04 (*0.03) 3.8 f 0.1 (fO.l) 0.91 f 0.03 (f0.05) 0.43 f 0.01 (f0.02) 0.32 f 0.04 (f0.03) <0.1 15.5 f 0.3 (f0.6) co.1 3.3 f 0.2 (fO.l) <0.05 1 NBS or literature value 0.08 f 0.01 0.27 f 0.05 0.06 f 0.02*0 0.15 f 0.05 0.039 f 0.015a0 0.21 f 0.04 0.049 f 0.004''' 0.92 f 0.09'O 13.4 f 1.9 2.1 f 0.5 0.055 f 0.004 1.1 & 0.1 3.3 f 0.4 3.6 f 0.4 (0.006) 1.1 f 0.2 0.41 f 0.05 0.4 f 0.1 10 f 2 -* All values obtained by HEAA following the Chelex procedure.Values obtained for the high-silicate SRMs based on 1-3 samples per digestion. Values for NBS oyster tissue bovine liver, tuna wheat flour and rice flour based on 3-6 samples per SRM per digestion.For selenium little difference was seen between samples treated with hydrofluoric acid and those not so treated. Selenium values were therefore pooled under the two principal digestion procedures nitric - perchloric and nitric - perchloric - sulphuric acids. It is worth repeating that preliminary recoveries of inorganic arsenic spikes added to food samples and carried through the binary acid - Chelex procedure were quantitative. Good recoveries of arsenic were also obtained for analyses of NBS orchard leaves spinach tomato leaves pine needles and brewers' yeast using binary acid digestion. No significant differences were observed between Chelex - HEAA results for the binary or the ternary acid digestions for these SRMs.However for NBS tuna rice flour and oyster tissue severe depression or elim-ination of arsenic HEAA signals was observed for samples subjected to the binary acid diges-tion. The presence of refractory organic arsenic in marine samples that is resistant to con-ventional wet ashing is well d o c ~ m e n t e d ~ ~ ~ ~ ~ and is confirmed here by the inability to generate HEAA signals from the two marine sample types. The ternary digestion or an equivalently rigorous digestion is also required for HEAA measurement of arsenic in most samples such as the rice flour. Similarly for antimony in NBS orchard leaves (NBS value 3.0 pg g-l) we obtained values of 1.15 and 3.00 pg g-l for the binary and ternary acid digestions respectively, indicating that the ternary acid digestion may also be required for successful measurement of stibine by HEAA.Antimony could not be measured quantitatively in any other SRM by either digestion procedure i.e. the level was less than 0.2 pg g l . The principal advantage of collecting the pH 5 column effluent for HEAA measurement is conservation of the analytical solution. Using the proposed analytical scheme 95% of the original solution is available for HEAA measurements and the Chelex ICAP measurement. Conversely a single HEAA measurement for the three hydride-forming elements discussed here would require up to 60% of the original digest solution if the arsenic selenium or antimony aliquots were withdrawn before the column step April 1982 AND HYDRIDE EVOLUTION AAS FOR PLANT AND ANIMAL TISSUE 375 Non-Chelexed results The non-Chelexed results shown in Table I11 illustrate that the alkali and alkaline earth metals and phosphorus are readily measured by direct ICAP after the various acid digestions.The effective dilution volume of these solutions was 500 ml ie. 1-3 g of digested sample is diluted to 100 ml then 5 ml is diluted to 25 ml. For this relatively large dilution factor the total dissolved salt concentration was usually below 500 pg ml-l For these low salt con-centrations no adverse nebuliser effects associated with needle crusting were observed. Further the 500-ml dilution did not inhibit ICAP measurement of soluble iron manganese and zinc in the non-Chelexed fraction. The measurement precision for these elements was slightly poorer in the dilute non-Chelexed samples but was generally not a serious limiting factor in the analysis.An exception was manganese in the Albacore Tuna Research Material where the measurement precision was relatively poor for the very low manganese solution concentration (2-3 ng ml-l) . For iron aluminium and manganese silicate occlusion represents a major source of error for several SRMs if hydrofluoric acid is not used in the digestion. Proposed Scheme The analytical scheme presented in Fig. 1 outlines the approach among those examined which appears most reliable when using the separation procedure for analysis of biological samples. The ternary acid digestion is suggested because of the apparent requirement of complete digestion in fuming sulphuric acid for arsenic and vanadium determinations.This digestion may not be the optimum procedure for some elements in all biological matrices e.g. high-calcium samples. The binary acid digestion and subsequent hydrofluoric acid treatment of residual silicates from the two digestions can be used for some sample types. Iron and manganese must be measured in the non-Chelexed fraction because these elements are not consistently well behaved on the resin. Chromium is not included in the scheme. The dependence of the separation procedure on oxidation state and pH must be further investigated to determine if these three elements can be isolated quantitatively in the Chelexed fraction. Alumin-ium can apparently be determined semi-quantitatively if insoluble silicates are absent and if contamination is reduced e.g.by use of quartz digestion flasks. Semi-quantitative lead estimates can be obtained on most samples. Cobalt and antimony appear to be determinable by the scheme but further study is required for these elements. Elements in parentheses are amenable to the scheme under favourable conditions. Dilute to 25 ml Chelex 100 Separate Procedure of trace elements [As Se (Sb)] Fig. 1. Proposed analysis scheme 376 JONES et a,?. MULTI-ELEMENT SCHEME USING PLASMA EMISSION Analyst VoZ. I07 Quantitation Limits These values were obtained from estimates of the solution detection limits (micrograms per millilitre) in the Chelexed non-Chelexed and HEAA fractions. Detection limits were multiplied by 5 to obtain solution quantitation limits which were then multiplied by the appropriate dilution factors i.e., 15 500 and 160 ml for the Chelexed non-Chelexed and HEAA fractions respectively.Table V lists approximate quantitation limits in micrograms. TABLE V APPROXIMATE MINIMUM ANALYTE MASSES REQUIRED FOR ACCURATE ANALYSIS BY PROPOSED MULTI-ELEMENT SCHEME Element Alt . . As . . Ca . . Cd c o . . cu . . Fe . . K Mg . . Minimum required masslClg 2.0 0.20 2.5 0.15 0.30 0.15 5.0 250 5.0 Measurement Minimum required step* Element mass/Clg ICAP-C Mn . . 2.5 HEAA Mo . . 0.90 ICAP-NC Ni 0.15 ICAP-C P . . 250 ICAP-C Pb . . . . 1.5 ICAP-C Se . . 0.20 ICAP-NC Sb . . 0.20 ICAP-NC V . . . . 0.15 ICAP-NC Zn 0.15 Measurement step* ICAP-NC ICAP-C ICAP-C ICAP-NC ICAP-C HEAA HEAA ICAP-C ICAP-C * C = Chelexed fraction; NC = non-Chelexed fraction.t A1 appears to be amenable to the scheme provided that glassware contamination can be avoided. Conclusions The Chelex 100 separation procedure of Kingston et aZ.1° can be used for the separation and concentration of several elements from the inorganic matrices of digested tissue samples. Several elements including cadmium copper molybdenum nickel and zinc react quantita-tively with the resin under most of the conditions examined and can be successfully isolated and determined by ICAP in several diverse matrices. Other elements such as chromium, manganese and iron are subject to losses from the resin possibly because of ammonium acetate elution or mixed oxidation states.The analytical scheme using the Chelex 100 separation followed by ICAP and HEAA measurement is influenced by the type of acid digestion procedure used to destroy the organic matrix. A thorough digestion using the ternary acid appears necessary to ensure quantitative recoveries of arsenic vanadium and perhaps other elements. Other factors including the presence of insoluble silicates which occlude several elements, and ICAP measurement imprecision near detection limits can drastically influence the accuracy of the scheme. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. References Gorsuch T. T. in Belcher R. and Frieser H. Editors “The Destruction of Organic Matter,” Dahlquist R. L. and Knoll J . W. AppZ. Spectrosc. 1978 32 1. Novak J . W. Jr. Lillie D. E. Boorn A. W. and Browner R. F. Anal. Chem. 1980 52 576. Burman J. 5th Annual Meeting of Fed. Anal. Chem. Spec. SOC. Boston MA October 30th-Larson G. F. Fassel V. A. Winge R. K. and Kniseley R. N. Appl. Spectrosc. 1976 30 384. Taylor C. E. and Floyd T. L. Appl. Spectrosc. 1980 34 472. Baetz R. A. and Kenner C. T. J . Assoc. Ofl. Anal. Chem. 1974 57 14. Agarwal M. Bennett R. B. Stump I. G. and D’Auria J . M. Anal. Chem. 1975 47 924. Barnes R. M. and Genna J. S. Anal. Chem. 1979 51 1065. Kingston H. M. Barnes I . L. Rains T. C. and Champ M. A. Anal. Chem. 1978 50 2064. Sturgeon R. E. Berman S. S. Desaulniers A. and Russell D. S. Talanta 1980 27 85. Fiorino J . A. Jones J . W. and Capar S. G. Anal. Chem. 1976 48 120. Bio-Rad Laboratories Product Bulletin No. 2020 July 1978. Capar S. G. and Dusold L. R. Am. Lab. 1978 10 17. Nut. Bur. Stand. ( U S ) Spec. Publ. 1977 No. 492. Nadkarni R. A. and Morrison G. H. J . Radioanal. Chem. 1978 43 347. Van der Sloot H. A. Wals G. D. Weers C. A. and Das H. A. Anal. Chem. 1980 52 112. Pergamon Press Oxford 1970 Chapters 4 and 5. November 3rd 1978 Paper No. 121 Abril 1982 AND HYDRIDE EVOLUTION u s FOR PLANT AND ANIMAL TISSUE 1 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 377 Nadkarni R. A. Radiochem. Radioanal. Lett. 1977 30 329. Munter R. C. Grande R. A. and Ahn P. C. ICP Inf. Newsl. 1979 5 368. Gladney E. S. Anal. Chim. Acta 1980 118 385. Maienthal M. J. J . Assoc. Off. Anal. Chem 1972 55 1109. Lad J . C. and Rancitelli R. A. J . Radioanal. Chem. 1977 38 461. Evans W. H. Dellar D. Lucas B. E. Jackson F. J. and Read J. I. Analyst 1980 105 629. Matsumoto K. and Fuwa K. Anal. Chem. 1979 51 2355. Blotcky A. J. Falcone C. Medina V. A. Rack E. P. and Hobson D. W. Anal. Chem. 1979 51, Myron D. R. Givand S. H. and Nielson F. H. J . Agric. Food Chem. 1977 25 203. Riley J. P. and Taylor D. Anal. Chim. Acta 1968 40 479. Leyden D. E. Channell R. E. and Blount C. W. Anal. Chem. 1972 44 607. Kumpulainen J. T. Wolf W. R. Veillon C. and Mertz W. J . Agric. Food Chem. 1979 27 490. Lunde G. J . Sci. Food Agric. 1973 24 1021. Hoffman I. and Gordon A. D. J . Assoc. Off. Anal. Chem. 1964 47 629. 178. Received April lst 1981 Accepted August 24th 198

ISSN:0003-2654    

DOI:10.1039/AN9820700353

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

378 Analyst April 7982 Vol. 107 $9. 378-384 Determination of Mercury in Pharmaceutical Products by Atomic-absorption Spectrophotometry Using a Carbon Rod Atomiser Pamela Girgis Takla and Victor Valijanian Welsh School of Pharmacy University of Wales Institute of Science and Technology King Edward V I I Avenue Cardiff CF1 3NU An electrothermal atomisation procedure using a carbon rod atomiser is described for the determination of mercury after its extraction with dithizone (diphenylthiocarbazone) into chloroform. The increased stability of mercury after extraction allows drying and ashing to be carried out adequately without loss prior to atomisation. The sensitivity to mercury in the carbon rod atomiser is 1.1 x 10-lo g to give 1% absorption a t 253.7 nm. Calibration graphs are linear over the range 0.2-2.0 pg ml-l of mercury.The method can be applied directly without a preliminary digestion procedure to solutions containing organic mercurial preservatives or bactericides [phenylmercury (11) acetate or nitrate or thiomersal (sodium ethylmercurithiosalicylate)] and to trace determination in basic and some neutral and acidic compounds soluble in water or in dilute acids. A standard additions procedure is recommended t o overcome possible matrix effects and to allow the detection of contamin-ation errors. The results compare favourably with those obtained by the conventional cold vapour technique. Keywords Mercury determination ; dithizonate ; phavmaceuticals ; atomic-absor$tion spectrophotometry ; electrothermal atomisation A rapid and specific method is needed to determine low mercury levels that may be present in drug substances or formulations either from the use of mercury compounds as additives or due to contamination.The British Pharmacopoeia gives no procedure for the control of the organomercurial compounds phenylmercury(I1) acetate or nitrate or thiomersal (sodium ethylmercurithiosalicylate) in injections eye drops or eye lotions although their use as bactericides in such preparations is officia1.l Little information is available concerning the likely extent of mercury contamination but the only general limit test provided by the British Pharmacopoeia is the non-specific thioacetamide test for heavy metals,l which is considerably less sensitive for mercury than for lead which is used as the standard for comparison.The determination of mercury by atomic-absorption spectrophotometry has been widely used since the introduction of the cold vapour technique,2 which involves the reduction of ionic mercury with tin(I1) chloride and is sensitive and specific. Matrix effects can interfere with the release of mercury from the test solution and the use of a standard additions method is re~ommended.~ The main disadvantage of the procedure is that a preliminary digestion step is generally required for organically combined although a combined tin(I1) chloride - cadmium chloride reagent5 has been found to release mercury from phenylmercury. Reliable wet-digestion procedure^^,^ that prevent losses of the volatile mercury are tedious and potentially hazardous and reviews7.* of the copious literature on mercury analysis show that the numerous more rapid procedures that have been developed are usually of limited applicability.Thus various partial digestion procedures recommendedg-ll for organomercurial preservatives have been used only for certain specified medicinal products and not all will produce cleavage of the mercury - sulphur bond of thiomersal.12 For a procedure to be generally applicable yet avoid a digestion stage it seemed advisable to separate the mercury as far as possible from the product being tested. For this purpose, dithizone (diphenylthiocarbazone) extraction was seen to be a well established method,s as mercury(I1) ions in acidic or neutral solution will react13 with excess of dithizone (H2Dz) in an organic solvent to give a 1 2 complex Hg(HDz),.Alkyl- or aryl-organomercury(I1) salts will react14 to form 1 1 complexes RHg(HDz) which in some instances are decomposed at acidic pH values to give mercury(I1) dithizonate. Of the other metals only copper an GIRKIS TAKLA AND VAL1 JANIAN 379 the noble metals will react appreciably with dithizone from 0.1 N acid solution. As the basis of a spectrophotometric procedure the dithizone method requires a preliminary wet digestion to ensure that only mercury(I1) dithizonate is obtained as different organic com-plexes can vary in their absorption spectra.l41l5 Also the colour reaction can suffer reversible photochromic changes.16 In this work it was found that although direct electrothermal atomisation procedures for mercury are usually hampered by mercury losses even during drying,17 the mercury content of extracted dithizone complexes can readily be determined using a carbon rod atomiser.Pre-atomisation temperatures of up to 250 “C can be used without loss of mercury. Back-ground absorbances are low and no digestion is needed even for thiomersal as recoveries of mercury from organic and inorganic combinations are in agreement. Measurements also are unaffected by photochromic effects. Experimental Reagents Mercury(I1) nitrate stock standard solution 1.00 mg ml-1. A solution of mercury as mercury(I1) nitrate in approximately 1 M nitric acid. This solution is used to prepare dilute standard solutions of mercury using 0.1 M hydrochloric acid as diluent. Phenylmercury(I1) acetate.Phenylmercury(I1) nitrate. Thiomersal. Dithizone. Dithizone stock solution 0.05y0 m/V in chloroform. Store in a dark-glass bottle in a refrigerator .6 Dilute dithizone solution 0.005~0 m/V in chloroform. Prepare by dilution of the stock solution. This solution may be kept in a dark-glass bottle a t room temperature for at least 1 week. Solutions of mercury in dilute dithizone solution. Dilute 1.0 ml of a suitable dilution in absolute ethanol of the mercury(I1) nitrate stock standard solution to 10.0 ml using dilute dithizone solution as diluent. Hydrochloric acid 0.1 M. Chloroform. Absolute ethanol. Water. Containing 100 -+ 0.5% m/m of Hg calculated as C,H,HgO,. Containing 97.6% m/m of Cl,HllHg,N04 determined by the procedure of the British Phannac0poeia.l Minimum content 97% m/m of C,H,HgNaO,S.Use analytical-reagent grade material or purify if necessary.6 Prepare from acid supplied as “low in lead.” De-mineralise using an Elgastat or other suitable de-mineraliser. Apparatus A Varian Techtron CRA-90 carbon rod atomiser was used in conjunction with a Varian Techtron Model AA-175 atomic-absorption spectrophotometer. The CRA-90 was kept cool with water flowing at a rate of 2 1 min-l and set up using standard-size cup atomisers and with nitrogen as the inert gas at a pressure of 10 lb in-,. A single-element mercury hollow-cathode lamp (Varian Techtron) was used as a line source with a lamp current of 3 mA to determine mercury at 253.7 nm. A spectral band-width setting of 0.5 nm was used for total absorbance measurements and of 1 .O nm for corrected absorbance measurements.Atomisation peak absorbance signals were registered on a digital readout with the spectro-photometer set in the peak readout mode. Peak signals obtained during the temperature programme were recorded with an Oxford 3000 chart recorder which was connected to the test socket of the spectrophotometer. Simultaneous measurements of corrected absorbance were made using a hydrogen lamp (Varian Techtron) in conjunction with the mercury lamp. Liquid samples were injected into the cup of the carbon rod using a 5-p1 Autopette injection syringe (Excalibur Laboratories Ltd.) fitted with disposable polypropylene tips. Pre-cautions against mercury contamination and loss were observed at all times.After washing, glassware was soaked overnight in 50% nitric acid and rinsed well with water prior to use.l8 Procedures ExtractioFt of mercury using dithizone Procedure A for aqueous solutions containing organic mercuriai compounds as preservative 380 Analyst Vol. 107' or foy sterilisation. Accurately dilute a suitable volume of the solution with 0.1 M hydro-chloric acid to obtain 100 ml of test dilution containing the equivalent of about 0.05 pg ml-1 of mercury. Pipette 20.0-ml aliquots of the test dilution into each of four separators to which have been added 0 2.0, 3.0 and 4.0ml of a dilute standard solution containing 0.5pgml-1 of mercury and dilute to 25 ml with 0.1 M hydrochloric acid. Determine also a normal calibration graph by pipetting 2.0 3.0 and 4.0 ml of the dilute standard solution containing 0.5 pg ml-l\of mercury into individual separators and dilute each solution to 25 ml with 0.1 M hydrochloric acid.Measure 25 ml of 0.1 M hydrochloric acid into another separator for a blank determination. Extract the solution in each separator by shaking once for 3 min with 2.0 ml of dilute dithizone solution. Allow the layers to separate and run the lower layer into a glass-stoppered tube. NOTE-The dilute dithizone solution should normally show no change from its original colour (blue -green) after extraction has been carried out. The appearance of a mixed colour or the characteristic orange colour of mercury(I1) dithizonate indicates that the concentration or the volume of the test dilution used should be reduced.Interference from copper is not usual but the presence of relatively large amounts of copper could cause some red copper(I1) dithizonate to form after the reaction with mercury is complete.8 GIRKIS TAKLA AND VALIJANIAN HG IN PHARMACEUTICAL Carry out the method of standard additions as follows. This should not invalidate the determination. Procedure B for solid samples liable to contain mercury as a n impurity. Accurately weigh a suitable amount of the solid and dissolve it in 0.1 M hydrochloric acid to obtain a test dilution containing 0.003-0.005 pg ml-l of mercury. As a preliminary trial 10.0 g of solid can be made up to 100ml of solution but the mass needed will depend on the mercury content and may be limited by the solubility of the sample. Carry out the method of standard additions as described for procedure A using aliquots of 100 ml of the test dilution and adding 0 2.0 3.0 and 4.0 ml of a dilute standard solution containing 0.2 pg ml-1 of mercury.Pipette also 2.0- 3.0- and 4.0-ml volumes of the dilute standard solution con-taining 0.2 pg ml-l of mercury into individual separators for the determination of a normal calibration graph and dilute each volume to 100 ml with 0.1 M hydrochloric acid. Measure 100 ml of 0.1 M hydrochloric acid into another separator for a blank determination. Extract the solutions as described in the second paragraph in procedure A. Selection of carbon rod atomiser control unit settings Adjust the CRA-90 control unit settings to the following dry 80 "C for 45 s; ash 220 "C for 15 s ; and atomise 1600 "C with a hold time of 3 s and a ramp rate of 600 "C s-1.The sensitivity to mercury under these conditions is about 1.1 x 10-lo g to give 1% absorption at 253.7 nm. Table I shows that with higher atomisation temperatures or ramp rates there is no improvement in sensitivity. At lower atomisation temperatures and ramp rates there is some loss of sensitivity although for some of the work described here an atomisation temperature of 1200 "C was used to prolong the life of the atomiser cup. TABLE I IN THE CRA-90 CARBON ROD ATOMISER EFFECT OF ATOMISATION CONDITIONS ON MERCURY ABSORBANCE Injection standard extract (2.0 pg ml-l of Hg) in dilute dithizone solution (5 pl) Atomisation temperature/OC (hold time 1200 1600 1800 3 s ) . . . . . . w -7 & Ramp rate/OC s-' .. . . . . 300 600 800 600 800 600 800 Absorbance (corrected) . . . . . . 0.232 0.287 0.337 0.397 0.404 0.390 0.397 Preliminary trials are advisable for establishing suitable settings so that loss of mercury during drying or ashing is avoided. With the instrument used in this study absorbance readings for mercury extracted with dithizone in chloroform remained constant with the dry setting at 80 "C maintained for 30 s up to at least 90 s. Table I1 shows that mercury(I1) dithizonate in chloroform could be dried and ashed up to a temperature of 250 "C without loss of the metal and possessed considerably greater thermal stability than mercury dissolve April 1982 PRODUCTS BY AAS USING A CARBON ROD ATOMISER TABLE I1 EFFECT OF ASH TEMPERATURE ON MERCURY ABSORBANCE USING THE CRA-90 CARBON ROD ATOMISER FOR SOLUTIONS OF MERCURY CONTAINING 2 pg ml-1 OF Hg IN DILUTE DITHIZONE SOLUTION 0.1 M NITRIC ACID AND 0.01 M HYDROCHLORIC ACID Atomisation temperature 1200 "C for 3 s; ramp rate 300 "C s-1.381 Dry settings Ash settings 7- -- Temperature/ Temperature/ Diluent "C Time/s "C Tim+ Dilute dithizone solution . . . . 80 60 150 15 200 30 200 60 250 30 300 30 0 . 1 ~ H N 0 ~ . . 130 90 150 30 200 30 250 30 0.01 Y HCl . . . . . . . . 130 90 150 30 200 30 250 30 Mean Hg absorbance (uncorrected) 0.255 0.244 0.254 0.247 0.058 0.093 0.149 0.088 0.020 0.051 0.017 Standard deviation (9 degrees of Blank freedom) absorbance 0.011 0 -01 5 0.005 O.oi5 0.012 0.015 0.008 0.015 0.013 0.010 0.018 0.020 0.024 0.000 0.012 0.002 - 0.000 - 0.000 - 0.000 in acidified aqueous solution.A further advantage is that high ashing temperatures are unnecessary for the dithizonate or for excess of dithizone as the background absorption produced during atomisation is low even though ashing conditions are unlikely to have been efficient for the destruction of organic matter. The ash temperature setting can however, affect the release of mercury during the atomisation stage. Thus with an ash setting of 200 "C for 15 s lower absorbance readings were obtained for mercury extracted with dithizone from phenylmercury(I1) acetate or phenylmercury(I1) nitrate than for equivalent amounts of mercury extracted from mercury(I1) nitrate or thiomersal and this effect was indepen-dent of the atomisation temperature used.The responses only became equal for each substance when the ash temperature setting was increased to 220 "C for 15 s. Determination of mercury Measure the corrected absorbance of the dithizone extracts adjusting the sensitivity (scale expansion) so that the readings lie within the range 0.2-0.8 absorbance unit. In this work, procedure A was carried out with a scale expansion of about x 2 and procedure B with a scale expansion of x4. Determine the mean response for at least five absorbance measure-ments for each extract. The response for the blank dithizone extract should be zero. Plot the standard-additions calibration graph of corrected absorbance against concentra-tion of mercury added.The absorbance reading at the intercept of the plot 6n the absorbance axis should equal the corrected absorbance reading obtained for the test dilution extract (with no added mercury). The concentration of mercury in the test dilution can be read from the graph in the usual way taking the intercept on the concentration axis as zero concentration. Where the slope of the standard-additions graph is seen to be the same as that of the normal calibration graph it becomes unnecessary to use a standard-additions graph and the concentration of mercury in the test dilution can be determined by reference to a normal calibration graph. Results and Discussion Calibration graphs (Fig. 1) of corrected absorbance zleysus concentration obeyed Beer's law over the range 0.2-2.0 pg ml-l of mercury both for extracts and for direct solutions of mercury made with dithizone in chloroform.At higher concentrations a pronounced negative deviation was observed which could not be prevented either by changing the atomisation temperature over the range 1000-1 800 "C or by altering the dithizone con-centration over the range 0.005-0.02% m/V. The use of 0.005:/ m/V dithizone solution in the recommended procedure for the extraction of mercury from 0.1 M acid solution was found to be convenient as any change from the normal blue - green colour of the extract to purple - green or eventually to orange served to indicate not only that the extraction of mercury might be incomplete but also that the mercury concentration was too high for measurements to be within the linear range of the calibration graph.To determine it 382 GIRKIS TAKLA AND VAL1 JANIAN HG I N PHARMACEUTICAL Analyst VOt!. 107 A 0 2 4 6 8 10 1 e 1 1 , B'O 0.2 0.4 0.6 0.8 1.0 Mercury concentrationipg ml- ' Fig. 1. Calibration graphs for mercury in (A) 0.02% 'm/V and (B) 0.005% m/V dithizone in chloroform determined with a CRA-90 carbon rod atomiser. Absorbance scale expansion : (A) x l and (B) x4. Atomisa-tion temperature 1200 "C. precision the dithizone extraction procedure was carried out in triplicate and absorbance measurements made as shown in Table 111. The mean relative deviation from the mean of the absorbances determined for the three extractions is 2%. The recovery of mercury from organic mercurial preservatives was tested using freshly prepared solutions containing accurately weighed amounts of each of the compounds thio-mersal phenylmercury(I1) acetate and nitrate which were analysed for their mercury content after dilution and extraction according to procedure A.A dithizone extract of mercury(I1) chloride containing 0.5 pg ml-l of mercury was used as reference standard. The results (Table IV) show a good recovery of mercury from each of the compounds tested, and signify that a preliminary wet-digestion stage can be avoided in the determination. A relative standard deviation of 3.9% was calculated for the procedure using five extracts of separate aliquots of the same dilution containing thiomersal. The results are shown in Table V. TABLE I11 PRECISION OF DITHIZONE EXTRACTION AND MERCURY ABSORBANCE MEASUREMENTS USING THE CRA-90 CARBOX ROD ATOMISER Extraction of 4.0 pg of mercury from 100 ml of 0.1 M hydrochloric acid into 2.0 ml of dilute dithizone solution.Atomisation temperature 1600 "C. Deviation from Extraction Mean total Degrees of Standard mean of 3 N O . absorbance freedom deviation extractions 1 0.396 9 0.006 0.012 0.002 2 0.382 9 0.004 3 0.374 9 0.008 0.010 Mean . . 0.384 Mean . . 0.00 April 1982 PRODUCTS BY AAS USING A CARBON ROD ATOMISER TABLE IV DETERMINATION OF MERCURY IN DITHIZONE EXTRACTS OF STANDARD SOLUTIONS OF ORGANIC MERCURIAL PRESERVATIVES WITH THE CRA-90 CARBON ROD ATOMISER USING PROCEDURE A Compound Thiomersal . . . . . . . . . Phenylmercury(I1) acetate . . . . Phenylmercury (I1 ) nitrate (97.6% pure) Mercury content of extractlgg Sample mass/ Sample mass r-v mg extracted/yg Found Calculated 202.3 198.1 167.4 168.5 83.50 87.35 2.023 1.981 1.674 1.685 1.670 1.747 0.988 1.034 0.953 0.998 0.988 1.092 1.002 0.982 0.997 1.004 1.031 1.078 383 Mean Recovery recovery yo 98.6 102 105.3 95.6 98 99.4 101.3 99 95.8 TABLE V PRECISION OF DETERMINATION OF MERCURY IN DITHIZONE EXTRACTS OF A SOLUTION OF THIOMERSAL USING PROCEDURE A Test dilution thiomersal sample mass of 0.201 8 g made up t o 100 ml in water and an aliquot of the solution diluted in 0.1 M hydrochloric acid to contain 2.018 p g in the volume extracted (20.0 ml) into 2.0 ml of dilute dithizone solution.Corrected absorbance ( x 2) * I 7 Mean of Extraction 5 readings Range 1 0.195 0.1 76-0.20 1 0.196 0.1'73-0.2 10 0.189 0.186-0.192 0.179 0.162-0.182 4 0.197 0.174-0.222 0.188 0.166-0.219 2 0.190 0.1 70-0.2 12 3 0.184 0.175-0.195 5 0.201 0.191-0.207 0.203 0.196-0.2 10 Standard deviation Mercury (9 degrees of freedom) recovery,? % 0.016 101.0 101.6 0.010 98.4 97.9 0.009 95.3 92.7 0.021 102.1 97.4 0.006 104.1 105.2 Mean 99.6 Relative standard deviation 3.9% * Absorbance scale expansion x 2.t Mean corrected absorbance determined for three extracts containing 0.5 pg ml-l. of mercury from a dilute standard solution of mercury(I1) chloride = 0.193. Various compounds were tested for the presence of mercury as an impurity as well as for their effect on the recovery of a standard addition of mercury.Recoveries of mercury were quantitative from samples of lactic acid citric acid sodium chloride potassium chloride, potassium citrate zinc sulphate iron(I1) sulphate nicotinamide sulphacetamide and glycerol using 100 ml of a 10% m/V solution of each compound in 0.1 M hydrochloric acid (sulphacetamide was dissolved with the aid of an extra equivalent volume of hydrochloric acid). Recoveries of mercury were also quantitative from a 2% m/V solution of nicotinic acid and a 5% w/V solution of calcium lactate. However recovery was inhibited by potassium bromide and potassium iodide both of which will also interfere in the cold vapour t e c h n i q ~ e . ~ Mercury was detected as an impurity only in samples of potassium chloride, sulphacetamide nicotinamide and citric acid.The background (molecular) absorbances due to these compounds in the extracts were 0 0.9 0.6 and 0.03 respectively. The compounds were then assayed for mercury by procedure B taking 10-g amounts of sample per 100 ml of test dilution and making three standard additions of mercury. The results (Table VI) were compared with those obtained by the cold vapour m e t h ~ d ~ * l ~ which, with the exception of potassium chloride was carried out after a preliminary wet digestions of the sample with nitric acid and sulphuric acids. Reasonable agreement of results was obtained by the two procedures for potassium chloride and sulphacetamide. With the nicotinamide and citric acid samples repeated attempts to digest the samples and subse-quently to determine the mercury by the cold vapour technique or by using dithizon 384 GIRKIS TAKLA AND VAL1 JANIAN TABLE VI COMPARISON OF THE CRA-90 (PROCEDURE B) AND COLD VAPOUR METHODS FOR THE DETERMINATION OF MERCURY IN VARIOUS CHEMICALS Mercury found p.p.m.Sample ’ CRA-90 method Cold vapour method ’ Potassium chloride . . 0.052 0.058 Sulphacetamide . . . . 0.072 0.066 Nicotinamide . . . . 0.030 No recovery Citric acid . . 0.044 No recovery extraction were unsuccessful. The standard-additions calibration graph obtained by the electrothermal atomisation procedure B had the same slope as the normal calibration graph for each of the four compounds. The determination of a standard-additions calibration graph does even in the absence of matrix effects help to prevent errors due to contamina-tion of solutions which is a problem that requires constant attention in mercury analysis.For the determination of mercury in some water-soluble compounds it is not always convenient to adjust the test dilution to an acidic pH. In testing for mercury in sodium hydrogen carbonate for example an aqueous 10% m/V solution was extracted directly with dilute dithizone solution. The omission of acid made other metals also liable to extraction and the sodium hydrogen carbonate caused the extract to show a colour change to red. With the sample examined no mercury could be found in the extract but standard additions of mercury were recovered. In such instances more than one extraction with dithizone should be made until complete extraction of any mercury is assured.Work is in progress to evaluate further applications for the determination of mercury by this procedure. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. References “British Pharmacopoeia 1980,” H.M. Stationery Office London 1980 Volume I pp. 345 and 452, Poluektov N. S. Vitkun R. A. and Zelyukova Y . V. Zh. Anal. Khim. 1964 19 873. Mercury Analysis Working Party of the Bureau International Technique du Chlore Report Anal. Magos L. and Cernik A. A. Br. J . I n d . Med. 1969 26 144. Magos I. Analyst 1971 96 847. Analytical Methods Committee Analyst 1965 90 515. Manning D. C. A t . Absorpt. Newsl. 1970 9 97. Ure A. M. Anal. Chim. Acta 1975 76 1 . Woodward P. W. and Pemberton J. R. Appl. Microbiol. 1974 27 1094. Calder I. T. and Miller J. H. McB. J . Pharm. Pharmacol. 1976 28 25P. May J . C. Sih J. T. C. and Mustafa A. J. J . Biol. Stand. 1978 6 339. Meakin B. J. and Khammas 2. M. J . Pharm. Pharmacol. 1979 31 653. Briscoe G. B. and Cooksey B. G. J . Chem. SOC. A 1969 205. Irving H. and Cox J. J. J . Chem. Soc. 1963 466. Litman R. Williams E. T. and Finston H. L. Anal. Chem. 1977 49 983. Meriwether L. S. Breitner E. C. and Sloan C. L. J . A m . Chem. Soc. 1965 87 4441. Kunert I. Komarek J. and Sommer L. Anal. Chim. Acta 1979 106 285. Gladney E. S. Owens J. W. and Perrin D. R. Los Alamos Scienti=fic Report New i’lfexico LA-Mercury Analysis Working Party of the Bureau International Technique du Chlore Report Anal. and Volume 11 pp. A87 and A196. Chim. Acta 1976 84 231. 7865-MS National Technical Information Service 1979 12pp. Chim. Acta 1974 72 37. Received October 142h 1981 Accepted November loth 198

ISSN:0003-2654    

DOI:10.1039/AN9820700378

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, April, 1982, Vol. 107, PP. 385-391 385 Gravimetric Determination of Nickel with Th iosemicar bazones D. Rosales and J. M. Cano-Pavbn Department of Analytical Chemistry, Faculty of Chemistry, University of Seville, Seville-4, Spain A systematic study of the use of thiosemicarbazones in the gravimetric determination of nickel is reported. The compounds tested were furfural thiosemicarbazone (FAT), thiophen-2-aldehyde thiosemicarbazone (TAT) and furfural 4-phenyl-3-thiosemicarbazone (FAPT) . FAPT is the most appropriate reagent owing to its sensitivity and selectivity. FAPT has been applied to the gravimetric determination of nickel in diverse standard samples. The reagents tested in this work were compared with classical vic-dioximes used in the gravimetric determination of nickel.Keywords: Furfural thiosemicarbazone; thiophen-%aldehyde thiosemi- carbazone ; furfural 4-Phenyl-3-thiosemicarbazone ; nickel determination ; gravimetry Thiosemicarbazones have been used as gravimetric reagents for some metal ions : nickel with ,B-resorcylaldehyde thiosemicarbazonel ; copper with dih ydroxybenzaldehyde, p-dimethyl- aminobenzaldehyde and cinnamaldehyde thiosemicarbazones2 ; cadmium with salicylalde- hyde thiosemicarbazone3 and p-resorcylaldehyde thiosemi~arbazone~ ; mercury with p - ethylsulphonylbenzaldehyde thio~emicarbazone~ ; and palladium with p-ethylsulphonyl- benzaldehyde thiosemicarbazone,6 furfural thiosemicarba~one,~ thiophen-2-aldehyde and benz- aldehyde thiosemicarbazoness and furfural4-phenyl-3-thio~emicarbazone.~ With palladium, the precipitates formed (pH 3-5) are easy to filter off and stable, and can be dried at 170 "C; their compositions are PdL, (HL being the undissociated thiosemicarbazones, which are weak acids).The complexes are slightly soluble and their high relative molecular masses give appreciably higher sensitivities than classical reagents such as dimethylglyoxime, furfur- aldoxime and thiophenaldoxime. In this work, a systematic study was made of the use of thiosemicarbazones in the gravi- metric determination of nickel. The reagents tested were furfural thiosemicarbazone (FAT), thiophen-2-aldehyde thiosemicarbazone (TAT) and furfural-4-phenyl-3-thiosemicarbazone (FAPT). FAPT reagent was applied to the determination of nickel in real samples. The results are compared with those obtained by the use of vic-dioximes.Interferences of foreign ions were examined. Experimental Reagents Thiosemicarbazones were obtained by the methods previously described .7-9 Stock solutions of the reagents were prepared in ethanol at concentrations of 0.9% m/V for FAT, 0.5% m/V for TAT and 0.2% m/V for FAPT. The solutions of nickel were prepared from the nitrate salt and were standardised with dimethyl- glyoxime before use. Acetate buffer of pH 3.7 was prepared by mixing 450 ml of 2.0 M acetic acid with 50 ml of 2.0 M sodium acetate and diluting to 1 1 with water. All chemicals were of analytical-reagent grade. Apparatus calomel electrode. balance. potassium bromide discs). Perkin-Elmer R-12B spectrometer. balance. pH measurements were made using a Philips PW 9408 potentiometer with a combined glass - Thermogravimetric curves were obtained with an Adamel 93 thermo- Infrared spectra were recorded with a Beckman Acculab-2 spectrophotometer (in Nuclear magnetic resonance (NMR) spectra were obtained with a Magnetic susceptibilities were measured with a Gouy386 ROSALES AND CANO-PAV~N : GRAVIMETRIC Analyst, Vol.107 Procedures Determination of nickel with F A T Take a sample solution containing 10-50 mg of nickel, adjust the pH to 8.0-11.0 with an ammonia buffer solution and dilute to 100-150 ml. Add 10-50 ml of 0.9% FAT solution with constant stirring and heat the mixture in a water-bath for 30 min. After 1 h, collect the precipitate on a weighed G-4 sintered-glass crucible, wash with 200-300 ml of 10% ammonia solution in 1+5 ethanol - water, and then with 100-200 ml of 1+5 ethanol - water.Dry at 70-90 "c and weigh. The conversion factor for Ni(C6H6N,S0)2 to Ni is 0.1486. Determination of nickel with TAT Place an aliquot of the nickel solution (10-50 mg of nickel) in a beaker, adjust the pH to 5.0-1 1 .O with acetate or ammonia buffer solution, dilute to 100-150 ml and add 20-100 ml of 0.5% TAT solution. Heat the mixture in a water-bath for 30 min. After 1 h collect the precipitate on a weighed G-4 sintered-glass crucible, wash with 200-300 ml of 1004 ammonia solution in 1 +4 ethanol - water, and then with 100-200 ml of 1 +4 ethanol - water. Dry at 70-90 "C and weigh. The conversion factor for Ni(C,H,N,S,), to Ni is 0.1374. Determination of nickel with FAPT Take a solution containing 5-50 mg of nickel, adjust the pH to 3.3-11.0 with acetate or ammonia buffer solution and dilute to about 100 ml.Add 30-300 ml of 0.2% FAPT solution with constant stirring and heat the mixture in a water-bath for 30 min. After 30-60 min collect the precipitate and wash it with 300400 ml of a 0.1 M sodium hydroxide solution, and then with 200-300 ml of 2+3 ethanol - water. The conversion factor for Ni(C,,H,$J,SO), to Ni is 0.107 3. Dry at 70-90 "C and weigh. Analysis of Real Samples Determination of nickel in Raney nickel A 1-g amount of sample was dissolved in 50 ml of dilute hydrochloric acid (1 +l), the solution was evaporated to dryness, the residue was dissolved in water and the solution was diluted to 100 ml with water. Aliquots of 10 ml were used in the gravimetric determination of nickel with dimethylglyoxime in accordance with the literature.1° A similar procedure was used with FAPT, with 25 ml of acetate buffer at pH 3.7.Determination of nickel in alloy cast iron A 10-g amount of sample was treated with 100 ml of a mixture of sulphuric and ortho- phosphoric acids (20 ml of 96% sulphuric acid plus 10 ml of 85% orthophosphoric acid and 70 ml of water), and boiled for 1 h nearly to dryness. I t was then treated with 200-300 ml of distilled water and acidified, drop by drop, with concentrated sulphuric acid to dissolve the precipitate. The silicic acid that was precipitated was filtered off and the solution was diluted to 500 ml with water. Aliquots of 50 ml were treated with 50 ml of dilute ammonia solution (1+1) and then filtered on Albert 238 filter-paper (equivalent to Whatman 41).The pre- cipitate was dissolved by adding hot hydrochloric acid dropwise and the solution was collected in a beaker, reprecipitated with 50 ml of ammonia (1 + 1) and filtered on the same filter-paper. The solution was concentrated to 300 ml and nickel was determined with FAPT at pH 3.7 with 25 ml of acetate buffer. Alloy Cast Iron 33b had the following certificate composition: iron 91.72%, carbon 2.24y0, silicon 2.00y0, sulphur o.035y0, phosphorus O.llyo, manganese 0.64y0, nickel 2.24y0, chromium 0.61y0 and molybdenum 0.40%. Determination of nickel in chromium nickel steel peroxide. water. addition of ammonia. acetate buffer. nickel 11.84y0, molybdenum 2.16y0 and carbon 0.09%.A 2-g amount of sample was dissolved in 80 ml of aqua regia and a small amount of hydrogen The solution was evaporated to dryness, redissolved and diluted to 200 ml with Aliquots of 25 ml were then treated using the above procedure commencing with the Nickel was then determined using FAPT at pH 3.7 with 25ml of The standard composition of the steel was iron 6S.15($(0, chromium 17.77%April, 1982 DETERMINATION OF NICKEL WITH THIOSEMICARBAZONES 387 Determination of nickel in nickel stone A 1-g amount of sample was dissolved with 100 ml of aqua regia - bromine (3+1) and the solution was evaporated to dryness. The residue was dissolved in 25 ml of concentrated hydrochloric acid, the solution evaporated to dryness, the residue dissolved in water and the solution diluted to 100 ml with water.Aliquots of 5 ml were treated with hydrogen sulphide at pH 0-1 to remove copper. After filtration on Albert 240 filter-paper (equivalent to Whatman 44), nickel was determined using FAPT with 25 ml of acetate buffer at pH 3.7. The standard composition of the stone was nickel 55.89%, copper 30.03y0, sulphur 8.53%, arsenic 0.04%, silica 0.53% and the sum of iron(II1) and aluminium oxides 1 .58y0. Silicic acid was filtered off. Determination of nickel in aluminium bronze A 10-g amount of sample was dissolved in 100 ml of aqua regia and a small amount of hydro- gen peroxide. The solution was evaporated to dryness, the residue was dissolved in 25 ml of concentrated hydrochloric acid, the solution again evaporated to dryness, the residue re- dissolved in water and the solution diluted to 100 ml with water.Aliquots of 25 d were treated with hydrogen sulphide to remove copper, and nickel was determined a t pH 3.7 using FAPT in the presence of 400 mg of tartrate and 25 ml of acetate buffer. Aluminium Bronze 32a had the following certificate composition : copper 85.9y0, iron 2.67y0, aluminium 8.8%, nickel l.l6%, manganese 0.27% and zinc 0.94%. Determination of nickel in aluminium alloy Aliquots of 25 ml were treated first with ammonia, then with hydrogen sulphide following the methods described above. Nickel was determined using FAPT with 25 ml of acetate buffer at pH 3.7. Aluminium Alloy 20b had the following certificate composition : aluminium 91.45%, copper 4.10y0, nickel 1.93%, iron 0.43%, manganese 0.19%, silicon 0.29% and magnesium 1.61%.A 10-g amount of sample was treated in the same manner as the aluminium bronze. Results and Discussion Properties of the Reagents The thiosemicarbazones studied are poorly soluble in water, but their solubilities increase in alkaline medium owing to the deprotonation of the thiol group, which possesses a weak acid character. Solubilities in ethanol are 11.5,8.0 and 2.5 g 1-1 for FAT, TAT and FAPT, respect- ively. Solutions in ethanol stored at room temperature were stable for at least 1 month. Values of the apparent ionisation constants are 2.0 x 10-lf for FAT and TAT1' and 1.6 x for FAPT.g Slow hydrolysis of the reagents to aldehyde and amine occurs in dilute solutions (10-4-10-5 M) at pH 1. The compounds tested appear to be bidentate ligands with a convenient steric arrangement of their donor group, and contain a conjugate system of n-electrons connected with the donor group.The chelates are uncharged. Study of Nickel - Thiosemicarbazone Complexes Elemental analysis of the nickel complexes confirmed the formula NIL,. A study was made to elucidate the structures. The infrared spectra show an appreciable increase in the C=N stretching vibration bands in relation to the reagents at 1600-1620 cm-l, and double bands appear; on the other hand, bands due to the C=S group disappear and C-S vibration bands show up. NhlR spectra in deuterated dimethyl sulphoxide show multiplets between 6.5 and 8.0 p.p.m. due to the aromatic and amine groups. The ultraviolet spectra show a broad band in each instance, with maximum absorption at 335, 340 and 386 nm for FAT, TAT and FAPT, respectively.A study of the magnetic susceptibilities showed that the nickel - thiosemicarbazone com- plexes tested are diamagnetic at 293 K. From these results it can be concluded that the complexes studied are square planar. In general, thiosemicarbazones usually react as chelating ligands with transition metals ions by bonding through the sulphur and hydrazine nitrogen atoms,12 so that the structure of the388 ROSALES AND CANO-PAV~N : GRAVIMETRIC Anahst, Vol. I07 nickel complexes could be as shown in Fig. 1, or the alternative one with the two sulphur and the two nitrogen atoms occupying mutually trans-positions. +2 2 UJ -1 -2 R R - - ( a ) - ( b ) - (c) I+::\ 1- l 8 - P - - I I I , \ / II II CH CH R' R' Fig.1. Structure of the nickel complexes. Thermal Analysis Thermogravimetric curves for the nickel complexes were recorded (Fig. 2). are stable up to 180 "C; a sharp decrease occurs at 200 "C and also at 450 "C. conditions were investigated. carried out at 70-90 "C. The complexes Suitable drying In all subsequent work, drying of the precipitates was always 200 e, u) ln E 150 . 5 100 50 I I 0 200 400 600 800 1000 Temperature/"C Fig. 2. Thermogravimetric curves for the nickel complexes: A, Ni - FAT; B, Ni - TAT; and C, Ni - FAPT. Determination of the Optimum Experimental Conditions for Precipitation The influence of various solvents for washing the precipitates was studied. Pure water is unsatisfactory and, in general, mixtures of ethanol and water are more suitable.The optimum percentage of ethanol is different in each instance (Fig. 3) and must therefore be controlled. The volume of washing solutions can be reduced if the precipitates are previously treated withApril, 1982 DETERMINATION OF NICKEL WITH THIOSEMICARBAZONES 389 an alkaline solution: ammonia-water (1+9) solution for FAT and TAT and 0.1 M sodium hydroxide solution for FAPT. Owing to the acidic character of thiosemicarbazones, small amounts of coprecipitated reagents can be easily removed. The effect of increasing amounts of reagents was investigated in each instance. In general, excesses of 3040% over the stoicheiometric amounts are adequate. The influence of temperature on the precipitation of nickel complexes was investigated (Fig.4). With FAPT, precipitation can be carried out at room temperature or by heating to 100 "C. With FAT and TAT, precipitation must be effected by heating to over 50 "C in a water-bath ; incomplete precipitation occurs at room temperature with both of these reagents. 0 ( a ) Fig. 4. Influence of temperature. Amount of nickel, 24.2 mg. Reagents: (a) FAT; (b) TAT; and (c) FAPT. +1 _i;, Lu -3 /f-: -4 To determine the optimum pH, complexes were precipitated from solutions a t different pH values, adjusted by means of chloroacetate, acetate, phosphate and ammonia buffers, using the methods previously outlined. For Ni - FAT the optimum pH is 8.4-11.0, for Ni - TAT 5.0-1 1 .O and for Ni - FAPT 3.4-1 1 .O (Fig. 5). I b) I ( c ) -4 I , , , 4 6 8 1 0 1 2 0 2 4 6 8 10 12 0 2 4 6 8 10 12 Fig.5 . Influence of pH. Amount of nickel, 25.9 mg. Reagents: (a) FAT; (b) TAT; and (c) FAPT. Evaluation of the Relative Error of the Methods results are shown in Table I. Twelve determinations were made in each instance with the same amount of nickel. The TABLE I RELATIVE ERROR OF THE METHODS Reagent FAT . . . . TAT . . .. FAPT .. .. Standard Standard deviation Relative error, yo Amount of Ni/mg deviation/mg of the meanlmg (P = 0.05) 48.5 0.091 0.026 kO.1 48.5 0.135 0.039 5 0 . 2 47.4 0.074 0.021 k0 1390 ROSALES AND CANO-PAV~N : GRAVIMETRIC Analyst, Vol. 107 Interferences Amounts up to at least 0.1 g of the following ions do not interfere: alkali and alkaline earth metals, chromium(III), antimony(III), alu- minium, bismuth, manganese(II), tin(II), lead(II), phosphate, fluoride, citrate, oxalate, tartrate, thiosulphate, thiocyanate, arsenate and arsenite.With FAT reagent, serious inter- ferences were observed with iron(III), copper(II), cobalt(II), mercury(II), silver, zinc, cyanide and EDTA; 10 mg of cadmium does not interfere. With TAT, iron(III), zinc and cadmium do not interfere in acetate buffer; nor does 10 mg of cobalt in the same medium at pH 5.0-5.2, but it is necessary to wash the precipitate with ethanol - water (2+3). With FAPT, iron(III), zinc, cadmium and small amounts of mercury(I1) or silver (<5 mg) do not interfere in acetate buffer medium. Cobalt up to 20 mg does not interfere at pH 3.4-3.6 if the precipitate is washed with ethanol-water (1+1 or 3+2).Copper interferes at low concentrations, but it can be previously precipitated as the sulphide. Serious interferences were observed with cyanide and EDTA. Interference by foreign ions was investigated. Copper(II), mercury(II), silver, cyanide and EDTA interfered. Applications The gravimetric determination of nickel with FAPT reagent was applied to standard samples of Raney nickel, alloy cast iron, chromium nickel steel, nickel stone, aluminium bronze and aluminium alloy. The results obtained are given in Table 11. TABLE I1 DETERMINATION OF NICKEL WITH FAPT IN STANDARD SAMPLES Nickel content, Sample % m/m (standard value) Nickel found, yo m/m* Raney nickel . . .. .. 30.44t 30.41 Alloy cast iron . . .. .. 2.24 2.14 Chromium nickel steel . . .. 11.84 11.87 Nickel stone .. .. .. 55.89 55.85 Aluminium bronze . . .. 1.16 1.14 Aluminium alloy . . .. .. 1.93 1.95 * Each result is the mean of three determinations. t Determined with dimethylglyoxime. Mean result of three determinations. Conclusion We have examined gravimetric procedures for the determination of nickel using three thio- semicarbazones. FAPT is the preferred reagent because precipitation occurs at a lower pH, which gives greater selectivity in its behaviour with metal ions. It also exhibits the most favourable gravimetric factor, and the Ni - FAPT complex is the least soluble, The justification for a new gravimetric method for nickel is difficult because vic-dioximes behave as excellent reagents for this metal nevertheless, the results obtained with FAPT reagent are comparable to those obtained with the use of vic-dioximes, so FAPT can be used as an effective quantitative precipitant of nickel ion. Although FAPT reagent is less soluble in ethanol than vic-dioximes, it exhibits a most favourable gravimetric factor (Table 111), which implies greater sensitivity.The cost of FAPT is similar to that of dimethylglyoxime (Table 111), but the operating time TABLE I11 GRAVIMETRIC FACTOR AND RELATIVE COST OF SOME REAGENTS FOR NICKEL Reagent Gravimetric factor for nickel Dimethylglyoxime . . .. 0.2032 0.1722 Nioxime .. .. . . . . Heptoxime .. .. .. 0.1590 FAT .. .. .. .. 0.1486 TAT .. .. . . . . 0.1374 a-Benzyldioxime . . .. .. 0.1093 FAPT .. .. .. .. 0.107 3 a-Furyldioxime . . .. .. 0.1181 Relative cost 1 32 160 0.6 1.4 55 5 16 1.8April, 1982 DETERMINATION OF NICKEL WITH THIOSEMICARBAZONES 391 is shorter.A further advantage of the reagents tested includes the excellent mechanical properties of the precipitates, which are crystalline and very easy to filter. The authors express their appreciation to Professor F. Pino for his interesting and helpful comments. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. References Stankoviansky, S . , Carsky, J., and Beno, A., Chem. Zvesti, 1969, 23, 589. Guha Sircar, S. S., and Satpathy, S., J . Indian Chem. Soc., 1954, 31, 450. Hovoks, Z., and Holzbecher, Z., Bull. Int. Acad. Tche‘que Sci. Cl. Math. Nut. Med., 1953, 51, 43. Stankoviansky, S., Carsky, J., Beno, A,, and Dolnikova, E., Chem. Zvesti, 1968, 22, 50. Komatsu, S., and Hiroaki, Z., Nippon Kagaku Zasshi, 1957, 78, 715. Komatsu, S., Kida, T., and Hiroaki, Z., J . Chem. SOC. Jpn., 1956, 77, 1537. Cano-Pavbn, J. M., and Pino, F., Talanta, 1973, 20, 339. Cano-Pavbn, J. M., Martinez Aguilar, M. T., and Garcia de Torres, A., An. Quim., 1978, 74, 915. Cano-Pavbn, J. M., and Pino, F., Anal. Lett., 1974, 7, 159. Erdey, L., “Gravimetric Analysis, Part 11,” Pergamon Press, Oxford, 1965, p. 385. Muiioz Leyva, J. A., Cano-Pavh, J. M., and Pino, F., An. Quim., 1973, 69, 251. Campbell, M. J. M., Coord. Chem. Rev., 1975, 15, 279. Peshkova, V. M., and Savostina, V. M., “Analytical Chemistry of Nickel,” Israel Program for Bassett, J., Leton, G. B., and Vogel, A. I., Analyst, 1967, 92, 279. Jain, P., Singh, R. B., Garg, B. S., and Singh, R. P., J . Sci. Ind. Res., 1980, 39, 219. Scientific Translations, Jerusalem, 1967, p. 53. Received August 3 4 1981 Accepted October 9th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700385

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

392 Analyst, April, 1982, Vol. 107, pp. 392-397 Precipitation of Selenium and Tellurium from Homogeneous Solutions by Dimethyl Sulphite B. V. Narayana* and N. Appala Raju Department of Chemistry, S. V . University, Tirupati-517502, India Elemental selenium and tellurium are precipitated from homogeneous solutions by employing dimethyl sulphite for the in situ production of sulphur dioxide. Results obtained show that both selenites and selenates as well as tellurites and tellurates are quantitatively reduced by this reagent. The advantages of this precipitation over the conventional precipitation procedure are out- lined. Based on the different conditions under which the two elements are precipitated, a procedure for the separation of selenium and tellurium from one another is reported.Keywords : Selenium determination ; tellurium determination ; homogeneous precipitation ; dimethyl sulphite ; separation Most of the gravimetric methods for determining selenium and tellurium are based on reduction to their elemental forms in hydrochloric acid. Various reducing agent~1-l~ have been used for this purpose. Of these, sulphur dioxide was considered to be the most suitable reagent; it reduces selenium when it is in solutions greater than 3 N in hydrochloric acid and tellurium in 2 4 ~ solutions of the same acid. Lenher and Homberger,l5 however, reported that the reduction of tellurium with sulphur dioxide alone is too slow and that the precipitated tellurium is so very finely divided that it oxidises readily, in spite of washing with alcohol and diethyl ether.They therefore recommended the use of both sulphur dioxide and hydrazine hydrochloride for effecting the reduction. However, when tellurium is precipitated by this method using sulphur dioxide and hydraziue hydrochloride, the precipitated tellurium was found to stick to the walls of the beaker and to the bottom of the watch-glass, making its complete transfer to the filter very difficult. As precipitation from homogeneous solution would produce purer precipitates with good filtration characteristics, and as there are not many methods for the homogeneous precipita- tion of selenium and tellurium, it was thought worthwhile to see whether sulphur dioxide generated under homogeneous conditions would give improved precipitates. We also investigated whether tellurium could be quantitatively reduced under these conditions without the necessity of a further reducing agent, as in the conventional method.Dimethyl sulphite,l6 which hydrolyses to give sulphur dioxide according to the reaction CH,O.SO.OCH, + H,O + SO, + 2CH,OH was employed for this purpose and the results obtained are presented in this paper. Experimental Reagents Selenizm(IV) solution. A 10.0-g sample of sodium selenite (Riedel, Germany) was dissolved in water, transferred into a 1-1 calibrated flask and made up to the mark with water. SeZenium( V I ) solution. A 4.7032-g sample of sodium selenate (BDH Chemicals Ltd.) was dissolved in water, transferred into a 250-ml calibrated flask and made up to the mark with distilled water. The selenium contents of the two solutions were determined by independently analysing the respective salts by the standard rneth0d.l' TeZZwiztm(1V) soZution.A 1.9632-g sample of sodium tellurite (BDH Chemicals Ltd.) was dissolved in about 50ml of concentrated hydrochloric acid by gently warming. The * Present address : Department of Metallurgy, Indian Institute of Science, Bangalore-560 012, India.NARAYANA AND APPALA RAJU 393 solution, after cooling, was transferred into a 1-1 calibrated flask and diluted to the mark with water. A 4.8893-g sample of sodium tellurate (BDH Chemicals Ltd.) was dissolved in 25 ml of concentrated hydrochloric acid and made up to the mark with water in a 250-ml calibrated flask. The two solutions were analysed for their tellurium contents by the standard method.17 Dimethyl suZPphite.All the other chemicals used were of analytical-reagent grade. Tellurium( V I ) solution. E. Merck, Darmstadt, Germany, was used as supplied. Procedure Determination of selenium Known volumes of the standard selenium solutions were treated with concentrated hydro- chloric acid, 80 ml for the selenium(1V) and 100 ml for the selenium(VI), followed by 40 ml of methanol. The contents were diluted to 150 ml with water and cooled to room tempera- ture. The beaker containing these reagents was left for about 2 h, with occasional stirring. The red selenium precipitate was filtered through a previously weighed sintered-glass crucible (porosity 4), washed thoroughly with water and finally alcohol, dried at 100 "C for 1 h, cooled and weighed as elemental selenium. A 2-ml aliquot of the dimethyl sulphite was then added, with stirring.Determination of tellurium Concentrated hydrochloric acid [20 ml for tellurium(1V) and 30 ml for tellurium(V1) solutions] was added to known volumes of the tellurium standard solutions; 25 ml of methanol were then added to each mixture. The solutions were diluted to about 1OOml with water and treated with about 2 ml of dimethyl sulphite, with stirring. The beaker was then kept over a boiling water-bath for about 1.5-2 h, with occasional stirring. Losses of solution due to evaporation were not replaced. The precipitate was filtered through a weighed 1G4 sintered-glass crucible (porosity a), washed with hot water and alcohol, dried at 105- 110 "C for 1 h, cooled and weighed as elemental tellurium. Results and Discussion Determination of Selenium compounds with reasonable accuracy. conventional rneth0d.l' standard deviation and the relative mean error were 0.155 mg and O.13y0, respectively.Table I gives results for the determination of selenium in selenium(1V) and selenium(V1) The results agree well with those obtained by the For a series of six determinations on 45.64mg of selenium, the TABLE I DETERMINATION OF SELENIUM Selenites Selenates A I \ I A I Selenium(1V) Selenium(IV) Selenium(V1) Selenium(V1) takenlmg found/mg Error, takenlmg foundlmg Error, % 22.82 22.70 - 0.53 15.86 15.90 + 0.25 45.64 45.60 - 0.09 39.64 39.80 + 0.40 68.46 68.30 - 0.23 79.28 79.70 +0.53 91.28 91.40 +0.13 118.92 119.10 +0.15 114.10 114.40 + 0.26 Determination of Tellurium Results for the determination of tellurium in tellurium(1V) and tellurium(V1) compounds are given in Table 11.The standard deviation and the relative mean error of six deter- minations on 45.40 mg of tellurium were 0.061 mg and 0.46y0, respectively. Effect of Hydrochloric Acid on the Reduction of Selenium and Tellurium the reduction of selenium and tellurium were 30% and 15-16y0, respectively. Mellor and ThompsonlS reported that the concentrations of hydrochloric acid suitable for Experiments394 NARAYANA AND APPALA RAJU: PRECIPITATION OF SE AND TE Analyst, VoZ. I07 TABLE I1 DETERMINATION OF TELLURIUM Tellurites Tellurates r I f > Tellurium(1V) Tellurium(1V) Tellurium(V1) Tellurium(V1) takenlmg found/mg Error, % takenlmg found/mg Error, yo 18.16 18.20 + 0.22 22.70 22.60 36.32 36.50 + 0.49 45.40 45.50 + 0.22 45.40 45.50 + 0.22 90.80 91.30 + 0.55 90.80 91.30 +0.55 113.50 114.20 + 0.62 A A - 0.44 108.96 108.60 - 0.33 were conducted on the effect of this acid on the reductions using the proposed method, by adding varying amounts of acid to known amounts of selenium and tellurium containing 25 ml of methanol and 5 ml of dimethyl sulphite in a total volume of 100 ml.The precipi- tates were filtered and weighed as described previously. The results are recorded in Tables I11 and IV. From the Tables, it is clear that selenium(1V) is quantitatively reduced in solutions of greater than 5 N, while selenium(V1) is reduced in greater than 6 N acid solutions. The reduction of tellurium(1V) was found to be complete in 1-2 N hydrochloric acid solution while tellurium(V1) was quantitatively reduced only at a higher acid concentration of 3 N.TABLE I11 EFFECT OF HYDROCHLORIC ACID ON THE REDUCTION OF SELENITES AND SELENATES Acidity/N 1 .o 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Selenites* A Selenium( IV) found/mg 7.0 24.7 31.3 44.3 45.4 45.6 45.5 45.5 45.6 - Selenium( IV) recovery, yo 15.3 54.1 90.5 97.1 99.5 99.9 99.7 99.7 99.9 Selenatest r Selenium( VI) Acidity/N found/mg 5.0 30.2 6.0 39.8 7.0 39.8 8.0 39.7 9.0 39.5 7 Selenium(V1) recovery, Yo 76.2 100.4 100.4 100.2 99.7 * Amount of selenium(1V) taken, 45.64 mg. t Amount of selenium(V1) taken, 39.64 mg. TABLE IV EFFECT OF HYDROCHLORIC ACID ON THE REDUCTION OF TELLURITES AND TELLURATES Tellurites* Tellura tes t A I Tellurium(1V) Tellurium(1V) ’ -(VI) Tellurium’ Acidity/N found/mg recovery, Yo Acidity/N found/mg recovery, yo 0.5 44.9 1.0 45.4 1.5 45.5 2.0 45.5 2.5 45.2 3.0 45.1 98.9 1.0 10.5 23.1 100.0 2.0 42.3 93.1 100.2 3.0 45.6 100.4 100.2 4.0 44.9 98.9 99.5 99.3 * Amount of tellurium(1V) taken, 45.40 mg. t Amount of tellurium(V1) taken, 45.40 mg.April, 1982 FROM HOMOGENEOUS SOLUTIONS BY DIMETHYL SULPHITE 395 Effect of Reagent Concentration Although theoretical calculations showed that 0.1 ml of the dimethyl sulphite could bring about a quantitative reduction of 100 mg of both selenium and tellurium, it was found from experiments that a minimum of 1 rnl of the reagent has to be used to obtain satisfactory results.Effect of Methanol Concentration As the reagent was not soluble in aqueous solutions, water-miscible organic solvents, such as methanol, were emploj ed to bring about homogeneous conditions for the precipitation. It was found that the reagent was completely soluble when the methanol concentration was 25% V/V or greater.Hence a methanol concentration of 25% was employed in all the investigations. Effect of Associated Metal Ions Selenium and tellurium are usually found associated with metals such as copper, nickel, cobalt, iron, lead and antimony. Hence, the effect of these metals on the determination of both selenium and tellurium was studied and the results obtained are presented in Table V. The results in Table V show that selenium could be conveniently determined in the presence of these metals without any interferences. The results also show that, except for copper, all the other metals did not interfere in the determination of tellurium.When lead was present as well as tellurium high results were obtained, owing to the formation of lead sulphate, thereby contaminating the tellurium precipitate. This interference could be over- come by washing the precipitate with hot ammonium acetate solution before washing with hot water and alcohol. Lead chloride precipitation will not take place owing to the fonna- tion of complex chloro compounds of lead.lg TABLE V EFFECT OF ASSOCIATED METAL IONS ON THE DETERMINATION OF SELENIUM AND TELLURIUM Amount of selenium taken was 46.64 mg and the amount of tellurium taken was 44.50 mg. Amount Foreign metal added/mg Copper . . .. . .100 Nickel . . .. .. 200 Iron . . .. . . 100 200 Cobalt . . .. . . 200 Lead . . .. . . 100 Antimony .. .. 50 100 Selenium found/mg 45.3 45.5 45.3 45.6 45.9 45.9 - - Selenium recovery, yo 99.3 99.7 99.3 99.9 100.6 100.6 - - Tellurium found/mg 40.5 44.4 44.7 44.6 44.7 44.6 - Tellurium recovery, yo 91.0 99.8 100.4 100.2 100.4 100.2 - - Separation and Determination of Selenium and Tellurium From the results, it is seen that selenium was quantitatively precipitated from hydrochloric acid solutions when the acid concentration was greater than 5 N, while tellurium was com- pletely precipitated from acid solutions of below 3 N. Attempts were therefore made to separate selenium and tellurium at these acid concentrations. However, it was observed from experiments that at both of these acid concentrations, both of the metals were either wholly or partially precipitated. The extent of coprecipitation of one metal in the presence of the other is indicated in Tables VI and VII.From the results, it is evident that tellurium was not precipitated when the acid concentra- tion was 9 N with respect to hydrochloric acid, at which concentration selenium was com- pletely precipitated. Thus, to achieve a perfect separation of selenium and tellurium from one another, experiments were conducted by taking known amounts of selenium(1V) and tellurium( IV) solutions, adjusting the acid concentration to 9 N, when selenium was precipitated. The precipitate was filtered, washed, dried and weighed as described previously. This is in accordance with results reported by Kolthoff and396 NARAYANA AND APPALA RAJU : PRECIPITATION OF SE AND TE Analyst, VO,?.I07 TABLE VI Tellurium takenlmg 44.50 44.50 44.50 44.50 Selenium takenlmg 45.64 45.64 45.64 45.64 45.64 45.64 45.64 45.64 EXTENT OF COPRECIPITATION OF SELENIUM WITH TELLURIUM Total mass of Selenium Amount of Selenium precipitate/ precipitated with selenium added/mg Acidity/N mg tellurium/mg coprecipitated, % 143.0 1 .o 186.8 142.3 99.5 143.0 1.0 181.9 137.4 96.1 143.0 2.0 186.7 142.2 99.5 143.0 2.0 186.8 142.3 99.5 TABLE VII EXTENT OF COPRECIPITATION OF TELLURIUM WITH SELENIUM Tellurium added/mg 100.0 150.0 100.0 150.0 100.0 150.0 100.0 200.0 Acidity/N 6.0 6.0 7.0 7.0 8.0 8.0 9.0 9.0 Total mass of precipitate/ mg 128.8 170.6 84.0 103.1 48.7 50.0 45.4 45.6 Tellurium precipitated with selenium/mg 83.16 124.96 38.36 57.46 3.06 4.36 - - Amount of tellurium coprecipitated, yo 83.2 83.3 38.4 38.3 3.1 2.9 - - To determine tellurium, the filtrate was quantitatively transferred into a separate beaker and diluted with water to reduce the acidity to 2 N.Precipitation of black tellurium was observed immediately on dilution. The determination of tellurium was then completed by the procedure described previously. The results are shown in Table VIII. TABLE VIII RESULTS FOR SEPARATION AND DETERMINATION OF SELENIUM AND TELLURIUM Selenium Tellurium Selenium Tellurium Selenium Tellurium takenlmg takenlmg found*/mg foundt /mg recovery, "/o recovery, % 45.64 44.50 45.40 44.70 99.5 100.4 46.64 44.50 45.50 44.60 99.7 100.2 * At the initial acid concentration of 9 N with respect to hydrochloric acid.t Determined after dilution of the filtrate to 2 N with respect to hydrochloric acid. Conclusions The advantages of the proposed procedure include production of a dense precipitate of selenium and tellurium with good filtration characteristics. The precipitates obtained did not show any tendency to stick to the beaker walls or watch-glass, thus making their transfer to the filter very easy. The results obtained clearly show that, under homogeneous conditions of precipitation, the reagent brings about a quantitative reduction of tellurium, thus obviating the necessity of using another reducing agent, as is required in the conventional method. Moreover, this reagent has the added advantage of bringing about a quantitative reduction of selenium(V1) and tellurium(VI), which was not possible with sulphur dioxide alone in the conventional precipitation procedure.References 1. 2. 3. 4. 5. 6. Keller, E., J . Am. Chem. SOG., 1897, 19, 771. Keller, E., J. Am. Chem. SOG., 1900, 22, 241. Goto, H., and Kakita, Y., Sci. Rep. Res. Inst. Tohoku Univ., Ser. A , 1952, 4, 28. Schoeller, W. R., Analyst, 1939, 64, 318. Simon, V., and Grim, V., Chem. Listy, 1964, 48, 1774. Nadkami, R. A., and Haldar, B. C . , J . Indian Chem. SOC., 1964, 41, 813.A$ril, 1982 FROM HOMOGENEOUS SOLUTIONS BY DIMETHYL SULPHITE 397 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. Nadkarni, R. A., and Haldar, B. C., J . Indian Chem. SOC., 1966, 43, 429. Misra, G. J., and Tandon, J . P., Indian J . Chem., 1967, 5, 560. Pierson, G. G., Ind. Eng. Chem. Anal. Ed., 1934, 6, 437. Ripan, R., Marcu, G., and Pascu, N., Comun. Acad. Rep. Pop. Romine, 1958, 8, 467. Tomicek, 0.) Bull. SOC. Chim. Fr., 1927, 41, 1399. Hecht, F., and John, L., 2. Anorg. Allg. Chem., 1943, 251, 14. Meyer, J., 2. Anal. Chem., 1914, 53, 145. Ooba, S., and Uneo, S., Talanta, 1975, 22, 51. Lenher, V., and Homberger, A. W., J . Am. Chern. SOC., 1908, 30, 387. Soundar Rajan, S. C., and Appala Raju, N., Anal. Chim. A d a , 1978, 102, 237. Vogel, A. I., “A Text Book of Quantitative Inorganic Analysis,” Third Edition, Longmans, London, Mellor, J . W., and Thompson, H. V., “A Treatise on Quantitative Inorganic Analysis,” Charles Vogel, A. I., “A Text Book of Micro and Semimicro Qualitative Inorganic Analysis,” Fourth Edition, Kolthoff, I. M., and Elving, P. J., “Treatise on Analytical Chemistry, Part 11,” Volume 7, Inter- 1962, p. 508. Griffin & Co. Ltd., London, 1938, p. 483. Longmans, London, 1964, p. 208. science, New York, 1961, p. 162. Received March 12th. 1981 Accepted October 8th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700392

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

398 Analyst, April, 1982, "01. 107, $9. 398-402 Automatic Potentiometric Titration of Thiocyanate = Cyanide Mixtures in Hydrometallurgical Effluents G. F. Atkinson Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada, N2L 3GI J. J. Byerley and B. J. Mitchell Department of Chemical Engineering, University of Waterloo, Waterloo, Ontario, Canada, N2L 3GI Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada, N2L 3GI Mixtures of cyanide and thiocyanate in hydrometallurgical effluents heavily clouded with particulates are titrated quickly and successfully with silver nitrate solution by using a potentiometric automatic titrator fitted with a silver working electrode and a glass reference electrode. When thiocyanate is to be determined, cyanide is masked with formalin.Titrations over a wide range of concentration and ratio of the two species require minimum pre-treatment of the samples and give sharp end-points and good replication. Keywords Thiocyanate - cyanide mixtures; silver - glass electrode p a i r ; hydrometallurgical efluents Recovery of gold and silver from sulphur-containing ores, tailings and residues by cyanidation results in the formation of thiocyanate. Although the conversion of cyanide into thiocyanate represents a significant reagent loss to the leaching process, the major concern is the disposal and detoxification of many thousands of cubic metres of cyanide - thiocyanate effluent gener- ated in the course of precious metal extraction. Our research has been directed at electro- chemical methods of treating cyanidation effluents.I t is necessary to measure both thio- cyanate and cyanide in solutions clouded with a suspension of fine particles. This suspension consists of entrained particles from leaching and additional eroded material from the electrode surfaces of the reactor. For successful reactor development a rapid and reliable analytical method capable of determining thiocyanate and cyanide in cloudy solutions is required. Many methods exist for the determination of thiocyanate and cyanide separately. Thio- cyanate may be determined gravimetrically as silver thiocyanatel or as copper thiocyanate. Schulek3 developed a method in which thiocyanate was oxidised to cyanogen bromide, then potassium iodide was added and the liberated iodine was titrated with thiosulphate.There are also numerous argentimetric methods such as the Volhard t i t r a t i ~ n , ~ and adsorption indi- cator titrations with eosia6 Cyanide may be determined gravimetrically as silver cyanide,6 by Schulek's method3 or by titrations such as the Liebig - Deniges method6 or direct titration with Rhodamine 6G as indicator.' These methods suffer from severe drawbacks when the thiocyanate and cyanide are in solu- tion together. In many instances, sample pre-treatment such as volatilisation of cyanide from an acidic medium is prohibitively time consuming, and even then difficult to carry to comple- tion confidently. The methods used originally in this laboratory were chosen principally to minimise sample pre-treatment times.A 25.0-ml aliquot of standard 0.01 or 0.1 M silver nitrate solution was pipetted into a 250-ml Erlenmeyer flask, treated with 5 ml of 6 M nitric acid and 1 ml of ammonium iron(II1) sulphate indicator, and titrated using the thiocyanate sample as titrant to a faint brown end-point. Vigorous shaking is necessary throughout. Cyanide was titrated with silver nitrate solution using Rhodamine indicator.' The cyanide titration showed definite end-points even at concentrations below 10 p.p.ni. However, the thiocyanate titration was less satisfactory for concentrations of thiocyanate below 500 p.p.m. in the clouded reactor samples because of indefinite visual end-points. In seeking to improve this determination, a more satisfactory method was developed, which is reported here.Thiocyanate was titrated using a modified Volhard titration.ATKINSON, BYERLEY AND MITCHELL 399 The new procedure involves titrating the sample with silver nitrate solution while observing the potential difference between a glass electrode used as reference and a silver indicator electrode. When titrating thiocyanate, the cyanide was masked with formalin.8 Kolthoff and Linganeg investigated the accuracy of the potentiometric titration of thio- cyanate with silver nitrate solution and reported it to be good (0.08%). A single titration a t this accuracy took more than 1 h. ConradlO investigated the silver - silver sulphide ion- selective electrode for the potentiometric titration of cyanide. In general, such electrodes show strong mutual interferences of cyanide and thiocyanate as well as sulphide, and are thus inappropriate for use in the liquors under study. Experimental A Sargent-Welch recording potentiometric titrator, Model DG, was fitted with a Radio- meter P401 NH silver electrode and a Sargent-Welch S-30050-15A glass electrode that func- tioned as the reference.Silver nitrate solutions (0.01 M and 0.001 M) were prepared as re- quired by accurate dilution of Harleco Volumetric concentrate. Other reagents were supplied by the J. T. Baker Chemical Company and used as received. Formalin solution (37%) was used to mask cyanide. Stock solutions of potassium thiocyanate and potassium cyanide were prepared from analytical-reagent grade salts and standardised by Schulek’s bromimetric method., In order to determine thiocyanate in the thiocyanate - cyanide mixture, a suitable aliquot was placed in a 250-ml beaker, diluted to about 125 ml with de-ionised water and treated with 5 ml of formalin to mask the cyanide.The mixture was stirred and allowed to stand for 10 min. Nitric acid (6 M, 5 ml) was added to give a total volume of 135 ml and the sample was titrated with silver nitrate solution. For cyanide determination, a similar aliquot was taken and diluted to 135 ml with de-ionised water as before. The potentiometric titration was then carried out to the first sharp end-point. A second end-point was also observed if the titration was continued. The concentrations of thiocyanate and cyanide were calculated as follows : [SCN-] (mg 1-l) = AgNO, molarity (moll-l) x AgNO, titration volume (1) x 58.08 x 1000 (mg mol-l) Sample volume (1) [CN-] (mg 1-l) = AgNO, molarity (moll-l) x AgNO, titration volume (1) x 2 x 26.018 x 1000 (mg mol-l) Sample volume (1) The silver electrode was cleaned after approximately every five titrations by wiping it with a fine metal polishing cloth and rinsing it with de-ionised water.Results and Discussion The potentiometric titration technique was applied to a number of synthetic solution mixtures of cyanide and thiocyanate prepared from appropriate stock solutions. Standard addition experiments were performed on those synthetic solution mixtures that had undergone partial electrochemical oxidation and therefore contained considerable turbidity due to sus- pended electrode erosion material. Addition of chloride was also carried out in order to check for possible interference.All data presented are based on the average of three determinations using an initial volume for titration of 135 ml. Table I gives selected data for the potentiometric determination of thiocyanate in the pres- ence and absence of cyanide. The cyanide added appears to be without effect on the thio- cyanate determination when it is masked by formalin. Fig. 1 (curve A) shows a representative potentiometric titration of 10 mg of thiocyanate. The sharp end-point represents the complete formation of silver thiocyanate. A trace superimposable on curve A is obtained when 10 mg of thiocyanate are titrated in the presence of 20 mg of cyanide masked with formalin. If no masking agent is present the thiocyanate end-point is indistinct and non-reproducible, with the problem increasing in severity as the cyanide level is raised.Fig. 2 (curve A) shows similar behaviour for 0.25 mg of thiocyanate. This curve is unchanged when the 0.25 mg of thio- cyanate is titrated in the presence of 100 mg of cyanide masked with formalin.400 ATKINSON et al. : AUTOMATIC POTENTIOMETRIC TITRATION Analyst, VoZ. I07 TABLE I POTENTIOMETRIC DETERMINATION OF THIOCYANATE IN THE PRESENCE AND ABSENCE OF CYANIDE > E Itl 2 600 Y- U .? 400 800 0 - al 0 4- ," 200 > 0 al .- c - ! = o An initial titration volume of 135 ml was used. Thiocyanatelmg Taken 25.18 25.18 12.50 12.50 9.47 9 47 12.58 12.58 6.25 6.25 4.74 4.74 1.26 1.26 0.625 0.625 0 474 0.474 0.252 0.252 0.125 0.125 Found' 25.46 25.50 12.65 12.66 9 52 9 58 12 65 12.63 6.28 6.28 4.76 4.76 1.14 1.13 0.565 0.578 0.449 0.459 0.241 0.249 0.121 0.109 Cyanide presentlmg 10.0 19 0 20.0 50.0 50.0 100 0 50.0 50 0 100.0 100.0 100 0 - - - - - - - - - - - Absolute errorlmg 0.28 0.32 0.15 0 16 0.05 0.11 0.07 0.05 0.03 0.03 0.02 0 01: - 0.12 -0 13 -0 06 - 0.047 -0 025 -0.015 -0 011 -0 003 - 0 004 - 0 016 Recovery. % 101.1 101.3 101.2 101.3 100.5 101 2 100.6 100.4 100 5 100.5 100.4 100 4 90.5 89.7 90 4 92 2 94 7 96.8 95.6 98.8 96.8 87 2 > 400 $ u 0, 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 Volume of 0.01 N silver nitrate solutioniml 300 z U - 200 .: C Q, 4- 100 > m .- 4- 0 % a Fig.1. Potentiometric titrations for high concentrations of thiocyanate and cyanide. A, 10mg of thiocyanate and B, 5 mg of cyanide.Both initial volumes were 135 ml. > E 8 600 Itl . S 400 73 m - .- 2 200 CI 0 0. al .z 0 4- m a, - 1 I I I I 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 Volume of 0.001 N silver nitrate solutioniml > E Itl 1 300 200 E - m .- 4- 100 0 n W rn W 0 .g - a a Fig. 2. Potentiometric titrations for low concentrations of thiocyanate and Both initial cyanide. volumes were 135 ml. A, 0.25 mg of thiocyanate and B, 0.50 ing of cyanide.April, 1982 OF THIOCYANATE - CYANIDE MIXTURES 401 Table I1 presents data for the potentiometric determination of cyanide in the presence and absence of thiocyanate. The data show that cyanide may be titrated effectively in the pres- ence of thiocyanate. Two end-points are observed representing the formation of Ag(CN),- and that of silver cyanide.Additions of thiocyanate had no detectable effect on either end-point. Fig. 2 (curve B) shows the first end-point for a titration of 0.50 mg of cyanide. Even at this low level, addition of thiocyanate had no effect on the shape of the curve at the end-point or on its reproducibility. Fig. 1 (curve B) shows a typical titration of 5.0 mg of cyanide. TABLE I1 POTENTIOMETRIC DETERMINATION OF CYANIDE IN THE PRESENCE AND ABSENCE OF THIOCYANATE An initial titration volume of 135 ml was used. Cyanidelmg Taken 9.73 9.73 4.88 4.88 0.488 0.488 0.098 0.098 Found * 9.72 9.70 4.88 4.95 0.433 0.413 0.125 0.070 Thiocyanate Absolute presentlmg errorlmg - -0.01 12.6 -0.03 - 0.0 62.5 + 0.07 - -0.005 62.5 -0.075 125.0 -0.028 - +0.027 Recovery, % 99.9 99.7 100.0 101.4 88 7 84.6 127.6 71.4 The data in Table I11 represent the results of standard addition experiments performed on thiocyanate - cyanide partially oxidised effluents.These solutions were products of electro- chemical oxidation and were very discoloured and turbid. The data indicate no difficulty in determining both cyanide and thiocyanate potentiometrically over a wide range of con- centrations. Replicate experiments showed a typical precision of * 1 %. TABLE I11 STANDARD ADDITION DATA FOR THIOCYANATE - CYANIDE EFFLUENTS Effluents were selected samples of electrochemically oxidised laboratary-simulated waste solutions, Efiluent composition/mg 1-’ Amount taken/- Standard additionlmg Amount found/mg -- r- 1 Thiocyanate Cyanide Thiocya- = C y a n i d e ‘Thiocyanate’ Cyanid: 1482 6.1 7.41 0.031 4.735 - 12.23 - 990.8 99.5 24.27 2.49 - 9.73 - 12.12 762 95.6 3.81 0.478 4.735 - 8.49 - 213 79.2 6.33 1.98 4.735 - 10.08 - 4.78 - - 0.485 - 9.73 - 9.835 8.2 19.4 0.205 2.09 1.4 0.052 0.038 4.735 2.09 1.5 0.052 0.038 - 9.73 - 9.679 Thiocyanate or cyanide recovery, % 100.7 99.2 99.4 100.1 96.3 99.9 99.1 The interference of chloride on both the titration of cyanide and of thiocyanate was examined.Table IV gives selected data for both titrations. Generally it appears that chloride causes no interference. Fig. 3 (curve A) depicts a titration of thiocyanate in the presence of chloride. Addition of cyanide masked with formalin has no effect. The second TABLE IV CHLORIDE NON-INTERFERENCE Thiocyanate Amount taken/mg Chloride Amount found/mg or cyanide ,-=-, added/ Absolute recovery, Thiocyanate Cyanide mg Thiocyanate Cyanide errorlmg % 6.26 - 12.53 6.40 - 0.15 102.4 - 4.88 12.53 - 4.85 -0.03 99.4402 > 800 E ? 600 .0 C z -0 - .: 400 Q) c a : 200 .- c - al P= 0 ATKINSON, BYERLEY AND MITCHELL 2.0 4.0 6.0 8.0 10.0 12.0 14.0 ” 40.0 42.0 44.0 400 > E . 0 300 !s -u m - 200.g P) 0 a c 100 .; m Q) n - 0 Volume of 0.01 N silver nitrate solution/ml Fig. 3. Graphs showing non-interference of chloride. A, 5.0 mg of thiocyanate in the presence Both initial of 12.5 mg of chloride; B, 2.5 mg of cyanide in the presence of 12.5 mg of chloride. volumes were 135 ml. end-point shows completion of silver chloride formation. Curve B shows the two charac- teristic cyanide end-points and a final chloride end-point. The determination of cyanide was found to be unaffected. This new method is now used routinely in our laboratory. We believe it merits considera- tion by others dealing with turbid solutions of cyanide and thiocyanate. The support of the National Sciences and Engineering Research Council of Canada is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. References Williams, W, J., “Handbook of Anion Determinations,” Butterworths, London, 1979, p. 229. Newman, E. J., Analyst, 1963, 88, 500. Schulek, E., 2. A n d . Chem., 1923, 62, 337. Williams, W. J., “Handbook of Anion Determinations,” Butterworths, London, 1979, p. 231. Williams, W. J. , “Handbook of Anion Determinations,” Buttenvorths, London, 1979, p. 232. Williams, W. J., “Handbook of Anion Determinations.” Butterworths, London, 1979, p. 73. Ryan, J. A., and Culshaw, G. W., Analyst, 1944, 69, 370. Charlot, G., “Analyse Quantitative MinQale,” Fifth Edition, Masson, Paris, 1966, p. 675. Kolthoff, I. M., and Lingane, J. J., J. Am. Chem. SOC., 1935, 57, 2126. Conrad, F. J., Tdunta, 1971, 18, 952. Received JuZy 21d, 1981 Accepted October 12th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700398

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, April, 2982, Vol. 107, pp. 403-407 403 Dead-stop Determination of EDTA and NTA in Commercially Available Detergents* R. Calapaj and L. Ciraolo Istituto d i Merceologia, Universith d i Messina, 98 100 Messina, Italy F. Coriglianot Istituto di Chimica Industriale, Universith di Messina, 98 100 Messina, Italy and S. Di Pasquale Istituto d i Chimica Analitica, Universith d i Messina, 981 00 Messina, Italy A rapid and selective method for the determination of etliylenediaminetetra- acetic acid (EDTA) and/or nitrilotriacetic acid (NTA) in commercially avail- able detergents has been developed. I t is based on a titration with a standard solution of copper(I1) in acetate buffer, where the end-point is revealed by means of a “dead-stop” system with two polarised copper electrodes.Furthermore, it is possible to determine in the course of the same titration the relative amounts of both chelating agents by using 4-(2-pyridylazo)- resorcinol (PAR), which changes colour a t the end-point of the first reaction (copper - EDTA) . Most detergent constituents, including polyphosphates, have been observed to have no effect on the determination ; interference from some constituents (perborates and zeolites) can easily be removed. The method has been shown to give good results in the analyses of different commercially available products. Keywords : EDTA and NTA determination ; copper(1I) sulphate titrant; dead-stop indicator ; PAR indicator ; detergents Ethylenediaminetetraacetic acid (EDTA) and/or nitrilotriacetic acid (NTA) are frequently used in detergent formulations at low concentrations to support the activities of poly- phosphates (e.g., sodium tripolyphosphate) and to inhibit trace amounts of free metal ions from catalysing the decomposition of perborates and depressing the bleaching performance of optical brighteners.l In some countries, such as Canada and Finland, higher percentage contents of NTA (up to 10%) are allowed as polyphosphate substitutes in order to reduce the phosphorus content of waste water, which is a source of eutrophication.In other countries the environmental convenience of this substitution has been questioned for a long time and alternatives to polyphosphates have been sought in other directions, e.g., zeolites. However, the environmental and health risks from the use of NTA in laundry detergents have recently been re-evaluated in the US by the Environmental Protection Agency and judged to be generally low.2 Thus, the use of NTA and other chelating agents in detergent formulations as well as the interest in establishing practical methods of determination will probably increase in the future.Present methods are founded on complexometric titrations with various metal ions and with both optical and instrumental end-point One of the most recent methods consists of a titration with a standard iron(II1) solution using a redox potentiometric end-point detector, but the procedure is complicated on account of the interference of poly- phosphates, which have to be converted into orthophosphates and removed as ammonium magnesium pho~phate.~ Another recent method, based on ultraviolet spectrophotometric titration with copper(I1) and on the difference in absorptivity between the NTA complex and the aquo complex, is without serious interferences, but the absorbance break is in- sufficiently sharp to afford precise results and may be obscured by other components in the formulation that absorb in the ultraviolet region.1° * Presented a t the 1st International Symposium on Technological, Environmental and Economic Trends t To whom correspondence should be addressed.in Detergency, Rome, October 22-24th, 1980.404 CALAPAJ et d. : DEAD-STOP DETERMINATION OF EDTA Analyst, Vd. I07 This paper is concerned with the proposal of a dead-stop amperometric system with two polarised copper electrodes, suitable for the selective detection of the end-point of an EDTA plus NTA titration with a standard copper(I1) solution in acetate buffer.Moreover, the addition of 4-(2-pyridylazo)resorcinol (PAR), which changes in colour at the end of the first reaction [copper(II) - EDTA], allows the relative amounts of both chelating agents to be determined in the course of the same titration. Hence the results are obtained very rapidly and the process can easily be automated. Experimental Reagents All chemicals were of analytical-reagent grade. All of the reagent solutions were standardised complexometrically : a standard solution of zinc(II)ll was used to standardise 0.1 M solutions of EDTAll and NTA,12 as the disodium salts; then the EDTA standard solution was employed to standardise a 0.1 M solution of copper(I1) sulphate.11 PAR was used as a saturated methanolic solution.The acetate buffer of pH 4.4 & 0.2 was prepared by mixing 33.6g of acetic acid (sp. gr. 1.0499) and 7.0g of sodium hydroxide and then diluting the mixture to 1 1. Apparatus Fig. 1 shows the apparatus necessary for the amperometric titration with two identical polarised electrodes. The polarising tension (50-200 mV) is taken from a common 1.5-V battery through a potential divider and is measured with a voltmeter; the circuit current and the volume of titrant are measured with a 25-pA full-scale microammeter and a 5-ml full-scale microburette, within 0.2 pA and 0.01 ml, respectively; the solution is stirred at a constant velocity by a synchronous motor. Both electrodes consist of wires of electrolytic copper (about 100 mm long and 1 mm diameter) dipped into a small glass tube (80 mm long) by sucking in melted paraffin and then cooling quickly under water. The copper wire emerges 10 mm from either side of the tube.Before each titration both electrodes must be dipped into 2.5 M nitric acid for a few seconds and then washed with distilled water. Different pairs of electrodes gave reproducible titration curves. Fig. 1. Dead-stop device. Titration Tests Fig. 2 shows typical conditions and titration curves; only a residual current (less than 5 PA) passes on the addition of copper(I1) ions as long as free EDTA and/or NTA are present in the solution. After the equivalence point has been reached, the current rises in proportion to the excess of added copper ions.The break is so sharp that its graphical location isApril, 1982 AND NTA IN COMMERCIALLY AVAILABLE DETERGENTS 405 G P 0.6 I 0.8 ~~~ 1 .o 1.2 Cu2+/ rnequiv Fig. 2. Dead-stop titration graphs for: A, 0.34 mmol of EDTA; B, 0.53 mmol of NTA; C, 0.34 mmol of EDTA + 0.53 mmol of NTA; D, 0.68 mmol of EDTA + 0.21 mmol of NTA; E, 0.13 mmol of EDTA + 1.05 mmol of NTA; F, 1.00 mmol of NTA in the presence of 0.54 mmol of STP; and G, 0.50 mmol of EDTA in the presence of 1.63 mmol of STP. superfluous. Moreover, when a few drops of PAR are added to the solution, its colour changes from yellow to full red at the equivalence point of the copper(I1) - EDTA complex formation only, without any interference from the successive NTA reaction or on the amperometric end-point.Although a larger pH range (4-8) is tolerated when observing the amperometric end-point, the optical end- point needs a buffered solution at pH 4.4 & 0.2. This enables the titration to be easily automated. This pH range was therefore selected. Effects of Detergent Constituents Titration tests have also been performed in the presence of the main constituents of detergent formulations. Surfactant concentrations of 0.5-15% m/V, for example, lauryl sulphate, sodium dodecylbenzenesulphonate and isooctylphenoxypolyethoxyethanol (approxi- mately 10 mol of ethylene oxide) and 0.2-0.8 g of sodium tripolyphosphate have been found to have no effect on either equivalence point. No effects have also been observed from pyro- and orthophosphates, silicates, sulphates and carbonates.Zeolites release aluminium ions in acidic solutions containing EDTA or NTA.13 They should therefore be filtered before adding the acetate buffer. Finally, perborates affect the amperometric titration. Nevertheless, a prior reduction with an excess of sodium sulphite in warm solution for 5 min is sufficient to eliminate their interference. Procedure About 2 g of liquid or powdered detergent are carefully weighed and dissolved in about 20 ml of distilled water in a beaker. If insoluble matter persists, even after slightly warming, the solution is centrifuged and the solid discarded. Then, 30 ml of acetate buffer and 4-5 drops of PAR solution are added to the solution, the stirrer is set in motion, the electrodes are dipped and polarised with a low tension (so as to maintain the current under 5 PA), and the titration with the standard copper( 11) solution is commenced. The colour change of the optical indicator may take place either immediately (at the beginning of titration), at a point when the same volume of titrant has been added as gives the amperometric end-point, or at an intermediate volume. These three events mean that only NTA, only EDTA or both are present, respectively. In the first two instances, the addition of PAR is superfluous.In the third instance, the optical equivalence point is relative to the amount of EDTA (in milliequivalents) while the amperometric end-point considers EDTA plus NTA milliequivalents ; the milliequivalents of NT,4 will be obtained from the difference.The following procedure is proposed for commercially available detergents.TABLE I EDTA AND/OR NTA DETERMINATION IN COMMERCIALLY AVAILABLE DETERGENTS Sample 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Both chelating agents were added t o all the detergent samples, except samples 3 and 8 (only EDTA) and samples 5 and 13 (only NTA). Use of detergent (about 2 g) General purpose cleanser. . .. .. General purpose cleanser.. .. . . General purpose cleanser. . .. .. Dishwashing, by hand . . .. .. Dishwashing, by hand . . .. .. Clothes washing, by hand . . .. Clothes washing, by hand . . .. Clothes washing, by hand . . .. Clothes washing, by hand . . . . Delicate clothes washing . . .. .. Delicate clothes washing . . .. * . Machine clothes washing . . .. .. Machine clothes washing .. .. . . Machine clothes washing . . .. .. Machine clothes washing . . 8 . * . GT/ mg 98.9 98.9 98.9 39.5 98.9 158.2 98.9 59.3 98.9 118.6 79.1 98.9 197.7 - - EDTA P Found/ Recovery, 94 95.0 96 97.1 98 99.1 37 93.7 95 96.1 154 97.3 97 98.1 58 97.8 93 94.0 112 94.4 77 97.3 95 96.1 193 97.6 mg %* - - - - NTA f A -I Added/ Found/ Recovery, mg mg %t 100.5 103 102.5 100.5 98 97.5 160.9 167 103.8 100.5 103 102.5 100.5 101 100.5 40.2 43 106.9 60.3 64 106.1 100.5 100 99.5 120.6 129 106.8 80.4 78 97.0 160.9 159 98.8 100.5 107 106.5 201.1 196 97.5 - - - - - - EDTA + NTA Added/ Found/ Recovery, mequiv mequiv yo$ 0.864 0.860 99.5 0.864 0.841 97.4 0.338 0.335 99.2 0.977 0.999 102.3 0.526 0.539 102.5 0.864 0.854 98.8 0.751 0.746 99.3 0.338 0.332 98.2 0.518 0.533 103.0 0.864 0.843 97.6 1.037 1.058 102.0 0.692 0.671 96.9 0.842 0.832 98.8 0.864 0.886 102.5 1.728 1.686 97.6 A I 5 * PAR end-point.t Determined by difference. $ Dead-stop end-point. bA j M , 1982 AND NTA IN COMMERCIALLY AVAILABLE DETERGENTS 407 Only if perborates are present, as in some types of European laundry detergents, should the procedure be modified as follows: about 2 g of detergent are carefully weighed, poured into a beaker and mixed with 1 g of sodium sulphite and 25-30 ml of distilled water. After 5 min of stirring and slight warming, the solution is centrifuged, 30 ml of acetate buffer are added and the solution is titrated with copper(I1) solution as before. Results and Discussion Known amounts of EDTA and/or NTA were added to various commercially available detergents, which were known to be free of them, and were then determined according to the above procedure.The recovery of either or both chelating agents, as given by the amperometric equivalence point, is very good, being complete to within &3%. Although the optical equivalence point shows a higher error (*7%) due to a heavier matrix effect the determination of the relative amounts of both chelating agents is still largely satisfactory for technical analysis purposes. Thus, with respect to other recent methods, this method produces comparatively precise analytical results with greater rapidity and selectivity. The results are shown in Table I. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References Benedetti, L., “I1 Problema dell’Inquinamento delle Acque da Detersivi,” Quaderni dell’Istituto di Chem. Eng News, 1980, 58 (22), 6. Farrow, R. N. P., and Hill, A. G., Analyst, 1965, 90, 210. Farrow, R. N. P., and Hill, A. G., Analyst, 1965, 90, 241. Siggia, S., Eichlin, D. W., and Rheinhart, R. C., Anal. Chem., 1955, 27, 1745. Clinckemaille, G. G., Anal. Chim. Acta, 1968, 43, 520. Kump, K. J., Palocsay, F. A., and Gallaher, T. N., J . Chem. Educ., 1978, 55, 265. Longman, G. F., “The Analysis of Detergents and Detergent Products,” John Wiley, New York, Taddia, M., Lippolis, M. T., and Pastorelli, L., Microchem. J . , 1979, 24, 102. Harju, L., and Sara, R., Anal. Chim. A d a , 1977, 91, 393. Flaschka, H. A., “EDTA Titrations,” Pergamon Press, Oxford, 1959, pp. 63, 75 and 78. Meites, L., Editor, “Handbook of Analytical Chemistry,” McGraw-Hill, New York, 1963, pp. 3-207. Kerr, G. T., J . Phys. Chem., 1968, 72, 2594. Ricerca sulle Acque, No. 8, C.N.R., Rome, 1971, p. 122. 1975, pp. 403-408 and 487-488. Received May llth, 1981 Accepted November 4th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700403

出版商:RSC    

年代:1982

数据来源: RSC