ISSN:0003-2654    

DOI:10.1039/AN98106FX033

出版商:RSC    

年代:1981

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98106BX035

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

iv SUMMARIES OF PAPERS IN THZS ISSUE September, 1981Summaries of Papers in this IssueComparison of Flame, Electrothermal and Inductively CoupledPlasma Atomisation Techniques for the Direct Analysis of SlurriesA comparison is made between the use of flame, electrothermal and induc-tively coupled plasma atomisation techniques for the analysis of slurries.For flame atomisation, pulsed nebulisation must be used and for inductivelycoupled plasma atomisation a high-solids cross-flow nebuliser is essential.The results show that atomisation efficiency in nebuliser-based systems isdependent on sample transport efficiency, particle size, atomisation tempera-ture and sample matrix. For analytical determinations it is necessary togrind the sample to a particle size of less than 10 pm and to use standardsthat are closely matched to the samples.For electrothermal atomisation,particle-size effects become significant only above 25 pm when samplingbecomes the major source of error.Keywords : Slurry techniques ; ore analysis ; flame, electrothermal and induc-tively coupled plasma atomisation ; atomic-absorption spectrometryC. W. FULLER, R C. HUTTON and B. PRESTONTioxide International Limited, Central Laboratories, Stockton-on-Tees, Cleveland,TS18 2NQ.Analyst, 1981, 106, 913-920.Investigation of the Determination of Tin Tetraalkyls and AlkyltinChlorides by Atomic-absorption Spectrometry after Separationby Gas - Liquid or High-performance Liquid - LiquidChromatographyA variety of sample presentation and atomisation methods have beenexamined to obtain sufficient sensitivity for the trace determination of tintetraalkyls and dkyltin chlorides (R,,SnCl,; R = methyl or ethyl; n = 14)after their separation by gas - liquid or by high-performance liquid - liquidchromatography.Detection limits range from 10-20 p g (flame atomisation)to 1 p g (direct pyrolysis) and 2-2Opg (hydride reduction followed by directpyrolysis). The construction of systems including an automated hydridegenerator is described.Keywords : Gas - liquid and high-performance liquid - liquid chromatography ;tin tetraalkyls and alkyltin chlorides ; pyrolysis ; hydride generation ; tubefurnaceD. THORBURN BURNS, F. GLOCKLING and M. HARRIOTTDepartment of Chemistry, The Queen’s University of Belfast, Belfast, BT9 5AG,Northern Ireland.Analyst, 1981, 106, 921-930vi SUMMARIES OF PAPERS IN THIS ISSUE September, 1981Inexpensive, Simple Hydride Generation System with MinimumInterferences for the Atomic-absorption Spectrophotometryof ArsenicA system for the hydride generation atomic-absorption spectrophotometryof arsenic is described that uses ordinary boiling-tubes to hold the acidifiedsamples into which sodium tetrahydroborate(II1) solution is injected and theevolved hydrogen carries the generated arsine directly through the spectro-photometer nebuliser to a nitrogen-supported hydrogen flame.The totalarsenic evolved is measured by integrating the absorbance for 16 s. Apartfrom transition metals of the cobalt, nickel and copper groups, the arsenicresponse appears to be insensitive to interferences, and the effects of thesemetals may be overcome by adding 0.5 g of thiourea to each sample. Thelimit of detection (95% confidence) is about 6 ng of arsenic and the repro-ducibility is about 1% for 0.5 p g of arsenic.Keywords : A rsenic determination ; hydride generation ; atomic-absorptionspectrophotornetry; interferencesC. J.PEACOCK and S. C. SINGHDepartment of Chemistry, University of Lancaster, Bailrigg, Lancaster, LA1 4YA.Analyst, 1981, 106, 931-938.Determination of Trace Amounts of Barium in Calcium- containingMatrices by Atomic-absorption Spectrophotometry FollowingSolvent Extraction with Polyethylene GlycolPolyethylene glycol (PEG) 1540 shows a high efficiency (almost 1OOyo) forthe extraction of barium from a calcium matrix under appropriate conditions.This has permitted trace amounts of barium in a calcium matrix to be deter-mined by flame atomic-absorption spectrophotometry without interferencefrom calcium.The effectiveness of PEG is noteworthy because of its readyavailability and cheapness compared with the 18-crown-6 used in existingprocedures. Further, the extraction process may be completed successfullyin one step without subsequent washing of the organic phase.Keywords A tomic-absorption spectrophotometry ; solvent extraction ; calciuminterference ; Polyethylene glycol; 18-crown-6A. M. Y. JABER and M. Y. S. EL-ISSAChemistry Department, Faculty of Science, University of Jordan, Amman, Jordan.Analyst, 1981, 106, 939-943September, 1981 SUMMARIES OF PAPERS IN THIS ISSUEExtraction - Spectrophotometric Determination of Tantalum(V)with 2- (2-Thiazoly1azo)- 5-dimethylaminophenol and1,3 - DiphenylguanidineTantalum(V) - 2-( 2-thiazolylazo)-5-dimethylaminophenol (TAM) chelateanion is extracted quantitatively into benzyl alcohol with 1,3-diphenylguani-dine (DPG) to form a ternary complex. In the organic phase, the complexhas an absorption maximum at 605nm.The optimum pH range for theextraction is 4.1-4.9 and the ternary complex is stable for at least 50 min.Beer’s law is obeyed over a concentration range of 2-18 pg of tantalum(V) in10 ml of the organic phase. The molar absorptivity of the ternary complexis 4.1 x lo4 1 mol-l cm-l a t 605 nm.The composition of the ternary complexis considered to be (TaO,),(TAM),(DPG+).Keywords : Spectrophotometry ; tantalum( V ) determination ; 2-( 2-thiazolylazo) -5-dimethylaminophenol ; 1,3-diphenylguanidineviiCHIKAO TSURUMIDepartment of Industrial Chemistry, Shibaura Institute of Technology, Shibaura,Minato-ku, Tokyo 108, Japan.KEIICHI FURUYADepartment of Applied Chemistry, Faculty of Science, Science University of Tokyo,Kagurazaka, Shinjiku-ku, Tokyo 162, Japan.and HITOSHI KAMADADepartment of Industrial Chemistry, Faculty of Engineering, University of Tokyo,Hongo, Bunkyo-ku, Tokyo 113, Japan.Analyst, 1981, 106, 944-948.Spectrophotometric and Titrimetric Determination ofCatecholamines Using Organic Brominating AgentsTitrimetric and spectrophotometric methods for the determination of catechol-amines as pure substances and in their dosage forms were investigated andfound to offer an impravement in ease, speed and accuracy.Both methodsare based on the formation of adrenochromes and employ organic bromi-nating agents. Data obtained for several commercial samples are reportedand compared with those obtained using official procedures.Keywords ; Catecholamine determination ; organic brominating agents ; spectro-photometry ; titrimetryA. ABOU OUF, M. I. WALASH and F. B. SALEMFaculty of Pharmacy, Mansoura University, Mansoura, Egypt.Analyst, 1981, 106, 949-954September, 1981 SUMMARIES OF PAPERS I N THIS ISSUEExtraction - Spectrophotometric Determination of Sub- microgramAmounts of Nitrite Using 4-Nitroanaline and Naphth- 1-01ixNitrite ion reacts with 4-nitroaniline in hydrochloric acid medium to form4-nitrophenyldiazonium chloride, which couples with naphth- 1-01 in alkalinemedium to give a purple azo dye (Amax.= 580 nm, E = 3.7 x lo4 1 mol-1cm-I). This extractionprocess is used for the determination of sub-microgram amounts of nitritespectrophotometrically. Of the alcohols studied (butanol, pentanol, iso-pentanol, hexanol and octanol), hexanol was found to be the most suitableextractant. The blue hexanol extract exhibits a maximum absorption a t605-615 nm. Beer’s law is obeyed over the nitrite concentration range0.02-0.14 p.p.m. The molar absorptivity and Sandell’s sensitivity werefound to be 5.24 x lo4 1 mol-1 cm-I and 0.00088 pg cm-2, respectively.The optimum reaction conditions for diazotisation, full colour developmentand the effect of variables have been studied.Several common ions do notinterfere with the method if appropriate masking agents are used. Themethod has been applied successfully to the determination of nitrite insamples of river water.The dye can be extracted into higher alcohols.Keywords : Nitrite determination ; spectrophotometry ; 4-nitroaniline ;naphth- 1-01 ; hexanol extractionANIL K. BAVEJA, JAGADEESAN NAIR and V. K. GUPTADepartment of Chemistry, Ravishankar University, Raipur-492 010 (M.P.), India.Analyst, 1981, 106, 955-959.Spectrophotometric Determination of Thiamine HydrochlorideUsing Orthogonal PolynomialsA direct spectrophotometric method for the determination of thiaminehydrochloride in the presence of its degradation products is presented.Themethod depends upon the use of the combined polynomial method, a modifi-cation of Glenn’s method of orthogonal functions, to eliminate interferencesby degradation products to the absorption spectrum of thiamine hydro-chloride. Thus, the coefficient (PIP) of the combined polynomial, Pw, calcu-lated over the wavelength range 226-292 nm a t 6-nm intervals, is independentof degradation products, is linearly related to the concentration of thiaminehydrochloride and is reproducible, with a relative standard deviation of0. 13-0.54y0. The mean recovery for five mixtures of thiamine hydrochloridewith degradation products was found to be 100.1 f 0.6%. When appliedto commercial tablets, the results were in good agreement with the USPfluorimetric method. Further, a graph of logarithm of concentration againsttime for a solution in borate buffer (pH 10) gave a straight line with a slopeof -0.022 6 h-l, showing that the method may be used t o monitor stability.Keywords : Thiamine hydrochloride determination ; orthogonal polynomials ;spectrophotometryA. M. WAHBI, SAIED BELAL, H. ABDINE and MONA BEDAIRFaculty of Pharmacy, University of Alexandria, Pharmaceutical AnalyticalChemistry Department, Alexandria, Egypt.Analyst, 1981, 106, 960-967

ISSN:0003-2654    

DOI:10.1039/AN98106FP109

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

SUMMARIES OF PAPERS IN THIS ISSUE September, 1981Determination of Tqptophan in Feeds and Feed Ingredients byHigh-performance Liquid ChromatographyA high-performance liquid chromatographic procedure for the determinationof tryptophan has bCen developed using the native fluorescence of the mole-cule as a means of detection. Proteins were hydrolysed with 6 M sodiumhydroxide solution in the presence of maltodextrin, which has been demon-strated to prevent tryptophan degradation. Comparison of results from theproposed method with those from an existing procedure developed by Slumpand Schreuder showed the two methods to be statistically equivalent. Awide range of feeds and feed ingredients were analysed, and the measuredtryptophan levels were within the range expected from published data.Theproposed method is rapid, sensitive and specific, and is recommended forroutine laboratory use.Keywords : Tryptophan determination ; high-performance liquid chromato-graphy ; animal feedsA. D. JONES, C. H. S . HITCHCOCK and G. H. JONESUnilever Research, Colworth House, Sharnbrook, Bedford, MK44 1LQ.Analyst, 1981, 106, 968-973.Barium - Polyethoxylate Complexes as Potentiometric Sensors andTheir Application to the Determination of Non-ionic SurfactantsElectrodes having PVC matrix membranes consisting of 2-nitrophenyl phenylether and sensors of the tetraphenylborate salts of barium complexes withAntarox CO-730, Antarox CO-430, Lutensol A07 and Monflor 51 are com-pared with the corresponding electrodes of barium complexes with AntaroxCO-880.Generally, they are inferior to the Antarox CO-880 electrodes forsensing barium ions. However, all of the electrodes respond to non-ionicsurfactants in the manner described previously for the Antarox CO-880system.The electrode based on Antarox CO-430 is considered to be superior in itsresponse to non-ionic surfactants and shows some selectivity towards thesematerials compared with its response towards cationic and anionic surfactants.The electrode has been evaluated for determining Dobanol 25-7, Synperonic7 and Lutensol A07 in detergent powders with good recoveries of the non-ionic surfactants and relative standard deviations ranging between 0.8 and4.0%. The recommended procedure is based on known (standard) additionwhch, although taking up to 2 h to complete, is faster than analate addition.Keywords : Non-ionic surfactants ; ion-selective electrode response by non-ionic surfactants ; barium ion-selective electrodes ; polyethylene glycolsDILYS L.JONES, G. J. MOODY and J. D. R. THOMASChemistry Department, Redwood Building, University of Wales Institute of Scienceand Technology, Cardiff, CFl 3NU.and B. J. BIRCHUnilever Research, Port Sunlight Laboratory, Bebbington, Wirral, Merseyside,L63 3JW.Analyst, 1981, 106, 974-984September, 1981 SUMMARIES OF PAPERS IN THIS ISSUEErrors in the Direct Potentiometric Electrode Method of FluorideDetermination : Adsorption, Illumination and Temperature EffectsPre-soaking of clean plastic bottles for 1 day in dilute hydrochloric acidreduces adsorption of fluoride from dilute fluoride solutions, stabilisingstorage. Changes in ambient illumination cause an electrode potential drift,implying an increase in the fluoride concentration on commencement ofirradiation.The potential does not necessarily return to the original valueon cessation of irradiation. The extent of temperature dependence of theelectrode is directly proportional to the state of deterioration (“age”) of theelectrode membrane. Older electrodes show greater potential drift perdegree of temperature change than newer electrodes; this is often large in the15-30 “C range. Although the extent of drift is variable, the direction ofdrift is uniform, a temperature rise producing an apparent fall in the fluorideconcentration. A temperature hysteresis exists on sample heating andcooling.xiiiKeywords : Fluoride-selective electrode ; adsorption effects ; illumination effects ;temperature dependence ; errorsK.NICHOLSON and E. J. DUFFTurner Dental School, University of Manchester, Bridgeford Street, Manchester,M16 6FH.Analyst, 1981, 106, 985-991.Differential-pulse Polarography of Trichothecene Toxins :Detection of Deoxynivalenol in CornThe polarographic behaviour of T-2 toxin and deoxynivalenol has beeninvestigated. I t was found that the electrochemical activity of trichothecenetoxins is related to the presence of an a,p-unsaturated keto group and conse-quently the polarographic method was selective for trichothecenes having ketogroups.Differential-pulse polarography gave a detection limit for a puresolution of deoxynivalenol of 0.029 pmol 1-1 (8.6 ng ml-I). This technique,coupled with an appropriate extraction procedure, was applied to the deter-mination of deoxynivalenol in Fusarium infected corn. A detection limit ofabout 50 ng g-’ (for 50 g of the original sample) was estimated.Keywords : Deoxynivalenol determination ; T-2 toxin ; differential-pulsepolarography ; food analysisF. PALMISANOIstituto di Chimica Analitica dell’Universit& degli Studi, Via Amendola 173, Bari,A. VISCONTI, A. BOTTALICO and P. LERARIOCentro di Studio del C.N.R. su le Tossine e i Parassiti Sistemici dei Vegetali, Istitutodi Patologia Vegetale dell’universiti degli Studi, Via Amendola 165/A, Bari, Italy.and P. G.ZAMBONINIstituto di Chimica Analitica dell’Universit& degli Studi, Via Amendola 173, Bari,Analyst, 1981, 106, 992-998.Italy.ItalySeptember, 1981 SUMMARIES OF PAPERS I N THIS ISSUEDetermination of Meptazinol in Plasma by High-performanceLiquid Chromatography with Fluorescence DetectionXVShort PaperKeywords : Meptazinol determination ; high-performance liquid chromato-graphy ; Puorescence detectionT. FROSTWyeth Laboratories, Huntercombe Lane South, Taplow, Maidenhead, Berkshire,SL6 OPH.Analyst, 1981, 106, 999-1001.Determination of Diazepam and Its Major Metabolites UsingHigh-performance Liquid ChromatographyShort PaperKeywords : Diazepam ; benzodiazepines ; metabolites ; drugs in serum ; high-performance liquid chromatographyN. RATNARAJ, V.D. GOLDBERG, A. ELYAS and P. T. LASCELLESDepartment of Chemical Pathology, Institute of Neurology, The National Hospital,Queen Square, London, WClN 3BG.Analyst, 1981, 106, 1001-1004.Separation and Extractive Spectrophotometric Determination ofPalladium and Platinum with 6-Aminoquinoxaline-2,3- dithiolShort PaperKeywords : Separation of palladium from platinum ; platinum and palladiumdetermination; 6-aminoquinoxaline-2,3-dithiol; extractive spectrophoto-metryC. K. BHASKARE and R. G. PAWASHEDepartment of Chemistry, Shivaji University, Kolhapur 416 004, India.Analyst, 1981, 106, 1005-1009.Determination of Methylmercury in Tissue Using EnzymeProteolysisShort PaperKeywords ; Methylmercury determination ; tissue analysis ; subtilisin proteo-lysis ; electron-capture gas chromatography ; internal standardisationG. I. CALLUMDepartment of Clinical Physics and Bio-Engineering, West of Scotland HealthBoards, Glasgow, G4 9LF.M. M. FERGUSONUniversity Department of Oral Medicine, Dental Hospital, Glasgow, G2 3 JZ.and J. M. A. LENIHANDepartment of Clinical Physics and Bio-Engineering, West of Scotland HealthBoards, Glasgow, G4 9LF.Analyst, 1981, 106, 1009-1013

ISSN:0003-2654    

DOI:10.1039/AN98106BP117

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

SEPTEMBER 1981 The Analyst Vol. 106 No. 1266 Comparison of Flame, Electrothermal and Inductively Coupled Plasma Atomisation Techniques for the Direct Analysis of Slurries* C. W. Fuller,? R. C. Hutton; and B. Preston Tioxide International Limited, Central Laboratories, Stockton-on-Tees, Cleveland, TS18 2NQ A comparison is made between the use of flame, electrothermal and induc- tively coupled plasma atomisation techniques for the analysis of slurries. For flame atomisation, pulsed nebulisation must be used and for inductively coupled plasma atomisation a high-solids cross-flow nebuliser is essential. The results show that atomisation efficiency in nebuliser-based systems is dependent on sample transport efficiency, particle size, atomisation tempera- ture and sample matrix. For analytical determinations it is necessary to grind the sample to a particle size of less than 10 p m and to use standards that are closely matched to the samples.For electrothermal atomisation, particle-size effects become significant only above 25 pm when sampling becomes the major source of error. Keywords : Slurry techniques ; ore analysis ; flame, electrothermal and induc- tively coupled plasma atomisation ; atomic-absorption spectrometry Flame and electrothermal atomic spectrochemical methods of analysis have achieved great popularity over the last two decades. However, a disadvantage is that they require, almost exclusively, the sample to be in solution. This requirement means that, for refractory materials, the potential speed of these techniques has not been realised fully, as complex, time- consuming dissolution procedures are often required.Investigations into the advantages and possibilities of the direct analysis of solids have been reviewed by Langmyhr.l The review concluded that the problems of sample homogeneity, the need for repeated micro-weighings and the slow rates of analysis prevented this mode of analysis from being universally acceptable. Several workerss4 have demonstrated that by using samples ground to a fine particle size and suspended in an aqueous or organic medium by agitation, it was possible to nebulise slurries directly into an analytical flame. WilW outlined the limitations of this technique as being that the samples had to be ground to a particle size of less than 10 pm and that the slurries required constant agitation in order to achieve a homogeneous suspension.Previous papers5ys from these laboratories have described briefly methods of overcoming these limitations. These methods showed that the use of electrothermal atomisation, where larger particle sizes can be tolerated, together with a thixotropic thickening agent, to form stable suspensions, was preferable to those methods using aqueous suspensions and flame atomisation. This paper compares the use of flame, electrothermal and inductively coupled plasma (ICP) atomisation methods for spectrochemical analysis using this slurry technique. Experimental Instrumentation Flame atomisation measurements were carried out using a Perkin-Elmer, Model 360, atomic- absorption spectrometer.Electrothermal atomisation measurements were carried out using a Perkin-Elmer, Model 306, atomic-absorption spectrometer fitted with an HGA-74 graphite furnace atomiser. Both instruments were used under standard analytical conditions unless stated otherwise. * Presented a t the 5th SAC International Conference on Analytical Chemistry, Lancaster, July 20-26th, 1980. Present address : Central Electricity Generating Board, Scientific Services Department, Ratcliffe-on- To whom correspondence should be addressed. Soar, Nottingham, NGIl OEE. 913914 Analyst, VoZ. 106 Inductively coupled plasma measurements were carried out using an instrument constructed from a Radyne R50P generator and a Spex 1704 monochromator of 1-m focal length. The plasma generator was rated at a maximum of 5 kW forward power at 27.13 MHz.The plasma torch system used was a demountable, nitrogen-cooled unit of the Greenfield type. The nebuliser was a cross-flow slot type based on the design of Wolcott and Sobel.' Slurries were fed to the nebuliser using a Watson Marlow 501 variable-speed peristaltic pump operating at 3-4 ml min-l. An EM1 9568 QB photomultiplier tube was used, operated at 1.3 kV by an EM1 28B power supply. The photomultiplier signals were monitored using a Bentham 210E current amplifier and a Chessel flat-bed recorder. FULLER et al. : FLAME, ELECTROTHERMAL AND ICP Milling of Samples In order to study the effect of particle size on atomisation efficiency, samples were ground in a ball mill for various periods of time to achieve a final sample that was 90% m/m in the desired particle size range.Samples required for analysis were wet-milled in a McCrone micronising mill for 30 min with acetone; with this procedure it was possible to achieve samples with a mean particle size below 6 pm. The measurement of particle size was carried out by using either standard sieves or a Coulter Counter. Preparation of Slurried Weigh 1 g of the milled sample into a beaker, then wet the powder by adding 3 ml of 0.1% m/V sodium hexametaphosphate solution. Add 1.5 ml of Viscalex HV30 (Allied Colloids Ltd., Bradford), dilute to approximately 50 ml with water and then add 0.1 ml of NOPCO 'NPZ (Diamond Shamrock Process Chemicals Ltd., Kirkstall Rd., Leeds). Adjust the pH of the slurry to between 6 and 7 with a few drops of ammonia solution and stir constantly with a glass rod while the gel forms.Either dilute the sample to 100 ml in the beaker with water or transfer it into a 100-ml calibrated flask prior to dilution. It is necessary to dilute some solutions for electrothermal atomisation to lo00 ml. Suspensions in this form are stable for several days6 Results and Discussion Information on the following aspects of this technique has been reported previously,596 and (i) the effect of varying concentrations of Viscalex HV30 on the stability of the slurry; (ii) the addition of a de-foaming agent; (iii) the reproducibility of sampling Viscalex HV30 gels; and (iv) the use of a micro-sampling cup for flame atomisation. is not repeated here: Effect of Particle Size on Atomisation Efficiency Relative atomisation eflciency The notation used by Willis2 is employed in order to compare the atomisation efficiency, E , obtained from a slurry with that obtained from an aqueous solution containing the equivalent concentration of the same element.The value of E is given by the equation E = signal from slurry/signal from aqueous solution. Flame atomisation Conventional flame nebulisation systems are prone to blocking when slurries of this type are continuously nebulised : therefore, pulsed nebulisation, using 100-pl aliquots, was used to avoid this pr~blern.~ It was also observed during initial experiments that significantly higher absorbance readings were obtained for slurries, with only a slight worsening of precision, if the flow-spoiler in the spray chamber was removed.All results reported here, therefore, were obtained with the flow-spoiler removed. The effects of variations in particle size were 'determined for chromium and vanadium in samples of ilmenite and rutile ore using a dinitrogen oxide - acetylene flame (Table I). It can be seen that the highest values for E (0.3-0.5) were obtained for slurries containing particles less than 10 pm in diameter. When the particle size increased to above 25 pm the915 September, 1981 ATOMISATION TECHNIQUES FOR ANALYSIS OF SLURRIES TABLE I EFFECT OF PARTICLE SIZE ON E FOR CHROMIUM AND VANADIUM USING A DINITROGEN OXIDE - ACETYLENE FLAME Chromium Vanadium t A * # A . Particle size rangelpm Below 6 Below 26 2644 44-63 63-76 76-90 90-106 Ilmenite & Sample A Sample B 0.42 0.49 0.18 0.24 0.0s 0.16 0.03 0.03 0.01 0.01 - - - - Rutile Ilmenite iample A Sample 6 0.28 0.26 0.10 0.14 0.06 0.07 0.01 - - - - - - - r 1 Sample A Sample B 0.31 0.33 0.26 0.20 0.16 0.10 0.09 0.03 - - - - - - Rutile 1 Sample A Sample B 0.11 0.13 0.07 0.12 0.03 0.06 0.01 0.03 - - - - - - value of E fell to below 0.1.This effect can be partially attributed to sample transport effects. In general, for the nebulisers/spray chambers used in commercial, atomic-absorption spectro- meters, only droplets less than 10pm in diameter are efficiently transported through the system.6 Therefore, the large, solid particulates and agglomerates generated from slurries with large particle sizes will be similarly rejected as they pass through the spray chamber. The poor analytical response with slurries is also explained, in part, by incomplete atomisation of the particulates when they reach the flame.It can be seen from Table I that the values of E vary significantly for each element and also for each matrix. It can be concluded that the release of an element from the solid particles is dependent on the matrix and the form in which the individual elements are present in the matrix, in addition to the particle size of the sample present in the flame. It would be expected therefore that flame composition and flame temperature would have a significant effect on the observed atomisation efficiency. This is illustrated for manganese by the results given in Table 11. Although under aqueous condi- tions manganese would normally be determined in an air - acetylene flame, it can be seen that the hotter dinitrogen oxide - acetylene flame increases the values of E , for slurries with small particle size, significantly.TABLE I1 EFFECT OF FLAME TYPE ON E FOR MANGANESE E Particle size rangelpm Below 6 Below 26 25-44 44-63 63-75 75-90 90-106 , I Ilmenite Rutile Air - C,H, 0.36 0.33 0.16 0.04 0.03 0.02 0.01 -7 0.62 0.18 0.39 0.16 0.19 0.11 0.06 0.03 0.04 0.02 0.02 - 0.01 - N,O - C,H, Air - C,H, - N,O - CSH, 0.49 0.24 0.10 0.03 0.02 - Inductively cou9led plasma atomisation It would be expected that the hotter thermal environment of the ICP would give rise to improved signal response for slurries compared with the cooler air - acetylene and dinitrogen oxide - acetylene flames. However, the inability of the available concentric and cross-flow nebulisers to cope with slurries resulted in a lack of success in the initial attempts to use an ICP source, even when using pulsed nebulisation with discrete samples of 50-100 pl.To overcome this problem a cross-flow slot type nebuliser, similar in design 40 that described by Wolcott and Sobe1,'was fabricated. Tests on this nebuliser showed that, by using a peristal- tic pump, continuous nebulisation of slurries of up to and over 10% m/V could be achieved. A response curve for the increasing zirconium spectral emission from 2 to 10% m/V rutile samples is illustrated in Fig. 1. The reproducibility with 10% m/V slurries is still excellent (relative standard deviation = 0.02%) and typical results are shown for zirconium in Fig. 2.916 .c 250 a L.2 200 e CI .- 150 .- v) Q) .- E 100 .- w 5 0 E C v) .- Lu 0 FULLER et al.: FLAME, ELECTROTHERMAL AND ICP Analyst, Vol. 106 I 2 4 6 8 1 0 Rutile suspension, % m N Fig. 2. Consecutive scans across the Zr 343.82-nm erriission line while nebu- lising a sample containing 10% m/V of rutile in suspension. The sixth scan is a solution blank. Fig. 1. Effect of slurry concentration on emission intensity, for Zr, with an ICP. This nebuliser enabled a study of particle-size effects to be carried out in a similar manner to the other atomisation techniques. The results in Table I11 for vanadium and zirconium show the same trends as those observed with flame atomisation. As expected, the atomisation efficiency, for samples of small particle size, is significantly higher than that observed using the dinitrogen oxide - acetylene flame (compare with Table I).However, the predominant effect, as with the flame, is still one of sample transport efficiency. TABLE I11 EFFECT OF PARTICLE SIZE ON E FOR VANADIUM AND ZIRCONIUM USING ICP ATOMISATION Particle size rangelpm Below 6 Below 26 26-44 44-63 63-75; 76-90 90-1 06 Vanadium 5 i z r 7 G z 0.43 0.24 0.14 0.10 0.06 0.04 0.01 - - - Zirconium 5zrYG-z 0.45; 0.29 0.18 0.07 0.06 0.04 0.01 - Electrothermal atomisation The experiments were repeated using electrothermal atomisation, with 20-pl aliquots of the slurry, and the results are given in Table IV. In this instance, the values of E do not show the same variation with particle size as that observed with flame and ICP atomisation.Instead, there is a wide scatter in the values of E about unity, over the range of particle sizes studied. As it has been shown previouslys that the error in volume involved in pipetting 20-pl sample aliquots of the slurry is negligible compared with the errors observed here, it suggests that it is the relative standard deviation of the analytical measurement that is affected by particle size. The main sampling error associated with analysing a 20-4 aliquot of a 0.1% m/V slurry is illustrated in Table V. The theoretical numbers of particles were calculated by assuming spherical particles with an average diameter (for the range covered) and a density of 5 g ml-1. The actual numbers of particles were determined by pipetting ten consecutive 20-4 aliquots on to a microscope slide, counting the numbers of particles in each instance and calculating the mean and relative standard deviations.The experimentally observed results for the number of particles agree well with the calculated results, and show that as the particle sizeSeptember, 1981 ATOMISATION TECHNIQUES FOR ANALYSIS OF SLURRIES TABLE IV EFFECT OF PARTICLE SIZE ON E FOR CHROMIUM AND VANADIUM USING A GRAPHITE FURNACE 917 Particle size rangelpm Below 6 Below 26 26-44 44-63 63-76 76-90 90-106 Chromium I A ! Ilmenite Rutile -,, Sample A Sample B Sample A Sample B 1.04 1.01 1.12 1.13 0.88 1.2 1.08 1.07 0.64 1.4 0.96 0.97 0.63 1.06 1.01 1.36 0.92 0.86 0.81 0.76 0.73 0.83 0.76 0.76 0.64 1.0 1.07 0.86 Vanadium f A \ Ilmenite Rutile && Sample A Sample B Sample A Sample B 1.2 1.16 0.87 0.87 1.2 1.16 0.93 0.89 0.96 0.84 0.88 0.87 0.90 1.14 0.92 0.80 1.2 0.91 1.08 0.82 0.97 1.06 0.96 1.01 0.93 0.66 1.02 0.96 decreases the number of particles in a sample aliquot increases dramatically.These results are also in general agreement with results reported by Langmyhrl for the determination of mercury(I1) sulphide in an iron(I1) sulphide matrix. The relative standard deviations were also calculated for the determination of chromium and lead by electrothermal atomisation, for each of the particle size ranges used (Table V). It can be seen from the results that generally the sampling error represents 50% of the over-all analytical error. Therefore, the main requisite for electrothermal atomisation methods is that it is necessary to reduce the sampling error to an acceptable level to achieve the best analytical precision.TABLE V ERRORS ASSOCIATED WITH SAMPLE DELIVERY OF A RUTILE ORE SLURRY, USING A GRAPHITE FURNACE Aliquots (20 pl), containing 20 pg of sample, of a 0.1% wz/V slurry were used. Particle size rangelpm Below 6 Below 26 26-44 44-63 63-76 76-90 90-106 Mean number of particles 7 Calculated Measured 61200 About los 2260 About 1Oa 179 148 49 46 23 31 14 22 8 6 Relative standard deviation, % I L % Measured number of particles ,* -* 14 18 20 48 64 Determination of Pb 10 18 28 39 44 61 136 Determination of Cr 11 11 17 26 46 67 44 * Not determined. Analytical Determinations Flame and inductively coupled plasma atomisation The successful application of the slurry technique for practical analyses, using nebuliser- based systems, depends on satisfying two criteria.As the signal response is strongly depend- ent on particle size, some method of reproducibly grinding the sample to particle sizes of less than 10 pm (and preferably less than 6 pm) must be used. In the work described here, a McCrone micronising mill was found to be satisfactory (see under Milling of Samples and also Table IX). The second criterion is that as signal response varies with the sample matrix and gives only 30-50% of an aqueous solution response, then standardisation is important. The standard-additions procedure is not applicable to slurry analysis as the elements in the sample and the added standards are present in different physical states. They would therefore have different atomisation efficiencies.The alternative, but less versatile, approach is to use matrix matched standards. Results obtained by flame atomisation for the determination of trace elements in United States Geological Survey (USGS) silicate rocks, using one other standard for calibration, are shown in Table VI. The results obtained, by ICP atomisation, for the918 FULLER et d. : FLAME, ELECTROTHERMAL AND ICP A d J & , VOl. 106 TABLE VI DETERMINATION OF TRACE ELEMENTS IN STANDARD USGS SILICATE ROCKS USING FLAME ATOMISATION Concentration/ pg g-1 Sample No. 2098 2612 2613 2614 2616 2616 r -i Cr cu Mn Zn 5 E z - k z Z - m - 80 120 - - 2300 1600 - - - 1200 1400 66 60 2600 2660 - 3700 a430 - - 1800 1400 60 66 - 126 120 96 106 210 206 320 310 1400 1600 130 120 60 66 105 110 - - - - - - - determination of trace and minor elements in USGS silicate rocks and rutile ores, using the same calibration technique, are shown in Tables VII and VIII.In all instances a comparison of the determined values and the certified values showed good agreement. TABLE VII DETERMINATION OF TRACE ELEMENTS IN STANDARD USGS SILICATE ROCKS USING ICP ATOMISATION Gmmtration/pg g1 Ti Zr Ba Cr V Y Mn Sample -*- Certi- M i - ---- No. Found Certified Found fied Found fied Found fied Found fied Found fied Found fied 2514 1900 2000 120 120 70 80 3200 3400 60 65 10 5 1000 1400 2515 10500 11600 130 150 110 90 170 205 300 320 30 22 1500 1400 2266 13000 12000 130 200 250 290 570 630 170 180 10 - 1000 1400 1500 1700 2487 15000 13000 140 180 60 66 75 77 220 200 10 - TABLE VIII DETERMINATION OF TRACE ELEMENTS IN RUTILE ORES USING ICP ATOMISATION Concentration/pg g l A1 Cr Mn Nb V Zr Sample 5GzzZzl 5&z-ziG 5zsziz GzTzzz KLrZizl 30 2500 2900 3000 3500 2800 3500 A.. ..800 800 100 90 25 B .. .. 700 800 120 90 30 40 2600 2800 5000 5600 4600 4900 Electrothermal atomisatiovt With electrothermal atomisation, the problem of standardisation is not as critical as with the nebuliser-based systems. As the signal response is within &20% of the aqueous response it would be feasible for practical analyses to employ aqueous standards, provided that the samples have been ground to a particle size of less than 25 pm. The results, given in Table IX, illustrate the reproducibility of values of E , for alummiurn, chromium and vanadium, TABLE IX EFFECT OF SAMPLE GRINDING ON THE REPRODUCIBILITY OF E USING ELECTROTHERMAL ATOMISATION E v Cr 1 0.47 0.26 0.10 2 0.45 0.30 0.13 3 0.40 0.28 0.12 4 0.46 0.28 0.16 6 0.46 0.30 0.12 Sample No.Se$tember, 1981 ATOMISATION TECHNIQUES FOR ANALYSIS OF SLURRIES 919 obtained by electrothermal atomisation, using five samples of separately milled rutile ore.The reproducibility for the three elements is acceptable (relative standard deviation = 0.05- 0.15%). The elements cobalt, chromium, copper and vanadium were determined, by electrothermal atomisation, in a range of titaniferous ores and silicate rocks, and the results are presented in Figs. 3-6. The line drawn in each graph is the line of slope equal to 1, corresponding to 100% correlation between the concentrations determined and the accepted values.The actual slope values obtained were 1.05, 1.00, 0.98 and 0.95, respectively. 100 c 80 5. 0 5 60 . ,o - 6 40 20 0 20 40 60 80 100 120 Co (certified)/pg g-l Fig. 3. Correlation of cobalt concentrations, determined for a variety of samples by electro- thermal atomisation, against the certified values. Solid line shows 100% correlation. 120 100 c b 80 aj 40 ./ 2 0 y , , , , 1 0 20 40 60 80 100 120 C u (certified)/pg g-l Fig. 5. Correlation of copper concentrations, determined in a variety of samples by electro- thermal atomisation, against the certified values. Solid line shows 100% correlation. 250 200 c b t 50 0 50 100 150 200 250 Cr (certified)/pg g-l Fig. 4. Correlation of chromium concentra- tions, determined in a variety of samples by electrothermal atomisation, against the certi- fied values.Solid line shows 100% correla- tion. - 300 1 b 1 'c > i-l 100 I I I I 0 100 200 300 400 V (certified)/pg g-l Fig. 6. Correlation of vanadium con- centrations, determined in a variety of samples by electrothermal atomisation, against the certified values. Solid line shows 100% correlation.920 FULLER, HUTTON AND PRESTON Conclusions The results reported here allow an assessment to be made of the possibility of using slurries for spectrochemical analytical determinations. The slurry procedure is rapid, reproducible, has acceptable accuracy and is applicable to flame, electrothermal and ICP methods of atomisation. For nebuliser-based systems the variation in signal response with particle size and matrix means that, for practical analyses, closely matched standards must be used. How- ever, with electrothermal atomisation it is possible to use aqueous standards and obtain accept- able results, as there is a greater tolerance to particle size. This technique is therefore of use in situations that do not require the highest accuracy and precision, in which trends, rather than absolute results, are required, in which large numbers of samples are to be analysed and in which the samples can be ground to a fine powder. This work is published by permission of the Directors of Tioxide International Limited. References 1. 2. 3. 4. 6. 6. 7. 8. Langmyhr, F. J., Andyst, 1979, 104, 993. Willis, J . B., Anal. Chem., 1976, 47, 1762. Harrison, W. W., and Juliano, P. O., Anal. Chem., 1971, 43, 248. Kashiki, M., and Oshima. S., A n d . Chim. Ada, 1970. 51, 387. Fuller, C. W., Analyst, 1976, 101, 961. Fuller, C. W., and Thompson, I., Analyst, 1977, 102, 141. Wolcott, J . F., and Sobel, C. B., A@#. Spectvosc., 1978, 32, 691. Novak, J. W., Jr., and Browner, R. F., A n d . Chem., 1980, 52, 792. Received February 26th, 1981 Accepted April 13th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600913

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, September, 1981, Vol. 106, PP. 921-930 921 Investigation of the Determination of Tin Tetraalkyls and Alkyltin Chlorides by Atomic-absorption Spectrometry after Separation by Gas - Liquid or High-performance Liquid = Liquid Chromatography* D. Thorburn Burns, F. Glockling and M. Harriott Department of Chemistry, The Queen's University of Belfast, Belfast, BT9 5AG, Northern Ireland A variety of sample presentation and atomisation methods have been examined to obtain sufficient sensitivity for the trace determination of tin tetraalkyls and alkyltin chlorides (R,,SnCl,,; R = methyl or ethyl; n = 1-4) after their separation by gas - liquid or by high-performance liquid - liquid chromatography. Detection limits range from 10-20 pg (flame atomisation) to 1 pg (direct pyrolysis) and 2-20 pg (hydride reduction followed by direct pyrolysis).The construction of systems including an automated hydride generator is described. Keywords : Gas - liquid and high-performance liquid - liquid chromatography ; tin tetraalkyls and alkyltin chlorides ; pyrolysis ; hydride generation ; tube furnace The low sensitivities of standard detectors was a major problem in earlier chromatographic studies of tin tetraalkyls and alkyltin ch1orides.l Atomic-absorption spectrometry is suitable for the determination of single organometallic compounds,2 but it must be interfaced with a separation system in order to deal with mixture^.^-^ As with inorganic tin,6p7 organotin compounds are best atomised using a dinitrogen oxide - acetylene flame.2 Considerable problems exist in interfacing the sensitive carbon tube furnace atomisation system with a high- performance liquid chromatographic (HPLC) system,8 hence initial experiments were carried out using direct interfacing of a liquid - liquid chromatograph with flame atomisation and an atomic-absorption spectrometer.*v9 To improve sensitivity, electrothermal atomisationl0gll coupled with gas - liquid chromatography (GLC) was studied and pyrolysis conditions were optimised.Considerable improvements in sensitivity can be obtained for tin and other post-transition elements by conversion into gaseous hydrides prior to a t o r n i s a t i ~ n . ~ ~ ~ ~ ~ A further advantage is that most chemical interferences are eliminated because of the phase separation between sample matrix and the hydride vapour.13-16 In this way, detection limits for germanium, tin, selenium, antimony and arsenic have been improved by several orders of magnitude by gas- chromatographic separation of the individual hydrides using either a balloon or cold trap to collect separated material prior to atomisation.These time-consuming procedures can be avoided by using discrete sample automated method^,^^^^^ which also provide greater control over reaction conditions. A continuously operated hydride generation system, which can be interfaced with either a gas chromatograph or a liquid - liquid chromatograph and a low- temperature gradient quartz tube furnace, has been developed. Low detection limits and linear calibration graphs were obtained for the organotin compounds examined.Experimental Samples, Reagents and Solutions Reagents. All reagents were of analytical-reagent grade unless stated otherwise. Solvents. Acetone (laboratory grade) and pentane (spectroscopic grade) were used. Synthesis of organotin compounds. Tetramethyl- and tetraethyltin, prepared by Grignard reactions, were converted into the alkyltin chlorides by standard reactions19 and purified by recrystallisation or distillation under nitrogen as appropriate, until their elemental analyses agreed to within 0.3% of the theoretical values for each element. * Presented at the 5th SAC International Conference on Analytical Chemistry, Lancaster, July 20-26th, 1980.922 THORBURN BURNS et al. : TIN TETRAALKYLS AND Analyst, Vol. 106 Samples (20-30 mg) of each compound were weighed, transferred into calibrated flasks (100 cm3) and dissolved in and made up to volume with acetone or the appro- priate mobile phase (3 + 2 acetone - pentane for the methyl series; 7 + 3 acetone - pentane for the ethyl series).Stock inorganic tin solutiora 1 O00 pg ~m--~. Dissolve 1.0oO g of tin in 100 cm3 of concentrated hydrochloric acid and dilute to 1 dm3 with water. Dilute standard inorganic tin soZution, 1 pg ~ m - ~ . Dilute 1 to 1 O00 with 0.46 M hydrochloric acid. This solution must be prepared fresh daily. Sodium tetrahydroborate(II1) solution, 1 yo m/V. Dissolve 10 g of sodium tetrahydroborate- (111) (Aldrich) in 1 dm3 of 0.2% m/V potassium hydroxide solution. Vacuum filter through a 0.45-pm membrane filter crucible (Jobling Glassware).This solution is stable for at least 1 week. Organotin solutions. Apparatus Gas chromatograph oven A Perkin-Elmer F11, modified to take 6 mrn i.d. glass columns, was used. Liquid ch~omatograph This was equipped with a constant-flow pump (Pye Unicam, Model LC3), a syringe injection head (Pye Unicam), a column water jacket and a circulating water thermostat (Grant Instru- ments, SMOl), and was used with a 50-p1 high-pressure Hamilton syringe (V. A. Howe & Co.). Atomic-absorption s$ectrometer A Perkin-Elmer, Model 403, spectrometer fitted with a deuterium arc background corrector and a 5-cm single-slot dinitrogen oxide - acetylene burner head was used. Optimum operating conditions for the flame atomisation of organotin effluents were as follows: tin lamp, 8 mA (Southern Spectral Sources Ltd.); wavelength, 286.3 nm; slit width, 4 (0.7 nm) ; burner height, 2 mm; aspiration rate, 1 cm3 min-l; acetylene flow-rate, 5.5 dm3 min-l; dinitrogen oxide flow- rate, 3.3 dm3 min-l; recorder, 1 absorbance unit f.s.d., range 0-10 mV; noise suppressions, 1.Quartz T-tube, asbestos-wrapped farnace The dimensions and constructional details of the quartz T-tube furnace are shown in Fig. 1. The atomiser was heated to 1 OOO "C with a single filament 1.5-ft Brightray C wire (20 SWG, 1.681 ohm m-l), powered by a 10-V a.c., 30-A transformer and a Variac (RBS-MT, Claude Lyons Ltd.). Asbestos cord, thickness 1 cm (James Walker & Co. Ltd.), was used to cover the Brightray C wire H 10 rnm i.d. Fig. 1. Quartz T-tube furnace for GLC.September, 1981 ALKYLTIN CHLORIDES BY ATOMIC-ABSORPTION SPECTROMETRY 923 winding.The Pyrex transfer line was wrapped with a 1-m, 100-W heating tape (Gallenkamp) and heated to 500 "C using a second Variac. Temperatures were monitored using platinum - rhodium thermocouples and a temperature indicator. Air was introduced at 300 em3 min-l to support combustion of the solvent and organic pyrolysis fragments and to reduce the residence time of materials in the tube furnace. The furnace was mounted in a metal cradle and aligned in the optical axis of the atomic-absorption spectrometer. 1r 1 T 1% NaBH4, 1.5 cm3 min-' Hydride to furnace and atomic-absorption 1 spectrometer . v A - 16 cm I Air, 1.9 cm3 min-' I 0.5cm H 12 i cm NaBHo HCI Gas - liquid separator 1-4 cm i Fig.2. Hydride generator. I Air, 1.9 cm3 min-I 1 v 1 Proportioning pump Fig. 3. Hydride generation AutoAnalyzer manifold flow diagram.924 Analyst, Vol. 106 Quartz tube $rebrick fccrrtace The dimensions and constructional details are shown in Fig. 4. The thermal insulation was provided by an 11 x 6 x 5 cm Selfrac firebrick (GR-Stein Refractories) cut and channelled to fit the quartz tube, and the top and sides were covered with Supdux boards (Cape Boards and Panels). The quartz tube was heated with a single-filament 0.5-m Brightray C wire, powered by a transformer and a Variac regulator as for the quartz T-tube furnace (see above). A hole was drilled for the platinum - rhodium thermocouple to be in close proximity to the furnace. THORBURN BURNS et al. : TIN TETRAALKYLS AND Inlet I-2.5 cm4 Fig.4. Quartz tube firebrick furnace. Chromatographic Conditions GLC A glass column (2 m x 6 mm i.d.) was used, the stationary phase being 3% silicone rubber SE-30 (E301) on a Chromosorb G AW DMCS support. The column temperature was 120 "C for the methyltin compounds and 180 "C for the ethyltin compounds. The carrier gas was nitrogen at a flow-rate of 16 cm3 min-l for the methyltin compounds and 50 cm3 min-l for the ethyltin compounds.September, 1981 ALKYLTIN CHLORIDES BY ATOMIC-ABSORPTION SPECTROMETRY 925 HPLC The column dimensions were 250 x 3.0 mm i.d. and the stationary phase was ODS Spheri- sorb S5W (Phase Separations Ltd.). The mobile phases were acetone - pentane (3 + 2) for the methyltin compounds and acetone - pentane (7 + 3) for the ethyltin compounds; the flow- rates were 1.0 cm3 min-l for the methyltin compounds and 1.2 cm3 min-l for the ethyltin compounds.Results and Discussion HPLC-Direct Flame Atomisation, Atomic-absorption Detection The flame and observation heights were optimised by spraying solutions of each organotin compound into air - acetylene and dinitrogen oxide - acetylene flames at 4 cm3 min-l and obtaining absorbance - height and absorbance - acetylene flow-rate profiles. The dinitrogen oxide - acetylene flame was twice as sensitive as the air - acetylene flame and hence was used for further studies. Effects of the structure of the tin compounds on sensitivity, based on tin, were noted in the preliminary studies and in the HPLC results (Table I). A reduction in sensitivity with increasing halogen content is to be expected because of the relative strengths of Sn-C and Sn-Cl bonds, mean valuesz0 being about 200 and 340 k J mol-l, respectively.The high-performance liquid chromatograph was coupled directly to the atomic-absorption spectro- meter via 5 cm of 1 mm i.d. PTFE tubing and the nebuliser uptake rate was adjusted from 4 to 1.2 cm3 min-l to be compatible with the HPLC flow-rate. Base-line separations were obtained1 within each series, calibration graphs based on peak height were linear for each compound up to 50 pg of tin and relative standard deviations were consistent at about 4% (Table I). The limits of detection ( 2 4 were lower than those attain- able by spectrometry using dithizone or fluorimetrically using 3-hydro~yflavone,~~ but are not sufficient for environmental or toxicological analysesz2 and a more sensitive means of detection was sought.The temperature was 23 & 0.1 "C and the chart speed was 10 cm min-l. TABLE I HPLC - FLAME ATO~ISATION DATA FOR ORGANOTIN COMPOUNDS Relative standard deviation (u), yo 1% absorbance/ Detection limit (2u)/ Compound tB/min (10 results) CLg "g Me4% . . .. . . 1.0 4.3 0.9 11 MeSnC1, .. . . 1.8 4.6 1.6 19 Me,SnCl . . . . 1.2 4.6 1 .o 16 Me,SnCl, . . . . 1.6 4.4 1.1 17 Et4Sn . . .. . . 1.1 4.4 Et,SnCl . . . . 1.4 4.3 Et,SnCl, . . . . 1.7 4.7 EtSnC1, .. . . 2.1 4.1 0.9 1 .o 1.1 1.5 11 16 17 19 GLC-Quartz Tube Pyrolysis - Atomisation, Atomic-absorption Detection Samples were injected on to the column using a 10-pl long needle calibrated syringe and chromatograms were recorded; a typical example is shown in Fig.5. Base-line separations were obtained within each series except for methyl- and dimethyltin chlorides, which were not resolved using a 2-m column. Owing to on-column rearrangement reactions1 it is not possible to examine all four methyltin compounds in admixture; either methyltin trichloride or tetra- methyltin must be absent in order to avoid redistribution. Calibration graphs based on peak height were linear for each compound for up to 400 ng of tin. The variations in slope were less marked than for flame atomisation. Relative standard deviations (100 ng of tin, 10 replicate determinations) were reasonably consistent in the range 2-3.5% ; detection limits were 1-2 ng (Table 11).The use of the background corrector completely removed the solvent peak. The system also gave satisfactory results with butyl- and phenyltin chlorides. HPLC-Hydride Generation, Quartz Tube Atomisation, Atomic-absorption Detec- tion Inorganic tin A variety of tube heating and insulation systems were examined; the final design (Fig. 4) is926 THORBURN BURNS et al. : TIN TETRAALKYLS AND Artalyst, Vd. 106 TABLE I1 GLC - QUARTZ TUBE ATOMISATION DATA FOR ORGANOTIN COMPOUNDS Relative standard deviation (u), y-, 1% absorbance/ Detection limit (20) / Compound ts/min (10 results) ng ng Me4Sn . . .. . . 1.0 2.6 10 1.0 Me,SnCl . . . . 2.1 2.8 15 1.2 Me,SnCl, . . . . 3.2 3.3 10 1.0 MeSnC1, .. . . 3.2 3.5 11 1.1 Et4Sn . . .. . . 1.1 2.7 Et,SnCl, . . . . 2.2 3.4 EtSnC1, ... . 3.4 3.5 Et,SnCl . . . . 1.6 2.9 12 11 10 10 E J b) a) B C J ,-CJ 1.4 1.2 1.1 1.0 5 4 3 2 1 0 5 4 3 2 1 0 4 3 2 1 0 Retention time/min Fig. 5. GLC - quartz tube pyrolysis atomisation of methyl- and ethyltin compounds containing 200ng of tin: (a) 120"C, without background correction; (b) 100 "C, with background correction; and (c) 180 "C, with background correction. A, Me4Sn; B, Me,SnCl; C, Me,SnCl,; D, Et,Sn; E, Et,SnCl; F, Et,SnCl,; and G, EtSnC1,. robust and has a temperature variation along the length of the tube of less than 10 "C at the optimum operating temperature (950 "C). The effects of varying chemical and physical parameters using inorganic tin (10 pg) were examined sequentially (Fig. 6). The effects of carrier-gas flow-rate and acid and sodium tetrahydroborate( 111) concentrations are interdependent, as was noted by Vijan and Chanl' and Subramanian and Sastri.= The final conditions adopted could be subject to improvement if a complete multivariable optimisation study was carried out.= However, the sensitivities achieved were adequate for the proposed applications.The use of filtered sodium tetrahydro- borate(II1) increased the useful life of the solution from 1 day to 1 week, and also increased theSeptember, 1981 ALKYLTIN CHLORIDES BY ATOMIC-ABSORPTION SPECTROMETRY 0.5 (a) 0.5 0.4 al 0.3 e a j 0.2 0.1 0.2 0.4 0.6 0.8 1.0 1 2 3 4 5 Concentration of acid/ M NaBH4 concentration,% by mass 0.5 0.4 0.3 0.2 0.1 1 2 3 4 5 100 200 300 400 500 Air flow-ratekrn3 min-1 Nitrogen flow-rate/cm3 min-' 927 10 20 30 40 50 200 400 600 800 1000 Coil lengthkm Tube temperaturePC Fig.6. Variation in operating conditions of the hydride system. precision of results by a factor of 10, in accordance with the findings of Knechtel and Fra~er.~S Maximum sensitivity is achieved when the acid concentration of samples matched that in the manifold. Calibration and detection limit data were obtained for inorganic tin under optimised conditions as described in Table 111. Calibration graphs were linear for 0-10 and 0-100 ng of tin, relative standard deviations (10 replicate determinations) were 3.7% at 1 ng of tin and 1.0% at 10 ng of tin, the 20 detection limit was 1.8 pg of tin and the 1% absorption sensitivity was 5 pg of tin. Alkyltin compounds During preliminary studies of organotin compounds made by injection of samples in organic solvents, it was noticed that large solvent signals were caused by molecular absorption by928 THORBURN BURNS et al.: TIN TETRAALKYLS AND TABLE I11 OPERATING CONDITIONS FOR THE DETERMINATION OF INORGANIC Sodium tetrahydroborate(II1) concentration . . 1 % m/V Air flow-rate . . .. .. .. . . 1.9 cma min-1 Nitrogen flow-rate. . .. .. .. . . 180 cms min-1 Coil length .. .. .. .. .. 10cm Tube temperature .. .. .. .. 960°C Hydrochloric acid concentration .. . . 0.6 M Analyst, Vol. 106 TIN(II) water-immiscible solvents such as pentane, diethyl ether, benzene and chloroform. These interferences were eliminated by the use of background correction. For tetramethyl- and tetraethyltin the response was similar to but not identical with that of inorganic tin, but for other compounds (Table IV) the response, although linear, was a function of the thermal stability and volatility of the alkyltin hydrides produced, namely R3SnH, R,SnH, and RSnH,.TABLE IV VARIATION IN SENSITIVITY WITH STRUCTURE OF ORGANOTIN COMPOUNDS Alkyltin compound 1 yo sensitivity /pg Alkyltin compound 1 % sensitivitylpg Me4Sn . . .. .. 7.6 Et4Sn . . .. .. 8.0 Me,SnCl . . .. 7.6 Et,SnCl . . .. 8.0 MeSnC1, . . . . 6.0 EtSnC1, .. .. 6.4 Me,SnCl, . . .. 6.0 EhSnC1, .. .. 6.3 To avoid the use of separate calibration graphs for each compound, a variety of sample decomposition methods were examined. Results for Bernas's method26 using 1-2 mg of sample and 1 cm3 of aqua regia, digestion for 40 min at 110 "C followed by appropriate dilution and conversion into hvdride, and those of Kohama's2' wet oxidation to tin(1V) oxide, gave identical results (Table* V). .ANALYSIS OF ORGANOTIN Compound Me4Sn Me,SnCl Me,SnCl, MeSnC1, Et4Sn Et,SnCl EhSnCl, EtSnC1, Bu4Sn Bu,SnCl Bu,SnCl, BuSnC1, Ph4Sn Ph,SnCl Ph,SnCl, PhSnC1, TABLE V COMPOUNDS BY THE METHODS OF BERN AS,^ AND KOHAMA2' Tin content, % I A > Found I A \ Calculated Bernas's method Kohama's method .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . . . .. .. 66.37 69.66 64.03 49.43 60.67 49.18 47.91 46.60 34.46 36.46 39.06 42.06 27.78 30.79 24.62 39.28 66.1 60.0 64.1 60.3 60.3 48.7 47.8 46.7 33.8 36.3 39.2 42.2 27.7 30.6 34.7 39.2 66.1 60.1 64.1 49.4 60.6 48.9 47.8 46.8 33.9 36.3 39.2 42.3 27.8 30.6 34.8 39.2 The hydride system proposed is thus convenient for the determination of tin in organotin compounds after mineralisation.As it was shown that the detector can tolerate organic solvents immiscible with water, it was coupled to a high-performance liquid chromatograph. Under the conditions used, base-line separations were obtained for the methyl- and ethyltin chlorides (Fig. 7). The HPLC system had the advantage over the GLC system for the methyltin series in that redistribution reactions did not take p1ace.l The GLC system may be coupled to the hydride system with enhanced sensitivity owing to a reduction in dilution of acid in the manifold (Table VI); the detection limits (Table VII) show less enhancement, owing to in- creased base-line noise.September, 1981 ALKYLTIN CHLORIDES BY ATOMIC-ABSORPTION SPECTROMETRY 929 4 D C A B I' J Q J I I 5 4 3 2 1 0 3 2 1 0 Retention time/min Fig.7. HPLC - hydride generation - quartz tube pyrolysis atomisation of methyl- and ethyltin compounds containing 10 ng of tin. (a) A, Me,Sn; B, Me,SnU; C, MeaSnCIa; and D, MeSnC1,. (b) A, Et4Sn; B, Et,SnCl; C, EtaSnCla; and D, EtSnC1,. Conditions as in text. TABLE VI COMPARISON OF HPLC AND GLC RESPONSES USING HYDRIDE GENERATION AND PYROLYSIS ATOMISATION Compound (10 pg of tin) Me4Sn .. .. .. Me,SnCI . . .. .. Me,SnCI, . . .. .. MeSnC1, .. .. .. Et4Sn .. .. .. EGSnC1 .. .. .. EtaSnC1, . . .. .. EtSnC1, .. .. .. Absorbance r HPLC - hydride A 3 GLC - hydride generation generation 0.03 0.30 0.03 0.30 0.05 0.36 0.07 0.40 0.02 0.11 0.02 0.11 0.03 0.17 0.06 0.30 Conclusion Atomic absorption has been shown to be a valuable detection system for tin in both GLC and HPLC studies of organotin compounds. The best method of atomisation depends on the sensitivity required and to some extent on the compounds to be detected.Direct flame atomisation from HPLC effluents is more sensitive than current spectrometric or930 THORBURN BURNS, GLOCKLING AND HARRIOTT fluorimetric methods.21 Direct pyrolysis of GLC effluents in a quartz tube is, in general, less sensitive than hydride pyrolysis after GLC or HPLC, but is useful for compounds that give, under the conditions studied, little or no response, e.g., tetrabutyl- and tetraphenyltin. The latter system has a sensitivity comparable to the best previously recorded for the modified flame photometric GLC detection of tetrapropyltin ; however, results were reported only for this compound.28 The sensitivities of the various systems examined are summarised in Table VII.The low detection limits for tetramethyl- and tetraethyltin are difficult to account for, be- cause neither will react with tetrahydroborate(II1) during the short contact time involved. It is possible that the hydrogen evolved from acidification of tetrahydroborate(II1) may modify the decomposition processes and enhance the concentration of tin atoms in the gas phase. TABLE VII SUMMARY OF DETECTION LIMITS Detection system* I A AA (H - ETA) Mixture in order Katharometer - RI - AA (FA) - AA P -ETA) - of elution GLC/wg Sn HPLC/pg Sn HPLC/ng Sn GLC/ng Sn ‘ HPLC/pg Sn A GLC/pg sn\ Me,Sn .... 8.25 60 11 1.0 9.6 a.o Me,SnCl . . .. 8.25 80 16 1.2 9.5 2.0 Me,SnCl, . . . . 835 90 17 1.0 8.6 1.5 MeSnC1, .. .. Redistribution 100 19 Redistribution 8.2 Redistribution 60 80 90 100 11 16 17 19 1.4 1.2 1.1 1.0 14.0 13.5 12.0 10.0 6.5 6.3 4.6 2.a RI = refractive index; AA = atomic absorption; FA = flame atomisation; P - ETA = pyrolysis - electrothermal atomisation; H - ETA = hydride generation - electrothermal atomisation. 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. 14. 16. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. References Bums, D. T., Glockling, F., and Harriott. M., J. Chromatogv., 1980, 200, 306. Petree, 1.-B., and Smith, B. E. F., Mikrochim. Acta, 1974, 301. Van Loon, J. C., Anal. Chem., 1979, 51, 1139A. Fernandez, F.J., Chromutogv. Newsl., 1977, 5 (2), 17. Van Loon, J . C., Radziuk, B., Khan, N.. and Lichwa, J., At. Absorpt. Newsl., 1977, 16 (4), 79. Pickett, E. E., and Koirtyohann, S . R., Specfrochim. Acta, 1969, 24B, 325. Juliano, P. O., and Harrison, W. W., Anal. Chem., 1970, 42, 84. Vickrey, T. M., Howell, H. E., Harrison, G. V., and Ramelow. G. J., Anal. Chem., 1980, 52, 1743. Stoveken, J., and Fernandez, F. J., “Liquid Chromatography Technical Note,” No. 64, October Van Loon, J. C., and Radziuk, B., Can. J. Spectrosc., 1976, 21, 46. Parris, G. E., Blair, W. R., and Brinckman, Anal. Chem., 1977, 49, 378. Robbins, W. B., and Caruso, J. A., Anal. Chem., 1979, 51, 889A. Gadden, R. G., and Thomerson, D. R., Analyst, 1980, 105, 1137. Belcher, R., Bogdanski, S. L., Henden, E., and Townshend, A., Anal. Chim. Acta, 1977, 92, 33. Kadeg, R. D., and Christian, G. D., Anal. Chim. Acta, 1977, 88, 117. Braman, R. S., and Tompkins, M. A., Anal. Chem., 1979, 51, 12. Vijan, P. N., and Chan, C. Y., Anal. Chem., 1976, 48, 1788. Thompson, M., Pahlavanpour, B., Walton, S. J., and Kirkbright, G. F., Analyst, 1978, 103, 568. Dub, M., Editor, “Organometallic Compounds, Methods of Synthesis, Physic$ Constants and Compounds of Germanium, Tin and Lead, Second Edition, Neumann, W. P., “The Organic Chemistry of Tin,” John Wiley, Chichester, 1970. Aldridge, W. N., and Street, B. W., Analyst, 1981, 106, 60. Smith, P. J., “Toxicological Data on Organotin Compounds,” ITRI Publication No. 638, Inter- Subramanian, K. S., and Sasti, V. S., Talanta, 1980, 27, 469. Massart, D. L., Dijkstra, A., and Kaufman, L., “Evaluation and Optimisation of Laboratory Methods and Analytical Procedures,” Elsevier. Amsterdam, 1978. Knechtel, J. R., and Fraser. J. L., Analyst, 1978, 103, 104. Bemas, B., Anal. Chem., 1968, 40, 1682. Kohama, S., Bull. Chem. Soc. Jpn., 1963, 36, 830, Aue, W. A., and Flinn, C. G., Anal. Chem., 1980, 52, 1537. 1976, Perkin-Elmer, Norwalk, Conn., USA. Chemical Reactions, Volume 11. Springer-Verlag, Berlin, 1967. national Tin Research Institute, Greenford, 1978. Received February 23rd, 1981 Accepted April 7th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600921

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, September, 1981, Vol. 106, PP. 931-938 931 Inexpensive, Simple Hydride Generation System with Minimum Interferences for the Atomic-absorption Spectrophotometry of Arsenic* C. J. Peacock? and S. C. Singh Department of Chemistry, University of Lancaster, Bailrigg, Lancaster, LA 1 4 Y A A system for the hydride generation atomic-absorption spectrophotometry of arsenic is described that uses ordinary boiling-tubes to hold the acidified samples into which sodium tetrahydroborate(II1) solution is injected and the evolved hydrogen carries the generated arsine directly through the spectro- photometer nebuliser to a nitrogen-supported hydrogen flame. The total arsenic evolved is measured by integrating the absorbance for 16 s. Apart from transition metals of the cobalt, nickel and copper groups, the arsenic response appears to be insensitive to interferences, and the effects of these metals may be overcome by adding 0.5 g of thiourea to each sample.The limit of detection (95% confidence) is about 6 ng of arsenic and the repro- ducibility is about 1% for 0.5 pg of arsenic. Keywords : A rsenic determination ; hydride generation ; atomic-absorption spectrophotometry ; interferences Occasionally we have had a wish to determine arsenic at levels where hydride generation atomic- absorption spectrophotometry is appropriate, but the need has never been sufficient to justify the purchase of a commercial generation kit. However, Reesl has demonstrated a simple system consisting of a boiling-tube with a wide-bore tap through which sodium tetrahydro- borate(II1) pellets could be dropped.We originally tried to use his design but found difficulty adapting it as the pressure build-up was too great and the joints in the system tended to blow. We have designed a similar but even simpler system using 23 mm i.d., 150 mm long boiling- tubes holding the samples, on to which is fitted a rubber bung through which passes firstly a glass tube connected by a small rubber bung and a length of 1 mm i d . plastic capillary to the spectrophotometer nebuliser inlet, and secondly a short length of similar plastic capillary reaching to the bottom of the boiling-tube down which sodium tetrahydroborate(II1) solution may be injected from a plastic syringe (Fig. 1). To nebuliser 23 x 150 mm Fig. 1. Schematic diagram of hydride * Based on a Workshop Demonstration held at the 5th International Conference on Analytical Chemistry, t To whom correspondence should be addressed.generation apparatus. Lancaster, July 20-26th, 1980.932 PEACOCK AND SINGH: HYDRIDE GENERATION SYSTEM FOR Analyst, VoZ. 106 Whilst examining the optimum operating conditions, the evolution of arsine with time was recorded (Fig. 2). The normal method of measuring response by peak height is not appropriate with the variable manual injection we use; as the evolution of arsine fitted within the 16-s integration time we can use on our spectrophotometer, and as we were more interested in accuracy than in sensitivity, it was convenient to integrate the absorbance from the whole of the arsine evolved. This led to two great advantages : the between-sample reproducibility was improved to better than 1% at 0.5 pg of arsenic per sample, and virtually all of the many inter- ferences to which the hydride generation method is normally subject2J [arsenic(III) or arsenic- (V), other hydride-forming elements, strong acid or salt solutions, etc.) disappear because changes in the rate of evolution of hydride do not affect the total integrated signal.The only interferences still apparent were from transition metals of the cobalt, nickel and copper groups, probably because these form stable complexes with arsine, so limiting the amount evolved. Various suppressants are reported that allow this problem to be overcome. 0.5 0.4 6 0.3 Q) f! 8 0.2 a 0.1 0 5 10 15 20 Injection injection started finished t Time/s t Fig. 2.Variation of absorbance with time from arsine evolved Injection of 200mg of sodium tetrahydro- The from 1 pg of arsenic. borate(II1) started at t = 0 s and finished at about t = 10 s. trace shows the absorbance averaged over l/lbth-s steps. Experimental An Instrument ation Laboratory IL25 1 double-beam a tomic-absorp tion spec tropho tome t er was used with a Rank-Hilger slotted-cathode arsenic hollow-cathode lamp generally operating at 193.7 nm. Background correction was not used. The hydride generator was connected directly to the nebuliser inlet, feeding the evolved arsine directly into a nitrogen-supported hydrogen flame. To convert from normal solution analysis with an air - acetylene flame to hydride generation or back took less than 5 min.The spectrophotometer has an ultraviolet detector to sense whether the flame is lit, but as it has difficulty in noticing the very low emission from the hydrogen flame it was by-passed by soldering a resistor across the terminals. As the instrument has an automatic gas control system, the ultraviolet detector is redundant and is not included in later Instrumentation Laboratory instruments where automatic gas control and air and fuel gas sensing are standard fittings. Arsenic standards were prepared from analytical-reagent grade arsenic( 111) oxide ; inter- ferences were tested with analytical-reagent grade reagents when available (reagent grade for gold, platinum, zirconium, titanium, antimony, bismuth and calcium chloride) ; analytical- reagent grade mineral acids were used.Sodium tetrahydroborate( 111) solutions were pre- pared as required by dissolving reagent-grade material in de-ionised water in which one pellet of sodium hydroxide was dissolved for every 20ml. These solutions were stable and any excess could be kept in unsealed containers in a refrigerator for many days, but as they were so simple to prepare they were generally made as required. In the tables below, the term “response” is used to represent “absorbance averaged over 16 s x 1OOO.” On all occasions the fume extraction system over the flame was used.September, 1981 ATOMIC-ABSORPTION SPECTROPHOTOMETRY OF ARSENIC Results and Discussion Reaction Conditions 933 As the amount of sodium tetrahydroborate(II1) injected into the acidified sample was in- creased the response first rose, then reached a plateau (Table I).As there was sufficient sensitivity to dilute samples of more than 5 pg, 300 mg of sodium tetrahydroborate(II1) per injection was taken as normal although, if the arsenic content was known to be low, less could be used with no effect on the integrated response. TABLE I EFFECT OF AMOUNT OF SODIUM TETRAHYDROBORATE(III) ON SENSITIVITY TO ARSENIC All samples in 10 ml of 10% V/V hydrochloric acid. Amount of sodium tetrahydroborate(II1) addedlmg . . .. 60 100 160 200 300 400 540 672 697 695 Response from 600 ng of arsenic . . .. .. .. 46 70 75 76 75 - - - Response from 5 pg of arsenic . . .. .. .. .. Similarly, the response to increasing acidity of the sample reached a plateau and higher acidities had no effect (Table 11).Provided that there is more than the theoretical amount of acid present for complete reaction with the tetrahydroborate(III), the total acidity or type of acid is immaterial. An amount of 300 mg of sodium tetrahydroborate(II1) requires more than 0.008 mol of hydrogen ion for complete reaction. This insensitivity to type or concentra- tion of acid makes the digestion of the samples much easier as there is no need to evaporate off large volumes of acid with possible consequent losses of arsenic, and the acid regime most con- venient for sample decomposition may be used. TABLE I1 EFFECT OF ACIDITY AND TYPE OF ACID ON SENSITIVITY TO ARSENIC All samples in 10 ml of solution, injected with 2.0 ml of 16% m/V sodium tetrahydroborate(II1) solution.Hydrochloric acid concentration/M . . 0.3 0.6 1.2 1.8 2.6 1 + 1* Response from 600 ng of arsenic . . . . 69 69 75 74 75 76 Other acids . . .. .. .. . . Sulphuric acid Nitric acid Perchloric acid Response from 600 ng of arsenic . . .. 76 74 76 76 75 76 acid. & r - - - - - V Concentration . . .. .. .. 0 . 6 M 1 + 1* 1.2M 1 + 1* 1.2M 1 + 1 1 + 1 acids were prepared by mixing equal volumes of de-ionised water and the concentrated The volume of the solution is again relatively immaterial, the same response being obtained for volumes varying between 5 and 20 ml. Below 5 ml mixing seems to cause a slightly vari- able response, and above 20 ml the volume becomes too large for the boiling-tube. Larger boiling tubes cannot be used satisfactorily because the arsine evolution is then too slow to be completed in 16 s.Flame Conditions The gas flow-rates affected the sensitivity markedly: the lower the hydrogen flow-rate the higher was the response, but the blank increased because the excess of hydrogen evolved made the flame more opaque. Similarly, lower nitrogen flow-rates also increased the sensitivity. We standardised on 30mls-1 for hydrogen and 9Omls-l for nitrogen as a compromise between lower flow-rates giving longer arsenic residence times and higher sensitivity, and higher flow-rates giving a stiffer, less draught-sensitive flame. The response did not vary much with the height in the flame at which measurements were made. Normally a height of 6 mm above the burner was used. The use of argon as the support gas in place of nitrogen gave no measurable improvement.934 Reproducibility and Limit of Detection standard deviation of about 1% for 0.5 pg of arsenic. PEACOCK AND SINGH: HYDRIDE GENERATION SYSTEM FOR Analyst, VoZ.106 Table I11 shows the reproducibility of the method under normal conditions, giving a relative TABLE I11 REPRODUCIBILITY Two separate repeat determinations of 500 ng of arsenic. hydrogen, 30 ml s-1. of 4% sulphuric acid solution. Gas flow-rates: nitrogen, 90 ml s-l; 2.0 ml of 15% m/V sodium tetrahydroborate(II1) solution added to 10 ml Relative standard Mean deviation, % Average absorbance . . 385, 379, 382, 377, 383, 376, 381 380 0.9 over 16 s x 5000 {it\ . . 438, 428, 433, 431, 428, 431, 435, 431 432 0.8 Table IV gives the results of repeated determinations of blank and 20-ng arsenic samples, using injections of solutions containing 100 mg of sodium tetrahydroborate(II1) to lessen the flame disturbance and blank from arsenic in this reagent.Under these conditions the 95% confidence limit for detection of arsenic in a single sample is about 6 ng. With injections of solutions containing 300 mg of sodium tetrahydroborate(II1) this limit increases about 3-fold. TABLE IV LIMIT OF DETECTION FOR ARSENIC Alternate blank and 20-ng arsenic samples in 10 ml of 4% V/V sulphuric acid, injected with 2.0 ml of 6.0% m/ V sodium tetrahydroborate(II1) solution. Standard Average absorbance over 16 s x 6000 Mean deviation Blank . . .. . . 57, 58, 55, 55, 56, 54, 54, 65, 57. 56, 66 55.7 1.3 20 ng of arsenic . . . . 76, 73, 74, 72, 76, 76, 77, 77, 76, 70 74.6 2.4 Net response: 18.9 (standard deviation 2.7). Linearity With our equipment the linearity of the response for atomic-absorption measurements always becomes poor at absorbances above about 0.4, irrespective of the element being deter- mined.With samples containing more than about 0.7 pg of arsenic, unless the initial rate of injection of sodium tetrahydroborate(II1) solution is very slow, absorbances greater than 0.4 are instantaneously exceeded and linearity is lost. Further, with the irreproducibility of the manual injection, once linearity is lost then reproducibility is also lost because the total absorb- ance recorded depends on the proportion that was above 0.4 absorbance unit. If the sodium tetrahydroborate(II1) solution could be added slowly enough, then linearity could be extended to about 3pg of arsenic using the 193.7-nm arsenic line.However, normally the limit is about 1 pg (Table V) with the sodium tetrahydroborate(II1) solution being added at a more or less constant rate over about 10 s. Addition of the sodium tetra- hydroborate(II1) solution slowly at first whilst watching the intensity meter on the spectro- photometer and attempting to keep the transmittance above about 50% can be seen to increase sensitivity and linearity at higher concentrations. Interferences Samples of 0.5 pg of arsenic were tested for interferences by adding 2 mg of various cations and anions. It was found that 0.6 pg of arsenic is well within the linear part of the response graph, so minimising variations in total response caused by the rate of evolution of the arsine, which would affect the maximum absorbance reached.In all instances a determination of the blank response (interferent without added arsenic) was also made, because some samples contained appreciable amounts of arsenic.September, 1981 ATOMIC-ABSORPTION SPECTROPHOTOMETRY OF ARSENIC TABLE V LINEARITY OF RESPONSE FOR ARSENIC 935 Samples in 10 ml of 4% V / V sulphuric acid, injected with 2.0 ml of 16% m/V sodium tetrahydroborate(II1) solution. Arsenic added/pg 0 0.5 1 .o 2.0 3.0 5.0 10.0 A* 16 91 166 310 416 666 766 Relative sensitivity B* Relative sensitivity - 16 - 1 .oo 92 1.01 1 .oo 168 1.01 0.98 318 1.01 0.89 463 0.99 0.73 616 0.80 0.49 905 0.59 * Column A records the response obtained when sodium tetrahydroborate(II1) solution was injected at a normal, nearly constant rate over about 10s; and column B records the response obtained when the initial addition of sodium tetrahydroborate(II1) solution was kept as low as possible to try and keep the transmittance always above 60%.From Table VI it can be seen that only cobalt, nickel, platinum, copper, silver and gold interfered appreciably with the determination of 0.5 pg of arsenic. No interferences were caused by other hydride-forming elements, although the antimony and iron samples studied had fairly high blank arsenic values; also, no difference was noted between the normal arsenic- (111) or the arsenic(V) that would have been present in the chromium, manganese, vanadium or molybdenum samples, which were added as their oxoanion salts [K,Cr,O,, KMnO,, NH,VO, and (NH4),Mo,0,,.4H,O, respectively].Interference Suppressants In order to allow the determination of arsenic in any matrix, it was necessary to find a means of overcoming the interferences from the softer transition metals. All of these gave black precipitates of the metal after reaction with sodium tetrahydroborate(II1) (although this in itself is not sufficient to cause interference, for so do bismuth, iron and others). It was felt that arsine was probably complexing with the metal formed and was therefore not being carried into the flame. Various complexing agents were tried to compete for the arsine and/or to keep the metals in higher oxidation states, which have a lower tendency for arsine complexation.Some improvements were found with most complexing agents [e.g., dimethyl glyoxime, ethylenediaminetetraacetic acid (EDTA), triphenylphosphine and trimethyl phosphite). Some ligands were found to be very effective for certain of the elements: cyanide and thio- cyanate overcome completely the suppressive effect of the first-row transition metals, fluoride is equally effective for the coinage metals (Table VII), but only thiourea seemed to overcome the interferences from all of these elements. Table VIII gives the results of a repeat of the experiment in Table VI for those elements that exhibit interference but with 500 mg of thiourea added to each sample. It can be seen that this amount of thiourea removes all of these interferences. Brief experiments were carried out in order to find the maximum amounts of some of the interfering elements that could be tolerated.It was found that 0.7 g of thiourea overcame the interferences from 100 mg of copper, silver, cobalt and nickel, although there is a tendency at these levels for the solutions to foam on adding the sodium tetrahydroborate(II1) and occa- sional analyses had to be repeated because foam was swept into the flame, altering the response by several per cent. As the limit of detection is about 10 ng, it seems possible to determine down to 0.1 p.p.m. of arsenic in these metals without any concentration or separation steps. Recommended Procedure The exit capillary of the hydride generation apparatus (Fig. 1) is connected to the nebuliser inlet of the atomic-absorption spectrophotometer and the nebuliser flow-rate is turned up to the maximum (about one full turn on the IL251 instrument). A hydrogen cylinder is connected as fuel supply and nitrogen in place of air, the flow-rates being set to 30 ml s-l (4 s.c.f.h.on the IL251 flow meter) and 9Omls-l (12 s.c.f.h.), respectively. For the most sensitive work a936 PEACOCK AND SINGH: HYDRIDE GENERATION SYSTEM FOR AMbSt, T/'Ol. 106 wavelength of 193.7 nm is used with the beam passing through the flame about 6 mm above the burner. Better reproducibility is obtained if the flame is shielded from draughts. Samples prepared in the normal way (acid digests, extractions, etc.) are placed in 150 x 23 mm i.d. boiling-tubes with each sample containing at least 0.008 mol of hydrogen ion (e.g., 10 ml of 10% V/V hydrochloric acid or 4% V/V sulphuric acid are sufficient).If metals of the cobalt, nickel or copper groups are present, 0.5g of thiourea should be added to each sample. A total volume of 5-20 ml is required. TABLE VI EFFECT OF VARIOUS ANIONS AND CATIONS ON RESPONSE FOR ARSENIC Solutions contained 2 mg of each element tested in 12 ml of 4% V/V sulphuric acid, except for silver, barium and lead, for which 10% V/V nitric acid was used, and tin, which was in 20% V/V hydro- chloric acid. 2.0 ml of 16% m/V sodium tetrahydroborate(II1) solution were injected. Response Element Li .. Na . . K .. c s .. Mg . . Ca .. Sr . . Ba .. B .. Al .. Ga .. In .. Si . . Sn .. Pb .. P .. Sb .. Bi .. Se .. La .. Ti .. Zr .. v .. Nb . . Ta .. Cr ..Mo .. w .. Mn . . Fe .. co .. Ni .. Pt .. cu .. Ag .. Au . . Zn .. cd * . Hn . . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. No additives Form added (aqueous solution) LiCl NaCl KCl cs,so, MgCl, CaCl, Srcl, HSBO, AlCl, BaCl, GaCl, InCl, Na,SiO, SnCl, Pb"032 K,HPO, K(Sb0)tartrate BiCl, Na,SeO, KTi (oxalate) , LaCl, zr(s04h Ta (NOS) 0 KaCr207 NHAVO, NbCl, (NH36M07024 Na,WO, KMnO, FeC1, coc1, Ni (NOS), cu (NOS), &NOS K,PtC16 HAuCl, ZnC1, CdSO, HgC1, - Standard deviation without additives .. .. .. Mean with non-interfering additions . . .. Standard deviation . . Blank 24 21 23 22 21 22 21 21 24 21 21 22 22 28 22 22 64 22 23 21 28 28 23 22 22 24 26 26 24 40 9 11 12 19 22 11 22 22 23 22 +600 ng of arsenic 106 106 106 106 104 106 106 104 108 104 105 106 106 112 106 106 137 106 107 106 111 112 106 104 106 106 109 108 107 122 20 11 23 93 80 16 104 106 106 106 Net for 600 ng of arsenic 81 86 82 83 83 83 84 83 84 83 84 84 83 84 83 83 83 84 84 86 83 84 82 82 83 82 84 83 83 82 11 0 11 74 68 4 82 84 82 83 1.4 83.1 1.0 Remarks Some black precipitate Slowed evolution by several seconds Gave pale blue solution Slowed evolution by several Gave yellow solution Gave dark blue solution Slowed evolution by several Gave black precipitate Gave black precipitate Gave black precipitate Gave black precipitate Gave brown - black precipitate Gave yellow - black precipitate Gave black precipitate seconds seconds Gave grey precipitate Mean for 14 samplesSe$tember, 1981 ATOMIC-ABSORPTION SPECTROPHOTOMETRY OF ARSENIC 937 TABLE VII EFFECTIVENESS OF VARIOUS ADDITIVES IN OVERCOMING THE SUPPRESSION OF ARSENIC RESPONSE BY VARIOUS TRANSITION METALS Conditions.as in Table VI.Net response after adding 0.6 g of- , 1 Diethylammonium Interferent KCN KCNS KF Dipyridyl o-Phenanthroline dithiocarbamate Thiourea 83 co .. . . 82 84 - 66 Ni .. . . 83 83 0 84 62 82 83 62 30 30 82 83 80 a4 86 64 76 82 lo 86 Pt .. .. 11 26 cu .. . . 83 83 Ag .. . . 48 87 86 66 83 Au .. . . 13 24 84 2 - - - - The rubber bung with a capillary inlet and outlet is fitted on to the sample tube and the plastic syringe is filled with 2.0 ml of 15% m/V sodium tetrahydroborate(II1) solution, wiped and connected to the inlet capillary. The signal may be integrated for 16 s to give the zero, although this does not vary much once the lamp, burner and gas flow-rates have settled down.A check after about every five samples have been processed is generally adequate. The IL251 atomic-absorption spectrophotometer has a so-called “manual” mode, which gives a 4-s delay before beginning the signal integration that conveniently allows the operator to take hold of the boiling-tube and bung in one hand and the syringe in the other before begin- ning to inject the sodium tetrahydroborate(II1) solution once integration starts. If arsine evolution can be watched on the instrument intensity meter, this allows the operator the possibility of keeping the initial injection rate low in order to avoid the transmittance falling into the non-linear region (below 40%), and of seeing that all the arsine has been evolved within the integration time.After evolution has ceased the syringe is disconnected and wiped. At this stage some residual sodium tetrahydroborate(II1) solution in the capillary may squirt back, so it is advis- able to hold the syringe nozzle in a tissue to catch any such drops and the operator should wear plastic or rubber gloves and suitable eye protection. The rubber bung is then disconnected from the sample tube and wiped, and is then ready for the next sample. After about every ten samples it is useful to blot away any condensation from the exit capillary and tube, although small drops of water carried into the flame do not appear to have any significant effect on the absorbance recorded. However, large amounts of carryover do limit the samples to those that do not foam much, which means that organic matter should be totally destroyed in an acid digest.This is relatively easy as the method is insensitive to the acid in which the sample is present, so optimum wet ashing conditions can be used. Normally an injection time of about 10 s is convenient. TABLE VIII EFFECTIVENESS OF THIOUREA IN OVERCOMING THE SUPPRESSION OF ARSENIC RESPONSE BY VARIOUS TRANSITION METALS Conditions as in Table VI, with 0.6 g of thiourea added to each sample. Response Interferent co .. .. Ni . . .. Pt .. .. c u .. .. Au . . .. No additions . . Ag . . .. Blank 20 23 19 20 20 19 19 +600 ng of arsenic 107 109 106 107 106 107 106 Net for 600 ng of arsenic Remarks 87 Gave slightly blackish solution 86 Gave blackish solution 86 87 86 Gave blackish solution 88 87 Solution stayed clear and colourless Solution stayed clear and colourless Gave blue - black solution938 PEACOCK AND SINGH Conclusion An inexpensive, simple system for the hydride generation atomic-absorption spectrophoto- metric determination of arsenic has been developed.Because it uses the injection of sodium tetrahydroborate(II1) solution into the acidified sample solution held in simple boiling-tubes and measures the average absorption by the arsenic atoms generated from the total arsine evolved in a nitrogen-supported hydrogen flame, there are several advantages : it takes less than 5 min to change from normal solution analysis with an air - acetylene flame to the hydride system or back; analyses are relatively rapid (about 30 samples per hour); the ordinary glass boiling-tubes that hold the samples are convenient to handle and clean and cheap to buy in quantity; integrating the total absorbance means that there are none of the normal interfer- ences that depress the signal when peak height is measured, except those from the softer transition metals, which may be suppressed by adding thiourea to the samples; and the method is sensitive and reproducible, having a relative standard deviation of 1% for 500-ng levels of arsenic and a limit of detection of 6 ng of arsenic.Preliminary work has already shown that other elements may be determined in the same way, apparently with the same advantages. It is hoped to report conditions for these in the near future. As there is no interference between the different hydride-forming elements, work is also in progress to find conditions for the simultaneous determination of pairs of elements by hydride generation. Although outside the criterion of being cheap or simple, an automatic injection system and furnace atom cell are being developed to improve reproducibility and linearity further. We acknowledge the many helpful discussions with those who attended the Workshop Session at the SAC 80 Conference held at the University of Lancaster in July, 1980, at which this system was demonstrated and preliminary results were presented. References 1. 2. 3. Rees, D. I., “Determination of Trace Metals in Foods by Atomic-absorption Spectrophotometry,” Circular No. 5, Lyons Central Laboratories, London, 1978. Evans, W. H., Jackson, F. J., and Dellar, D., Analyst, 1979, 104, 16. Arbab-Zavar, M. H., and Howard, A. G., Analyst, 1980,105, 744. Received September 23rd, 1980 Accepted April 22nd. 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600931

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, September, 1981, Vot. 106, j5$. 939-943 939 Determination of Trace Amounts of Barium in Calcium-containing Matrices by Atomic-a bsorption Spectrop hotomietry Following Solvent Extraction with Polyethylene Glycol A. M. Y. Jaber and M. Y. S. El-lssa Chemistry Defiartment, Faculty of Science, University of Jordan, Amman, Jordan Polyethylene glycol (PEG) 1640 shows a high efficiency (almost 100%) for the extraction of barium from a calcium matrix under appropriate conditions. This has permitted trace amounts of barium in a calcium matrix to be deter- mined by flame atomic-absorption spectrophotometry without interference from calcium. The effectiveness of PEG is noteworthy because of its ready availability and cheapness compared with the 18-crown-6 used in existing procedures.Further, the extraction process may be completed successfully in one step without subsequent washing of the organic phase. Keywords : A tomic-absorption spectrophotometry ; solvent extraction ; calcium interference ; polyethylene glycol; 18-crown-6 The determination of barium in environmental samples containing high levels of calcium, such as limestone, silicate rocks, marine organisms and bone, is of great importance. The deter- mination of barium by flame atomic-absorption techniques has suffered from interference by the large background due to cal~ium.l-~ Such interference has been attributed mainly to the formation of the CaOH molecular species, which has an emission band at 554 nm that overlaps the barium band at 553.5 nm. Rooney and Woolley* have attributed the interference to the type of instrument used.Although the interference has been reported to be overcome by using a dinitrogen oxide - acetylene flame,6 other workersl4Pg have found the interference to be present even with such flames, especially when the barium is present at trace levels. Hence the barium must be separated from the calcium matrix in order to avoid the interference. Separation by coprecipitation,l ion-exchange chromatography3 or liquid - liquid extraction6 using the crown ether 18-crown-6 have been used successfully. Organic solutions of polyethylene glycol (PEG) and its derivatives have shown selective efficiency towards alkali and alkaline earth metal ions in a similar manner to crown ethers.' The extraction efficiency of PEG increases as the dielectric constant of the organic solvent increases, e.g., it is high with nitrobenzene.Also, under such conditions PEG is more selective towards barium compared with calci~rn.~ The use of PEG as a quantitative extractant for trace amounts of barium from a calcium matrix prior to atomic-absorption spectrophotometric determination is described in this paper. Experimental Reagents reagent grade (BDH Chemicals). PEG 1540 was of GLC grade (BDH Chemicals). 6 was synthesised according to the Dale and Kristiansen method.* in an Elgastat de-ioniser. Barium and calcium nitrates, nitric acid, picric acid and nitrobenzene were of analytical- 18-Crown- All solutions were prepared in de-ionised water obtained by distillation followed by treatment Extraction and Measurement Procedure The aqueous phase consisted of barium nitrate of various concentrations (greater than 2.0 x 10-5 mol dm-3), calcium nitrate (5.0 x 1W2 mol dm--5) and picric acid (1.2 x or 1.8 x 10-2 mol dm-3).The organic phase of nitrobenzene contained either 18-crown-6 or PEG at different concentrations. Equal volumes (20 cm3) of each phase were mixed and shaken vigorously in a Quickfit test-tube for 15 min at room temperature with a flask shaker and940 JABER AND EL-ISSA : BARIUM IN CALCIUM-CONTAINING A d y S t , VOZ. 106 centrifuged. The extraction steps were completed according to the procedure of Takeda et aZ.6 An aliquot (18 cm3) of the organic phase was withdrawn, placed in another Quickfit test-tube with an equal volume of picric acid (1.8 x 1W2 mol dm3, shaken and centrifuged under the same conditions as above.An aliquot (16 cm3) of the washed organic phase was withdrawn and re-washed with an equal volume of picric acid (1.8 x 1W2 mol dm-3). The metal ions extracted into the organic phase after double washing with picric acid were stripped back into the aqueous phase by shaking an aliquot (14 cma) of this phase with an equal volume of nitric acid (1 mol dm-9. This work indicated that the double washing with picric acid could be avoided by using PEG in the organic phase. Thus, an aliqout (16 cm3) of the organic phase from the first extraction step was withdrawn, shaken with an equal volume of nitric acid (1 mol dm-3) and centrifuged. The nitric acid phase was used for the determination of barium with a Perkin-Elmer 305A atomic-absorption spectrophotometer in conjunction with a Varian strip-chart recorder using a dinitrogen oxide - acetylene flame.Calcium co-extracted with barium was also determined by atomic-absorption spectrophotometry but using an air - acetylene flame. Each extraction run was repeated three times for each barium and calcium concentration. As the triplicate results agreed very closely with each other, only average values are reported here. The calcium interference effect was studied with a range of calcium concentrations (1.0 x 10-5-2.8 x 10-3 mol dm-9 in the presence of a constant level of barium (5.0 x mol dm-3). Results and Discussion Interference of calcium in the determination of barium by flame atomic-absorption spectro- photometry is confirmed by the data in Fig.1. The onset of calcium interference on the barium absorbance signal of a solution containing 5.0 x mol drn4 of barium becomes appreciably evident at calcium concentrations above about 1.0 x 10"' mol dm-3 (Fig. 1, A). The re- sponse of the instrument for barium is overshadowed completely and becomes linear for cal- cium when the calcium-ion concentration exceeds 1.6 x 10-8 mol dm-3 (Fig. l , B). The extraction power of PEG in comparison with 18-crown-6 is assessed according to the method of Takeda et d . , 6 in which the complexing carrier concentration is set at three times the barium level. However, the recovery of barium was less than 25% (Table I). The poor 18.0 16.0 14.0 (D -e s 2 12.0 0 0.5 1 .o 1.5 A 10.0 20.0 30.0 40.0 B Concentration of Ca*+/mol dm-3 x 1 V Fig.1. Effect of calcium additions on the atomic-absorption absorbance signal of a solution containing 6.0 x 10-6moldm-* of barium at a wavelength of 663.6nm using a dinitrogen oxide - acetylene flame.September, 1981 MATRICES BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY 941 TABLE I SOLVENT EXTRACTION OF BARIUM FROM A CALCIUM MATRIX BY SOLUTIONS OF PEG 1540 IN NITROBENZENE FOLLOWED BY ATOMIC-ABSORPTION SPECTROPHOTOMETRIC DETERMINATION OF BARIUM AND CALCIUM [PEG]in.* = 3[Baa+]in.t; [Ca'+]in,t = 5.0 X lO-'mol dm-3; Cpicric acid]^,. t = 1.2 x lo-* mol dm-3 (subscript in. = initial). The nitrobenzene phase containing the extracted barium and calcium was separated after the first shaking and then washed twice with equal volumes of picric acid (1.8 x lo-* mol dm-3).[Ba'+Iin. t I [Bas+] * / [Gas+]*/ Barium recovery, moldm-3 x lo-* moldm-a x lo-* moldm-3 x % 1.200 0.298 5 1.700 24.97 1.600 0.3155 1.500 19.72 2.800 0.458 7 2.250 16.38 3.900 0.6189 2.250 15.87 4.400 0.677 2 2.130 15.39 In organic solvent phase. t In aqueous phase. barium recovery is attributable to the relatively low extraction efficiency in comparison with 18-crown-6,9 which is claimed to be effective at three times the barium concentration.6 Increasing the PEG concentration to 6 and 30 times that of barium in the aqueous phase and carrying out subsequent washings with picric acid also gave poor barium recoveries (Table I1 , a and b). A single extraction step with PEG at a concentration 30 times greater than that of barium and in the presence of 1.8 x mol dm3 of picric acid gave a good recovery of barium, of almost of 100% (Table 11, d).It is obvious from Table 11, d, that calcium co-extracted into the nitrobenzene phase is below the interference level (Fig. 1) even at this high concentration of PEG. The failure of the subsequent washings with picric acid (Table 11, a and b) is thought to be due to the stripping of barium by the picric acid from the nitrobenzene phase. Thus, a single extraction into nitrobenzene without subsequent washing with picric acid is sufficient, then it is followed by extraction into 1 mol dm-3 nitric acid for the atomic-absorption stage. When the extraction into nitrobenzene was carried out in the presence of 1.2 x rnol dm-3 of picric acid , the recovery of barium was less than 100~o (Table I1 , c) .This may be due to the high level of calcium co-extracted with barium. The effect of acidity of the aqueous phase in the initial extraction was tested in the presence of nitric acid (2.0 x lW3 mol dm-3), when a recovery of less than 100% was obtained (Table 11, e). Such behaviour is attributable to the protonation of PEG. TABLE I1 OPTIMISATION OF CONDITIONS FOR QUANTITATIVE EXTRACTION OF BARIUM FROM A CALCIUM MATRIX WITH SOLUTIONS OF PEG 1540 IN NITROBENZENE FOLLOWED BY DETERMINATION OF BARIUM AND CALCIUM BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY [Ca'+Iln-* = 6.0 x lo-* mol dm-3 (subscript in. = initial). [PEClin. = ~[Ba'+Itn. [PEGlin. = 30[Ba'+Iin. r A t r A t at bt ct dt et ---- [Baa+]in.+/ [Gas+]$/ Barium [Gas+]$/ Barium [Caa+]$/ Barium [Ca8+]$/ Barium moldm-' moldm-' recovery, moldm-' recovery, moldm-' recovery, moldm-a recovery, moldm-8 recovery, x 10-4 x lo-' % x 10-6 % x lo-' yo x 10-6 yo x lo-' % 0.2000 2.940 33.00 2.250 100.00 7.500 102.00 6.380 100.00 5.708 98.00 0 .4 0 0 2.310 41.75 2.310 100.00 10.90 83.25 6.250 97.00 9.375 82-00 - __--_ . ~ _ . ~ _ . ~... . ~. - _.._. .._ . - 0.6000 1.940 26.67 2.400 76.67 8.790 68.00 5.750 100.00 7.500 77.67 0.8000 5.130 28.25 2.400 78.75 7.670 74.38 5.750 98.75 6.938 85.75 1.0000 1.380 26.20 2.690 26.90 8.130 69.90 6.880 101.8 6.750 86.00 In aqueous phase. t (a) Extraction conditions as in Table I. (b) Eptraction conditions as in Table I, but only a single washing with picric acid (1.8 x lo-' mol dm-') was carried out. (c) Single-step extraction in the presence of 1.2 x lo-' mol dm-8 picric acid.(d) Conditions as in (c), but 1.8 x 10-a rnol dm-a picric acid was used. (e) Conditions as in (c), but nitric acid (2.0 x 10-8 mol dm-a) was added to the aqueaus phase. $ In organic solvent phase.942 JABER AND EL-ISSA: BARIUM IN CALCIUM-CONTAINING Analyst, Vol. 106 The precision of the optimised method (derived from Table 11, d) was checked with an initial barium concentration of 2.0 x lW5 mol dm-3 in the presence of 5.0 x rnol dm-3 of calcium ions, and good recoveries were obtained (Table 111). TABLE I11 PRECISION AND ACCURACY OF THE OPTIMISED METHOD (TABLE 11, d) FOR EXTRACTION OF BARIUM FROM A CALCIUM MATRIX WITH PEG 1540 [PEGIin.* = 30[Baa+]1n.t; [picric acid1in.t = 1.8 X 10-amoldm-3; [Caa+]in.t = 6.0 X 10-3 mol drn4 (subscript in.= initial). [Baa +I I . t / [Bas+] */ moldm-3 x lo-& moldm-3 x lo-' 0.2000 0.1990 0.4000 0.3960 0.6000 0.6080 0.8000 0.7980 1.000 1.000 1.200 1.191 1.400 1.406 Standard deviation/ moldm-3 x 0.1126 1.128 2.894 0.269 9 1.960 1.619 1.023 Coefficient of variation, % 0.6909 2.848 4.760 0.326 7 1.960 1.276 0.727 6 Average barium recovery, yo 99.50 99.00 101.3 99.76 100.00 99.26 100.4 * In organic solvent phase. t In aqueous phase. Each of the values reported is the mean from eight extractions, each followed by atomic-absorption spectrophotometric determination. For comparison, barium extractions were carried out with 18-crown-6 using the optimised conditions of Table 11, d, but with the crown ether concentration three times greater than that of barium (the concentration used by Takeda et d6).The recovery of barium was not satisfactory (Table IV). Hence, under the conditions used, PEG is a more efficient extractant than 18-crown-6. The low efficiency of 18-crown-6 may be due to the high concentration of co-extracted calcium, which increases the barium signal. PEG is effective as a quantitative extractant with a single extraction step because of its high concentration in the organic phase and its low efficiency in binding calcium compared with that of 18-crown-6. The results indicate that the concentration of PEG or crown ether in the organic phase, the concentration of picric acid in the aqueous phase and the number of subsequent washings with picric acid are decisive factors governing the extraction of barium in the presence of a certain level of calcium.TABLE IV EFFICIENCY: EXTRACTION OF BARIUM FROM A CALCIUM MATRIX WITH PEG 1540 AND 18- CROWN-6 IN NITROBENZENE USING A SINGLE EXTRACTION [PEG]in.* = 30[Ba'+]in.t ; [18-crown-6]in.* = 3[Ba*+]in.t; lpicric acid]in:t. = 1.8 x lo-$ mol dm-3; [Caa+]i,.t = 6.0 x lo-' mol drn4 (subscript in. = imhal). PEG 1640 18-Cr0m-6 [Baa+] I n./t mol dm-3 0.200 0 0.400 0 0.6000 0.8000 1.000 1.200 1.400 x 10-4 [Baa+]*/ [ha+]*/ mol dm-3 mol dm4 x 10-4 x 10-6 0.2000 6.380 0.3880 6.260 0.6000 6.760 0.7900 6.760 1.010 6.880 - - - Barium recovery, % 100.0 97.00 100.00 98.75 101.0 - [Baa+] / mol drn4 0.1460 0.306 0 0.5100 0.7140 0.874 0 1.080 1.280 x 10-4 [Caa+] */ mol dm-3 6.600 9.670 x 10-5 10.10 13.30 16.30 16.80 17.20 7 Barium recovery, 73.00 76.60 86.00 89.26 87.40 90.0 91.43 % In organic solvent phase.t In aqueous phase.September, 1981 MATRICES BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY 943 Conclusion PEG can be used for the extraction of trace amounts of barium from a calcium matrix prior to atomic-absorption spectrophotometry. In the recommended procedure, an aqueous phase of barium with a calcium matrix is adjusted to be 1.8 x 10-2mol dm-3 in picric acid and shaken with an equal volume of nitrobenzene containing 30 times the expected barium concentration of PEG 1540. An aliquot of the organic layer is then shaken with aqueous 1 mol dm-3 nitric acid and the aqueous extract submitted to atomic-absorption spectrophotometry, The University of Jordan is thanked for a research studentship to M.Y.S. El-Issa. Dr. J. D. R. Thomas (UWIST, Cardiff, UK) is thanked for his constructive comments on the manuscript. 1. 2. 3. 4. 6. 6. 7. 8. 9. References Bano, F. J., Analyst, 1973, 98, 655. Cioni, R., Mazzucotelli, A., and Ottonello, G., Analyst, 1976, 101, 966. Frache, R., and Mazzucotelli, A., Taluntu, 1976, 23, 389. Rooney, R. C., and Woolley, J. F., Analyst, 1978, 103, 1100. Capacho-Delgado, L., and Sprague, S., At. Absorpt. Newsl., 1965, 4, 363. Takeda, Y., Suzuki, S., and Ohyagi, Y., Chem. Lett., 1978, 1377. Jaber, A. M. Y., Moody. G. J.. and Thomas, J. D. R., J . Inorg. Nucl. Chem., 1977. 39, 1689. Dale, J., and Kristiansen, P. O., Actu Chem. Scund., 1972, 26, 1471. Jaber, A. M. Y., and El-Issa, M. Y. S., unpublished data. Received February 2nd, 1981 Accepted April 6th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600939

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

944 Analyst, September, 1981, Vol. 106, pp. 944-948 Extraction - Spectrophotometric Determination of Tantalum(V) with 2-(2-Thiazolylazo)-5- dimethylaminophenol and 1,3=Diphenylguanidine Chikao Tsurumi Japan Department of Industrial Chemistry, Shibaura Institute of Technology, Shibaura, Minato-ku, Tokyo 108, Keiichi Furuya Department of Applied Chemistry, Faculty of Science, Science University of Tokyo, Kagurazaka, Shinjiku-ku, Tokyo 162, Japan and Hitoshi Kamada* Department of Industrial Chemistry, Faculty of Engineering, University of Tokyo, Hongo, Bunkyo-ku. Tokyo 1 13, Japan Tantalum(V) - 2-( 2-thiazolylazo)-5-dimethylaminophenol (TAM) chelate anion is extracted quantitatively into benzyl alcohol with 1,3-diphenylguani- dine (DPG) to form a ternary complex. In the organic phase, the complex has an absorption maximum a t 605nm.The optimum pH range for the extraction is 4.1-4.9 and the ternary complex is stable for at least 50 min. Beer’s law is obeyed over a concentration range of 2-18 pg of tantalum(V) in 10 ml of the organic phase. The molar absorptivity of the ternary complex is 4.1 x 104 1 mol-l cm-l at 605 nm. The composition of the ternary complex is considered to be (TaO,),(TAM),(DPG+). Keywords : Spectrophotometry ; tantalum( V ) determination ; 2-( 2-thiazoZyluzo)- 5-dimethylaminofihenol ; 1,3-diphenylguanidine The ternary complexes formed by the ionic association of metal chelate anions and large organic cations such as diphenylguanidine (DPG) and quaternary ammonium salts have been investi- gated by many workers.1-5 2- (2-Thiazolylazo ) -5-dime t h ylaminophenol (TAM) reacts with various metal cations to form coloured complexes. A survey of the literature revealed that water-soluble metal - TAM - DPG and quaternary ammonium ternary complexes in micelle solutions have been applied to the spectrophotometric determination of several ions.&lO A tantalum complex has been reported for the systems Pyrocatechol Violet (PV) - cetylpyridin- ium bromide (CPB)ll and 4-(2-pyridylazo)resorcinol(PAR) .12 Tantalum - TAM chelate anions are extracted quantitatively into benzyl alcohol with DPG to form the ternary com- plex. Both the ternary complex and the chelate anion have an absorption maximum at 605 nm and the molar absorptivity of the ternary complex is 4.1 x lo4 1 mol-l cm-l. Extrac- tion into benzyl alcohol increases the stability of the complex and the sensitivity of the complex in the presence of DPG is nearly 1.3 times that of the complex in the absence of DPG.In this paper, the fundamental conditions for the extraction of tantalum(V) and the composition of the ternary complex will be discussed. A number of ions that interfere with the determination were masked by the addition of various masking agents. The combination of this method and extraction using a high relative molecular mass amine is of importance industrially for the determination of tantalum in the presence of niobium in metal alloys. Experimental Reagents All of the chemicals were of the guaranteed-reagent quality. Standard tantahm( V ) solution. Tantalum (0.125 g) was dissolved in hydrofluoric acid (10 ml) in a platinum crucible, and heated with 2 ml of concentrated sulphuric acid on a hot- * Present address : Department of Applied Chemistry, Faculty of Engineering, University of Yamagata, Jonan, Yonezawa-shi, Yamagata 992, Japan.TSURUMI, FURUYA AND KAMADA 945 plate until fumes appeared. The solution was cooled and diluted to 250 ml with sulphuric acid (1 + 1).Working standard solutions were prepared by further dilution with 2% tartaric acid solution. TAM solution, 5 x 104 M. Prepared by dissolving 0.124 g of the Dotite TAM reagent in 1 1 of re-distilled methanol. DPG solzltion, 2 x 10-1 M. Prepared by dissolving 10.6 g of DPG in 12 ml of 6 M hydro- chloric acid and diluting to 250 ml with water. Bufler solution. About 1 x 10-1 M monochloroacetic acid and 2 x 10-1 M sodium acetate solution were used for pH adjustment.Apparatus A Shimadzu 210 automatic recording spectrophotometer was used with a 1-cm quartz cell. The pH was measured using a Hitachi - Horiba M-7 pH meter with a glass electrode. The solution was shaken in an Iwaki, Type KM, reciprocating shaker. Standard Procedure Up to 10 pg of tantalum were placed in a 100-ml beaker to which were added 2 ml of TAM solution and 3 ml of DPG solution. After adjusting the pH to 4.5 with the buffer solution, the solution was transferred into a 100-ml separating funnel and diluted to 25 ml with water. The solution was allowed to stand for 15 min then shaken with 10 ml of benzyl alcohol for 60 s. The aqueous phase was removed, then the organic phase was transferred into a beaker contain- ing about 1 g of anhydrous sodium sulphate and swirled to remove droplets of water.The absorbance of the clear organic phase was measured at 605 nm against a reagent blank. Results and Discussion Absorption Spectra together with that of the reagent blank. The absorption spectra of the complex extracted into benzyl alcohol are shown in Fig. 1, The complex has an absorption maximum at 605 nm. Effect of pH The effect of pH on the formation of the complex was examined usingvarious buffer solu- tions. There was little difference between the absorbance of the complex when measured in hydrochloric acid - sodium acetate or monochloroacetic acid - sodium acetate buffer solutions. Wavelengthhm Fig. 1. Absorption spectra of TAM and its Ta- TAM complex at pH 4.6.A: Ta, 1Opg; TAM (6 x ~O-'M), 2.0ml; DPG (2 x 10-l M), 3.0 ml; and reference, benzyl alcohol. B: Conditions as for A except that the reference is a reagent blank. C: Con- ditions as for A except that the Ta is omitted. nn n B w - 0.0 - I I 1 4.0 4.5 5.0 5.5 PH Fig. 2. Effect of pH on absorbance of organic phase a t 605nm. A: Ta, 1Opg; TAM (6 x 10-4 M), 2.0 ml; DPG (2 x lo-' M). 3.0 ml; and reference, reagent blank. B : Reagent blank versus solvent.946 TSURUMI et d. : EXTRACTION - SPECTROPHOTOMETRIC AutabSt, Vd. 106 However, the lowest absorbance for the reagent blank was obtained in monochloroacetic acid - sodium acetate solution. The pH of an aqueous solution was adjusted to various values with the monochloroacetic acid - sodium acetate buffer solution and the solution was extracted according to the method given under Standard Procedure. As shown in Fig.2, the absorb- ance remained constant over the pH range 4.14.9. Organic Solvents The extraction of the complex was tested by using benzyl alcohol, benzene, toluene, xylene, carbon tetrachloride , chloroform and isobutyl methyl ketone as organic solvents. However, with the exception of benzyl alcohol, only small amounts of the complex were extracted by these solvents. The complex can be extracted quantitatively into benzyl alcohol, and it is more sensitive and stable in this than in aqueous solution. Effect of TAM Concentration complex was examined. more than 1.2 ml of 5 x 1W M TAM solution per 10 pg of tantalum(V) are used.The effect of the concentration of TAM in the aqueous solution on the extraction of the It was found that the absorbance of the extract is constant when Effect of DPG Concentration The effect of the concentration of the DPG in benzyl alcohol on the extraction of the complex was examined. The largest and most constant absorbance values were obtained at concentra- tions greater than 2.7 ml of 2 x 10-1 M DPG solution. The absorbance of the complex in the presence of DPG was nearly 1.3 times that of the complex in the absence of DPG. Effect of Shaking Time and Colour Stability The complex was observed to be stable for 10-30 min in aqueous solution and it was allowed to stand for 15 min. The absorbance of the organic extract was constant for a shaking time from 30 s to 30 min and stable for at least 50 min at room temperature. Extraction Efficiency and Molar Absorptivity The efficiency of the extraction of the ternary complex by the benzyl alcohol was confirmed by the following experiment in which the tantalum was extracted into the organic phase quantitatively. Tantalum(V) (10 pg) was extracted according to the method given under Standard Procedure, then the amount of tantalum remaining in the aqueous phase was examined by adding a further portion of TAM solution, extracting with another 1Oml of the organic solvent and measuring the absorbance of the organic phase.There was no difference between the absorbance of the second organic phase and the reagent blank. The molar ab- sorptivity of the ternary complex at 605 nm is 4.1 x lo4 1 mol-l cm-l.The method is more sensitive and reproduceable than those employing the PV - CPB ternary complex or the PAR binary complex. Calibration Graph A calibration graph for the determination of tantalum(V) was prepared under the optimum conditions. A good linear relationship was obtained over the concentration range from 2 to 18 pg of tantalum(V) per 10 ml of benzyl alcohol, and Sandell's sensitivity for an absorbance value of 0.001 was found to be 4.41 x 1W3 pg cm--2. The reproducibility of the method was examined for solutions containing 10 pg of tantalum(V) and the coefficient of variation of the absorbance values was calculated to be 0.73% for five determinations. Composition of Ternary Complex The composition of the ternary complex was determined by both the continuous variation and the molar ratio method.As shown in Fig. 3, the molar ratio of TAM to tantalum in the ternary complex was found to be 3: 2. The same result was also obtained by the molar ratio method. The molar ratio of tantalum(V) - TAM chelate anion to DPG was found to be 1 : 1 by the molar ratio method, as shown in Fig. 4. According to these results, the composition of the ternary complex is considered to be (TaO,),(TAM) 3(DPG+).September, 1981 DETERMINATION OF TANTALUM(V) 947 0.4 0.3 Q -2 0.2 2 0.1 0.0 0.5 1 .o 0.5 I 1 0 1 2 3 [DPG j/[Ta1 Fig. 3. Continuous variation method Fig. 4. Molar ratio method applied to the applied to the absorbance of the organic absorbance of the organic phase at 605nm phase at 605 nm and pH 4.6: pa] + and pH 4.5.A: [Ta] + [TAM] = 5 x [TAM] = 4 x M; and DPG (2 x M ; reference, reagent blank. B : Reagent lo-' M), 3.0 ml. blank versus solvent. TABLE I INFLUENCE OF COEXISTING IONS ON THE DETERMINATION OF 10 pg OF TANTALUM(V) Al(II1) Ba(I1) Bi (111) Cd(I1) Ce(II1) Co(I1) Cr(V1) Fe (111) Hg(W La(II1) 21::; 2(!g) Mo(V1) Ni(I1) Pb(I1) Ti(1V) W V ) Th(Iv) V(V) WIV) Ion .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. Amount addedlpg 15 300 16 16 2 60. 3 20. 15 2 9 2 16 600 3 000 250 6 4 60 30 30 170 6 2 goo* 3 60* 1400 3 20* 5 loot SO# Form added Chloride Chloride Nitrate Nitrate Nitrate Nitrate Sulphate Sulphate Potassium dichromate Ammonium iron(II1) sulphate Ammonium iron(II1) sulphate Chloride Chloride Nitrate Nitrate Chloride Sulphate Ammonium molybdate Sulphate Nitrate Nitrate Nitrate Chloride Nitrate Sulphate Nitrate Nitrate Nitrate Nitrate Sodium tungstate Chloride Chloride Chloride Ta(V) foundlpg 9.8 10.1 9.9 10.2 10.1 9.9 10.4 10.0 10.2 10.4 9.8 9.6 10.1 9.6 9.7 9.6 10.0 10.0 11.0 10.6 10.0 9.6 9.6 10.0 10.6 9.7 9.9 10.4 10.0 9.9 9.6 10.0 10.7 Relative error, % -2.0 + 1.0 - 1.0 + 2.0 + 1.0 - 1.0 -+ 4.0 0.0 + 2.0 +4.0 -2.0 -4.0 + 1.0 - 4.0 - 3.0 - 4.0 0.0 0.0 + 10.0 + 6.0 0.0 -4.0 - 4.0 0.0 + 6.0 - 3.0 - 1.0 + 4.0 0.0 - 1.0 -4.0 0.0 + 7.0 * 1 ml of 6 x lo-' M EDTA was added.1 ml of 1 x 10-a M sodium citrate was added. $ 1 ml of 1 x lo-' M a-cyanoacetamide was added.948 TSURUMI, FURUYA AND KAMADA Effect of Coexisting Ions The effects of coexisting ions on the tantalum(V) determination are summarised in Table I.Alkali metals, alkaline earth metals, manganese(II), molybdenum(VI), tungsten(VI), borate, chloride, phosphate, sulphate and sulphite ions did not interfere. Up to six-fold amounts of cerium(III), two-fold amounts of cobalt(II), 1.5-fold amounts of lanthanum(III), six-fold amounts of vanadium(V), five-fold amounts of nickel(II), 90-fold amounts of thorium(1V) and two-fold amounts of yttrium(II1) were masked by adding 1 ml of 1 x M EDTA solution as a masking agent. Ten-fold amounts of iron(II1) and six-fold amounts of mercury(I1) could be masked by adding 1 ml of 1 X M a-cyanoaceta- mide solution, respectively. Niobium(V), titanium( IV) and zirconium( IV) interfere seriously in the determination of tantalum(V), causing positive errors.The interfering ions can be removed by extraction with a solution of hydrofluoric and hydrochloric acids in a high relative molecular m a s amine - benzene mixture,13 and tantalum(V) can then be back-extracted into the aqueous phase with 1 M ammonium fluoride - 4 M ammonium chloride solution. M sodium citrate and 1 d of 1 X 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References Tolmachev, V. N., Goltsberg, I. M.. and Konkin, V. D., Zh. And. Khim., 1967, 22, 950. Otomo, M., and Wakamatsu, Y., Nippon Kuguku Znsshi, 1968,89, 1087. Wakamatsu, Y., and Otomo, M., Bull. Chem. SOC. Jpn., 1972, 45, 2764. Ueda, K., Bull. Chem. SOC. Jpn., 1979, 52, 1215. Ishii, H., and Watanabe, H., Bunseki Kuguku, 1977, 26, 86. Tsurumi, C., Furuya, K., and Kamada, H., Bunseki Kuguku, 1979, 28, 754. Tsurumi, C., and Furuya, K., Nippon Kuguku Kuishi, 1975, 10, 1738. Tsurumi, C., Bunseki Kuguku, 1977, 26, 260. Tsurumi, C., and Furuya, K., Bunseki Kaguku, 1975, 24, 666. Tsurumi, C., and Furuya, K., Bunseki Kuguku, 1977, 26, 149. Nakashima, R., Sasaki, S., and Shibata, S., Bunseki Kuguku, 1973, 22, 729. Kajiyama, R., and Watanabe, M., Bunseki Kugaku, 1966, 15, 163. Furuya, K., Fujimura, K., and Tsurumi, C., Bunseki Kuguku, 1971, 20, 535. Received March 27th. 1980 Accepted April 9tk, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600944

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

AnaZyst, September, 1981, Vol. 106, pp. 949-954 Spectrophotometric and Titrimetric Determination of Catecholamines Using Organic Brominating Agents 949 A. Abou Ouf, M. I. Walash and F. B. Salem Faculty of Pharmacy, Mansoura University, Mansoura, Egypt Titrimetric and spectrophotometric methods for the determination of catechol- amines as pure substances and in their dosage forms were investigated and found to offer an improvement in ease, speed and accuracy. Both methods are based on the formation of adrenochromes and employ organic bromi- nating agents. Data obtained for several commercial samples are reported and compared with those obtained using official procedures. Keywovds : Catecholamine detevmination ; organic brominating agents ; sfiectvo- photometry ; titrimetry Pharmaceutical preparations containing catecholamines have been available for many years, and methods for their analysis as pure substances and in dosage forms include titrimetric,l spectropho tometric, polarographic, polarime tric,* amperome t r i ~ , ~ fluorime tric,* gas-chroma- tographic' and ion-pair high-performance liquid chromatographics procedures.The purposes of the present investigation were to develop a simple assay for catecholamines using organic brominating agents as titrants or chromogenic agents and to apply the pro- cedures to various dosage forms. Experimental Apparatus No. 529560 and platinum and calomel electrodes Nos. 527653 and 529571 were used. A Unicam SP 1800 spectrophotometer, a Unicam, Model 292, pH meter, glass electrode Reagents Buflk~ solution.Phosphate buffer of pH 7.4 was used; 500 ml of 0.1 M potassium dihydrogen orthophosphate solution and 390 ml of 0.1 N sodium hydroxide solution were mixed and diluted to lo00 ml with distilled water. Bromamine-T (BAT). This was prepared according to Nair et aL9 An approximately 0.02 M solution was prepared by dissolving 6.53 g of BAT in 1 1 of water, kept in amber- coloured bottles and standardised iodimetrically. Freshly recrystallised. A 0.02 M solution was prepared by dissolving 3.56 g of NBS in the minimum volume of water and diluting to 1 1 with water. The solution was freshly prepared before use and standardised iodimetrically.1° N-Bromophthalimide (NBP). Freshly recrystallised. A 0.02 M solution was prepared by dissolving 4.52 g of NBP in the minimum volume of glacial acetic acid and diluting to 1 1 with glacial acetic acid.The solution was freshly prepared before use and standardised iodi- met rically.ll The compound was synthesised according to Burry et aZ.12 and was recrystallised from hot water. A 0.02 M solution was prepared by dissolving 5.72 g of DBH in hot water, cooling and diluting to 1 1 with water. The solution was freshly prepared before use and standardised iodimetrically. Aqueous solutions (0.1%) of amaranth, methyl red, methyl orange and indigotine were prepared and stored in amber-coloured bottles. The catecholamines used were adrenaline and nora- drenaline as the bases and isopropylnoradrenaline as the hydrochloride. Their purities were established by the British Pharmacopoeia method.2 All compounds had a purity of not less than 98.5% of the active ingredients present.Dosage forms were analysed for these contents. N-Bromosuccinimide (NBS). Dibromohydantoin (DBH). Indicator solutions. Catecholamines and dosage forms.950 ABOU OUF et al. : SPECTROPHOTOMETRIC AND TITRIMETRIC Analyst, Vol. 106 Stoicheiometric Study The quantitative nature of the reaction between adrenaline, noradrenaline and isopropyl- noradrenaline and the brominating agents was checked by titrating a 0.002 M solution with the reagent solutions. The stoicheiometry was calculated to be 1 : 4 with BAT, NBS and NBP and 1 : 2 with DBH. In titrations without indicators, the molar ratio was 1 : 8 for BAT, NBS and NBP and 1 : 4 for DBH. Procedure A for Pure Drugs-Titrimetric Method An accurately measured volume of solution containing 0.2-30 mg of the catecholamines was transferred into a 100-rnl beaker, 10 ml of concentrated hydrochloric acid were added and the solution was diluted to 50 ml with distilled water.The solution was titrated with standard solutions of the brominating agents with the aid of a magnetic stirrer; the end-point was determined by using three drops of indicator solutions, and titration was carried out until the colour of the solution was discharged. The end-point can also be determined in 0.01 N hydrochloric acid without using an indi- cator, as, during addition of the titrant, the solution changed from colourless to a red colour that gradually disappeared on continuous addition of the brominating titrant and shaking.Alternatively, titration can be followed potentiometrically using a combination of platinum and calomel electrodes. Also, known volumes of solutions containing 0.2-30 mg of the catecholamines were added to a measured excess of the organic brominating solution in an iodine flask and the mixture was kept at room temperature with occasional shaking for 20 min. About 10 ml of 10% potassium iodide solution were added and the iodine liberated was titrated with 0.02 N sodium thio- sulphate solution using starch as indicator. A blank experiment was carried out under identical conditions but omitting the active ingredients. Procedure B for Pure Drugs-Spectrophotometric Method An accurately measured volume of solution equivalent to about 1 mg of the catecholamines was introduced into a 10-ml calibrated flask, 0.5-1ml of one of the brominating agents (0.02 M solution) was added, the volume was adjusted with phosphate buffer (pH 7.4) and the solution was left for 20 min.The absorbance of the solution was measured at 485 nm against a blank prepared in an iden- tical manner but omitting the addition of catecholamines. The concentration of catechola- mines was calculated by reference to a calibration graph prepared in the same manner using concentrations of 0.04-1.4 mg of adrenaline, 0.04-1.27 mg of noradrenaline and 0.06-1.9 mg of isopropylnoradrenaline hydrochloride. Procedure C for Dosage Forms Tablets Twenty tablets were weighed and powdered, and a sample equivalent to approximately 20 mg of the active ingredient was weighed accurately, dissolved in the minimum amount of water, filtered into a Wml calibrated flask and diluted to volume with water. An aliquot of this solution was analysed according to procedure A or 3 above. Injections and soladions accurately pipetted into a 50-ml calibrated flask and diluted to volume with water.aliquot of this solution was analysed according to procedure A or B above. An aliquot of the liquid preparation containing about 10 mg of the active ingredient was An Results and Discussion As BAT, NBS, NBP and DBH are oxidising agents, catecholamines are oxidised to the corresponding adrenochromes ; hence the equivalence point in the titration corresponds to the addition of 4 mol of BAT, NBS or NBP or 2 mol of DBH per mole of catecholamines. The titrations were carried out in aqueous hydrochloric acid solution, and during the titra- tions the colour of the solution increased to a maximum after 4 equivalents of titrant had been added, then decreased in intensity and disappeared entirely after 8 equivalents had been added.TABLE I RESULTS FOR TITRIMETRIC DETERMINATIONS OF ADRENALINE, NORADRENALINE AND ISOPROPYLNORADRENALINE Recoveries by the official method' are 99.384 f 0.072, 98.67 f 0.073 and 98.81 f 0.316% for adrenaline, noradrenaline and isopropylnoradrenaline hydrochloride, respectively.Each result is the mean of six experiments with confidence interval (95%). Recovery with BAT, % Recovery with DBH, % U Titration Titration M Titration with without Titration with without w Catecholamine Range/mg indicator indicator Potentiometric Back-titration indicator indicator Potentiometric Back-titration 5 Adrenaline .. . . 0.28-22 99.63 & 0.309 99.66 f 0.349 99.69 f 0.343 99.474 f 0.422 99.892 f 0.11 99.65 f 0.361 99.69 f 0.12 99.63 f 0.39 5 Noradrenaline . . 0.26-20 99.37 f 0.463 99.67 f 0.16 99.63 f 0.374 99.43 f 0.344 99.662 f 0.234 99.68 f 0.241 99.69 f 0.011 99.37 f 0.366 2 Isopropylnor- adrenaline.HC1 . . 0.37-30 99.64 f 0.271 99.69 f 0.177 99.58 f 0.288 99.63 f 0.246 99.7 f 0.343 99.71 f 0.194 99.69 f 0.34 99.64 f 0.229 8 0 L A I \ r ' 3 t* 1.4 1.4 1.1 0.373 8.064 0.79 3.03 0.646 t* 2.6 9.67 3.96 3.781 7.22 7.21 4.963 3.288 Z t* 6.043 6.462 4.38 6.46 4.342 7.36 3.843 5.14 0 Fi A L X Recovery with NBS, % Recovery with NBP, % I ! I ' ? Ti tra tion Titration Titration with without Titration with without Catecholamine Range/mg indicator indicator Potentiometric Back-titration indicator indicator Potentiometric Back-titration Adrenaline .. . . 0.28-22 99.872 f 0.1 99.47 f 0.64 99.262 f 0.64 99.62 f 0.23 98.62 f 1.183 98.67 f 1.177 98.92 f 0.36 98.99 f 0.46 3 t* 8.62 0.28 0.4 1.79 1.12 1.19 2.23 1.48 v, Noradrenaline . . 0.26-20 99.62 f 0.371 99.632 f 0.32 99.06 f 0.36 99.3 f 0.363 98.64 f 0.964 98.8 f 0.894 98.67 f 0.62 99.03 & 0.64 t* 4.4 4.69 1.93 2.98 0.06 0.26 0 0.97 Isopropylnor- adrenaline.HC1 . . 0.37-30 99.69 f 0.291 99.46 f 0.38 99.464 f 0.362 99.4 f 0.33 98.97 f 1.266 98.91 f 1.282 99.03 f 0.42 99.36 f 0.3 t* 6.27 2.3 3.1 3.0 0.22 0.134 0.9 3.02 * Tabulated t with seven degrees of freedom is 2.36 at the 0.06 significance level.TABLE I1 RESULTS FOR TITRIMETRIC DETERMINATION OF ADRENALINE , NORADRENALINE AND ISOPROPYLNORADRENALINE IN PHARMACEUTICAL PREPARATIONS Each result is the mean of six experiments with confidence interval (96%).Sample Adrenaline* . . Arterenolt . . Epifrint . . Prenasmas . . Aleudrinq . . Sample Adrenaline*. . Arterenolf . . Epifrint . . Prenasmas . . Aleudriny . . Catecholamine . . Adrenaline . . Noradrenaline .. Adrenaline . . Isopropylnor- adrenaline.- HC1 . . Isopropylnor- adrenaline. - HCI Catecholamine . . Adrenaline . . Noradrenaline .. Adrenaline . . Isopropylnor- adrenaline.- HC1 . . Isopropylnor- adrenaline. HCl Recovery with BAT, % L I \ Ti tration Titration with without Potentio- Back- indicator indicator metric titration 99.39 f 0.46 99.65 f 0.42 99.074 f 0.82 99.56 f 0.39 99.47 f 0.27 99.44 f 0.34 99.31 f 0.84 99.58 f 0.374 99.49 f 0.293 99.36 f 0.412 99.69 f 0.364 99.6 f 0.622 Recovery with DBH, % Ti tration L I 1 Titration with without Potentio- Back- indicator indicator metric titration 99.66 f 0.472 99.73 f 0.176 99.39 f 0.29 99.35 f 0.62 99.472 f 0.396 99.64 f 0.401 99.492 f 0.362 99.34 f 0.68 99.8 f 0.302 99.66 f 0.278 99.622 f 0.275 99.76 f 0.294 99.69 f 0.248 99.78 f 0.23 99.662 f 0.41 99.66 f 0.394 99.77 f 0.11 99.67 f 0.36 99.42 f 0.68 99.76 f 0.371 99.694 f 0.22 99.722 f 0.26 99.52 f 0.69 99.662 f 0.375 99.732 f 0.22 99.89 f 0.19 99.64 f 0.483 99.77 f 0.28 Recovery with NBS, % Recovery with NBP, % & 3 1 & * Titration Titration with without Potentio- indicator indicator metric 99.6 f 0.194 99.67 f 0.276 98.67 f 0.68 99.62 f 0.28 99.44 f 0.24 98.692 f 0.923 99.62 f 0.43 99.64 f 0.311 99.18 f 0.72 99.68 f 0.202 99.66 f 0.261 99.66 f 0.37 99.49 f 0.274 99.662 f 0.29 99.61 f 0.323 Ti tration Back- Titration with without Po tentio- Back- titration indicator indicator metric titration 99.61 f 0.213 98.42 f 0.96 98.9 f 0.23 98.3 f 1.127 98.59 f 1.067 99.73 f 0.21 98.49 f 0.922 98.37 f 1.042 98.14 f 0.633 98.37 f 1.462 99.69 f 0.39 98.64 f 0.78 99.0 f 0.46 98.74 f 0.791 98.68 f 1 99.62 f 0.61 99.13 f 0.866 98.984 f 1.043 98.77 f 0.69 98.84 f 0.71 99.63 f 0.431 99.09 f 0.78 98.34 f 1.427 98.76 f 0.641 98.65 f 1.056 * Adrenaline ampoules, labelled to contain 1 mg ml-’ of adrenaline (Misr, Egypt).t Arterenol ampoules, labelled to contain 1 mg ml-’ of noradrenaline (Hoechst Laboratories).3 Epifrin ophthalmic solution, labelled to contain 2.0% of adrenaline base (Allergan Pharmaceuticals). f Prenasma tablets, labelled to contain 20.0 mg of isoprenaline sul hate per tablet (Misr, E . 7 Aleudrin tablets, labelled to contain 20.0 mg of isoprenaline sulpIate per tablet (Boehringqt)September, 1981 DETERMINATION OF CATECHOLAMINES 953 It appears, therefore, that a simple procedure could be devised in which this self-indicating property of catecholamine derivatives might be utilised for their determination. Accordingly, a series of titrations were carried out in which pure catecholamine derivatives in 0.01 N hydrochloric acid were titrated to a colourless end-point with standard solutions of the bromin- ating agents. Typical results are shown in Table I.The titrations were also carried out using visual indicators (methyl red and methyl orange). These indicators give a very sharp change at the equivalence point. A potentiometric method for the determination of catecholamines using platinum and calomel electrodes was also used. The visual titration and potentiometric methods were also applied to pharmaceutical dosage forms of those catecholamines investigated for which the best results were obtained. The results of these titrations are given in Tables I and 11. The absorption spectra of buffered catecholamine derivatives showed absorption maxima at 280 nm. The addition of brominating agents to a solution of catecholamines converted the spectra to those of the adrenochromes, which exhibited absorption maxima at 300 and 485 nm.The absorption due to adrenochrome species produced during the reaction can be measured without interferences. Spectrophotometric measurements of these compounds at 485 nm could be made. The results are shown in Tables I11 and IV. TABLE I11 RESULTS FOR SPECTROPHOTOMETRIC DETERMINATION OF ADRENALINE, NORADRENALINE AND ISOPROPYLNORADRENALINE USING BAT, DBH, NBS AND NBP WITH PHOSPHATE BUFFER (PH 7.4) Recoveries by the official method2 are 99.384 f 0.072, 98.67 f 0.073 and 98.81 f 0.316% for adrenaline, noradrenaline and isopropylnoradrenaline hydrochloride, respectively. Each result is the mean of six experiments with confidence interval (95%). Recovery, % r 7 Catecholamine R w & w BAT DBH NBS NBP Adrenaline .. . . . . . . 0.046-1.4 98.72 f 0.51 98.83 f 0.51 98.61 f 0.514 98.2 f 0.48 Noradrenaline . . . . . . 0.043-1.27 98.52 f 0.58 98.52 f 0.48 98.58 f 0.8 98.102 f 0.53 Isopropylnoradrenaline.HC1 . . 0.062-1.9 98.95 f 0.66 99.05 f 0.472 99.01 f 0.463 98.45 f 0.373 t* 2.29 1.88 2.62 4.253 t* 0.448 0.542 0.198 1.86 t* 0.353 0.863 0.738 1.4 Tabulated t with seven degrees of freedom is 2.36 at the 0.05 significance level. TABLE IV RESULTS FOR SPECTROPHOTOMETRIC DETERMINATION OF ADRENALINE, NORADRENALINE AND ISOPROPYLNORADRENALINE IN PHARMACEUTICAL PREPARATIONS Each result is the mean of six experiments with confidence interval (95%). Recovery, yo I A -I Sample* Ca techolamine BAT DBH NBS NBP Official method' Adrenaline . . . . Adrenaline 98.68 f 0.704 98.78 f 0.922 98.62 f 0.764 98.31 f 0.689 99.384 f 0.144 Arterenol .. . . Noradrenaline 98.56 f 0.771 98.582 f 0.829 98.74 f 0.61 98.4 f 0.49 98.604 f 0.175 Epifrin . . . . Adrenaline 98.852 f 0.502 99.102 0.3 98.87 f 0.61 98.48 f 0.334 99.174 f 0.33 Prenasma . . . . 1sopropylnoradrenaline.- HCl 99.02 f 0.329 99.08 f 0.233 99.13 f 0.342 98.62 f 0.64 98.08 f 0.273 Aleudrin . . . . 1sopropylnoradrenaline.- HCI 99.01 f 0.329 99.09 f 0.281 98.99 f 0.239 98.66 f 0.77 98.25 f 0.25 Samples as in Table XI. Thin-layer chromatography on silica gel showed that the spot corresponding to the starting material disappeared completely and new spots appeared corresponding to the adrenochromes. The solvent system used was as reported by Desimio,13 vix., butan-1-01 - acetic acid - water (66 + 17 + 17).The R, values for the adrenochromes were 0.615, 0.65 and 0.72 for adrenaline, noradrenaline and isopropylnoradrenaline , respectively. The study of the stoicheiometry of the reactions revealed that the reaction betwen catechola- mines and brominating agents can be represented by the following scheme :954 ABOU OUF, WALASH AND SALEM / - 2HBr Conclusion As a result of this work, catecholamines can now be determined spectrophotometrically and titrimetrically using organic brominating agents. Each of the methods was applied success- fully to the determination of catecholamines as raw materials and in their dosage forms and the results, as shown in Tables I and 111, were precise and statistically valid. 1. 2. 3. 4. 5. 6. 7 . 8. 9. 10. 11. 12. 13. References “United States Pharmacopeia,” Nineteenth Revision, Mack Publishing Co., Easton, Pa., 1975, “British Pharmacopoeia 1973,” HM Stationery Office, London, 1973, pp. 257 and 304. Theory, I., Cantin, D., Alary, J., and Coeur, A., Analusis, 1973-74, 2, 654. “United States Pharmacopeia,” Eighteenth Revision, Mack Publishing Co., Easton, Pa., 1970, Crovetlo Montoya, G., Thomas Gomez, J., and Crovetlo Montoya. L., Ars Pharm., 1976, 17, 307. Prasad, V . K., Ricci, R. A., Nunning, B. C., and Granatek, A. P., J . Pharm. Sci., 1973, 62, 1130. Arnold, E. L., and Ford, R., Anal. Chem., 1973, 45, 85. Maurice, W., and Gustave, H., J . Chromatogr., 1978, 160, 297. Nair, C. G. R., Lalithakumari, R., and Senan, P. I., Talanta, 1978, 55, 625. Barakat, M. Z., and Abdel-Wahab, M . F., Anal. Chem., 1954, 26, 1973. Saadia, M. M., Master’s Thesis, Mansoura University, 1980. Burry, S., Kwit, D., and Jawarska, R., Polish Pat., 51-580, 1966; Chem. Abstr., 1967, 67, 90811a. Desimio, M., Boll. SOC. Ital. Farm. Osp., 1962, 8, 155; cited an Pharm. Acta Helv., 1963, 38, 383. pp. 274 and 320. pp. 227-228. Received March 16th, 1981 Accepted March 30th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600949

出版商:RSC    

年代:1981

数据来源: RSC