ISSN:0003-2654    

DOI:10.1039/AN98106FX021

出版商:RSC    

年代:1981

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98106BX023

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

A202 for further information. See page xvii

ISSN:0003-2654    

DOI:10.1039/AN98106FP069

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

xii SUMMARIES OF PAPERS I N THIS ISSUESimple and Rapid Gas- chromatographic Determination of TraceAmounts of Lower Aliphatic Carbonyl Compounds in ExhaustGases from Some Odour SourcesJune, 1981A simple and rapid gas - liquid - solid chromatographic method for thedetermination of trace amounts of the lower aliphatic carbonyl compounds(C,-C,) in free forms in the exhaust gases from some odour sources was devel-oped using the cold-trapping method with liquid argon. In the main analyticalcolumn the conditions were as follows : stationary phase, 1,2,3-tris-2-cyano-ethoxypropane (TCEP), 5% ; support, Carbopack B (60-80 mesh) ; glass,column, 1.5 m x 3 mm i.d.; column temperature, 7 0 "C; carrier gas, nitrogen,flow-rate 50 ml min-l.In the cold-trapping pre-column the conditions were as follows : stationaryphase, TCEP (25%) ; support, Shimalite (AW, DMCS) (60-80 mesh) ; glasscolumn, 31 cm x 4 mm i.d., operating.temperature for trapping, - 186 "C(liquid argon) ; operating temperature for injection of the condensed sampleinto the gas chromatograph, increased from - 186 "C to 120 "C in 3.5 minand maintained a t this temperature for 30 s. The peaks due to lower ali-phatic carbonyl compounds were identified by the disappearance methodusing a reaction pre-column of 2,4-dinitrophenylhydrazine (0.05 g) - ortho-phosphoric acid (0.1 ml) on glass beads.The method has been applied to the determination of lower aliphaticcarbonyl compounds in real samples (the exhaust gases from a poultry manuredryer, a metal paint dryer and a pig manure dryer).The volume of samplegas required is as low as about 0.4 1 and the method is sensitive [detectionlimit about 10 p.p.b. (parts per log)] and rapid (for the concentration andanalysis of one sample about 30 min are required). The coefficient of vari-ation is less than 10%. Propionaldehyde, acrolein and acetone (compoundseach with three carbon atoms) were completely separated within about 7 minwithout tailing.Keywords : Lower aliphatic carbonyls determination ; gas chromatography ;cold-trapping ; pre-column concentration ; odour sourcesYASUYUKI HOSHIKAAichi Environmental Research Centre, 7-6, Tsuji-machi, Kita-ku, Nagoya-shiAichi, 462, Japan.Analyst, 1981, 106, 686-694.Formulation of Analytical Procedures Involving FlameAtomic-absorption SpectrometryDifferent atomic-absorption spectrometers may vary significantly in sensi-tivity, precision and shape of the analytical calibration graph, even when theinstrument is properly adjusted.Proposals are made for the formulation ofanalytical procedures that take account of existing differences betweeninstruments and criteria are given for their optimum adjustment of theinstruments. The use of only the linear part of the calibration graph isdiscouraged in favour of the formulation of a useful concentration range thatis derived from the acceptable precision. Suggestions are given for the step-wise dilution of the samples.Keywords : Flame atomic-absorption spectrometry ; formulation of analyticalprocedures ; useful concentration rangeHANS P.J. VAN DALEN and LEO DE GALANLaboratory for Analytical Chemistry, University of Technology, Jaffalaan 9, 2826 BXDelft, The Netherlands.Analyst, 1981, 106, 695-701June, 1981 SUMMARIES OF PAPERS IN THIS ISSUEDetermination of Anions in Atmospheric Precipitation byIon ChromatographyThe semi-automated ion-chromatographic system studied permitted analysesof 60 samples per day for the three major anions found in atmosphericprecipitation, viz., sulphate (0.06-10.0 mg l-l), nitrate (0.022-2.00 mg 1-1 as(nitrogen) and chloride (0.028-1.50 mg I-I). The sensitivity was too low todetermine nitrite, orthophosphate, bromide and sulphite under the selectedoperating conditions ; however, the expected concentrations of these specieswere less than 0.1 mg I-'.Fixed point calibrations were invalid as cali-brations were non-linear. When the subject system was compared withalternative procedures, the resultant data for sulphate, nitrate and chlorideagreed to within 2%. Ion-chromatographic data for fluoride were un-acceptable, as shown by the inter-comparison with two reference procedures.Keywords : A tmospheric precipitation analysis ; ion chromatography ; sulphatedetermination ; nitrate determination ; chloride determinationJOAN CROWTHER and JENIFER McBRIDEOntario Ministry of the Environment, Laboratory Services Branch, Water QualitySection, Box 213, Rexdale, Ontario, M9W 5L1, Canada.Analyst, 1981, 106, 702-709.Determination of Indigo Carmine in Boiled Sweets andSimilar Confectionery ProductsShort PaperKeywords : Indigo Carmine ; high-performance liquid chromatography ; con-fectionery productsN.P. BOLEY, N. T. CROSBY, P. ROPER and L. SOMERSDepartment of Industry, Laboratory of the Government Chemist, Cornwall House,Stamford Street, London, SE1 9NQ.Analyst, 1981, 106, 710-713.Determination of Bromoxynil and Ioxynil in Technical Bromoxyniland Ioxynil Esters by High-performance Liquid Chromatography,Gas - Liquid Chromatography and TitrimetryShort Pape-vKeywords : Bromoxynil determination ; ioxynil determination ; bromoxyniloctanoate ; ioxynil octanoate ; higJi-performance liquid chromatographyD. S . FARRINGTON,D. A. GEORGE, C. J. WOOLLAM and G. J. BRATTONDepartment of Industry, Laboratory of the Government Chemist, Cornwall House,Stamford, SE1 9XQ.Analyst, 1981, 106, 713-717.xiixiv THE ANALYST June, 1981The Royal Society of Chemistry-PublicationsAnnual Reports on Analytical Atomic SpectroscopyVol.9Edited by J. B. Dawson and B. L. SharpThis series provides the practising analytical chemist and spectroscopist with ahandbook of current practice and recent advances in instruments and methodsfor the determination of elements in the form of comprehensive, critical annualreports.“This is a worthwhile series, providing a survey of a tremendous bulk of originalliterature.”-Analytica Chimica Acta reviewing VoJ. 5Brief Contents : Atomization and Excitation; Instrumentation; Methodology;Applications; New Books; Reviews; Meetings; References; Author Index;Subject IndexHardcover 357 pp 83” x 6”RSC Members f 22.000 851 86 727 8 f 34.00 ($94.25)Selected Annual Reviews on the Analytical SciencesVOl.4Edited by L. S. BarkThe fourth volume continues the policy adopted in previous volumes ofpresenting critical reviews of selected topics in modern analytical science. Eachof these reviews embodies the work considered pertinent by a practisingchemist over the four or five years up to 1974.Softcover 80pp 83 x 6” 0 85990 204 8 f 14.50 ( $39.25)RSC Members f7.50Hazards in the Chemical laboratory3rd EditionEdited by L. BretherickHazards in the Chemical Laboratory has become established as a vital handbookin all types of laboratory environment. This third edition contains chapters on:The Health and Safety a t Work etc.Act 1974; Reactive Chemical Hazards; SafetyPlanning and Management; Fire Protection; Health Care and First-Aid; Precautionsagainst Radiations; Hazardous Chemicals; Chemical Hazards and Toxicology.The section dealing with hazardous chemicals provides detailed information onthe properties, warning phases, injunctions, toxic effects, hazardous reactions,first-aid treatments, fire hazards and spillage disposal procedures for all commonlaboratory chemicals. Short notes are given on the hazardous properties andreactions of several hundred other less common chemicals.Protective PVC cover 578pp 0 851 86 41 9 8RSC Members f9.75fl5.00 ($39.50)RSC Members should send their orders to: The Membership Officer, The RoyalSociety of Chemistry, 30 Russell Square, London WC1 B 5DT.All other ordersshould be sent to: The Royal Society of Chemistry, Distribution Centre,Blackhorse Road, Letchworth, Herts, SG6 1 HNJune, I981 SUMMARIES OF PAPERS I N THIS ISSUEDetermination of Therapeutic Concentrations of Indoramin byLiquid Chromatography with Fluorimetric DetectionShort PaperKeywords : Indoramin determination ; high-pevformance liquid chrovnato-graphy ; fluorimetric detectionA. J. SWAISLANDWyeth Laboratories, Huntercombe Lane South, Taplow, Maidenhead, Berkshire,SL6 OPH.Analyst, 1981, 106, 717-719.4,5-Diamino-2,6- dimercaptopyrimidine as a SpectrophotometricReagent for the Determination of Selenium in Semiconductorsand Animal FeedsShort PaperKeywords : Selenium determination ; 4,5-diamino-2,6-di~ercaptopyrimidine ;semiconductovs ; animal feeds ; spectvophotovtzetvyA. IZQUIERDO, M.D. PRAT and L. ARAGONESDepartment of A4nalytical Chemistry, University of Barcelona, Barcelona- 14, Spain.Analyst, 1981, 106, 720-723.Selective Spectrophotometric Method for the Determinationof Uranium(V1)Shovt PapevKeywovds : Uranium( V I ) determination ; avsenazo I I I ; spectvophotovnetvy ;rock analysis ; diethylenetriavninepentaacetic acidB. V. KADAM, B. MAITI and R. M. SATHEBhabha Atomic Research Centre, Analytical Chemistry Division, Trombay,Bombay-400 085, India.Analyst, 1981, 106, 724-726.Spectrophotometric Determination of Diethylcarbamazine andCentperazine in UrineShort PaperKeywords : iWicrojilavicida1 dvugs ; diethylcarbavnazine ; centperazine ;spectrophotonzetry ; picvolonic acidS.K. BAVEJA and K. V. RANGA RAODepartment of Pharmaceutical Sciences, Panjab University, Chandigarh-160 014,India.Analyst, 1981, 106, 726-728.Xxvi THE ANALYST June , 1981~~ Analytical Proceedings-Following the development and expansion of the former Proceedings of theAnalytical Division of The Chemical Society to include a wide range of topics ofgeneral interest to analytical chemists, a change of title toANALYTICAL PROCEEDINGScame into effect in January 1980.Recent and forthcoming issues include the following:Lecture Summaries-2-3 page technical papers based on lectures presented at meetings of theRSC Analytical Division, describing research and development studies.Special Articles and Editorials-Safety, recent legislation, controversial topics, etc.Equipment News-Information on the latest equipment, instruments and products.Conferences, Meetings and Courses-Announcements of forthcoming meetings and courses of interest toanalysts.The regular list of recent analytical books and publications now includesmini-reviews.Correspondence-Letters to the Editor appear regularly.B i og rap h i es-Information on and biographies of medallists, award winners anddistinguished analytical chemists visiting the U K.Diary-Full details of all forthcoming meetings of the Analytical Division and itsRegions and Subject Groups are listed every month.Advertising-Advertisements are accepted in Analytical Proceedings: full advertisements,classified, situations vacant, etc., are published.Books-For information on subscriptions and advertising rates, please return the formbelow. From 1981, Analytical Proceedings can be purchased separately from TheAnalyst, Price f30 (UK), $70.50(USA), f31.50 (elsewhere)..................................................................................................................................................................................Analytical ProceedingsTo : The M a r ket i ng Department * /Advertise men t M a nag er*("delete as appropriate)The Royal Society of Chemistry, Burlington House, Piccadilly,London, W1V OBN, UK.0 Subscriptions to Analytical Proceedings.0 Advertisement rates.Please send me details of:Name: .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Address: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . June, 1981 SUMMARIES OF PAPERS I N THIS ISSUEDetermination of Caffeine Using Sodium N-Chloro-p-toluene-sulphonamideShort PaperKeywords : Caffeine determination ; chloramine-T ; oxidation ; pharmaceuticalproducts ; back-titrationS . M. MAYANNA and B. JAYARAMDepartment of Chemistry, Central College, Bangalore University, Bangalore-560 001,India.Analyst, 1981, 106, 729-732.xvi

ISSN:0003-2654    

DOI:10.1039/AN98106BP077

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

JUNE 1981 The Analyst Vol. 106 No. 1263 Measurement of Heavy Water Concentration with a Density Meter Yoshikazu Inoue, Kiriko Tanaka and Yosihiko Kasida National Institute of Radiological Sciences, Division of Environmental Health, 9-1, 4-chorne, A nagawa, Chiba-shi, Japan 260 A densitometric technique involving the use of a high-sensitivity density meter is described for the determination of a wide range of deuterium con- trations in water. The effects of temperature and solute impurities on the method were determined and the purification of water by sublimation and anion-exchange techniques were examined as a means of overcoming the latter. Standard heavy water samples of various concentrations were analysed by this densitometric method and by mass spectrometry, and good agreement between the results was found.The application of the proposed method to the determination of the electrolytic separation factor, a, for deuterium - hydrogen is described. Keywords : Heavy water determination ; density meter ; water purification ; tritium electrolytic enrichment The determination of deuterium in water over a wide range of concentrations is of continuing interest to both laboratories and industries such as those with heavy water moderated reactors. Several methods have been proposed for the determination of the isotopic composition of a mixture of D,O and H,O such as mass spectrometry, infrared spectroscopy, neutron activation, neutron t hermalisationl and densit ometry . Although conventional methods such as the pyknometric, drop and float methods are still used on account of their simplicity and cheapness, they require experienced workers and are time consuming, especially when applied to the determination of low deuterium concentrations.The aim of this work was to ascertain the applicability of a high-sensitivity density meter for the determination of the deuterium content of water, the determination of the separation factor, a, in the electrolytic enrichment process for deuterium and to remove the drawbacks associated with conventional methods of density measurement. Principle of Density Meter and Calculation of Deuterium Content The high-sensitivity density meter measures density electronically, making the determina- tion both accurate and rapid. A hollow, glass, bending oscillator is excited into undamped oscillation and, when the system is filled with sample, the resonance frequency of the oscillator is influenced by the mass, and therefore by the density, of the sample.The volume V taking part in the oscillation is defined by the two fixed oscillation points, which are indicated in Fig. 1 . The period of a system with mass M and spring constant C of the hollow body, and density d of the sample inserted into the system, is Excitation Alcohol to thermostat Alco h o I from thermostat Y Phototransistor Fixed oscillation points Fig. 1. Illustration of an oscillator. Sample out - Thermistor Sample in in 609610 INOUE et al. : MEASUREMENT OF HEAVY WATER Analyst, “01. 106 T = 2 r d ( d V + M)/C * .. .. - - (1) Taking the square and substituting A = 4r2V/C and B = 4r2M/C, we obtain ..- * (2) P = A d + B .. .. .. The difference in the densities of two samples is where K = 1/A. The apparatus constant, K , which is dependent on temperature, can be determined by two calibration measurements with samples of known density. The density of unknown samples, ds, will be determined by substituting the measured period Ts into the equation ds = K (Ts2 - 7‘1,) + d , . . .. .. * * (4) In this study, every sample reached thermal equilibrium within 10 min and was counted for about 77.7 s until its oscillation reached the pre-set number of 4 x lo4. The measure- ment of this time elapsed, T , which is sometimes called “period” hereafter, was repeated five times and the mean value was used for the density calculation according to equation (4).There is a relationship between the molar fraction of deuterium, N,, and density, d, or specific gravity, S , of the sample2 : where A S = S - 1 .. .. .. .. . . d .. .. .. d s = - = dL dN - 0.0000160 F = .. . . . . .. dL = dN - 0.0000160 . . .. .. .. . . M , = relative molecular mass of light water (lH,O) (18.015 050) ; M , = relative molecular mass of heavy water (D,O) (20.027 604) ; d , = density of heavy water (g d-l) ; d, = density of natural water (g ml-l) (0.015 mol-% of deuterium); and d, = density of light water (g ml-l). Densities of natural water (H,O) and heavy water (D20), dN and d,, as a function of temperature were calculated by the equation proposed by Kell,3 and are given in Tables I and 11, respectively. The molar fraction of deuterium in water can be calculated from the measured density of water according to equation (5), and vice versa.Experimental Apparatus Meter, Model 02D. Density measurements were carried out with an Anton Paar Digital Precision Density In order to avoid thermal disturbances from the surroundings, theJune, 1981 CONCENTRATION WITH A DENSITY METER Tempera- ture/' C 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 TABLE I DENSITIES OF NATURAL WATER (g ml-l) AGAINST A SERIES OF TEMPERATURES I N THE NORMAL WORKING RANGE Next dt xima1 place in temperature value 61.1 0 0.9999924 0.9999939 0.999 995 3 0.9999965 0.9999975 0.999 998 3 0.999 999 0 0.9999996 1.000000 0 3 .oooooo 2 1.000000 2 1 .ooo 000 1 0.999 999 9 0.9999994 0.999 998 9 0.999 998 1 0.999 997 2 0.9999962 0,9999949 0.9999936 1 0.999 992 6 0,9999941 0.9999954 0.999 996 6 0,999995 6 0.999 9984 0.999 999 1 0.9999996 1 .ooo 000 0 1.0000002 1.0000003 1 .ooo 000 1 0.999 999 8 0.9999994 0.999 998 8 0.9999980 0.9999951 0,9999960 0.999 994 8 0.9999934 2 0.9999927 0.9999942 0.999995 5 0.9999967 0.999 997 7 0.9999985 0.999 999 2 0.9999997 1.0000000 1.0000002 1 .ooo 000 2 1.0000001 0.999 999 8 0.999 999 3 0.999 998 5 0.9999959 0,9999950 0.999 995 9 0.999 994 5 0.9999933 3 0.999 992 9 0.9999943 0.999 995 6 0.999 996 8 0.999 997 8 0.999 998 6 0.9999992 0.9999995 1.0000000 1.0000002 1.000000 2 1.0000001 0.9999997 0.999999 3 0.999 9986 0.999 997 9 0.999 996 9 0.9999958 0.999 9946 0.9999931 4 0.9999930 0.9999945 0.9999958 0.999 996 9 0,9999979 0.999 998 6 0,9999993 0.999 9998 1.0000001 1.0000002 1.000 000 2 1.000 000 0 0.999 9997 0.999 999 2 0.999998 6 0.999 997 8 0.9999968 0.999 995 7 0.9999944 0.9999930 5 0.9999932 0.999 9946 0.999 9959 0.9999970 0.999 997 9 0.999 998 7 0.999 999 3 0.9999998 1.0000001 1 .ooo 000 2 1 .ooo 000 2 1.000 000 0 0.999 999 5 0.999 999 2 0.999 998 5 0.9999977 0.999996 7 0.9999956 0.9999943 0.999 9928 6 0.999 993 3 0.9999948 0.999 9960 0.999 995 1 0.999 9980 0.999 998 8 0.999 999 4 0,9999998 1 .ooo 000 1 1.000 000 2 1.0000002 1.000 0000 0.999 9996 0,9999991 0.999 9984 0.9999976 0.999 996 6 0.999 9955 0.9999941 0.999 9927 7 0.9999935 0.999 9949 0.999 996 1 0.9999972 0.999 9981 0.9999989 0.999 999 4 0.9999999 1.000 000 1 1 .ooo 000 2 1 .ooo 000 2 1.000 000 0 0.999 9996 0,9999990 0.999 998 3 0,9999975 0.999 996 5 0.999 995 3 0.999 9940 0.999 992 5 8 0.9999936 0.9999950 0.999 9962 0.9999973 0.9999982 0.999 9989 0.999 999 5 9.999 9999 1.0000001 1.0000002 1 .ooo 000 2 0.9999999 0.999 999 5 0,9999990 0.999 998 3 0.9999974 0.999 9964 0.999 995 2 0.999 993 9 0.999 992 4 9 0.9999938 0.999 995 1 0.999 996 4 0.999 9974 0.9999983 0.999 999 0 0.999 999 5 0.9999999 1 .ooo 000 2 1.000 000 2 1.0000001 0.999 9999 0.999 9995 0.9999989 0.999 998 2 0.999 997 3 0.999 996 3 0.999 995 1 0.999 9937 0.999 9922 component of the oscillator was separated from the density meter and installed in a container with polystyrene foam (heat insulator) walls.The temperature of the oscillator was kept constant by circulating thermostatically controlled ethanol between the centre and inner wall of the oscillator cylinder by means of a Haake, Model FK, circulator with a proportional control with zero voltage switching, and a platinum resistance temperature sensor instead of a contact thermometer.The temperature-measuring system consisted of a thermistor, whose electrical resistance was measured by a Hewlett-Packard multimeter, Model 3465A, and a reference Beckmann thermometer (reading to 0.01 "C) calibrated using the triple point of pure water (+O.OlO "C). The variation in the temperature of the sample water with time was monitored by recording the output of the thermistor on a chart recorder through a Wheatstone bridge circuit and a micro voltmeter (Ohkura Electric, Model AM-1001). The temperature was adjusted to 3.98 The apparatus was arranged as shown in Fig.2. In order to check the purity of the water, its electrical conductivity was measured in a Pyrex glass cell (3-ml volume, cell constant 0.26 at 17.2 "C) using platinum electrodes connected to an electrical conductivity meter (Yanagimoto, Model MY-8). The deuterium content of the water was determined by means of a deuterium mass spectrometer (Hitachi, modified Model RMS-4) after conversion of the water to hydrogen on heated uranium metal. The radioactivity of the tritium was measured by means of a 0.006 "C throughout the experiments. TABLE I1 DENSITIES OF HEAVY WATER (g ml-l) AGAINST A SERIES OF TEMPERATURES I N THE NORMAL WORKING RANGE Next decimal point in temperature value Tempera- c 1 turelo C 0 1 2 7 4 5 6 7 8 9 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 1.105 354 6 1,105371 8 1.1 05 38 8 5 1.105 405 4 1.103421 9 1.1054381 1.1054541 1.105469 9 1.105 485 4 1.105 500 5 1.10.5 515 7 1.105530.3 1.1055451 1.105 5.394 1.1055735 1.105 5874 1.105 601 0 1 .105 6144 1 .lo5 62 7 6 1.105 640 6 1.105 356 3 1.105 373 5 1.1053904 1.1054071 1.105423 5 1 .lo5 439 7 1.1054555 1.1054514 1.105486 9 7 .lo5 502 2 1.105 515 2 1.1055320 1.10.3 546 5 1.105 560 8 1.105 5749 1.1055888 1.1056024 1.105 6 15 8 1.1056289 1.105 641 9 1.105358 0 i i o 5 3 7 5 i 1.105 392 1 1.1054087 1.1034252 1.105441 3 1.1054573 1.105473 0 1.105488 5 1.105 503 7 1.1055185 1.105 5334 1.1056480 1.1055622 1.105556 3 1.105 590 1 1.1036037 1.105 617 1 1.105 630 2 1.1056431 1.105 359 8 1.105 3'76 9 1.105 393 8 1.1054104 1.1054268 1.105442 9 1.1054589 1.105 474 5 1.105 490 0 1.1055052 1.1035202 1.1 I 6.3 3 4 9 1.105 5494 1.1055635 1.1055577 1.10.5591 5 1.103 605 1 1.105 618 4 1.105631 5 1.1056444 1.1053615 1.1053586 1.1053954 1.105412 0 1.1054284 1.1054445 1.1054604 1.1054761 1.1054915 1.105 506 7 1.1055216 1.1055364 1.1055508 1.105 565 1 1.105 579 1 1.105592 9 1.105 606 4 1.105619 7 1.1056328 1.1056457 1.105 3632 1.1053649 1.105 3666 1.1053803 1.1053820 1.1053835 1.1053971 1.1053988 1.1054004 1.105413 7 1.105415 3 1.105 4170 1.1054300 1.105 446 1 1.1054620 1.1054776 1.105493 0 1.105 508 2 1.1055231 1.105537 8 1.105552 3 1.1055665 1.105 5805 1.1055942 1.109 6078 1.105621 1 1.105 634 1 1.1036450 1.1053684 1.1053854 1.1054021 1.1054186 1.105484 9 1.105 4509 1 .105 466 7 1.105 5566 1 .lo5 -5 50 7 1.10.3.584 6 1.105 598 a 1.1056118 1.1056250 1.105 638 0 1.105650X 1.1053701 1.105 3870 1.1054038 1.105420 3 1.1054315 1.10.54333 1.105436 5 1.105447 7 1.105449 3 1.1054525 1.1054636 1.1054652 .. ~ ~ . . 1.105468 3 1.1054792 1.105480 7 1.105482 3 1.1054838 1.105 4946 1.105 496 1 1.195 497 6 1.105499 1 1.105 509 5 1.105 511 2 1.105 512 7 1 . l o 5 514 2 1.1055246 1.1055261 1.10552'76 1.1055290 1.105 5x1 3 1.105541) 5 1.105 5422 1.1055436 1.105 565!l 1 .lo9 969 3 1.105 552 1 1.105553 7 1,10555.51 1.1055580 l.lO55X10 1.10.55832 1 . m 586 o 1.1055956 1.1055950 1.103'5997 1.1036091 1.1056104 1.105 6131 1.105 6224 1.105623 5 1.1056263 1.10563.54 1.1056367 1.1056393 1.1056483 1.1056495 1.1056521612 Density meter A v Printer L r INOUE et d.: MEASUREMENT OF HEAVY WATER Analyst, V d . 106 Constant-temperature ci rcu I ator Multimeter liquid scintillation counter (Japan Radiation and Medical Electronics, Model Aloka 600 LB) . The apparatus used for the purification of the water by a sublimation technique consisted of 10 sublimation flask units connected to a vacuum line. The sublimation flask unit, shown in Fig. 3, was made of Pyrex glass with Viton O-ring connections and a PTFE tap. In order to avoid contamination by impurities that would affect the density, a long PTFE tube connected to the tip of a syringe was used throughout the experiments to transfer or inject the purified water into the electrical conductivity cell, density meter or sublimation-flasks. Unless stated otherwise, all glassware was made of Pyrex 7740 glass.PTFE containers, purchased from Tokyo Material Ltd., were used for storing purified water. PV Wheatstone Reagents Purified water used for washing glassware and as the density standard was prepared by passing distilled water through a Millipore Milli-Q water purification system. All chemicals were of analytical-reagent grade. Preparation of Standard Heavy Water and Determination of Deuterium Content Heavy water (99.7 mol-%), which was purchased from Aldrich Chemical Co., was purified by repeated sublimation. Seven working standards of heavy water whose concentrations ranged from 0.015 to 1 .O mol-yo were prepared gravimetrically by diluting the purified Vacuum e PTFE tap , Viton O-ring 1 cm diameter Pyrex glass Fig. 3. Sublimation flasks unit.June, 1981 CONCENTRATION WITH A DENSITY METER 613 99.7 mol-o/, heavy water with purified natural water.The concentration of deuterium in these standards was determined by both mass spectrometry and the densitometric method described here. PTFE containers were used in the preparation process and for the storage of the standards. Determination of Apparatus Constant, K calibration (standard) samples. tube of the density meter and the temperature was allowed to equilibrate (10 min). the periods of the sample were measured at three points around 4 "C. washed and refilled with another standard sample of 1 mol-yo heavy water. were measured a t the same temperature in the same way. calculated at each temperature and the dependence of K on temperature was determined Natural water (0.0150 mol-%) and heavy water (1.014 mol-%) were used as the density About 1 ml of purified water was placed in the sample Then The sample tube was The periods The apparatus constant, K , was Measurement of Electrical Conductivity of Water Containing Impurities Trace amounts of salts and carbon dioxide may still remain in the water after purification.Therefore, the influence of these impurities on the density of the water was examined by addition tests. Sodium chloride was selected as a representative salt, and was added to both purified natural water and purified 0.101 mol-yo heavy water. The electrical con- ductivities and densities of the sodium chloride solutions were measured. Carbon dioxide was also dissolved in the water samples and the relationship between electrical conductivity and density was determined in the same way.Determination of Variation in Deuterium Concentration Accompanied by an Anion- exchange Process Variations in heavy water concentration were checked for heavy water samples in the range 0.05-0.5 mol-yo before and after passing them through an anion-exchange resin. The deuterium concentration was determined by mass spectrometry. Determination of Deuterium Content of Electrolytically Enriched Natural Water The electrolysis method for the enrichment of hydrogen isotopes was based on literature proced~res.~~5 Electrolysis of tap water, containing known amounts of deuterium and tritium, was performed in sodium hydroxide solution using nickel - nickel or nickel - iron electrodes in a cooling bath.Enriched samples were neutralised with carbon dioxide, and purified by repeated sublimation and anion exchange. After checking the purity by measuring the electrical conductivity, the density of the water was measured using the density meter. Purification of Water by Sublimation and Anion Exchange Samples examined in the purification test contained 1.5 g of sodium peroxide in 10 ml of distilled water. This composition closely resembled that of the residual solution of the electrolytic enrichment of hydrogen isotopes in natural water. Purification procedures for the samples are described below and the time required for each step is shown in parentheses. In order to decompose any organic substances that might be present, the solution was mixed with 2 g of potassium permanganate in a flask and refluxed (1 h).After neutralisation with carbon dioxide,, the solution was distilled under atmospheric pressure (1 h). The distillate was transferred into a sublimation flask, attached to the sublimation unit with another empty flask and frozen by means of liquid nitrogen. The sublimation unit was evacuated using a rotary pump, then the PTFE tap was closed and the empty flask was cooled with liquid nitrogen so that all of the ice sublimed into it (4 h). After melting, the water was passed through a column (diameter 1 cm, resin length 3 cm) packed with dry anion-exchange resin (15 min) The anion-exchange resin [Dowex 1-X8 (100-200 mesh)] had been con- ditioned previously by passing 2 N hydrochloric acid and 2 N sodium hydroxide solution through it three times, washing with purified water several times and then drying at a low temperature.In order to confirm the effect of sublimation and anion exchange on the purification of the water, a Finally, the de-ionised water was re-purified by sublimation (4 h).614 INOUE et al.: MEASUREMENT OF HEAVY WATER Analyst, VoZ. 106 series of steps consisting first in sublimation then anion exchange and then a second sub- limation, were carried out, the individual steps being repeated five times, three times and once, respectively and the electrical conductivity of the purified water was measured in each step. Results and Discussion Deuterium Content of Working Standard Heavy Water The deuterium content of working standards was determined by mass spectrometry and densitometry and the results obtained from both of the methods are shown in Table 111.The results obtained using densitometry agreed well with those from mass spectrometry within a relative error of 3% or less for deuterium concentrations above 0.1 mol-%. This good agreement resulted from the high purity of the samples, which had contained only trace impurities originally. This suggests that the densitometric method is applicable to the determination of a wide range of deuterium contents in purity-controlled water, such as heavy water used in a heavy-water-moderated reactor. TABLE I11 COMPARISON BETWEEN RESULTS OBTAINED BY DENSITOMETRY AND MASS SPECTROMETRY Densitometry ( A ) , mol-% . . . . 0.0207 0.0456 0.0995 0.200 0.304 0.506 1.012 Mass spectrometry ( B ) , mol- yo .. 0.0197 0.0499 0.101 0.202 0.303 0.504 1.014 Error [I00 ( A - B)/B], %. . . . +5.2 -8.7 -1.4 -0.5 +0.6 f0.3 -0.2 Dependence of Apparatus Constant, K, on Temperature The temperature dependence of the “difference of square of periods” and the resulting apparatus constant, K , is shown in Fig. 4. The result shows a good linear relationship between the apparatus constant, K , and temperature in the 4.0 “C temperature region. The apparatus constant, K , is temperature dependent owing to the temperature dependence 3.27 N I 0 X c N c-” I K ui N m .- L 3.26 rc 2 s: 8 2 E n 3 \c 3.25 : 3.80 3.90 4.00 4.10 TemperaturePC 0 I 0, k- X *- m 3.25 8 v) 3 * E P 9 3.24 Fig. 4. Dependence of (A) periods and (B) apparatus constant on temperature.June, 1981 CONCENTRATION WITH A DENSITY METER 615 of both the modulus of elasticity and the thermal coefficient of expansion of the measuring system.In addition, over a long period of time, ageing of the oscillator may change the value of the constant slightly. Therefore, it is necessary to re-plot the graph of K as a function of temperature from time to time for maximum precision. 1 .OOO 8 1 .OOO 6 r > 4- .- ! 1.0004 .- 'c 0) Q v) 1 .ooo 2 0.001 0.01 0.1 Specific resistance/M& cm 1 Fig. 5. Relationship between specific gravity and specific resistance of natural water containing sodium chloride. Influence of Impurities on Densitometry The relationship between density and water purity is shown in Figs. 5 and 6 for natural water and 0.1 mol-% heavy water containing sodium chloride in the concentration range 1000-0.001 p.p.m., calculated from dilution data.Specific gravity as defined in equation (7) is given instead of density. The sample purity is represented in units of specific resistance (MQ cm) which is the reciprocal of electrical conductivity (pS cm-l). The specific gravity of natural water and 0.1 mol-% heavy water containing sodium chloride decreased rapidly I 1 1.0008 >. .+z 1.0006 z L [J, 0 != .- x 1.0004 v, 1 .ooo 2 1 .OOO 117 1 .ooo 1 0.001 0.01 0.1 1 Specific resistance/MQ cm Fig. 6. Relationship between specific gravity and specific resistance of 0.1 mol-% heavy water containing sodium chloride.616 INOUE et al. : MEASUREMENT OF HEAVY WATER Analyst, voj. 106 with a decrease in salt concentration from 1000 to 10 p.p.m., which corresponded to a specific resistance of about 0.5 to 50 kQ cm, respectively.Thereafter, it decreased gradually and approached a value of 1.000016, the specific gravity of pure natural water, and 1.000117, the specific gravity of 0.101 mol-% pure heavy water in the specific resistance region greater than 0.5 MQ cm, which corresponds to a sodium chloride concentration of 1 p.p.m. or less in each instance. The same tendency was observed for natural water containing carbon dioxide, as shown in Fig. 7, suggesting a smaller effect of carbon dioxide, as compared with sodium chloride, on the density. 1.0002 > > 4- .- 5 1.0001 0 Y- Q) P v) .- 1 .OOO 01 6 1 .ooo 0 t 0.02 0.05 0.1 0.5 1 Specific resistanceIMR cm Fig.7. Relationship between specific gravity and specific resistance of natural water containing carbon dioxide. Determination of Error of Densitometry Temperature affects substantially both the density of water and the stability of the measuring system. However, variation of the density due to temperature drift need not be considered as, under the operating conditions used, the temperature remained constant at 3.98 -+ 0.006 "C. Instrumental error was estimated from the experimental data relating to the small variation of density both for natural water and 0.1 mol-% heavy water. Impurities in the water sample also led to a positive error in the determination of density. This error was estimated from the data shown in Figs. 5-7. For instance, from the data in Fig.5, the difference in specific gravity between pure natural water and natural water containing sodium chloride (less than about 1 p.p.m. or of specific resistance above approximately 0.3 Mi2 cm) was about 7 x which corresponded to 0.007 mol-74 deuterium or a relative error of 46% in the former, because a density difference of 0.000016 g ml-l corresponds to a difference in deuterium content of 0.015 rnol-%. Errors estimated in this way are summarised in Table IV. For natural water whose purity was 0.7 MR cm, relative errors due to the instrument and the purity were about &9.3 and +ZO%, respectively. For 0.1 mol-% heavy water, these errors were k1.4 and +2.8%, respectively. Consequently, even if much TABLE IV SUMMARY OF ERRORS Error r A \ Purification Instrumental A -I (f0.006 "C) 0.3 MR cm 0.7 Mi2 cm Densitylg ml-' .. ... . . . . f 1.4 x 10-6 $7 x 10-6 +3 x 10-6 Natural water (0.015 mol-yo), % . . f9.3 + 47 4- 20 Heavy water (0.1 mol-yo), yo . . . . k1.4 + 6.6 + 2.8J u n e , 1981 CONCENTRATION WITH A DENSITY METER 617 care is taken in the purification of the water, the total relative error will not be less than 10% for the determination of heavy water at levels below 0.1 mol-%. Therefore, the method described here is suitable for the accurate determination of deuterium above 0.1 mol-%. Recently, another type of high-sensitivity density meter, which was equipped with a pair of oscillators, one for the reference (standard) and the other for the sample, was reported.6 Simultaneous operation of the two oscillators compensated for the influence of temperature drift on the density and on the stability of the measuring system.It has been confirmed experimentally that the error in the measurement of the density of water decreased from 6 x g ml-l when a single oscillator was used to 0.3 x g ml-l when using the pair of oscillators for a temperature drift of 0.01 "C. A relationship between electrical con- ductivity and water purity, which was represented by the difference in the density of various purified waters from that of the standard pure water, was also described, and it was indi- cated that purified natural water with an electrical conductivity of 0.5 AIL! cm causes a density difference of 5 x Effect of Sublimation and Anion Exchange on Purification of Water The results of the purification of water using sublimation and anion-exchange procedures are shown in Fig.8. Repeating the sublimation five times did not increase the specific resistance by more than 0.6 Ill2 em. Purification using an anion-exchange resin lowered the specific resistance, but a further sublimation step raised it to 0.7 M a cm. These facts suggest that the anion-exchange resin removes ionic impurities effectively, especially volatile impurities such as chloride ions and carbon dioxide, which might not be effectively eliminated by a sublimation procedure. Impurities passing through the resin can be separated easily by a subsequent sublimation step. Therefore, purification using a procedure such as sub- limation followed by anion exchange and then a second sublimation step is recommended for effectively increasing the purity of the water by more than 0.5 MQ em (the purification may also be effective if, to save time, the initial sublimation is discarded).A quartz flask was used instead of the Pyrex one for a final sublimation but no improvement in the specific resistance was observed. g ml-l. This result agreed very closely with ours. Variation of Deuterium Concentration Accompanied by an Anion-exchange Process The variation in the deuterium content of the water resulting from passing the water through anion-exchange resin are shown in Table V. No variation was observed in the 0.6 - 0 0.4 - A .- Y- i o..) v) 0 A 0 A A 0 A A 0.05} , , , 8 0 2 3 4 5 1 2 3 1 No. of steps -- I I I I Purification process Fig. 8. Effect of I, sublimation and 11, anion 0.First run and exchange on purification of water. A, second run.618 INOUE et d. : MEASUREMENT OF HEAVY WATER Analyst, voz. 106 TABLE V VARIATION IN HEAVY WATER CONCENTRATION DUE TO THE ANION-EXCHANGE PROCESS Before treatment, mol-% After treatment, mol-% Relative error, % 0.0492 f 0.0008 0.0498 & 0.0008 + 1.2 0.5160 f 0.0079 0.5107 f- 0.0079 - 1.0 0.1007 f 0.0015 0.1003 f 0.0015 - 0.4 deuterium concentration in the range 0.05-0.5 mol-yo within the analytical error. There- fore, anion exchange can be used as a purification technique for low concentration heavy water analysis. Electrolytic Separation Factor for Hydrogen Isotopes Determined by Densitometry Electrolytic enrichment of hydrogen isotopes is widely applied to water samples before radiation counting of low levels of tritium.A correction is usually made for the loss of tritium occurring during the process by measuring the loss of deuterium at the same time and using the following relationship4 between the separation factors a of deuterium - hydrogen and /3 of tritium - hydrogen to calculate the tritium recovery: ln/3/lna = 1.4 . . .. .. .. . . (12) where V represents the volume of water and D and T represent the concentration of deuterium and tritium, respectively. The subscripts i and f indicate before and after electrolysis, respectively. After enrichment of natural water, the deuterium concentration usually rises above 0.1 mol-%. Therefore, electrolytically enriched water is adequate for the present technique. Determination of the deuterium concentration, Df, calculated by inserting the density value into the equation ( 5 ) , allows the calculation of the separation factors a and /3calc.according to equations (11) and (12). Alternatively, the tritium concentration of spiked samples before and after electrolysis can be measured by means of a liquid scintillation counter, and the separation factor Pobs. can be calculated from equation (11). Parameters used to calculate a and /3, and the resulting separation factors a, /3cslc. and Pobs. are given in Table VI. The separation factor flcalc., which was derived indirectly from the factor, a , agreed with the factor pobs. within experimental error under each electrolytic condition. This indicated that the separation factors for deuterium can be determined correctly using densitometry, and that the present method is applicable to the accurate determination of the electrolytic separation factor a for deuterium.Conclusion The densitometric method, using a high-sensitivity density meter, proved to be applicable to the accurate determination of levels of deuterium in water samples above 0.1 mol-%. Electrical conductivity or specific resistance proved to be a good measure of the purity of the TABLE VI ELECTROLYTIC SEPARATION FACTORS a AND jl DETERMINED BY THE DENSITY METER Error shows the spread of observed data as one standard deviation. Number Electrode of runs Vj/ml Vdml Df/mol-% Df/Dj* Tf/Ti a PCalC. Boba. (-1 - ( + I 200 10 0.15 9.6 12.8 4.4 f 0.3 7.9 f 0.7 7.8 f 0.7 N i - N i . . ..{ 2 Fe-Ni . _ 4 200 13 0.18 12.0 13.7 9.3 f 0.3 22.7 + 0.9 23.3 f 2.6 Di = 0.015 mol-%. 300 10-27 0.12-0.24 8-16 8-22 5.1 f 0.3 9.8 f 0.7 10.0 f 2.4Jzdne, 1981 . CONCENTRATION WITH A DENSITY METER 619 water to be used for the densitometry. Purification of the water to a specific resistance value of 0.5 MR cm is achieved easily using a combination of anion exchange and sublima- tion procedures. These techniques are relatively simple and rapid in comparison with conventional purification methods and the time required for the purification may be shortened by as much as 5 h per sample; also, the simultaneous handling of many samples is easy. The density meter used allowed density measurements to be completed within 10 min. In general, this method is superior to the conventional methods with respect to sample size, measuring time, sensitivity and accuracy. We thank Dr. Sadao Matsuo, Tokyo Institute of Technology, for generously allowing us the use of a mass spectrometer. We are indebted to Mr. Tetsuo Iwakura for providing the computer-calculated density table and to Ms. K. Takahashi and Ms. 11. Ida for technical assistance. References 1. 2. Schtenstein, A. I., in Krell, E., Editor, ‘‘Isotopenanalyse des Wassers. Physikalisch-chemische Trenn- und Messmethoden,” VEB Deutscher Verlag der Wissenschaften, Berlin, 1960, Volume 4, p. 65. Wada, N., Nippon Genshiryoku Gakkaishi, 1979, 21, 434. 3. 4. 5. 6. Kell, G. S., J . Chem. Eng. Data, 1967, 17, 66. Takahashi, T., Hamada, T., and Ohno, S., Radioisotopes, 1968, 17, 357. Allen, R. A., Smith, D. B., Otlet, R. L., and Rawson, D. S., Nucl. Instrum. Methods, 1966, 45, 61. Senda, O., and Inamatsu, T., Keisoku Jido Seigyo Gakkai Ronbanshu, 1979, 15, 112. Received January 7th, 1980 Accepted January 27th 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600609

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

620 Analyst, June, 1981, Vol. 106, $9. 620-624 Multi-purpose Automatic Solvent Extractor V. M. Bhuchar and A. K. Agrawal National Physical Laboratory, New Delhi-110012, India A multi-purpose solvent extractor has been modified to be automatic and continuous in operation. This modification has the added advantages of requiring less solvent than the earlier apparatus and of minimising decomposition and charring of the extracted constituent. A sampling port is provided for on-line control sampling of the extract. The apparatus has been used for de-fatting whole groundnut kernels with hexane, and for the extraction of caffeine from aqueous solution with trichloroethylene. Keywords : Multi-purpose automatic solvent extractor ; groundnuts ; caffeine ; solvent extraction ; purification of chemicals In earlier papers from this laboratory;lS2 a multi-purpose solvent extractor was described, which could perform many types of extraction (e.g., solid - liquid and liquid - light or heavier liquid) at a defined temperature (from just above ambient up to the boiling-point of the solvent), a t a known pressure and under the desired atmosphere.The apparatus was used to purify sulphamphthalein indicators to chromatographic purity.ls3 The apparatus was suitable for multi-purpose extractions, but manual operation of the three-way proportionating stop-cock was necessary for a very short period after each extraction, necessitating the attention of the operator. It was therefore considered desirable, especially for large-scale extractions, to modify the apparatus so that even this limited atten- tion was unnecessary and the apparatus could operate completely automatically.The modified apparatus is described in this paper. Apparatus A standard joint is provided at the inlet of the vapour-lead tube a t the bottom of the extraction chamber (Fig. lA), and a duck-settling bulb with an angular two-way stopcock (Fig. 1C) replaces the siphon outlet in the original apparatus.l The exit from the extractor (Fig. 1B) is extended almost to the bottom of the duck-settling bulb (Fig. 1C). Another opening in the socket of the stop-cock of the duck-settling bulb is provided for sampling in order to monitor the progress of the extraction. The vapour-lead inlet (Fig. 1A) and the exit (Fig. 1B) leading to the settling bulb (Fig.1C) are positioned almost diametrically opposite to each other. The function of the proportionating stop-cock in directing the incoming vapour into the extraction solvent and in fixing the temperature of extraction has been explained ear1ier.l The extraction processes in the earlier and modified extractors are similar. The mode of extraction with the modified extractor in liquid - liquid extraction, however, differs from that in the earlier apparatus in that the three operations of inter- mixing, separation of the phases and removal of the extract take place simultaneously in the former, at different levels in the extraction chamber. This is achieved, for liquid - heavier liquid extraction by fixing a tube at the vapour inlet (Fig. 1A) so that the vapours are led into the extracting chamber at a level somewhat higher than its bottom.The mixing then takes place at a level above the top of this tube, and the settling zone is between the top of this tube and the bottom of the extraction chamber. In liquid - lighter liquid extraction, a long tube is fixed at the exit standard joint (Fig. lB), the major mixing zone then exists between the bottom of the extraction chamber and a level below the open end of the inverted tube, and the settling zone exists at a level above it. The lighter extract ascends through the inverted tube to the top exit of the long tube. Accompanying heavier solution is left behind during this ascent. The lighter extract passes into the bottom of the duck-settling bulb (Fig. lC), where any of the heavier solution that may have been carried over settles and only lighter extract then passes over into the distillation flask.In order to automate the extraction, the stop-cock in the settling bulb is adjusted so as to equalise the volumes of the incoming solvent and the outflowing extract. In liquid - lighter liquid extraction, it is not possible to achieve continuous extraction if the multi-purposeBHUCHAR AND AGRAWAL 62 1 automatic extractor has the same dimensions as the earlier multi-purpose solvent extractor1s2 or a Soxhlet extractor.' In the modified apparatus the length of the column between the closed end of the extraction chamber and the exit in the socket of the stop-cock of the duck- settling bulb (Fig. 1C) is increased. Thereby a greater liquid head is obtained for continuous outflow of the extract against the back-pressure of vapour being produced in the distillation flask.d B A J \ B Fig. 1. Multi-purpose automatic solvent extractor. Assembly: A, for liquid - denser liquid extraction; and B, for liquid - lighter liquid extraction. This apparatus has been fabricated in five capacities ranging from 400 ml to 10 1 using glass tubing of 50, 68, 86, 110 and 150 mm internal diameter in accordance with the relation 'o = 7rd3, where 'o is the volume and d the diameter. The length of the extractor is therefore fixed at four times the internal diameter of the tube. Spherical Multi-purpose Automatic Solvent Extractor For low-volume extractors, a cylindrical shape with dimensions conforming to 'o = 7rd3 is suitable ; however, this becomes inconvenient for greater volumes.A spherical vessel (e.g., a boiling flask) is more appropriate for the extraction chamber when larger capacities are required. For example, for 7-, 12- or 22-1 extractors, spherical flasks of diameter of 220, 280 or 350 mm and of 5-, lo-, or 20-1 capacity, respectively, are required. By providing the flask with a tube of diameter 80 mm and length 160 mm at one end, and of diameter 110 mm and length 110 mm at the diametrically opposite end, a spherical extractor with a length about three quarters that of a cylindrical extractor of the same capacity is obtained. The former is easier to fabricate and to work with. The spherical portion of the extractor serves as the mixing zone. Separation of the two phases in the bottom and top tubular portions at both ends of the spherical mixing zone then becomes easier in liquid - liquid (respectively heavier or lighter) extraction.Fabrication is easy and there is less chance of residual strain in the apparatus after fabrication. Repairs to the tubular portions, which carry most of the joining points, are easy. The arrangement of tubes is similar to that described for a cylin- drical extractor; the arrangement for liquid - heavier liquid extraction is illustrated in Fig. 1A and for liquid - lighter liquid extraction in Fig. 1B.622 BHUCHAR AND AGRAWAL : MULTI-PURPOSE Analyst, Vol. 106 Applications Solid - Liquid Extraction of Lipids from Whole Groundnut Kernels Extraction of lipids from roast, blanched, de-skinned whole groundnut kernels was carried out with nearly boiling hexane4 in a multi-purpose automatic solvent extractor.The rate of extraction of the lipids from the kernels is shown in Fig. 2, curve A. It was observed that the extraction of the lipids from whole groundnut kernels takes place in discrete steps. First the solvent permeates the kernel and solubilisation of the lipids in the solvent then follows. A further step follows, which is associated with distention of the kernels, the vaporisation of moisture and further solubilisation of oil. This step becomes distinguishable if the moisture content of the kernels exceeds 2.4%. After about 7 h approximately 70% of the lipids are extracted; maceration at room temperature requires about 70 h for com- parable extraction.The rate of extraction from whole kernels in hexane (curve A, Fig. 2) in the multi-purpose automatic extractor was about 20% higher than with Soxhlet extrac- tion (curve C, Fig. 2). The extraction of kernels using the earlier multi-purpose solvent extractor is shown in Fig. 2, curve B. 4 7.0 E 8 6.0 2 5.0 4.0 L 3 L 0, -0 + g 3.0 .- : 2.0 + 0 1 .o 0 2 4 6 8 10 12 14 16 18 20 Time/h 2 Fig. 2. Extraction of lipids by hexane from whole groundnut kernels with a moisture content of 2% in: A, multi-purpose automatic solvent extractor ; B, multi-purpose manual solvent extractor ; and C, Soxhlet extractor. Horizontal bars indicate the observation periods. Liquid - Liquid Extraction of Caffeine from Aqueous Solution In an attempt to recover caffeine from tea waste and coffee-bean husk, the extraction of caffeine by trichloroethylene (b.p. 87 "C) from acidified aqueous solutions5 was examined.The progress of the extraction was followed from the absorbance at 275nm of the residual caffeine in the aqueous solution (diluted if necessary) remaining after extraction with boiling trichloroethylene (Fig. 3, curve A). The absorbance of the trichloroethylene extract was also measured after evaporating the organic extract to dryness and diluting the residue to a known volume with water (Fig. 3, curve B). Curves A and B in Fig. 3 are parallel and give a distribution factor of 0.602 calculated by the formula of Bewick et aZ.6 The distribution of caffeine between 800 ml of trichloroethylene and 500 ml of an acidified (0.5 N in sulphuric acid) aqueous solution of caffeine (10-2 M) was also studled in the automatic extractor. The trichloroethy- lene extract was dried and the residue dissolved in a known volume of water.The absorbance of this solution was measured at 275nm and the result is shown in Fig. 3, curve C. The distribution factor determined by the method of Bewick et is 0.355. The half-extraction A sampling port is provided in the automatic extractor.June, 1981 AUTOMATIC SOLVENT EXTRACTOR 623 volume required in the calculation was found from Fig. 3, curve C. The curves in Fig. 3 have different slopes at the lower concentration; here the retention forces in the acidified water for caffeine are likely to be stronger and cause slower extraction. 10.0 : In 1.0 4- a) c (II n n $ 0.1 a 0.01 I,,,,,, 0 2 4 6 8 1 0 1 2 1 4 No.of extractions Fig. 3. Progress of extraction with trichloro- ethylene, of an acidified (0.5 N H,SO,) aqueous solution of caffeine M) at 59-61 "C using: A, multi-purpose manual solvent extractor1 (residual caffeine in aqueous solution) ; B, multi-purpose manual solvent extractor (tri- chloroethylene extract of caffeine; 80 ml of trichloroethylene were used to extract 80 ml of aqueous solution) ; and C , multi-purpose auto- matic solvent extractor (trichloroethylene extract of caffeine). The arrows indicate the points from which half-extraction volumes7 were measured. The distribution coefficient for the extraction of caffeine from aqueous solution (lov2 M) by trichloroethylene is 0.850. The better extraction in the earlier multi-purpose extractor has to be weighed against the advantages of continuous extraction and minimal attention required when the automatic extractor is used.Small volumes can be more easily handled in the earlier apparatus whereas larger volumes can be continuously extracted in the automatic extractor. On-line control of extraction can also be maintained with the automatic extractor. Conclusions The multi-purpose automatic extractor described here is able to perform solid - liquid and liquid - liquid (heavier or lighter) extractions at any required temperature under desired pressures and atmospheres and requires little attention in operation. A further advantage of the automatic apparatus is that a greater proportion of the extrac- tion chamber can be utilised for extraction than with siphoned extractor^.^^^ In the latter the useful portion of the extraction chamber is about half of its volume (i.e., up to about the siphon top).The out-flow of the extract in a siphoned apparatus is intennittent with the result that the extract in the distillation flask occasionally becomes concentrated or even supersaturated, and the solute is then liable to decompose or char. This possibility is minimised with the automatic apparatus, because the volume of solvent in the distillation flask is kept constant. The volume of extractant used in the apparatus described is less than is required in a siphoned extractor. A volume sufficient to produce enough vapour to keep the solution stirred and to prevent decomposition of the extracted constituent is required. Consequently, less energy is required for extractions using the extractor described. For solid - liquid extraction, an additional volume is required to keep the solids submerged in the extraction solvent. In such an extraction, the advantages of both submersion and percolation are achieved.624 BHUCHAR AND AGRAWAL The authors thank Mr. D. P. Sharma of the Glass Technology Unit of this laboratory for fabricating prototypes of the apparatus. 1. 2. 3. 4. 5. 6. 7. References Bhuchar, V. M., and Agrawal, A. K., Anal. Chem., 1975, 47, 360. Bhuchar, V. M., Agrawal, A. K., Kiss, F., Vasisht, J. P., Parkash, D., and Srivastava, 0. N. L., U.S. Bhuchar, V. M., Technical News Service, Sarabhai M. Chemicals, Baroda, India, 1976, Volume 8, Bhuchar, V. M., Agrawal, A. K., and Sharma, S. K., J . Food Sci., 1981, in the press. Berry, N. E., and Walters, R. H., U.S. Pat., 2309092, 1943. Bewick, H. A., Currah, J . E., and Beamish, F. E., I n d . Eng. Chem., Anal. Ed., 1948, 40, 740. Soxhlet, F., and Szombathy, J., Dinglers Polytech. J., 1879, 232, 461. Pat., 4006062, 1977; Indian Pat., 141756, 1977. NO, 6, pp. 1-40. Received September 9th, 1980 Accepted December 15th, 1980

ISSN:0003-2654    

DOI:10.1039/AN9810600620

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, June, 1981, Vol. 106, pp. 625-635 625 Determination of Ammonium- and Nitrate-nitrogen by Molecular Emission Cavity Analysis* R. Belcher,? S. L. Bogdanski,; A. C. Calokerinos$ and Alan Townshendy Department of Chemistry, University of Birmingham, P.O. Box 363, Birmingham, B15 2TT Ammonium nitrogen is determined a t parts per million to percentage levels by injecting the sample solution on to solid sodium hydroxide in a small vial. The ammonia generated is swept by nitrogen into a molecular emission cavity analysis (MECA) oxy-cavity and the intensity of the NO-0 continuum is measured a t 500 nm. The method is rapid, precise and free from interferences and has b z n applied successfully to the determination of ammonium ions in fertilisers, river waters and sEwage samples.Nitrate ions can also be determined after reduction to ammonia by Devarda's alloy. Keywords : Molecular emission cavity analysis ; ammonium ions ; nitrate ions ; hydrogen $ame The direct determination of ammonia by atomic-absorption spectrometry has not been accomplished because the principal resonance lines of nitrogen lie in the vacuum-ultraviolet region. The atomic-absorption technique can be used indirectly by measuring the enhance- ment of the zirconium absorbance by ammonia (1 x 10-*-5 x M) in a dinitrogen oxide - acetylene flame at 360.1 nm but the method lacks sensitivity and selectivity.1 Ammonia can also be determined by the enhancement of the silver absorbance in an air - acetylene flame.2 Nitrate (1 x 10-"7 x M) has been determined by atomic-absorption spectrometry after extract ion of Cu(neocuproine) ,NO p i s (2,g-dimet h yl-1 , 10-phenan t hrolino) - copper(1) nitrate] , aspiration into an air - acetylene flame and measurement of the copper absorbance at 324.7 nm.3p4 Plasma spectroscopy has been proposed for the determination of ammonia (0.1-2000 p.p.m.); the method is based on the oxidation of the ammonium ions to elemental nitrogen by hypobromite.The nitrogen gas is introduced into an inductively coupled plasma source operated at 27 MHz for measurement of the NH emission at 336 n ~ . ~ Gas-phase molecular absorption has also been used for the determination of ammonia.s-8 The sample solution is treated with aqueous sodium hydroxide to liberate ammonia, which is passed through a long tube in line with the beam of a hydrogen lamp focused on the detector.This monitors the absorbance of ammonia (parts per million levels or greater) at 201 nm. Nitrate can be determined similarly after reduction to ammonia by titanium(II1) sulphate in a basic ~olution.~ Aspiration of a 50% ammonia solution in methanol into a hydrogen -nitrogen flame causes an emission showing a broad continuum with maximum intensity at about 520nm but the observation has not been used for analytical work.1° However, treatment of an ammoniacal solution with aqueous sodium hydroxide and transfer of the generated ammonia into a hydrogen diffusion flame allows measurement of the NH emission a t 336 nm,ll so that ammonia at the parts per million level can be determined. Molecular emission cavity analysis (MECA) is also a technique that uses a hydrogen-based flame for generating molecular emissions. Its use has been proposed recently for the deter- mination of amines and non-sulphur-containing amino acids, after formation of the corre- sponding reaction product with carbon disulphide, and measurement of the S, emission.12 We have also briefly reported the possibility of a direct method for ammonia determination based on a broad-band emission at about 500 nm.13 This paper gives a detailed account of such a method and describes how it can be extended to measure nitrate ions.* Presented a t the 4th SAC International Conference on Analytical Chemistry, Birmingham, 1977. t Present address : Department of Chemistry, University of Aston, Gosta Green, Birmingham, B47 ET.Present address: Doma Instruments Ltd., Unit 19, Forge Trading Estate, Halesowen, West Midlands. 5 Present address : Laboratory of Analytical Chemistry, University of Athens, 104 Solonos Street, 7 To whom correspondence should be addressed. Present address : Department of Chemistry, Uni- Athens (144), Greece. versity of Hull, Hull, HU6 7RX.626 Analyst, VoZ. 106 Preliminary Investigations When ammonia vapour, formed by treatment of an ammonium solution with sodium hydroxide, was carried by nitrogen gas into a MECA cavity situated in a hydrogen diffusion flame, a faint blue emission appeared. On changing to oxygen as the carrier gas a brighter, white emission was observed, but immediately after introduction of oxygen the cavity became hotter, showing gradually increasing incandescence, and sodium atomic emission appeared and intensified rapidly, masking the white emission of the ammonia.The sodium emission arises mainly from the presence of dust and/or the introduction of the alkaline water droplets into the cavity. The use of an aluminium water-cooled cavity decreased the cavity tempera- ture and, therefore, the sodium emission, allowing the ammonia emission to be measured. By using the water-cooled cavity and having a continuous flow of nitrogen introducing ammonia into the cavity, the blue emission was sustained within the cavity. Spectral examination (slit width 0.5 mm = 8.5 nm) revealed a broad continuum showing maximum intensity at about 450 nm [Fig. l(A)]. By using oxygen as a purge gas, the bright, white emission was obtained; it showed a broad continuum with maximum intensity at about 500 nm [Fig.l(B)]. The white emission was not obtained when only nitrogen and oxygen were introduced into the cavity, so the ammonia emission could be measured without any interference from nitrogen gas. The white emission was much more intense than the blue emission, and is probably the NO-0 continuum. BELCHER et al. : AMMONIUM- AND NITRATE-NITROGEN OH (firs i A order, I ! I 1 I I I 1 250 350 450 550 650 Wavelengthlnm Fig. 1. Emission spectrum of ammonia when introduced into the MECA cavity : A, without oxygen and B, with oxygen, as measured by an RCA IP28 photomultiplier tube (at x and x’ the amplification was increased 100 and 10 times, respectively). For the determination of ammonium-nitrogen via the white emission due to ammonia in a cavity supplied with oxygen, the ammonium ions must be converted into ammonia.The reaction employed should be fast and specific, and should not give other potentially emitting species, e.g., carbon dioxide that could generate CH and C, emissions in the flame. Selective generation of ammonia will allow a wide slit to be used to measure a large section of the broad emission spectrum, thereby increasing the sensitivity. Sodium hydroxide fulfils these requirements. The reaction is carried out in the reaction chamber from which the vapours are continuously swept into the cavity and the emission intensity is measured at 500 nm as a function of time.June, 1981 BY MOLECULAR EMISSION CAVITY ANALYSIS Experimental Reagents 627 All chemicals used were of analytical-reagent grade.Distilled water was used in all Stock solutions of concentration 1 000 p.p.m. were prepared, and low-concentra- instances. tion solutions were prepared daily. Apparatus Later it was modified so that the monochromator, burner and cavity stood on an optical bench. This arrangement allows better adjustment of the components relative to each other for maximum light collection. A modified MECA spectrometer was also used. The main difference from the MECA-22 model is that it incorporates a damping control that prevents changes of signal above a pre- selected frequency from being amplified. Thus, flame noise (mainly flame flicker) is greatly decreased, and improved signal to noise ratios are obtained.A digital read-out was incorporated that could be set to show peak height or area, and hold the maximum value. A circular slit (4 mm in diameter, equivalent to 68 nm maximum spectral band width) was used for all measurements. A Honeywell integrator was used for the nitrate determination. Initially,13 a MECA-22 spectrometer (Anacon Inc., Houston, Texas) was used. An Oxford, 3000 Series, potentiometric recorder was used for all determinations. Cavity The cavity design described earlier13 was altered to allow introduction of oxygen and nitrogen gases into the cavity via two different side ducts. The aluminium cavity used (Fig. 2) was mounted on a brass cooling chamber; the incoming cooling-water hits the back of the cavity and cooling of the cavity space is direct and very effective.PTFE O-ring 7 - O 2 / ....... .......... J) ~ IJ 9 -Water in I Water out Fig. 2. Cavity mounted on cooling chamber for the determina- tion of ammonium- nitrogen. Not to scale. An alternative aluminium cavity was also employed (Fig. 3), again mounted on a brass cooling chamber, but incorporating an aluminium screw, at the back of the inner space, that separates the cavity into a mixing chamber (m) and the actual cavity (d). The gases (oxygen, nitrogen and ammonia) entering the mixing chamber are well mixed before they pass into the actual cavity. Thus, a more homogeneous vapour phase is formed within the cavity space. The advantage of this cavity over other MECA cavities is that when flammable gases, e.g. , hydrogen, are generated in the reaction chamber of the vaporisation system, they burn in the mixing chamber before entering the cavity.This gives a cooler, less fuel-rich flame in the cavity itself.628 BELCHER et d. : AMMONIUM- AND NITRATE-NITROGEN Analyst, VOZ. 106 Water out Water in Fig. 3. Cavity mounted on cooling chamber for the determina- tion of nitrate. Not to scale. m = mixing chamber, d = cavity. Vaporisation System The carrier gas was nitrogen, and oxygen was supplied to the cavity via a second side duct (Fig. 4) to avoid any possibility of flashback when hydrogen is evolved. The bubble breaker is a stainless-steel wire that hangs from the cap of the vial and forms a spiral around the gas inlet. The vaporisation system described initiallyl3 was slightly modified. Recommended Procedure for the Determination of Ammonium-nitrogen Transfer 0.5-0.6 g (about three pellets) of solid sodium hydroxide into the reaction chamber of the vaporisation system, which is continuously purged with nitrogen.Inject exactly 0.2 ml of the sample solution through a silicone septum into the reaction chamber, record the emission intensity at 500 nm as a function of time and measure the peak height for quantitative purposes. TABLE I The instrumental conditions used are summarised in Table I. EXPERIMENTAL CONDITIONS FOR THE DETERMINATION OF AMMONIUM-NITROGEN Cavity position . . .. .. Wavelength . . .. .. Slit . . .. .. .. .. H, flow-rate .. .. .. N, flow-rate .. .. .. N, (carrier gas) flow-rate . . 0, (to cavity) flow-rate . . . . Cooling water flow-rate . .. . Sodium hydroxide. . .. .. Analyte solution . . .. .. .. Flame centre, lowest part 1.4 cm above burner head 500 nrn Circular, 4 mm diameter (68 nm maximum spectral 4.2 1 min-l 5.9 1 min-l 75 ml min-l 110 ml min-1 10 ml min-l 0.2 ml band width) 0.5-0.6 g Recommended Procedure for the Determination of Nitrate-nitrogen Use the same procedure as for ammonium-nitrogen but add 30mg of Devarda’s alloy (50% copper, 45% aluminium and 5% zinc; particle size (335 pm) to the sodium hydroxide in the reaction chamber. Use peak area for quantitative measurements. Optimisation of Instrumental and Reaction Parameters for Ammonia Determination The hydrogen and nitrogen flow-rates used were 4.2 and 5.9 1 min-l, respectively, which were found to give the best signal to noise ratio.Air was not introduced into the flame in order to reduce random appearance of sodium emission from dust and impurities that increased when a hotter flame was used. The hydrogen and nitrogen gases supplied to the flame were pre-mixed in a mixing chamber. Mixing of the flame gases was further improved by filling the burner body with glass capillary tubes, thus giving a steady, colourless flame.June, 1981 BY MOLECULAR EMISSION CAVITY ANALYSIS lniection Water chamber out u -NaOH pe 629 2 !Ilets Cavity Burner head Detector Fig. 4. Vaporisation system for the determination of ammonium and nitrate ions. Not t o scale. When 1 ml of aqueous ammonia was transferred by pipette into the reaction chamber with- out any sodium hydroxide and the solution slowly purged by nitrogen, a constant intensity white emission that lasted for 2-3 min was generated within the cavity. During that time, the cavity position was slightly altered while measuring the continuous white emission.The best position of the cavity in the flame was considered to be the one at which the emission was the most intense. This was at the centre of the flame, with the lowest part of the cavity 1.4 cm above the burner head. The effect of nitrogen and oxygen flow-rates supplied to the vaporisation system and cavity, respectively, on the emission intensity from ammonia generated from 0.2 ml of 1000 p.p.m. of nitrogen, as ammonium chloride, was investigated. The results obtained (Fig. 5 ) show that a flow-rate of 75mlmin-l for nitrogen gives maximum emission intensity regardless of the oxygen flow-rate.For a flow-rate of 150 ml min-l of oxygen, the difference in the emission intensity between flow-rates of 75 and llOmlmin-l of nitrogen is not significant. The increased intensity with increasing oxygen flow-rate is at least partly due to the increase in temperature of the flame inside the cavity (see below). The introduction of high nitrogen flow-rates into the cavity dilutes the flame, thus lowering the cavity temperature and decreasing the ammonia intensity. 0 50 100 150 200 Nitrogen flow-rate/ml min-' Fig. 5. Effect of nitrogen carrier gas flow-rate on the emission intensity from 0.2 ml of ammonium-nitrogen (1000 p.p.m.) with: A, 90 ml min-'; B, 110 ml min-'; and C , 150 ml min-' of oxygen supplied to630 BELCHER et al.: AMMONIUM- AND NITRATE-NITROGEN Analyst, VoZ. 106 When the oxygen flow-rate was 150 ml min-l or greater, the hydrogen - oxygen flame only just remained within the cavity space and very frequently it extended into the body of the hydrogen diffusion flame. However, there was no difficulty in restricting the flame to within the cavity when an oxygen flow-rate of 110 ml min-l was used, so all subsequent measurements were made at this flow-rate and a nitrogen carrier gas flow-rate of 75 ml min-1. Higher flow-rates are not recommended because they cause an increase in bubble formation in the solution. When the rate of bubble formation increases severely, the bubble breaker is not sufficient to obstruct all of the bubbles and solid sodium hydroxide blocks the vapour duct.The effect is more pronounced when foaming samples are injected. Cavity Temperature The cavity temperature can be altered by changing the cooling water flow-rate, while maintaining the same flame composition. Hence, any effect on the emission intensity will be due mainly to the cavity temperature change. The cavity shown in Fig. 2 changes temperature markedly with small changes of water flow-rate. Flow-rates of 10 ml min-1 allow the cavity to be very hot whereas flow-rates of more than 100mlmin-l make the cavity so cool that severe condensation of water vapour from the flame can be observed on the metal surfaces, even those in the flame. Measurements of the apparent cavity temperature were made by inserting the end of a chrome-alumel thermocouple into the cavity through one of the side ducts.The term “cavity temperature” is an expression of the relative thermocouple output when the thenno- couple end is introduced 1-3 mm into the cavity. In order to investigate the effect of cavity temperature on the response from 500 p.p.m. of nitrogen, as ammonium chloride, 0.2-ml volumes of the solution were injected into the reaction chamber containing sodium hydroxide and the emission intensity was measured at different flow-rates of water. The results obtained (Fig. 6) show that the ammonia emission is enhanced when a hot cavity is employed. A 25% increase in emission intensity was obtained by lowering the water flow-rate from 100 to 10 ml min-l. Therefore, the cooling water flow-rate was adjusted to 10 ml min-l for all subsequent determinations.Relative thermocouple outputlmv Fig. 6. Effect of cavity temperature (expressed as relative thermocouple output) on the emission intensity from 0.2 ml of 100 p.p.m. of nitrogen as ammonium chloride.June, 1981 BY MOLECULAR EMISSION CAVITY ANALYSIS 631 Effect of the Amount of Sodium Hydroxide and Analyte During the preliminary studies it was noticed that when fast and repeatable emissions were obtained while bubbling ammonia from the solution, solid sodium hydroxide was present in the reaction chamber after the measurement was finished. It was decided to use 0.5-0.6g of sodium hydroxide (three pellets) and optimise all other parameters for this amount. In order to optimise the volume of analyte required for each injection, different volumes, containing 50 pg of ammonia, as ammonium chloride, were injected in the reaction chamber containing 0.5-0.6 g of sodium hydroxide. The results are shown in Fig.7. It was observed that when 0.4ml or more was used, no solid sodium hydroxide was present in the reaction chamber after analysis. Large volumes of injected solutions tend to show low, broad peaks, owing to incomplete formation of ammonia and increased resistance to diffusion of the gas. When less than 0.05 ml was injected, the purge gas inlet tube was not completely in contact with the solution. Therefore, the optimum volume of analyte solution for the system described is 0.2 ml and this volume was used for all subsequent measurements. The recommended experimental conditions for ammonia are summarised in Table I.0- 0 1 I 1 I 1 I 1 0 1 0 1 0 1 2 Time after injection/min I I I 0 1 2 Fig. 7. Effect on the emission intensity of injecting various volumes of a solution containing 50 pg of nitrogen as ammonium chloride into the reaction chamber containing 0.5-0.6 g of sodium hydroxide. Volume injected: (a) 0.05 ml; (b) 0.1 ml; (c) 0.2 ml; (d) 0.3 ml; and (e) 0.4 ml. Results Results for Ammonia Nitrogen Injections of aqueous solutions of ammonia or an ammonium salt (e.g., chloride, nitrate or acetate) containing the same amount of ammonia gave equal emission intensities. The effect was confirmed throughout the entire concentration range studied. Therefore, a single calibration graph can be used for ammonium ion salts or ammonia injected on to the same amount of sodium hydroxide.By using peak heights, the limit of detection (signal to noise ratio = 2) was found to be 1 p.p.m. of nitrogen. The coefficient of variation for ten measure- ments of 100 p.p.m. of nitrogen was 2%. The calibration graph, prepared using peak heights, was linear from 10 p.p.m. up to the percentage level. Interferences Interferences were investigated by determining 100 p.p.m. of ammonium-nitrogen in the presence of other species. The responses were compared with those obtained from an uncontaminated ammoniacal solution. No effect was observed from cobalt(II), cadmium(III), chromium( I I I), manganese (11) , zinc( 11) , nickel( 11) , lead( 11) , copper( 11) , mercury( 11) , iron( 11) , iron(III), silver(I), sulphate, phosphate, nitrate and urea when present in 9-fold excess by mass.Although urea hydrolyses in alkaline solutions to liberate ammonia, the analysis time (5 min per sample) is too short for significant hydrolysis to occur. For example, urea will be hydrolysed by an excess of sodium hydroxide (1 : 8) by about 17% in 75 min.14632 BELCHER et d. : AMMONIUM- AND NITRATE-NITROGEN Analyst, VoZ. 106 Determination of Ammonium4trogen in Fertilisers, River-water Samples and Ammoniacal Liquors It was decided to use the proposed method for the determination of water-soluble ammonium-nitrogen in corn ercially available fertilisers. The results were compared with those obtained using the cla sical method (distillation of ammonia from alkaline solution into excess of standard hydrc chloric acid and titration of the unreacted acid) .15 The fertilisers (1.5-2 g of ach) were dissolved in water and diluted to 250 ml.Analysis by both methods was carriel out on the same solution. The results, obtained using the unmodified cavity, have beel published previ0us1y.l~ Ammoniacal nitrogen was also deter- mined directly in river-wat zr samples provided by the Severn - Trent Water Authority. Table I1 shows the results Ibtained using the MECA method compared with the quoted values. TABLE I1 DETERMINATION OF AM1 [ONIUM-NITROGEN IN CONTAMINATED RIVER-WATER SAMPLES The values given are in parts per million. Sample . . .. 1 2 3 4 5 6 MECA . . .. . . 25 26 26 35 35 36 Quoted* .. . . 28 26 28 35 35 35 * These value ; were provided by the Severn - Trent Water Authority, Coventry, and v ere determined by spectrophotometric measurements.The method has also betn applied directly for the determination of ammoniacal nitrogen in ammoniacal effluent lic uors (supplied by the British Gas Corporation) and coke-oven liquors (supplied by the E ritish Carbonization Research Association) ; the results are com- pared with the quoted vah ies in Table 111. For the determination 3f ammonium-nitrogen in river-water samples and liquors, the modified MECA spectromt ter was used under the same conditions as the MECA-22 instru- ment. Some of the amrLoniaca1 liquors contained foaming material that blocked the vapour duct. The probler! i was overcome, without any loss of sensitivity of the method, by using a trap vial connecte 1 in series with the reaction chamber.TABLE I11 DETERMINATION OF PMMONIUM-NITROGEN IN EFFLUENTS AND COKE-OVEN LIQUORS The values given are in parts per million. Sample . . .. 1* 2* 3* 4t 5t 6* 7 t 8t MECA .. . . 130 280 350 $80 1025 1500 2400 4500 Quoted .. .. 140 264 331 911 1066 1443 2231 4500 * Coke-oven liquo s provided by the British Carbonization Research Association. t Effluent liquors ?rovided by the British Gas Corporation. Determination of Nitr: Ite The determination of . iitrate by MECA can be accomplished by the generation of oxides of nitrogen or ammonia. The latter scheme was adopted in this work mainly because it was a logical extension to t i e generation of ammonia for the determination of ammonium- nitrogen. The reductani used was Devarda’s alloy, which is powerful under strongly alkaline conditions.l6 During the reduction c E nitrate by Devarda’s alloy, hydrogen gas is liberated and when it is introduced into the ca- ity it burns in the flame which probably becomes fuel-rich, causing an increase in the backlround emission.In addition, the temperature of the flame, and sequentially that of the c wity, increases. Hydrogen also adds to the flow-rate of the carrier gas and some sodium hyd .oxide is carried into the cavity. The characteristic sodium emission at 589 nm will therefore appear within the cavity and will interfere with the measurement because of the broad slit used.June, 1981 BY MOLECULAR EMISSION CAVITY ANALYSIS 633 In order to overcome the problems associated with the introduction of hydrogen gas into the cavity, the cavity shown in Fig.3 was employed. This cavity gave greatly decreased background and sodium emissions, because of the decreased temperature in the cavity. Effect of Particle Size and Amount of Devarda’s Alloy Devarda’s alloy was sieved through a 44-mesh sieve (= 355 pm) and the two fractions, one with particle size less than 355 pm, the other with particle size greater than 355 pm, were examined, in order to investigate the effect of particle size on the reduction of nitrate and generation-rate of ammonia. Fig. 8 shows the results when a solution containing lo00 p.p.m. of nitrate, as potassium nitrate, was injected into vials containing equal amounts of alloy from the two fractions. Analysis of the two fractions by atomic-absorption spectro- metry showed that their chemical compositions were identical.Hence the decreased emission resulting from the large-particle-size fractions of the alloy is due to the slower and incomplete reduction of nitrate. Therefore, it was decided to use Devarda’s alloy of particle size less than 355 pm for all measurements. The effect of the mass of the alloy on the emission intensity from a solution containing 1000 p.p.m. of nitrate as potassium nitrate injected into the reaction chamber containing 0.5-0.6 g of sodium hydroxide was investigated. The results, calculated using peak heights, are shown in Fig. 9. I t is obvious that the amount of alloy necessary for the fast and complete reduction of the injected nitrate is not less than 30 mg. For more than 30 mg of alloy, no increase in intensity is observed.With much larger amounts, the larger amounts of hydrogen evolved cannot be accommodated in the cavity, so that the ammonia emission occurs in the body of the flame and part of it is not viewed by the detector. Perhaps, by using a larger reaction chamber, the hydrogen gas will be diluted by the higher purge-gas flow-rate and the final effect in the cavity will be minimised. This aspect was not studied further. Results for Nitrate to give the same emission intensity within the concentration range examined. single calibration graph can be used for all nitrates. Solutions of different nitrate salts containing the same concentration of nitrate were found Therefore, a I I 0 1 I 0 1 Time after injection!min Fig. 8. Emission intensities from 0.2-ml injections of 1000 p.p.m.of nitrogen as potassium nitrate into the reaction chamber containing 0.5-0.6 g of sodium hydrox- ide and 20 mg of Devarda’s alloy: (a) <44 mesh; and (b) 2 44 mesh.634 BELCHER et aZ. : AMMONIUM- AND NITRATE-NITROGEN Analyst, VoZ. 106 The reproducibility of the determination was found to be much better when peak areas were used instead of peak heights; for ten measurements of 100 p.p.m. of nitrogen the relative standard deviation was 3.5%. The limit of detection, defined as the concentration of nitrate giving an integrated response equal to twice that of the blank when only water is injected, was found to be 1 p.p.m. of nitrogen. The calibration graph is linear from 10 to 400 p.p.m. and a less steep linear graph is obtained from 500 to 2500 p.p.m.0 10 20 30 40 Mass of Devarda's alloy/mg Fig. 9. Effect of the mass of Devarda's alloy (<44 mesh) used on the emission intensity from 0.2 ml of 1000 p.p.m. of nitrogen as potassium nitrate. Interferences No interferences were observed in the determination of 100 p.p.m. of nitrogen as nitrate in the presence of 9-fold excess by mass concentrations of copper(II), iron(II), cadmium(II), mercury( 11) , manganese( 11), zinc( 11) , nickel( 11) , lead( 11) , chromium( 111) , silver( I), cobalt (11) , chloride, sulphate, acetate and phosphate. Urea does not interfere for the reason given for the ammonium-nitrogen determination. Nitrite, cyanide, metal - cyanide complexes and thiocyanate also react with Devarda's alloy under the conditions used, with the release of ammonia, and are likely to interfere.Ammonium ions, when pre?ent, will also interfere and should be determined separately in the absence of Devarda's alloy. Arsenic and antimony oxy-anions are reduced to their hydrides and produce emissions in the cavity which interfere with the ammonia emissi0n.l' Conclusion The method described for the determination of ammonium-nitrogen is simple, fast and sensitive and provides linear calibration over a wide concentration range. The emission intensity is independent of the ammonium salt injected. The analysis time is less than 5 min per sample. Although not investigated in detail, the method for the determination of nitrate appears to be promising and provides a fast procedure for an ion which is difficult to determine. However, the method is likely to be subject to serious interferences when using real samples.The determination of nitrate and ammonium ions in mixtures is also feasible on the basis of peak-area measurements. Finally, the method proposed can be modified for the determination of other nitrogenous compounds capable of generating ammonia quantitatively, and for related compounds such as amines.June, 1981 BY MOLECULAR EMISSION CAVITY ANALYSIS 635 A.C.C. thanks the Bodossaki Foundation (Athens, Greece) for financial support and Mr. S. Travers, Mechanical Workshop (Chemistry Department), for making the cavities. The authors thank Dr. H. C. Wilkinson, British Carbonization Research Association, Dr. S. J. Wood, formerly of the British Gas Corporation, and the Severn-Trent Water Authority for the provision of the analysed effluent samples. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Bond, A. M., and Willis, J. B., Anal. Chem., 1968, 40, 2087. Mitsui, T., and Fujimura, Y., Bunseki Kagaku, 1974, 23, 449. Kumamaru, T., Tao, E.. Okamoto, N., and Yamamoto, Y., Bull. Chem. SOC. JPn, 1965, 38, 2204. Yamamoto, Y., Kumamaru, T., Hayashi, Y., and Otani, Y., Bunseki Kagaku, 1969, 18, 359. Alder, J. F., Gunn, A. M.. and Kirkbright, G. F., Anal. Chim. Acta, 1977, 92, 43. Cresser, M. S., and Isaacson, P. J., Talanta, 1976, 23, 885. Cresser, M. S., Anal. Chim. Acta. 1976, 85, 253. Cresser, M. S., Lab. Pract., 1977, 26, 19. Cresser, M. S., Analyst, 1977, 102, 99. Dagnall, R. M., Smith, D. J., Thompson, K. C., and West, T. S., Analyst, 1969, 94, 871. Butcher, J. M. S., and Kirkbright, G. F., Analyst, 1978, 103, 1104. Al-Tamrah, S. A., Belcher, R., Bogdanski, S. L., Calokerinos, A., and Townshend, A., A n d . Chim. Belcher, R., Bogdanski, S. L., Calokerinos, A. C.. and Townshend, A., Analyst, 1977, 102, 220. Fawsitt, C. E., Z . Phys. Chem., 1902, 41, 601. Belcher, R., and Nutten, A. J ., “Quantitative Inorganic Analysis,” Third Edition, Butterworths, Mertens, J., Van den Winkel, P., and Massart, D. L., Anal. Chem., 1975, 47. 522. Belcher, R., Bogdanski, S. L., Ghonaim, S. A., and Townshend, A., Anal. Chim. Acta, 1974, 72, 183. Received August 14t12, 1980 Accepted December 23vd, 1980 Acta, 1979, 105, 433. London, 1970, p. 174.

ISSN:0003-2654    

DOI:10.1039/AN9810600625

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

636 Analyst, June, 1981, Vol. 106, $9. 636-640 Extraction of Potassium with Trifluoromethyl- substituted Chromogenic Crown Ethers G. E. Pacey, Y. P. Wu and B. P. Bubnis Department of Chemistry, Miami University, Oxford, Ohio 46066, USA Two new chromogenic crown ethers with trifluoromethyl substituents have been synthesised and evaluated as reagents for the extraction and spectro- photometric determination of potassium in the presence of sodium. These crown ethers, 4'- (2", 6"-dinitro-4"-trifluoromethylphenyl)aminobenzo- 15- crown-5 (4TF) and 4'-(2",4"-dinitro-6"-trifluoromethylphenyl)aminobenzo- 15- crown-5 (6TF), exhibited large differences in the values of Am. and molar absorptivity for the complexed and uncomplexed species and also in their extraction efficiencies. 6TF showed the best extraction efficiency and was capable of extracting 5-700 p.p.m.of potassium ions in the presence of 3000 p.p.m. of sodium ions. Keywords : Potassium determination ; chromogenic crown ether complex ; alkali metal extraction ; spectro$hotometry Crown ethers have shown a remarkable ability to extract alkali and alkaline earth metal ions selectively,l but until recently the use of crown ethers for analytical determinations has not been possible. Recent work has produced a chromogenic crown-like 4'-picrylaminobenzo- 15- crown-5 that was found to be a selective extractant for potassium ions (10-800 p.p.m.) in the presence of sodium ions (2300 p.p.m.).2 This reagent suffered from a poor extraction efficiency and spectral overlap of the protonated (HL) and deprotonated (ML) species.Other workers have added bromo- and nitro-groups in the 5'-position in an effort to lower the pK, of the monobasic proton and, hopefully, increase extraction We have chosen to attack the problem of spectral overlap and, in some instances, attempted to improve aqueous solubility.5 4"-(4'-Cyano-2" ,6" -dinitrophenyl) aminobenzo-15-crown-5 accomplished the first task of decreasing spectral overlap by increasing AX,,,, ML-HL from 55 to 145 nm. This paper describes compounds in which trifluoromethyl groups were used as substituents, because of their electronic effects on the phenyl nucleus groups and the improvement in aqueous solubility produced in the organic reagent .6 Experiment a1 Apparatus Spectrophotometric measurements were carried out using a Hewlett-Packard 8450A reversed optics spectrophotometer with 10-mm glass cells.The pH measurements were carried out using a Corning, Model 12, pH meter. Characterisation of the new organic compounds was accomplished using a JEOL nuclear magnetic resonance spectrometer and a Perkin-Elmer 180 infrared spectrophotometer. All elemental analyses were performed externally by Galbraith Laboratory. Reagents Synthesis of 4'-(~'',6"-dinitro4"-t~i~uoromethyl$henyl)aminobenzo-l5-crown-5 (4TF) and 4'- (2 ' ',4" -dinitro-6"-tri_fl.uoromethylp henyl) amino benzo- 15-crown-5 (6TF) Benzo-15-crown-5 was nitrated according to the reported procedure,' to give 4'-nitrobenzo- 15-crown-5, which was catalytically reduced to 4'-aminobenzo-15-crown-5.5 A mixture of the amino compound (3.9 g ; 0.013 7 mol) , l-chloro-2,6-dinitro-4-trifluoromethylbenzene (3.7 g; 0.0137 mol) and sodium hydrogen carbonate (1.15 g; 0.0137 mol) was refluxed for 5 h in 200 ml of absolute methanol.The mixture was cooled and filtered and the methanol was removed using a rotary evaporator. The residue was dissolved in isopropanol, and an equal volume of light petroleum (boiling range 30-60 "C) was added in order to precipitate the impurities. The isopropanol - light petroleum mixed solvent was filtered and evaporatedPACEY, W U AND BUBNIS 637 N 0 2 4TF 6TF to give the chromogenic crown ether 4TF as a dark orange powder (melting-point 171 “C, yield 60%). Calculated for C,H,,N,O,F,: C, 48.74;-H, 4.92; N, 8.12; and F, 11.02%. Found: C, 48.51; H, 4.41; N, 7.93; and F, 11.06%.6TF, prepared in a similar way using l-chloro-4,6-dinitro-2-trifluoromethylbenzene, was obtained as a dark orange powder (melting-point 165 “C, yield 60%). Calculated for CNH,,N,O,F,: C, 48.74; H, 4.92; N, 8.12; and F, 11.02%. Found: C, 48.67; H, 4.36; N, 7.86; and F, 10.74%. Chloroform. Spectroscopic quality chloroform was prepared using the literature procedure.8 Triethylamine (TEA). TEA was freshly distilled from the shipping containers at 88.6 “C. Alkali metal chlorides. (Alfa). All of the alkali metal chlorides were of 99.9% purity. Procedure Visible spectra 40% acetonitrile - water mixed solvent that was 1 M in sodium chloride. measured at pH 2 and 11.2 (TEA buffered). The visible spectra were obtained by dissolving a known amount of the crown ether in a The spectra were Determination of the extraction constant and the complex stoicheiometry Crown ether - 1 M TEA solutions in chloroform (5 ml), of different crown ether concentra- tions, were mixed with 5 ml of 0.1 M alkali metal solutions (made up in de-ionised water) and extracted using chloroform.The chloroform layer was analysed spectrophotometrically over the wavelength range 250-800 nm. Results and Discussion Visible Spectra and Molar Absorptivities of HL and ML Species. The aqueous spectra of 4TF, 6TF and, for reference purposes, 4’-picrylaminobenzo-15- crown4 in their protonated, uncomplexed forms and in their dissociated complexed forms are shown in Fig. l(a-c). The maximum absorptions and the molar absorptivities for the 300 400 500 600 350 400 450 500 300 400 500 600 700 Wavelengthhm Fig.1. Ultraviolet - visible spectra of protonated (HL) and deprotonated (ML) crown ethers: (a) 4’-picrylaminobenzo-15-crown-5, 2 x M in dioxan - water (1 + 9) ; (b) 4’-(2”,4”-dinitro-6”- trifluoromethylphenyl)aminobenzo-l5-crown-5, 2 x M in acetonitrile - water (2 + 3) : and (c) 4‘-(2”,6”-dinitro-4”-trifluoromethylphenyl)am~obenzo-l5-cro~-5,2 x M in acetonitrile - water (2 + 3).638 PACEY et al. : EXTRACTION OF POTASSIUM WITH TABLE I WAVELENGTH MAXIMA AND MOLAR ABSORPTIVITIES FOR Analyst, Vol. 106 CHROMOGENIC CROWN ETHERS Reagent QI2d Species 1 mol-l cm-l Amax. qo/nm AAmax./nm 4'-Picrylaminobenzyl-15-crown-5 . . HL 13 000 390 55 ML* 20 000 446 ML* 4 400 585 ML* 20 800 460 4TF . . .. .. .. . . HL 6 400 426 150 6TF .... .. .. . . HL 13 250 380 80 * Depending on the alkali metal cation complexed in the crown cavity, the complexed form may be either ML or ML.HL reagent species are shown in Table I. An important feature is that there is a decrease in the spectral overlap between the HL and ML species for both 4TF and 6TF. This is important as the spectral overlap in the 4'-picrylaminobenzo-15-crown-5 is so great that the ML species had to be determined at a wavelength where, according to the molar absorptivities, considerably less than maximum absorption occurs (20000 versus 5000 1 mol cm-1). It is clear from these data that 6TF is the superior reagent for spectrophotometric determinations. Extraction Solvents At this point in our work only one solvent, chloroform, works sufficiently well, in terms of the acid - base character of the crown ether, to be of use in analytical determinations.Both toluene and dichloromethane were unsatisfactory because their background blanks were too large, owing to the complicated equilibrium established between the aqueous and non- aqueous acid - base chemistry of the chromogenic crown ether. Extraction Constants and Stoicheiometry As outlined previously, an experiment in which the pH and the metal concentration in the aqueous phase were kept constant whilst the crown ether concentration in the organic phase was varied was performed in order to determine the extraction constant and the stoicheio- metry. The equations used to define and evaluate the system have been developed previ- 0 u s 1 y .~ ~ According to equation (1) where E is the absorbance of a standard divided by the concentration of the crown ether, E" is the blank value of E , [HL]t is the total crown ether concentration, [L-II, is the concentra- tion of the deprotonated crown ether in the aqueous phase and [M+]& is the concentration of the alkali metal in the aqueous phase, a plot of the left-hand term versus E - E" should produce a straight line with a slope equal to AE, the apparent molar absorptivity for the system (taking into account all complexed species ML and ML.HL), and with the y-intercept equal to the extraction constant K&L. Table I1 shows the data for the extraction constants, which suggest that sodium and caesium ions will not interfere with the determina- tion of potassium ion but that rubidium ion will.Fortunately, as rubidium has a low natural occurrence it will not be a problem in real samples. From the same experiment we can determine the stoicheiometry of the complex. Equation (2) shows that a plot of log Qm versus log [HL]. will give a straight line whose slope is equal to the stoicheiometry of the complex: - AAlAe Qm = metal distribution = - W+laJune, 1981 TRIFLUOROMETHYL-SUBSTITUTED CHROMOGENIC CROWN ETHERS TABLE I1 EXTRACTION CONSTANTS OF 6TF AND 4 -PICRYLAMINOBENZO-15-CROWN-5 A 1 M TEA buffer solution was used. 639 4'-Picrylaminobenz yl- Ion 6TF 15-crown-5 Na+ . . . . NA* -10 K+ .. . . 7.5 f 0.2 7.5 f 0.2 Rb+ .. . . 8.3 f 0.2 8.5 5 0.2 cs+ . . . . NA -10 * NA = no appreciable extraction. The results for the 6TF compound show the formation of a complex with a 2 : l ratio of ligand to metal for potassium and rubidium.This is consistent with previously reported work using benz0-15-crown-5.~-5 Effects of pH and Shaking Time For chromogenic crown ethers, the pH was found to be crucial as the amine proton must dissociate in the aqueous phase before complexation and remain dissociated during and after the phase transfer. For these extractions, the pH of the aqueous phase after extraction had to be greater than 11, but, as a high concentration of TEA was present in the organic phase, this was easily accomplished. Other buffering systems did not give successful extractions. Fig. 2 shows the effect of shaking time on the equilibrium process of the extraction. We chose 5 min as the standard time as 80% of the eventual extraction had occurred after this time and good precision was observed.Detection Limit and Calibration The extractions were performed as outlined earlier. The reagent blank absorbance was measured and subtracted from all of the readings. Fig. 3 shows the calibration graph for 6TF, which exhibited a 5-700 p.p.m. linear range. Fig. 4 shows the effects of an increase in sodium ion concentration on a standard potassium ion concentration determination. The data show that sodium ion begins to interfere at about 3000 p.p.m. Our experience with real samples indicates that the interference levels may increase or decrease depending on the matrix of the sample, but this method of evaluation gives an adequate indication of inter- ference. Rubidium shows a large interference at a concentration of 1000 p.p.m.Caesium and alkaline earth metals do not interfere. Conclusion Two new chromogenic crown ethers, 4’- (2”,6”-dinitro-4”-trifluoromet hylphenyl) amino- benzo-15-crown-5 and 4’-(4”,6”-dinitro-2”-t~fluoromethylphenyl)aminobenzo-I5-crown-5 I I , , , I I I I 0 5 10 15 20 25 Time/rnin Fig. 2. absorbance for 2 x solution. Relationship between shaking time and maximum M 6TF in chloroform and 0.1 M KCl Chloroform was buffered with 1 M TEA.640 PACEY, WU AND BUBNIS 50 100 200 300 400 500 K+ concentration, P.P.m. Fig. 3. Standardisation for potassium in aqueous solution extracted with 6TF in chloroform which is 1 M in TEA. 6TF concentration, 2 x 10-3 M. have been synthesised. The latter compound showed significant improvement over the earlier chromogenic crown ethers. A calibration graph showed a linear range of 5-700 p.p.m. for potassium ion in the presence of 3000 p.p.m. of sodium ion. The complex con- sisted of a 2: 1 ratio of ligand to metal in the form of an ML.HL adduct. . 1. 2. 3. 4. 5. 6. 7. 8. 0 750 1 500 2 250 3 000 3750 Na+ concentration, p.p.m. Fig. 4. Interference studies on the determination cjf potassium ion in Potassium ion concentrations: A, 200; and the presence of sodium ion. B, 400 p.p.m. References Pedersen, C. J., J. Am. Chem. Soc., 1967, 89, 7017. Takagi, M., Nakamura, H., and Ueno, K., Anal. Lett., 1977, 10, 1115. Nakamura, H., Takagi, M., and Ueno, K., Talanta. 1978, 26, 921. Nakamura, H., Takagi, M., and Ueno, K., Anal. Chem., 1980, 52, 1668. Pacey, G. E., and Bubnis, B. P., Anal. Lett., 1980, 13, 1085. Vandeberg, J. T., Moore, C. E., Cassaretto, F. P., and Posvic, H., Anal. Chim. Ada, 1969, 44, 175. Ungaro, R., El Hag, R., and Smid, J., J. Am. Chem. Soc.. 1976, 98, 5198. Gordon, A. J., and Ford, R. A., “The Chemist’s Companion.” John Wiley, New York, 1972, p. 432. Received November 24th. 1980 Accepted January 19th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600636

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, June, 1981, Vol. 106, pp. 641-646 641 Spectrophotometric Determination of Uranium and Iron Using 2,4- Dinitrosoresorcinol K. Rajamoorthi, S. V. Mohan and G. R. Balasubramanian Fuel Reprocessing Development Laboratory, Reactor Research Centre, Department of Atomic Energy, Kalpakkam 603 102, Tamil Nadu, India A sensitive spectrophotometric method for the determination of uranium has been developed, based on measurement of the colour produced by the reaction of uranyl ions (UOZ2+) with 2,4-dinitrosoresorcinol (DNR) , The method is best suited to the determination of 0.10-2.00 mg of uranium in 50 ml of solution. The effects of many other ions on the coloured complex formed were studied and some interferences were eliminated by carrying out the complex formation in the presence of ethyl acetate.The precision of the method was 0.52% (relative standard deviation) and the accuracy was found to be within this value. The reagent was used to determine uranium and iron simultaneously by measuring the mixed colours formed with DNR. Keywords : Uranium and iron determination ; 2,4-dinitrosoresovcinol reagent ; spectrophotometry ; interfering ions The reaction of 2,4-dinitrosoresorcinol (DNR) with transition metal ions to form coloured chelate compounds was applied to the determination of iron1 and copper, either alone or in a mixture.2 In the present method, DNR is used as the colour reagent for the determination of uranium. Various methods are available for the determination of uranium in milligram3-5 and microgram In most of these methods iron is the main interfering species.When iron is present, therefore, it must be separated from the uranium, or the interference due to iron must be eliminated. In the method described here, iron is simultaneously deter- mined with uranium, by measuring the colour formed when DNR is added to a mixture of the two ions. Experimental Instrumentation using 1-cm cells. Absorbance measurements were made with a Beckman, Model 25, spectrophotometer All pH adjustments were made with an ECIL pH meter. Reagents Standard uranium solution. A standard solution containing about 0.1 mg ml-l of uranium (as U022+) was prepared by dissolving accurately about 0.21098 g of uranyl nitrate [U02(N0,),.6H20] (BDH Chemicals) in 1 1 of water. The solution was standardised volume t rically.9 A standard stock solution containing about 25 p.p.m.of iron was prepared by dissolving accurately about 0.072 61 g of analytical-reagent grade iron( 111) chloride in 1 1 of water containing sufficient hydrochloric acid to suppress hydrolysis. The iron content was determined by the method described by Vogel.lo A 0.08% m/V solution was prepared by dissolving 1 g of sodium carbonate followed by 0.8 g of 2,4-dinitrosoresorcinol in water, and then diluting the solution to 1 1 with water. A 10% m/ V solution of analytical-reagent grade hydroxylammonium chloride in water was prepared. Standard iron solzction. DNR reagent. Hydroxylammonium chloride solution. EZectrolyte solution. Solutions containing various ions. Various electrolyte solutions (0.01 M) were prepared.Several salt solutions were prepared by dissolving the These solutions were used for required mass of salt in an appropriate volume of water. interference studies.642 RAJAMOORTHI et al. : DETERMINATION OF URANIUM AND Analyst, VoZ. 106 Procedure To 30 ml of 0.01 M electrolyte solution in a 50-ml calibrated flask were added 5 ml of a 0.1 mg ml-l uranyl solution, followed by 2 ml of 0.08% DNR reagent. The pH of the solution was then adjusted to 7 and the solution diluted to 50 ml with the same electrolyte solution arid mixed well. A blank solution without uranium was prepared and the absorb- ance of the sample was measured against the blank. A calibration graph was then con- s t ructed. When studying uranium and iron mixtures the absorbances were measured at 370 and 650 nm.Results and Discussion Determination of Uranium The reaction of uranium with DNR was studied in various media (water, perchlorate, sulphate and chloride). Uranium was found to give maximum and constant absorbance in the perchlorate medium. The extent of uranium - DNR complex formation in perchlorate medium depends strongly on the pH of the medium. Complex formation begins at pH 3 and is a maximum at pH 7; at higher pH values it decreases, as shown in Fig. 1. The spectrum of the uranium complex at pH 7 has maximum absorbance in the wavelength range 350-370 nm (Fig. 2A). The complex is formed immediately and reaches maximum intensity after 40 min. The absorbance of the uranium - DNR complex was therefore measured at 370nm after 40min. If the DNR concentration is kept constant a t 4.3 x 10-3 M, the absorbance of the complex increases steadily with increasing uranium concentra- tion.A graph of absorbance versus uranium concentration is linear for uranium concentra- tions in the range 0.84 x 1W5-16.8 x Variations in the DNR concentration were also studied. The absorbance of the complex is largely dependent on the DNR concentration. An interference study was performed by adding various foreign ions to a solution con- taining 20 p.p.m. of uranium in the final 50ml of solution. Table I shows the effects of individual interfering ions; most of the cations affecting the colour intensity even at the 1 p.p.m. level were hydrolysed at pH 7, the optimum pH for colour formation. Thorium, aluminium and magnesium formed precipitates at higher concentrations. To minimise interferences due to cations, 5 ml of various solvents (acetone, 4-methylpentan-2-one and ethyl acetate) were added to the final 50ml of solution.Ethyl acetate in the reaction mixture eliminates interferences due to most of the cations (Table I). The masking agents that were examined in an attempt to eliminate cationic interference did not improve the . M. results. 2 3 4 5 6 7 8 9 1 0 PH Fig. 1. Effect of pH on uranium - DNR complex. 350 400 450 500 550 600 650 700 750 Wavelengthlnm Fig. 2. Absorption spectra of complexes in solution: A, uranium; B, iron; and C, mixture of uranium and iron.June, 1981 IRON USING 2,4-DINITROSORESORCINOL TABLE I EFFECTS OF INTERFERING IONS ON THE ABSORBANCE OF THE URANIUM - DNR COMPLEX Ion added Cations- ~ 1 3 + . .. . Mga+ .. Zr*+.. . . Th4+ . . Cr3+. . . . Mo6+ Sr2+ . . . . v4+ . . . . &a+. . . . Anions- NO,- .. c1- . . . . I- . . . . so,,- . . F- . . . . CNS- .. HP042- . . . . .. . . . . .. . . . . . . . . . . . . . . . . . . . . . . Concentration, Material added p.p.m. Remarks Precipitate was formed even in Ix } the presence of ethyl acetate AWO,):, MgSO, ZrOC1, CrCl, Th(NO,), (NH4) GMo70Z4 SrC1, VOSO, cuso, NaNO, KC1 KI Na,SO, NaF KCNS Na,HPO, Serious interference, but no interference in the presence of ethyl acetate 3: } 10 2 Slight interference Serious interference, even in presence of ethyl acetate 1 100 Slight interference 'g: }No interference 100 Serious interference, even in presence of ethyl acetate 100 Sensitivity One of the most sensitive colorimetric methods for the determination of uranium 643 the the reported by Sandellll is the thiocyanate method.For an absorbance reading of 0.001 in a 1-cm cell, in the organic solvent at 375nm, Sandell reported a sensitivity of 0.09 p.p.m. of uranium. Adopting the same method of reporting sensitivity for an absorbance reading of 0.001, the uranium - DNR method has a sensitivity of 0.03 p.p.m. of uranium. The molar absorptivity of the thiocyanate complex is about 3 x 103 and that of the DNR complex is about 8 X lo3 1 mol-l cm-l. The sensitivity of the DNR method is therefore 2-3 times that of the t hiocyanate method. Precision and Accuracy To study the precision and the accuracy of the DNR method, standard solutions containing three different concentrations of uranium were prepared from analytical-reagent grade uranyl nitrate and five absorbance measurements were made on each standard uranium solution.Table I1 indicates that the relative standard deviations of the precision and accuracy were of the same magnitude. The over-all relative standard deviation of 15 determinations was 0.52%. Simultaneous Determination of Uranium and Iron in a Mixture According to Zayan et aE.,2 the spectra of iron solutions at pH 3-8 show two maxima in the wavelength ranges 360-390 and 640-680 nm, respectively (Fig. 2B). The spectrum of a solution containing 2 p.p.m. of iron and 20 p.p.m. of uranium in the final 50 ml of solution was examined. It showed two bands, a band at longer wavelength due to the absorbance of the iron complex only, and another at shorter wavelength due to both iron and uranium complexes (Fig.2C). In a mixture of uranium and iron, therefore, the iron could be deter- mined by direct measurement. The uranium can then be determined if the contribution to the total absorbance due to the iron complex at the shorter wavelength is known. A , = absorbance of complex in mixture at the longer wavelength (650 nm); A , = absorbance of complex in mixture at the shorter wavelength (370 nm) ; El = molar absorptivity of the iron - DNR complex at 650 nm; If644 RAJAMOORTHI et d.: DETERMINATION OF URANIUM AND A?dySt, vd. 206 E2 = molar absorptivity of the iron - DNR complex at 370 nm ; E , = molar absorptivity of the uranium - DNR complex at 370 nm and then the concentrations of iron and uranium can be determined from the following equations : ..* * (1) .. .. .. A1 Concentration of iron = - El Concentration of uranium = [ A , - (A+)] (&) . . . . According to Zayan et aE.,2 the iron complex does not attain a constant absorbance at the longer wavelengths (650 nm) in less than 48-72 h. Rapid, complete colour development is obtained only by adding reducing agents. However, all reducing agents seriously affect the uranium complex. A solution containing iron, uranium and DNR was therefore adjusted to pH 7 and after 40 min the absorbance at the shorter wavelength (370 nm) was measured (A2). After measurement 1 ml of a 10% m/V hydroxylammonium chloride solution was added to the solution and the pH adjusted to 7. The absorbance was then measured at the higher wavelength (650 nm, Al).From these two absorbance values the concentrations of iron and uranium were determined from equations (1) and (2). TABLE I1 TESTS ON PRECISION AND ACCURACY OF THE METHOD Solution no. 6 7 8 9 10 11 12 13 14 15 The molar absorptivities solutions containing known Uranium/moIl-l x Relative r standard Taken Found Error, % deviation, % 8.40 8.44 + 0.47 8.40 8.37 - 0.36 8.40 8.39 -0.12 8.40 8.45 + 0.59 8.40 8.38 -0.24 } 0.44 } 0.58 ] 0.53 6.72 6.75 + 0.45 6.72 6.73 +0.15 6.72 6.68 - 0.69 6.72 6.69 - 0.45 6.72 6.77 +0.74 10.08 10.06 -0.20 10.08 10.@4 - 0.40 10.08 10.13 + 0.50 10.08 10.06 -0.20 10.08 10.16 +0.79 of the complexes can be determined from the absorbances of amounts of iron and uranium in the same medium used for the analysis of the m>ture.Representative results are given in Table 111. To study the TABLE I11 DETERMINATION OF URANIUM AND IRON IN A SOLUTION CONTAINING BOTH ELEMENTS Uranium(V1) concentration/mol 1-1 x 1 0-5 A I > r Added Found 8.40 8.31 16.80 16.10 12.60 12.6 8.40 8.37 16.8 16.0 12.6 12.0 12.6 12.1 16.8 16.1 16.8 16.7 Iron( 111) concentration/moll-' x Added Found 3.58 3.58 5.37 5.29 3.58 3.60 1.79 1.83 3.58 3.62 2.68 2.60 1.79 1.70 1.79 1.69 0.895 0.994 A \June, 1981 IRON USING 2,4-DINITROSORESORCINOL 645 precision of the method, 20 p.p.m. of uranium and 1 p.p.m. of iron were taken in the final 50ml of solution and ten measurements were recorded at 370 and 650nm. The relative standard deviations of the measurements were found to be o.92y0 for uranium and 1.74% for iron.The method is suitable when the concentration of iron in the solution is up to 10% m/m of the uranium concentration. The effects of several interfering ions were also studied in solutions containing 20 p.p.m. of uranium and 1 p.p.m. of iron, in the final 50ml of solution. The results are given in Table IV. The method can be applied to the determination of uranium in various process TABLE IV EFFECT OF INTERFERING IONS ON THE ABSORBANCE OF A MIXTURE OF THE URANIUM - DNR AND IRON - DNR COMPLEXES Ion added Cations- AP+. . Zr4+ . . Ce4+. . Co2+ Mos+ Cr3+. . Ba2+ Mn2+ Mg2+ Ni2+. . Anions- I- . . NO,- F- . . CNS- HP042- . . . . . . . . .. . . . . . . . . . . . . .. .. . . . . . . Concentration, Material added p.p.m. Remarks .. Al(NO,), 20 Precipitate was formed . . ZrOC1, . . Ce(SO,), } Serious interference } Slight interference . . CO(NO,), * . (NH4)6M07024 . . Crcl, 1 Serious interference . . BaC1, . . MnSO, . . MgSO, . . NiSO, . . Na,SO, . . K I .. NaNO, Error in absorbance was f 10% 2 2 1 100 -l Negligible interference 1:: ) . . NaF . . KCNS :8X }serious interference . . Na,HPO, 100 streams containing iron up to 10% of the uranium content. Iron is the main corrosion product. Various synthetic solutions of uranium, which represent the process sample solutions, were prepared containing iron at 5% mlm of the uranium concentration and various concentrations of chromium, nickel, aluminium, strontium, barium, zirconium, molybdenum and cerium, and TABLE V DETERMINATION OF URANIUM AND IRON IN A SYNTHETIC SOLUTION The final 50 rnl of solution contained 1 mg of uranium and 50 pg of iron.Amount of Error in absorbance, % Solution each ion no. Ions present addedlpg 370 nm 650 nm 1 NP+, Ce4+, 2 NP+, Ce4+, 3 Ni2+, Ce4+, 4 NP+, Ce4+, 5 6.1 6.00 5 5 9.4 10 10 10 10.2 12.1 10 10 15.4 14.3 20646 RAJAMOORTHI, MOHAN AND BALASUBRAMANIAN were studied by the procedure described. It was observed that the errors in the absorbances of these solutions increased with increasing concentrations of seriously interfering ions (Tables I and IV). Table V summarises the results. The method avoids the disadvantages associated with the use of tin(I1) chloride as the reducing agent to eliminate the interference of iron in the t hiocyanat e met hod. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References El Khadem, H., and Zayan, S. E., Anal. Chem., 1962, 34, 1382. Zayan, S. E., Issa, R. M., and Maghrabi, J. Y., Microchem. J., 1973, 18, 662. Currah, J. E., and Beamish, F. E., Anal. Chem., 1947, 19, 609. Crouthanel, C. E., and Johnson, C . E., Anal. Chem., 1952, 24, 1780. Nietzel, 0. A., and DeSesa, M. A., Anal. Chem., 1957, 29, 756. Yoe, J. H., Will, F., 111, and Black, R. A., Anal. Chem., 1953, 25, 1200. Burtenenko, L. M., and Polucktov, N. S., Zh. Anal. Khim., 1968, 23, 700. Singh, N. S. B., Mohan, S. V., and Balasubramanian, G. R., J . Radioanal. Chem.. 1979, 52, 319. Cherry, J., P. G. Report, 1968, 827 (W). Vogel, A. I., “Textbook of Quantitative Inorganic Analysis,” Third Edition, Longmans, London, Sandel, E. B., “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience, New 1961, p. 898. York, 1958, p. 917. Received December 29th, 1980 Accepted January 27th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600641

出版商:RSC    

年代:1981

数据来源: RSC

摘要:

Analyst, June, 1981, Vol. 106, pp. 647-652 647 Determination of Sub-microgram Amounts of Selenium in Geological Materials by Atomic- absorption Spectrophotometry with Electrothermal Atomisation after Solvent Extraction R. F. Sanzolone and T. T. Chao U S Geological Survey, Box 25046, Denver, Colo. 80225, USA An atomic-absorption spectrophotometric method with electrothermal atomisation has been developed for the determination of selenium in geo- logical materials. The sample is decomposed with a mixture of nitric, perchloric and hydrofluoric acids and heated with hydrochloric acid to reduce selenium to selenium(1V). Selenium is then extracted into toluene from a hydrochloric acid - hydrobromic acid medium containing iron. A few micro- litres of the toluene extract are injected into a carbon rod atomiser, using a nickel solution as a matrix modifier. The limits of determination are 0.2- 200 p.p.m.of selenium in a geological sample. For concentrations between 0.05 and 0.2 p.p.m., back-extraction of the selenium into dilute hydrochloric acid is employed before atomisation. Selenium values for reference samples obtained by replicate analysis are in general agreement with those reported by other workers, with relative standard deviations ranging from 4.1 to 8.8%. Recoveries of selenium spiked at two levels were 98-108%. Major and trace elements commonly encountered in geological materials do not interfere. Arsenic has a suppressing effect on the selenium signals, but only when its concentration is greater than 1000 p.p.m. Nitric acid interferes seriously with the extraction of selenium and must be removed by evaporation in the sample-digestion step.Keywords : Selenium determination ; geoanalysis ; geological materials ; carbon rod atomic-absorption spectrophotornetry ; solvent extraction Selenium in small amounts is widely distributed in geological materials, with an estimated crustal abundance of 0.05-0.075 p.p.rn.,ls2 and is generally concentrated in the sulphides. Selenium is useful as an indicator for several types of mineral deposits, including uranium - vanadium - copper - silver - molybdenum deposits of the “red bed” sandstone type, certain pitchblende - selenide deposits, gold - silver selenide deposits and various types of poly- metallic ores, especially those containing copper, mercury, bismuth and silver.3 Methods available for the determination of selenium are unsuitable for geochemical exploration purposes.They entail either a lengthy procedure, poor sensitivity or interference problems arising from the diverse and complicated nature of the matrices. Therefore, a rapid, sensitive and interference-free method for the determination of selenium would be of great value for geochemical prospecting work. The proposed method involves digestion with a mixture of nitric, perchloric and hydro- fluoric acids, after which the sample is heated with hydrochloric acid to reduce the selenium to selenium(1V). Iron solution and hydrobromic acid are then added and the selenium is extracted into toluene. For selenium levels of 0.2 p.p.m. and above, in the sample, the toluene layer is used for determination by atomic-absorption spectrophotometry with electro- thermal atomisation.For concentrations between 0.05 and 0.2 p.p.m. , back-extraction into dilute hydrochloric acid prior to determination is required. Experimental* Apparatus A Varian AA6 atomic-absorption spectrophotometer equipped with a Model 63 carbon rod atomiser, a simultaneous background corrector, a selenium electrodeless discharge lamp and * Use of trade-names in this paper is for descriptive purposes only and does not constitute endorsement by the US Geological Survey.648 SANZOLONE AND CHAO: SELENIUM IN GEOLOGICAL Analyst, Vol. 106 a Model 53 automatic sample dispenser were used. The instrument settings were as follows: wavelength, 196.0 nm; lamp, 8 W; slit width, 1.0 nm; nitrogen flow meter, 8.0; drying voltage, 6.0 (about 90 "C) for 25s for toluene solution and 6.5 (about 100 "C) for 30 s for aqueous solution; ashing voltage, 6.0 (about 400 "C) for 10 s; and atomising voltage, 8.5 (about 2500 "C) for 2 s (voltages are arbitrary settings, not volts).Reagents Iron solution, 10% m/V. Dissolve 10 g of Specpure iron powder in 53 ml of concentrated hydrochloric acid while heating on a hot-plate, then add 5 ml of 50% hydrogen peroxide. Heat to remove the excess of peroxide and dilute to 100 ml with water. The residual acidity of this solution is approximately 1 M as hydrochloric acid. Dissolve 4.95 g of Ni(N0,),6H20 in 50 ml of water, add 0.83 ml of concentrated hydrochloric acid and dilute to 11 with water.Dissolve 2.190 g of Na,SeO, in 50 ml of water, add 10 ml of concentrated hydrochloric acid and dilute to 1 1 with water. Nickel solution, 0.1% m/V. Nickel - hydrochloric acid solution, 0.1% m/V nickel in 0.01 M hydrochloric acid. Dissolve 0.495 g of Ni(NO,),.GH,O in 100 ml of water. Selenium stock solzltion, lo00 g ml-l. Selenium standard solutions, 1 and 10 g ml-1 of selenium in 1% hydrochloric acid. Procedures Solvent extraction for samples containing down to 0.2 p.9.m. of selenium Weigh 0.50 g of the rock, soil or stream-sediment sample (less than 100 mesh) into a 50-ml PTFE beaker. Moisten with 1-2ml of water and add 1Oml each of concentrated nitric, perchloric (70%) and hydrofluonc acids. Place the beaker on a hot-plate pre-set at 150 "C under a perchloric acid hood.Gradually raise the temperature over approximately 1.5 h to 220 "C. Remove the beaker from the hot-plate when the volume of the acids is reduced to between 1 and 2ml. Transfer the liquid into a 25 x 150mm tube using two 2.5-ml portions of 3 M hydrochloric acid. Prepare selenium working standards by micro-pipetting 0, 0.1, 0.5, 1, 3, 5 and 10 pg of selenium from the selenium standard solutions into 25 x 150 mm tubes containing 5 ml of 3 M hydrochloric acid. (Working standards need not be carried through the digestion step as no difference in selenium recovery has been observed as a result of digestion.) Place the tubes with samples and standards in an aluminium heating block pre-set at 100 "C for 5 min. After cooling to room temperature add 10 pl of the iron solution, mix, add 10ml of Concentrated hydrobromic acid and mix again.The final solution is equivalent to 1 M in hydrochloric acid and 6 M in hydrobromic acid, and contains 1000 pg of iron. Pipette 1 ml of the toluene extract into a 16 x 150 mm tube and cap the tube. Carefully remove the remainder of the toluene layer with an Eppendorf pipette and discard. Add a further 2nd of toluene, shake for 5 min and centrifuge. Again pipette out 1 ml of the toluene extract, combine it with the first extract and mix. Selenium in the organic extract is determined using a carbon rod atomiser. A 5-pl volume of the nickel solution is pipetted manually into the atomiser just prior to the injection of either a 5- or 2-pl aliquot of the toluene extract by the automatic sample dispenser.Add 2 ml of toluene, shake for 5 min and centrifuge. Start the atomisation cycle and record duplicate or triplicate readings. Back-extraction for samples containing between 0.05 and 0.2 9.p.m. of selenium Samples and standards (0.025, 0.1, 0.5, 1, 3 and 5 pg of selenium) are treated in an identical manner to the above until after the first centrifugation. At this point, transfer the entire toluene layer into a 16 x 150 mm screw-capped tube. Add 2 ml of toluene to the sample and standard solutions, shake for 5min and centrifuge. Remove all of the organic extract with a pipette and combine with the first extract. Pipette 1 ml of the nickel - hydrochloric acid (0.01 M) solution into the combined toluene extract, shake for 5 min and centrifuge. Transfer the aqueous back-extract with a pipette into the automatic sample dispenser for selenium determination.Hydrochloric acid with concentrations varying from 0.01 to 0.1 M is equally effective in back-extracting selenium.June, 1981 MATERIALS BY AAS AFTER SOLVENT EXTRACTION 649 When using the proposed method, the procedure may be interrupted after the addition of the 3 M hydrochloric acid. Once the solution has been heated for reduction, the analysis should be completed on the same day. A reagent blank should be carried through the entire procedure to check for possible contamination. Selenium standards in toluene are stable for approximately 10 h and selenium standards in nickel - hydrochloric acid solution are stable for at least 72 h. Thirty samples can be analysed for selenium per 8-h day.Results and Discussion Sample Decomposition The acid digestion employed is a simplified version of the procedure recommended by B a j ~ . ~ The use of hydrofluoric acid in the digestion is necessary for decomposing selenium minerals occluded in or coated by siliceous materials. Digestion with nitric acid - perchloric acid in the absence of hydrofluoric acid yields lower results (compare column 2 with column 5 in Table I). The digest should not be taken to dryness as a loss of selenium is possible through volatilisation. The digestion step is simple and efficient, producing a liquid residue that can be easily transferred from the PTFE beaker. Reduction of Selenium For selenium to be extracted into toluene from the hydrobromic acid - hydrochloric acid medium, it must exist in the +4 oxidation state. After the digestion with the mixed acids, the selenium does not have a well defined ~ a l e n c y .~ Strong reducing agents may reduce the selenium beyond the +4 state to the free elemental form. Hydrochloric acid has been used extensively at various concentrations and varying heating times and temperatures to ensure proper reduction .5-10 Heating the liquid residue from the mixed-acid digestion for 5 min with 5 ml of 3 M hydro- chloric acid in an aluminium heating block pre-set at 100 "C is a convenient procedure for reduction of selenium to the +4 state. Selenium occurs in the -2,0, +4 and +6 oxidation states. Solvent Extraction of Selenium Selenium( IV) can be extracted quantitatively from either hydrochloric or hydrobromic acid solution into toluene at certain specific concentrations.ll It has been established in this investigation that selenium(1V) can also be extracted into toluene from a solution of 1 M hydrochloric acid and 6 M hydrobromic acid.Iron, although not extractable into toluene, has been found to exert a synergistic effect on the extraction of selenium. A 1 OOO-pg amount of iron (corresponding to 0.2% of iron in a 0.50-g sample) added to solutions containing two levels of selenium results in an average extraction of 95% for two extractions. This compares with an extraction efficiency of less than 50% without iron (Fig. 1). Greater amounts of iron (up to 30%) in the sample have no observable additional effect on' the extraction of selenium. The sequence of adding iron solution first and then hydrobromic acid in the procedure is important, otherwise incomplete extraction of selenium will result. Mixing iron solution and hydrobromic acid before addition to the sample solution nullifies the enhancement effect of the extraction of selenium.Matrix Modification The addition of nickel solution together with selenium to the carbon rod atomiser greatly enhances the absorbance signals of selenium, as shown in Fig. 2. This effect may be due to the formation of a stable nickel selenide compound, which reduces the loss of selenium during the ashing stage and increases the efficiency of the atomisation.12,13 A similar enhancement of absorbance is obtained for the aqueous back-extracted selenium when the nickel is present in the 0.01 M hydrochloric acid solution.Sensitivity The sensitivity (1% absorption) of the carbon rod for selenium in either toluene or aqueous solution has been determined to be 6 x 10-l2g. When using the selenium standards in toluene, a linear relationship exists between absorbance and amount of selenium: 0.1-5 p g of selenium using 5-p1 ali'quots and 3-10 pg of selenium using 2 - 4 aliquots (A1 and B1 in650 SANZOLONE AND CHAO: SELENIUM IN GEOLOGICAL Analyst, VoZ. 106 Fig. 1. Synergistic effect of Fe (1 000 pg) on the extraction of Se from 1 M HC1- 6 M HBr into toluene. Al, 6 pg of Se with Fe; A2,0.5 pg of Se with Fe; B1, 5 p g of Se without Fe; and B2, 0.5 pg of Se without Fe. Fig. 2). These amounts of selenium correspond to 0.2-20 p.p.m.in the actual sample. For selenium levels between 20 and 200 p.p.m., a 1 + 9 dilution of the organic layer may be made. For levels greater than 200 p.p.m., a sample size smaller than 0.50 g should be used. To determine selenium concentrations lower than 0.2 p.p.m., back-extraction of the selenium into aqueous nickel - 0.01 M hydrochloric acid solution is necessary. A linear relationship exists between absorbance and amount of selenium in the aqueous back-extracted standards : 0.025-1 pg of selenium using 10-pl aliquots and 0.25-5 pg of selenium using 2-pl aliquots. These amounts of selenium correspond to 0.05-10 p.p.m. in the actual samde. O.’ I “1 0.5 Q) 2 0.4 + a 2 0.3 0.2 0.1 0 2 4 6 8 10 Selenium content/pg Fig. 2. Enhancement effect of Ni on the absorbance of Se in toluene.Al, 5-pl aliquots with Ni; A2, 5-p1 aliquots without Ni; B1, 2 - 4 aliquots with Ni; and B2, 2-pl aliquots without Ni.June, 1981 MATERIALS BY AAS AFTER SOLVENT EXTRACTION 651 Interferences There is no interference to the selenium determination from major and trace elements commonly encountered in geological materials. These include 30% of iron, 25% of calcium, 20% of aluminium, 10% of manganese, magnesium, potassium or sodium and 2000 p.p.m. of antimony, bismuth, cadmium, cobalt, copper, lead, nickel, silver, tin, tellurium or zinc, which may be present in the sample either individually or in various combinations. Back- ground absorption caused some elevated readings for blank and low levels of selenium, but the background corrector eliminated this problem.Arsenic suppresses the selenium signal, but only when its concentration in the sample is 1000 p.p.m. or higher (Fig. 3). It apparently hinders the extraction of selenium into toluene under the experimental conditions, as arsenic added together with selenium to the carbon rod atomiser causes no suppression. Samples containing more than 1000 p.p.m. of arsenic should be analysed using a smaller sample size (less than 0.50 g). Nitric acid causes serious interference by preventing the extraction of selenium, and must therefore be removed by evaporation during the digestion step. 0.5 0) 0 0.4 2 0.3 % a 0'7 I - A - - 0.6 1 I o*2 I B I I I 1 I I I 1 0 800 1 600 2 400 3 200 4 000 Content of arsenic in sample, p.p.m. Fig. 3. Effect of As on Se absorbance.A, 1 pg of Se; and B, 5 pg of Se. Results for Geological Samples The proposed method was applied to six US Geological Survey geochemical reference samples, GXR-1-GXR-6,l4 and to a glass standard, GSE.15 The results were compared with those reported in the literature (Table I). The results for selenium obtained by the proposed method are in general agreement with the reported values. Replicate analyses for selenium of eight geochemical samples with varying matrices gave relative standard deviations ranging from 4.1 to 8.8% (Table I). Four samples were spiked with two levels of selenium prior to sample decomposition. Recoveries ranged from 98 to 111% with an average of 103% (Table 11). Conclusion The method for the determination of selenium described in this paper is rapid, sensitive, interference-free and suitable for a wide variety of geochemical samples.The detection of selenium at crustal abundance (0.5-0.075 p.p.m.) is important in that selenium anomalies hitherto undetected by previous less sensitive methods may be disclosed. The synergistic662 SANZOLONE AND CHAO TABLE I REPLICATE ANALYSES OF SELENIUM IN VARIOUS SAMPLES AND COMPARISON WITH REPORTED VALUES HNOa - HCIO, - HF digestion a- > Relative Mean (n = 5), standard Sample p.p.m Range, p.p.m. deviation. % GXR-1 (jasperoid) . . 19.4 18.0-20.7 6.1 GXR-2 (soil) 0.57 0.54-0.62 6.4 GXR-3 (Fe - Mn deposit)' ' 0.19) 0.17-0.20 6.1 GXR-4 (Cu mill head) . . 6.28 5.95-6.50 4.1 GXR-6 Isoil) ._ .. 0.91 0.88-0.97 4.7 GXR-5 (soil) .. .. 1.03 0.96-1.08 5.9 TCOT-~O (hematite GSE (glass) .. . . 9.61 3.31-3.73 4.9 gossan) . . . . . . 0.139 0.12-0.15 8.8 HNOI - HClOd p.p.m.* digestion, ,- 15.4 <0.2 0.20 5.6 0.23 0.40 Reported values, p.p.m. Ref. 16 Ref. 1Tt Ref. IS$ -i 18.6 f 0.8 15.52 17.0 0.74 f 0.11 0.48 0.8 0.22 f 0.02 0.10 0.3 6.0 f 0.05 5.22 5.8 1.1 * 0.1 0.80 1.1 1.07 f 0.13 0.90 0.9 By back-extraction. TABLE I1 RECOVERY OF KNOWN AMOUNTS OF SELENIUM ADDED TO VARIOUS SAMPLES Each result is the average of duplicate analyses. Sample Presentlpg Added/pg GXR-2 (Soil) . . .. . . 0.30 0.20 GXR-4 (Cu mill head) . . . . 3.10 2.00 0.30 0.40 3.10 4.00 GXR-5 (soil) . . .. . . 0.54 0.50 0.54 1 .oo 79-0T-46 (tuff)* .. . . 0.03 0.05 0.03 0.10 By back-extraction. Found/pg Recovery, yo 0.49 98.0 0.69 98.5 5.31 104.1 7.90 111.3 1.07 102.9 1.58 102.8 0.08 100.0 0.14 107.7 effect of iron in the extraction of selenium into toluene is a unique property that doubles the selenium extraction.Arsenic causes a suppression of selenium values, but only at concentra- tions in the sample greater than lo00 p.p.m. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. References Lakin, H. W., and Davidson, D. F., U.S. Geol. Surv. Prof, Pap., 1973, No. 820, 573. Lee, T., and Yao, C., Int. Geol. Rev., 1970, 12, 778. Boyle, R. W., Geol. Sum. Can. Pap., 1974, No. 74-45, 31. Bajo, S., Anal. Ckem., 1978, 50, 649. Lansford, M., McPherson, E. M., and Fishman, J. J., A t . Absorpt. Newsl., 1974, 13, 103. Mickael, S., and White, C. L., Anal. Chem., 1976, 45, 1484. Kamada, T., Shiraishi, T., and Yamamoto, Y., Tulanta, 1978, 25, 15. Egaas, E., and Julshama, K., A t . Absorpt. Newsl., 1978, 17, 185. Analytical Methods Committee, Analyst, 1979, 104, 778. Belcher, R., Bogdanski, S. L., Henden, E., and Townshend, A,, Artal. Chim. Acta, 1980, 113, 13. Landsberger, S., and Boswell, G. G. J., AnaZ. Chim. Acta, 1977, 89, 281. Ediger, R. D., At. Absorpt. Newsl., 1975, 14, 127. Martin, T. D., Kopp, J. F., and Ediger, R. D., A t . Absorpt: Newsl., 1975, 14, 109. Allcott, G. H., and Lakin, I-I. W., Proc. 5th Int. Geochem. Explor. Symp., 1974, 659. Myers, T., Havens, R. G., Connor, J. J., Coklin, N. M., and Rose, H. J., Jr., US. Geol. Surv. Pvof. Gladney, E. S., Perrin, D. R., Owens, J. W., and Knab, D., AnaZ. Chem., 1979, 51, 1657. Ho, C. L., Mineral Studies Laboratory, University of Texas at Austin, personal communication. Crenshaw, G. L., and Lakin, H. W., J. Res. U.S. Geol. Surv., 1974, 2 , 483. Pap., 1976, No. 1013. Received December 22nd. 1980 Accepted January 14th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9810600647

出版商:RSC    

年代:1981

数据来源: RSC