ISSN:0003-2654    

DOI:10.1039/AN98207FX001

出版商:RSC    

年代:1982

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98207BX003

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

January, 1982 SUMMARIES OF PAPERS I N THIS ISSUEDetermination of Trace Amounts of Nitrite by Derivatisation andGas ChromatographyixA gas-chromatographic method for the determination of nitrite is described.Nitrite is converted into substituted benzene derivatives by reaction withsubstituted anilines and liypophosphorous acid, and the resulting benzenederivatives are determined by gas chromatography. The derivatisationyields of nitrite were studied by using anilines with various substituents thathave a high response to an clectron-capture detector. Nitroanilines givehigh yields and w-nitroaniline is tlie most suitable derivatisation reagent fordetermining trace amounts of nitrite. By using nz-nitroaniline and anelectron-capture detector, the detection limit and determination range ofnitrite are 0.5 ng ml-1 and LIP to 1.00 pg ml-l, respectivcly, which are muchlower than those of the widely used colorimetric method for the determina-tion of nitrite (detection limit 16 ng ml-1). River water samples containingnitrite were arialysed by both this g-as-chromatographic method and thecolorimetric method.Keywovds ; Nitvite detevmination ; tvace analjlsis ; gas chvonzatography ;derivntisation.KOICHI FUNAZO, KENJI KUSANO, MINORU TANAKA and TOSHIYUKISHONODepartment of Applied Chemistry, Faculty of Engineering, Osaka University,Yamada-oka, Suita, Osaka 565, Japan.Analyst, 1982, 107, 82-88.Determination of Germanium in Coal Ashes by Hydride Generationand Flame Atomic - absorption SpectrophotometryThis paper describes the determination of germanium by atomic-absorptionspectrophotomctry with direct iiitrocluctioii into a dinitrogen oxide - acetyleneflame of the germanium( 1 1 7 ) iiydride generated by reducing with sodiumtetrahydroborate(II1) solution.-1 comparative study of the sensitivity ofthe germanium determination lias been carried out and a study of interferencesfrom lead(II), arsenic(II1) and -(I7), iron(I1) and -(111), tellurium(IV),selenium(IV), antiniony(III), tin( I I ) , tartrates, oxalates and fluorides isdescribed. The sensitivity of the proposed method is 0.012 pg ml-l andthe detection limit is approximately 100 times better than obtained withsilica tube atoiiiisatioii a t a wa\-elengtli o f 265.14 nm. The method has beenapplied to tlie tletermiiiation of gcrmanium in lignite ashes with an averagegermanium recovery of 9 7 .O 5 O c , and a relative standard deviation of 2.87 yo.KBjlwoids : Hydride genevation ; diiaitvogeu2 oxide - acetylene flame ; atomic-crbsovption spectvopl~otoiiietv~l; geviiinniuiii deteviiiination ; con1 ashesJ. R. CASTILLO, J. LANAJA and J. AZNAREZDepartment of Analytical Chemistry, Science Faculty, University of Zaragoza,Zaragoza, Spain.Analyst, 1982, 107, 89-95X SUMMARIES OF PAPERS I N THIS ISSUEDetermination of Mercury Vapour in Air Using ElectrothermalAtomic-absorption Spectrometry with an ElectrostaticAccumulation FurnaceJanuary, 1982The “electrostatic accumulation furnace for atomic-absorption spectrometry”technique has been tested for the determination of mercury vapour in theatmosphere.Even if the electrostatic capture of mercury vapour is muchmore difficult than for particles, an efficiency higher than 90% can be achieved.A calibration procedure is proposed. Under appropriate experimental con-ditions, a detection limit (signal to noise ratio = 3) of 50 ng m--3 was obtained.Keywords : Electrothermal atomic-absovption spectrometry ; electrostatic precipi-tators ; particulate matter analysis ; mercury vapour determinationG. TORSI, E. DESIMONI, F. PALMISANO and L. SABBATINIIstituto di Chimica Analitica dell’Universit2 Degli Studi, Via Amendola 173, Bari,Italy.Analyst, 1982, 107, 96-103.Interactions of Major, Minor and Trace Elements on theCarbon Rod Atomic-absorption Spectrophotometric Determinationof Micro - amounts of Ytterbium, Holmium, Dysprosiumand ThuliumThe electrothermal atomisation of trace amounts of some heavy rare earthelements has been investigated using a carbon rod atomiser.The inter-ferences due to the components (major, minor and trace) of a commonsilicate matrix have been studied in order to obtain the best analyticalconditions for future determinations. The detection limits (using uncoatedgraphite rods) are 5 pg of dysprosium, 3 pg of holmium and thulium and0.4 pg of ytterbium.Keywords : Dysprosiunz, holmium, thulium and yttevbiunz detevminations ;silicate rocks; atomic-absovption spectvophotoutetvy with electrothermalatomisation ; interferencesAMBROGIO MAZZUCOTELLI and MARIO GALL1Istituto di Petrografia, Universiti di Genova, Genoa, Italy.and ROBERTO FRACHEIstituto di Chimica Generale ed Inorganica, Universita di Genova, Genoa, Italy.Analyst, 1982, 107, 104-109.Some Causes of Bias in the Measurement of Dissolved OxygenUsing Certain Modifications of the Winkler MethodShort PapevKeywords : Dissolved oxygen detenizilzation ; Winklev modification ; nitviteintevfevence ; ivon(III) ilztevfevence ; pliospliate modificationN.B. HETHERINGTON and J. HILTONFreshwater Biological Association, Ferry House, Far Sawrey, Ambleside, Cumbria,LA22 OLP.Analyst, 1982, 107, 110-112January, 1982 SUMMARIES OF PAPERS IN THIS ISSUEDifferential-pulse Polarographic Determination of theCoccidiostat Arprinocid in Feed Pre -mixesShort PaperKeywovds Avprinocid deteriiainatiom ; dijffeevential-pulse polarography ; feedpre-mixesJOHN D.STONG and DAVID W. FINKMerck Sharp & Dohme Research Laboratories, P.O. Box 2000, Rahway, N.J. 07065,USA.A.tzalyst, 1982, 107, 113-116.Proton Magnetic Resonance Shifts of the TrimethylsilylDerivatives of Silicate MaterialsCommunicationKeywovds ; Tviiizethylsilylation ; silicate analysis ; A S I S effect ; protonmagnetic vesonanceB. R. CURRELL, H. G. MIDGLEY, J. R. PARSONAGE and E. A. VIDGEONSchool of Chemistry, Thames Polytechnic, Wellington Street, London, SE18 6FF.Analyst, 1982, 107, 117-121.Rapid Determination of Albumin- bound Zinc in HumanSerum by Simple Affinity Chromatography andAtomic - absorption SpectrophotometryCowmunicationKeywords ; Z i n c ; albumin ; a f i n i t y chvowzatography ; atoiiaic-absovfition spectro-photoiizetvy ; electvotheviizal atomzsationJ. W. FOOTE and H. T. DELVESllniversity Department of Chemical Pathology and Human Metabolism, SouthamptonGeneral Hospital, Southampton, SO9 4x1’.Analyst, 1982, 107, 121-124.Technique for Reducing the Cycle Time in Atomic-absorptionSpectroscopy with Electrothermal AtomisationCoinnzunicationKeywovds : A tomic-absovption spectvoscopy ; electvotlievmal atomisatiom ; cycletimeM. H. BAHREYNI-TOOSI, J. B. DAWSON and D. J. ELLISDepartment of Medical Physics, University of Leeds, The General Infirmary, Leeds,LS1 3ES.Analyst, 1982, 107, 124-125.x

ISSN:0003-2654    

DOI:10.1039/AN98207BP007

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

12 Analyst, January, 7982, Vol. 107, pp. 12-16 Physico-chemical Study of Mixed-ligand Selenium( IV) Complexes: Ternary Complex of Selenium( IV) with Alizarin Maroon and Eosin K. A. Idriss, M. M. Seleim, M. S. Abu-Bakr and M. S. Saleh Department of Chemistry, Faculty of Science, University of Assiut, Assiut, Egypt The reaction of selenium(1V) with alizarin maroon (AZM) as a primary ligand and eosin as a secondary ligand was examined spectrophotometrically and potentiometrically a t 20 f 0.1 "C and an ionic strength of 0.1 M (perchloric acid). The solution spectra of the mixed-ligand complex formed is character- ised by an absorption band with h,,,. a t 560 nm within the pH range 6.5-7.2. The pink selenium - (AZM), - (eosin), association complex conformed to Beer's law over the concentration range 0.16-2.0 pgml-1 of selenium with a molar absorptivity of 2.5 x lo4 1 mol-l cm-l.Keywords : Selenium(I V ) mixed-ligand complexes ; alizarin maroon; eosin ; spectrop hotowetry ; potentiounetry Considerable work has been done on the formation of association complexes using fluorescein compounds as counter anions.l-ll The analytical applications of the ternary complexes of some of the transition and rare earth elements have been i n ~ e s t i g a t e d , l ~ ~ ~ ~ ~ 7 ~ ~ but little is known about the ternary complexes of selenium(1V) .ll In continuation of our work on the binary complexes of amino-substituted anthraquin- ones,12-14 we have studied the ternary complexes of alizarin maroon (3-amino-l,2-dihydroxy- anthraquinone) (AZM) with selenium( IV) using eosin as a secondary ligand.Selenium is usually determined by fluorimetry after complexation with diaminonaphthalene or diamin~benzidine,~~ but neutron-activation analysis,16 gas chromat0graphy,~7 colori- m e t r i ~ , ~ * ~ ~ ~ atomic-absorption spectroscopic20v21 and electrochemical have also been used. This paper describes spectrophotometric and potentiometric studies of the mixed ligand complex of selenium(1V) with alizarin maroon and eosin. The use of the coloured complex formed for the micro-determination of selenium( IV) has been investigated. The stability constant of the selenium(1V) - AZhl - eosin complex has been determined. Experimental Reagents preparations of all solutions. standard solutions. All chemicals were of analytical-reagent grade and doubly distilled water was used for the The solutions were diluted as necessary to prepare working AZM stock solution, 5 x M.Prepared from Merck reagent. Eosin stock solzdion, 5 x M. Prepared from Merck reagent. Sele?ziunz(lV) stock solution, M. Prepared using AnalaR sodium selenite and standard- The solution was diluted as necessary to prepare standard working ised as re~omrnended.~~ solutions. EDTA stock solution, 10-1 M. B@er solzttions. Sodium hydroxide standard solution 0.1 PIT. Perchloric acid staizdard solution 0.028 M. so dim^ pevchlorate staiidard soldion 1 .O M. Solzitioizs of ioizs. Prepared from BDH Chemicals reagent. Buffer solutions of pH 3-9 consisting of boric acid, borax, succinic acid and sodium sulphate were prepared as described by B r i t t ~ n .~ ~ Solutions of the diverse ions used for the interference studies were pre- pared by dissolving the calculated amount of each compound in doubly distilled water in order to give 1-10 mg ml-l of the particular ion.IDRISS, SELEIM, ABU-BAKR AND SALEH 13 Apparatus photometer in the range 350-650 nm using 1-cm matched silica cells. used as a blank solution. ated calomel - glass electrode system. the ionic strength of all solutions measured was adjusted to I = 0.1 (sodium perchlorate). The absorption spectra of solutions to be tested were recorded on a Unicam SP 8000 spectro- An aqueous buffer was pH measurements were carried out using a Radiometer pH-meter, Nodel 28h, with a satur- 0.1 “C and All measurements were performed at 20 Procedure An aliquot of a standard solution of selenium(I1’) containing 3.95-79.0 pg of selenium was introduced into a 25-ml calibrated flask then 1.0 nil of 10-1 11 EDTA solution and 0.5 ml of The pH was adjusted to pH 7, 5 ml of lop4 nr eosin were added and the solution was diluted to volume uith doubly distilled water.After thoroughly mixing the reaction mixture, the absorbance was measured at 560 nm against a reagent blank similarly preparcd but containing no selenium. For examining tlie effect of interfering ions, solutions of such ions were pipetted first, followed by the selenium(IT’), EDTA, AZM, buffer and eosin solutions. 11 AZM solution were added. The above procedure was followed. Results and Discussion Absorption Spectra The visible spectra of AZRl exhibit an absorption band at about 480-500 nm within the pH range 6.5-7.5.M) shows an absorption band at about 510 nm (Fig. 1) and there are no significant changes in colour or in the absorption spectrum in the presence of seleniurn(I1’) or ilZRI However, the solution containing AZRI and eosin undergoes a change in colour, from orange - yelloLv to pink, when mixed with 2.5 ml of 10F M selenium(1V). The spectrum of tlie reaction mixture against a blank solution containing the same concentra- tion of the two ligands shows an apparent decrease in the absorption at 510 nm and exhibits a new band at 560 nm. The latter is presumably due to the formation of a mixed-ligand com- plex of selenium(TV). The maximum colour development for the ternary system was attained at pH 6.5-7.2.I t is worth mentioning that the visible spectrum of a mixture containing eosin and selen- ium(1V) exhibited no absorption when the solution was scanned vcysus eosin. Eosin solution (1 x 31). 0.8 0.7 a, 0.6 ,$ 0.5 + g 0.4 a 0.3 u I] 0.2 0.1 0 380 460 540 620 700 Wavelengthinm Fig. 1. Absorption spectra of Se(IV) - AZN - eosin ternary complex in aqueous solution. (1) AZN; (2) eosin, Se - eosin, .\ZAI - eosin (all give essentially the same spectrum) ; (3) Se - eosin uevsus eosin ; (4) Se - ilZJI - eosin uevsus S e ; (5) Se - AZM - eosin uevsus AZilI - eosin. Effect of Masking Agents and Diverse Ions had no effect on the sensitivity of the method. The addition of EDTA as a masking agent in up to a 2500-fold molar excess over seleniurn(1V) The effect of diverse ions a t levels of 1.0-14 IDRISS et al.: PHYSICO-CHEMICAL STUDY OF Analyst, Vol. 107 14mg per 25 ml on the determination of selenium was studied for a sample containing 27.63 pg of selenium(1V) by the recommended procedure. Each ion was tested individually. The solutions were also 4 x 10-3 M in EDTA. There was no interference from 14 mg (about a 500-fold excess) of Li+, Na+, Ba2+, Pb2+, Th4+, U023+, C1-, I-, NO3-, SO,2- and HP0,2-, 8 mg (about a 300-fold excess) of Mg2+, Ca2+, Mn 2+ , Fe2+, Co2+, Cu2+, Zn2+, As3+, Br-, C0,2-, Cr,042-, NO2- and The presence of about 1 mg of aluminium(II1) caused a positive error corresponding to about 10 pg of selenium(1V). The interference due to aluminium(II1) was eliminated by increasing the concentration of EDTA to 1 x 10-2n~.The maximum amount that can be tolerated is about 2.5 mg. Of the anions investigated, cyanide caused a serious negative error even when present in only a 10-fold excess. The presence of about 1 mg of cyanide seems to prevent any reaction between selenium( IV) and AZM. Experiments to determine the stoiche- iometry of the reaction of cyanide with a solution containing AZM and eosin were inconclusive. or 4.0 mg (about a 100-fold excess) of Cr3+, Ni2+, F- and S2-. 0.7 0.6 s 0.5 * 0.4 C m Calibration Graphs and Reproducibility The molar absorptivity was 2.5 x lo4 1 mol-l cm-l. selenium(1V) concentration of 1.4 x cedure and their absorbances were measured. deviation of 0.003 absorbance unit. The system followed Beer’s law over the range 2 x 10-6-2.5 x M of selenium(1V).Ten identical samples, each with a final M, were treated according to the recommended pro- The mean absorbance was 0.35, with a standard - - - - Stoicheiometry of the Complex Job’s method of continuous v a r i a t i ~ n ~ ~ ~ ~ ~ was applied to establish the composition of the ternary complexes under investigation. The molar fractions of two of the components were varied continuously, keeping their combined concentration constant, and keeping the third component in a large constant excess for all solutions in the series. Under these conditions the ternary system was modified to a quasi-binary system. The results shown in Fig. 2 indicate that the over-all selenium(1V) - A211 - eosin composition is 1 : 2 : 2. The stoicheiometry of the ternary system was also determined by applying the molar ratio method27 (cf., Fig.3). In the formation of this complex, the first ligand (AZM), on entering B I I I I I I I I I I 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 Molar fraction of Se(lV) (A, B) or eosin (C) Fig. 2. Job plots: (A) eosin - Se(1V) (in excess of XZN), varying ratios of 1 x RI eosin and Se(I1’) plus 5 ml of bufier arid 1 nil of 31 A i Z I I per 25 nil; (B) A211 - Se(I\-) (in excess of cosin), x-arying ratios of 5 x lop4 31 A211 and Se(I\’) plu4 5 nil of buffer and 5 nil of 5 x lo--‘ 31 eosin per 25 nil; and (C) eoqin - LIZM [in excess of Se(I\-)], varying ratios of 5 x lo-* RI eosin arid A U I I plus 5 1111 of buffer and 5 rill of 5 x 51 Se(I\?) per 25 ml. 0.1 I .//”’ i I 4 I 4 0 1 .o 2.0 3.0 Molar ratio of (A) eosin to Se and (B) AZM to Se Fig.3. 3Iolar ratio plots: (,A) vari- ation of eosiii against constant [Se(I\-)], 0.5-12 nil of JI eosin added to 4.0 ml of 31 %(I\’), 1.0 nil of 10-3 31 AXZSI, 5 rill of buffer per 95 nil; (B) variation of A I Z l l against Se(I\-), 0 . 5 4 nil of 5 x 10-4 31 Se(IV), 7.0 nil of 5 x 1 0 - 4 31 eosiii and 5 nil of buffer per 25 r i l l .January, 1982 MIXED-LIGAND SELENIUM(IV) COMPLEXES 15 the coordination sphere of selenium(IV), fully satisfies it, but does so in a purely dative fashion, so that the complex ion still bears the over-all positive charge of the original central ion, (O=Se)2+, and is free to ion associate with a second ligand (eosin) of suitable anionic charge to form a ternary complex. Potentiometric Measurements Potentiometric titration of the following mixtures (total volume made up to 50ml) was studied: (A) 5 ml of 0.028 M perchloric acid + 5 ml of 1 M sodium perchlorate solution, (B) mixture A + 5 ml of 5 x M AZM solution, (C) mixture B + 2.5 ml of 5 x 10-3 M selenium- (IV) solution, (D) mixture B + 5 ml of 5 x M eosin solution and (E) mixture D + 2.5 ml of 5 x 31 selenium(1V) solution, all against carbon dioxide-free sodium hydroxide solution (10-1 31).The concentrations were No = 0.1 M, Eo = 0.0028 M, TC,o = 0.0005 and TC,o = 0.00025 M (symbols have their usual meanings). The titration curves obtained are shown in Fig. 4. 2' I 0 0.5 1 .o 1.5 2.0 Volume of NaOHiml Fig. 4. p H - titration graphs for Se(1T') - AZM - eosin. For identification of graphs see text. The titration curves were used to evaluate '?za (average number of protons associated with the ligand A4ZXf), ?i (average number of ligand AZJ1 molecules attached per metal ion) and pL (free ligand exponent).From these data, the values of the proton - ligand and metal - ligand stability constants were obtained (Table I ) . From curves A, B and C in Fig. 4, it can be seen that complete formation of Se - AZM species occurs at about pH 5.3, and they remain stable at higher pH values. In the mixture TABLE I IONISATION COXSTANTS OF AZM AND STABILITIES OF ITS BIXARY ASD TERXARl- COMPLEXES WITH SELESIU).I( IF7) ,4211 . . . . . . . . pK, = 9.1 pK, = 7.1 ~ B B = 16.2 Se(IV) - AZM . . . . . . Log K , = 6.9 Log K , = 6.0 LogB = 12.9 Se - AZhI - eosin . . . . . . Log K , = 11.3516 IDRISS, SELEIM, ABU-BAKR AND SALEH containing seIenium(1V) and the two ligands (curve E), the selenium(1V) - AZM complex is formed first (at low pH), then subsequently associates with the eosinate anion to form the mixed-ligand complex in a stepwise manner. The composite curve (curve D) is not super- imposable with the mixed-ligand titration curve (curve E), thereby confirming the formation of a selenium(1V) - AZM - eosin complex.The horizontal distances between curves D and E were measured and used for the calculation of %mix. {average number of secondary ligand molecules attached per [Se(IV) - (AZM),I2+ ion} using the equation - , . - - (1) . . (KJ-V,) (NO + EO) nmix. = ( Vo + V A ) fiA TCSe - AZW where V , and V , are the volumes of alkali consumed to reach the same pH value in curves D and E of the mixed-ligand system and TCSe.AZMo is the total initial concentration of the selenium - A214 complex, which is equivalent to the initial selenium( 1 1 7 ) concentration taken in mixture C; all other symbols have their usual meanings.28 Values of iiA a t different pH values were available from the data for the binary complexing system.From the values obtained for firnix., the free ligand exponent, pL,ix., was calculated using the equation where B: is the proton-ligand stability constant, B is the metal-ligand stability constant, TCLo is the total concentration of the ligand and n is the number of protons released from the ligand molecule. By plotting %mix. against PLmix., the formation curve was obtained and the formation con- stant of the ternary system was evaluated.1. 2. 3. 4. 5. 6 . 7. 8. 9. 10. 1 1 . 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. References Bailey, B. W., Dagnall, R. M., and West, T. S., Talanta, 1966, 13, 1661. Dagnall, R. M., El-Ghamry, M. T., and West, T. S., Talanta, 1968, 15, 1353. El-Ghamry, M. T., and Frei, R. W., Anal. Chem., 1968, 40, 1986. El-Ghamry, M. T., and Frei, R. W., Talanta, 1909, 16, 253. Poluektov, N. S., and Sandu, M. A., Zh. Anal. Khim., 1970, 25, 1510. Tananiko, M. M., and Gorenshtein, L. I., Ukr. Khim. Zh., 1974, 40, 275. Tananaiko, M. M., and Gorenshtin, L. I., Izv. Vyssh. Uckebn. Zaved. Khim. Khinz. Tekhnol., 1975, 18, Tananiko, M. M., and Bilenko, N. S., Zh. Anal. Khim., 1975, 30, 689. Idriss, K. A., Awad, A., Seleim, M.M., and Abu-Bakr, M. S., Anal. Chim. Acta, 1980, 116, 413. Idriss, K. A., Seleim, M. M., and Abu-Bakr, M. S., Proc. Indian Acad. Sci., 1980, 89, 519. Idriss, K. A., Seleim, M. M., and Abu-Bakr, M. S., Mikrochim. Acta, 1980, 11, 179. ldriss, I<. A., Issa, I. M., and Seleim, M. M., J . ,4ppl. Chem. Biotechnol., 1977, 27, 549. Idriss, I<. A,, Seleim, M. M., and Khalil, M., Monatsh. Chem., 1978, 109, 1383. Idriss, I<. A., Seleim, M. M., and Khalil, M., Cuvv. Sci., 1979, 48, 343. Olson, 0. E., J . Assoc. Off. Anal. Chem., 1969, 52, 627. Kronborg, 0. J., and Steinnes, E., Analyst, 1975, 100, 835. Shimoishi, Y . , Analyst, 1976, 101, 298. Sawicki, E., Anal. Chem., 1957, 29, 1376. Osburn, I<. L., Shendrikar, A. D., and West, P. W., Anal. Chem., 1971, 43, 594. Vijan, P. N., and Wood, G. R., Talanta, 1976, 23, 189. Verlinden, M., Baart, J., and Deelstra, H., Talanta, 1980, 27, 633. Bound, G. P., and Forbes, S., Analyst, 1978, 103, 176. Meites, L., “Handbook of Analytical Chemistry,” McGraw-Hill, New York, 1963. Rritton, H. T. S., “Wydrogcn Ions,” Fourth Edition, Chapman and Hall, London, 1952. Job, P., Ann. Chim. (Rome), 1928, 10, 113. Shirif, I;. G., and Awad, A. M., Inorg. Nucl. Chem., 1962, 24, 79. Yoe, G. H., and Jones, A. L., Ind. Eng. Chern. Anal. E d . , 1944, 16, 111. Irving, H. XI., and Rossotti, H. S., J . Chem. Soc., 1953, 3397; 1954, 2904. 893. Received March 9th, 1981 Accepted May Z W t , 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700012

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, Jan,uary, 1982, Vol. 107, pp. 17-24 17 Automatic Determination of Sulphur Dioxide by a Coulometric Method: Interferences and Reliability of Measurements Francesco Zilio Grandi Uqziversitci degli Studi di J'enezia, Facoltci di Chinzica Industviale, 30123 Venice, Italy Aldo Del Turco Eizte della Zona Indzistviale di Povto LWavglzeva, Dozsoduzo 1056, 30123 Venice, Italy and Marino Ravasio Montedipe S P A , Stabilimento d i Povto Mavgheva, 30175 Porto Mavghera, Venice, Italy The selectivity of filters used in coulometric sulphur dioxide analysers was evaluated for various volatile organic substances that can be present in the atmosphere of an industrial area with different chemical plants. The poor selectivity of chemical filters, mainly with respect to vinyl acetate and styrene, was ascertained by several tests a t various concentrations of the interfering substance.It is possible that the interferences arise from reactions with the bromine present in the analysis cell. The magnitude of the interference appeared to be related to the kinetics of addition of the bromine to unsaturated organic interferents. Pre-heating the air sample to 800 "C resulted in complete elimination of the vinyl acetate interference ; less promising results were obtained with styrene. Practical modifications of air pre-treatment on the analysers were consequently adopted. Keywords : S u l p J ~ u r dioxide determination ; coulometry ; aiv pollution contvol ; interferences On the basis of an evaluation1 of the various systems for the automatic determination of sulphur dioxide reported in the literature, the spectrophotometric method proved to be the most sensitive and interference free.The coulometric method appeared sensitive to oxidising and reducing substances, which therefore needed removal before introducing the air sample into the measuring cell. In spite of this, the need for automatic analysers for incorporation in complex monitoring networks (requiring low maintenance and giving signals processable by a computer) shifted interest again to monitors based on the coulometric defined as equivalent to the US Environmental Protection Agency reference n i e t h ~ d . ~ They have therefore found wide- spread use in the last 10 years.6 Though the use of analysers to control the air quality in urban areas ensures a satisfactory reliability of the measurements,' the situation could be very different in large industrial areas owing to the qualitative and quantitative composition of the atmospheric pollutants.The simultaneous presence in the atmosphere of various substances (such as nitrogen oxides, ozone, chlorine, hydrogen sulphide, thiols, saturated and unsaturated hydrocarbons and their derivatives) required a deeper investigation of the role they play in the redox process used in the coulometric method and a critical evaluation of the filters used for interfering sub- stances, del-eloped by the producers of the monitors. Exceptioiially high sulphur dioxide concentrations in some places and in particular exaniples of chemical pollution recorded by the monitoring stations of a network installed in a wide industrial area, could not in fact be correlated either with significant changes in sulphur dioxide emissions from the industries or with the meteorological conditions.Tliere \vas e\-idence for the presence of interferences in the sulphur dioxide determination during the analytical runs carried out by a mobile laboratory placed near the monitoring stations ; in some instances sulphur dioxide values higher than the total airborne sulphur concentrations value (Stat.) were recorded.18 ZILIO GRANDI et al. : AUTOMATIC DETERMINATION OF Analyst, Vvol. 107 We have therefore carried out a systematic investigation on the interferences on sulphur dioxide measurements in the ambient air in order to find the most adequate systems to eliminate them.Experimental Apparatus The monitors used in this work are parts of a monitoring network installed in an industrial area and in the surrounding urban area.*v9 A mobile laboratory is equipped with NO,, particulates, total hydrocarbons, hydrogen sulphide and total sulphur analysers and integrates the fixed monitoring system.1° The sulphur dioxide monitors of the network are connected by telephone cables to a Philips minicomputer, Model P 855 M, located at the network Control Centre, which converts the electrical signals into instantaneous concentration values and processes them, following suitable routine processes, into average values for 30-min periods. The total sulphur monitor, as well as the other analysers installed in the mobile laboratory, is linked to a Philips microcomputer, Model ECO IV, which processes the instantaneous electrical signals giving average concentration values over 30-min periods.A simple map, showing the location of the network monitors, is given in Fig. 1. Fig. 1. Simple map of the air-pollution control monjtoriiig network of the Industrial Board of Porto Distance between 1Iarghera (I‘enice) .8,9 monitors 1 and 2 = 1 km. 0, SO, monitors ; and 0, SO, plus meteorological monitors. Four coulometric monitors, Philips, Model PIT7 9‘700, equipped with selective filters, either heated silver gauze or “cheniical” type, were used for sulphur dioxide measurements. The “cheniical” filter would appear to consist, on the basis of X-ray diffractornetric analyses conducted by us, of actixTated carbon impregnated with potassium sulphate and natural zeolites and, according to the manufacturer, should be replaced el’ery 3 montlis.A coulometric monitor, Philips, Model P\IT 9755, supplied with a pre-heating furnace, was used for measuring total sulphur content (Stat.), determined as sulphur dioxide.January, 1982 SULPHUR DIOXIDE BY A COULOMETRIC METHOD 19 The concentration range selected for both sulphur dioxide and Stat. was 1.15 p.p.m. full scale. A measurement accuracy of better than 15% of the signal, a reproducibility of better than 1% and a detection level of lower than 4 p.p.b., for both sulphur dioxide and Stat., were quoted by the manufacturer. The calibrations of the monitors were made automatically daily by zeroing with air purified on activated charcoal and measuring the sulphur dioxide concentration resulting from a given amount of sulphur dioxide emitted in a purified air flow by a standard sulphur dioxide source.ll In an attempt to reduce the effect of some interfering substances, the PW 9700 monitor was equipped with a pre-heating furnace, consisting of a quartz tube, i.d.4mm, wound in 14 coils, each having a mean diameter of about 12 mm, and placed inside a thermostatically regulated heating unit. An air flow-rate of 150 ml min-I was used. The PIV 9700 monitor with the calibration apparatus and the pre-heating furnace is shown in Fig. 2. A selective filter was placed down-stream of the furnace. Flush ca pi I la ry Measuring capillary To vacuum pump 1 W Measuring cell Fig. 2. Diagram of the monitor PW 9700 with the calibration apparatusll and the pre-heating furnace.Procedure The average measurements over 30-min periods of sulphur dioxide and Stat. were com- pared at two positions in the industrial area, using sulphur dioxide monitors of the moni- toring network, supplied with “chemical” filters, and the Stat. monitor available in the mobile laboratory placed, for this purpose, near the measuring stations. In order to evaluate the efficiency of the various selective filters, two sulphur dioxide monitors were placed in a given position of tlie network in the middle of the industrial area, and the average values over 30-min periods were processed. The degree of selectivity of the filters for some substances was evaluated by means of sulphur dioxide monitors located at tlie network Control Centre.Calibration mixtures of the substances under investigation were prepared by a static method12 : the actual concentra- tions of the mixtures were controlled by gas chromatography or infrared spectrometry.20 ZILIO GRANDI et aZ. : AUTOMATIC DETERMINATION OF Analyst, VoZ. 107 The calibrated mixtures were sucked in by the analyser pump, simultaneously balancing the gas container by introducing purified air. The maximum value of the recorded signal was measured and the concentration of the substance fed in was calculated, taking into account the container volume, the time interval from feed start-up and the amount of air used for dilution. Results and Discussion Our investigation was originated, as already stated, by the high sulphur dioxide values not attributable to appreciable variations in the sulphur dioxide emissions in the industrial area or to particular meteorological conditions.I t was developed in stages as a result of the data reported and discussed in the tables. The Stat. values recorded by the mobile laboratory during analysis runs carried out near two sulphur dioxide monitoring stations of the network 11 ave been compared over three different time periods. The results obtained are reported in Table I. The mean value of the ratio [SO,]/[St,t,] is lower than unity when all the data collected are considered, hut it was found that the ratio becomes higher than unity when the data lower than a pre-fixed concentration [40 p.p.b. (lo9)] is rejected. In this last instance, which is less affected by errors in measurements and differences in instrument calibration, a decrease in the scattering of the ratio around the mean value was obtained.TABLE I COMPARISON OF THE [SO,]/[S,,t.] RATIOS AT GIVEN POSITIONS OF THE NETWORK [SO,]/[St,t.] ratio calculated from all data collected [SO,]/[St,t.] ratio calculated with values greater than 40 p.p.b. Station 7 r 7 number Measuring Standard Number of Standard Number of 4 April-May 0.85 0.50 2210 1.20 0.35 818 A I A (see Fig. 1) period Average deviation tests Average deviation tests 11 July-August 0.80 0.45 1824 1.25 0.35 319 11 November 0.76 0.46 1083 0.97 0.24 316 These anomalies could, in some way, be connected with the fact that the sulphur dioxide monitor is equipped only with a selective filter, while the Stat.monitor is also fitted with a pre-heating furnace suitable for oxidising all volatile sulphur compounds to sulphur dioxide. In order to verify if the anomalies mentioned above arise from the differences in the air pre-treatment, the average data for 30-min periods from two monitors placed a t a given position of the network and equipped with different filters, were collected. The experiment was conducted for 10-15 d, a time period that is long enough to cover different conditions of atmospheric pollution. The results are given in Table 11. TABLE I1 FREQUENCIES OF OCCURRENCE FOR DIFFERENT RANGES OF THE VARIATIONS I N THE RATIO BETWEEN THE DATA RECORDED BY TWO MONITORS PLACED AT A GIVEN POSITION OF THE NETWORK Range of data recorded for ratios of A/R 0.3-0.6 0-0.3 0.6-1.0 1.0-1.5 1.5-2.0 2.0- al Geometric average .. Frequency for A and B monitors equipped with “chemical” filters already in use, 76 23.7 16.3 35.6 5.5 2.0 17.0 . . 0.66 Frequency for A/B,* yo 30.5 22.0 38.6 0.6 0.4 7.9 0.50 Frequency for A/R,?- ?4 74.6 7.9 6.2 0.8 0.7 9.8 <0.10 * Monitor A with “chemical” filter already in use ant1 monitor B with a new 7 Monitor A with “chemical” filter already in use and monitor B with a heated “chemical” filter. silver gauze filter.Janaary, 1982 SULPHUR DIOXIDE BY A COULOMETRIC METHOD 21 For the “chemical” filters already in use (column 2), the geometric average of the ratios was 0.66, i.e., the values recorded by monitor B tended to surpass those recorded by monitor A, with a frequency of 75.6%. After replacing the “chemical” filter with a new one in monitor B, this frequency was expected to decrease, owing to the better efficiency of a new filter in comparison with one already in use and probably exhausted; in contrast, B values, higher than A values (column 3), increased up to 91.1%.This would be, partially at least, imputable to a general decrease in the pollutant concentrations, as it also appeared that a higher frequency of zero values were obtained from monitor A, and mostly to differences in the monitor calibration, which become more important in the low concentration ranges a t which the geometric average of the ratios decreases to 0.5. When replacing the “chemical” filter in monitor B with the silver gauze type, which was not supposed to be as effective (according to the indications of the manufacturer), the absolute value of B data with respect to A data was expected to increase; instead, monitor B went on recording a comparable number of values at concentra- tions lower than A (column a), equal to 88.7% compared with 91.1% as found previously.This could also have been due to the greater number of zero values recorded by monitor A. Therefore, if there is a difference between the “chemical” filter and the silver gauze filter, it did not emerge clearly during the tests. Table I11 reports the selectivity of the Philips, Model PW 9700, monitor including a heated silver gauze filter, as indicated by the manufacturer.ll No further information could be found in the literature. TABLE 111 SELECTIVITY OF THE PHILIPS MONITOR Pw 970011 In terferen t s,* % Nitrogen oxide .. .. <1 Nitrogen dioxide . . .. <5 Ozone . . .. . . <1 Chlorine . . . . . . <2 Hydrogen sulphide . . . . <1 Ethylene . . .. . . <2 Aldehydes . . .. . . <1 Benzene . . . . . . . . <1 Chloroform . . .. . . <1 Carbon disulphide . . .. <1 Methane thiols . . . . About 180 * S = signal from 0.5 p.p.m. of interferentlsignal from 0.5 p.p.m. of sulphur dioxide. Owing to the peculiar situation of the industrial area investigated, an attempt has been In the subsequent tables the interference from a chemical substance will be generally made to widen the knowledge on other possible interferences. expressed by the ratio signal measured for a given concentration of interferent signal measured for the same concentration of SO, x 100 The pollutant concentrations adopted in the tests were often at high levels, which are likely to occur under particular conditions and close to some production plants.The results obtained are given in Table IV. As can be seen, the most surprising fact is the lack of selectivity differences using the “chemical” filter and the silver gauze one, and, above all, the lack of selectivity with respect to the monitor without any filter. Another interesting aspect of the results is the poor selectivity of both filters for vinyl acetate and styrene. The same results were obtained for vinyl acetate and styrene using a wider concentration range and different analysers equipped with filters at various degrees of exhaustion.In Table V the mean values of the percentage interference, using monitors equipped with a silver filter, a “chemical” filter and without a filter, are reported. Mixtures of interferents in air were fed at different concentration values in the ranges indicated in column 1; the22 ZILIO GRANDI et aZ. : AUTOMATIC DETERMINATION OF TABLE IV EVALUATION OF FILTER SELECTIVITY FOR VARIOUS SUBSTANCES Analyst, VoZ. I07 Interfering substance Ethylene . . .. .. .. Toluene . . .. . . . . Vinyl acetate . . . . . . Vinyl chloride . . . . Acrylonitrile . . .. .. .. Hexane . . .. .. .. Nitrogen oxide + nitrogen dioxide Gasoline . . Dichloroethane . . . . . . Dimethylacetamide . . .. .. . . . . .. .. . . .. (2 + 3) Trichloroethylene . . . . . . Acetylene . . .. .. .. Cyclohexanone . . .. .. Acetaldehyde . . .. . . Ammonia . . . . . . . . Styrene . . . . .. .. .. .. . . .. .. .. . . .. . . . . . . . . . . . . .. .. Mean fed concentra- tion, p.p.m. V/V 6.5 0.87 0.16 60 24.9 10 8.4 24.3 8.6 6.2 19.4 7 2.3 11 VSTT VSTT 100 VSTT VSTT 114 93 76 0.84 1.09 Interference, % I m Ag filter 0.9 0.9 145 105 0 0.07 0.1 0.16 0 0 0 1 0 0 0 0 0 0.42 50 “Chemical” filter 0.6 0.8 130 105 0.2 0.3 0 0 0 0.9 0 0 0 0 0 0.34 33 No filter 0.7 0.7 130 105 0 0 0 0 0 ND* ND* 0 0 0 0 0.35 32 * ND = not determined. 7 VST = vapour-saturated air. nuiiiber of mixtures tested is reported in column 2. As can be seen, the selectivity of the t\vo filters is still low, although they show a wider degree of variability. The results, more- over, did not appear to be correlated n-it11 the concentration of interferent fed in.theoretical and experimental investigation on the reactivity of the analysed substances wit11 the bromine contained in the analysis cell in the presence of sulphuric acid was beyond the aims of our work. However, on the basis of kinetic data found in the literature13 on addition of broniine to alkenes, mainly ethylene, halosubstituted alkenes, styrene and its derivatil-cs, it could be reasonably stated that vinyl acetate and styrene interferences are to be attributed to oxidation reactions in the analysis cell. In contrast the absence of interferences, for instance for ethylene and vinyl chloride, could be attributed to the much lo\ver values of the coefficients of the rates of reaction of the bromine addition under the analytical conditions.TABLE IT SELECTIVITY EVALUATION USIXG DIFFEREST ISSTRUJIEXTS AXD FILTERS AT I7ARIOUS DEGREES O F EXHAUSTION Mean value of interference and standard deviation, 96 A Fed concentration Number 7 range, p.p.m. V / V of tests Ag filter “Chemical” filter No filter A . ‘Viizyl acetate- 0.42-1.38 3 33.7 f 1.18 0.17-1.05 6 61.0 & 11.8 0.34-0.S4 3 B. Sty~eize- 0.43-2.10 5 33.6 13.3 0.17-1.34 6 36.7 & 5.16 0.88-1.00 3 47.7 f 17.8 26.0 & 7.13January, 1982 SULPHUR DIOXIDE BY A COULOMETRIC METHOD 23 We consequently tried to eliminate or reduce the interferences of volatile organic substances by submitting the gas sample to a thermo-oxidation treatment, before its inlet to the analysis cell. IVhen using the thermo-oxidation treatment, possible sulphur-containing substances, not retained by the selective filters, are likely to give rise to apparent sulphur dioxide concentrations higher than those actually present in the atmosphere.On the basis of the knowledge of the chemical plants in the industrial area investigated, such occurrence, however, appears very unlikely. A set of tests at different temperatures were thus carried out by placing a pre-heating furnace up-stream of the selective filter and considering the main interferents investigated, that is, vinyl acetate and styrene. The air samples containing the interferents were fed to the monitor for 6 min; after this period, the maximum response of the analyser was obtained. As can be seen in Table VI, the treatment at 800 “C succeeds in eliminating vinyl acetate interference.For styrene, at the same furnace temperature, it appears that the interference could not be eliminated completely. TABLE VI EFFECT OF THERMO-OXIDATION ON VINYL ACETATE AND STYRENE INTERFERENCE ON DETERMINATION OF SULPHUR DIOXIDE Concentration of Concentration Furnace interfering substance, recorded as temperature/ “C p.p.b. V / V SO,, p.p.b. V / V Interference, % A . V i n y l acetate- 25 657 580 792 800 955 B. Styvene- 800 800 760 1180 990 888 14 114 306 151 112 1.5 15 26 Conclusion The influence of chemical interferences on the coulometric determination of sulphur dioxide, even in a complex industrial area, can be considered in most instances as being negligible and not affecting the reliability of the data collected. In particular positions and for very short periods, this statement, however, does not hold true.As a consequence of our findings, although on all analysers of the network the “chemical” filters have been replaced with the heated silver gauze filters, which show comparable performance and require lower operating costs, in some positions, where interfering substances are involved, the analysers have been equipped with a pre-heating furnace. In these instances the total or partial elimination of organic interferents was preferred to the possible higher sulphur dioxide values resulting from non-retained sulphur compounds. The first data obtained from the new arrangement of the analysers in the monitoring net- work agree with our expectations. 1 2. 3. 4. 5. 6. 7. 8. 9. 10. References Carcassoni, B., and Zilio Grandi, F., “L’Automazione nella Chimica Analitica,” Technicon Sym- Hollon.el1, C. D., Gee, G. Y., and Malaughin, R. D., Anal. Chenz., 1973, 45, 63A. IVerner, 1’. O., “Analysis of Air Pollutants,” John Wiley, New York, 1978, 178. Strauss, IV., “Air Pollution Control, Part 111,” John \Viley, New York, 1978, p. 325. Fed. Regzst., 1971, 36, 6884 and 1976, 41, 26252 and 34105. Syrota, M., Pollztt. Atmos., 1974, 62, 207. Marshall, G. I<., Clem2 A i v (Bvzghton), 1975, 15. Zilio Grandi, F., and Magri, R., I?zqzfzname?tto, Suppl., 1977, 6, 31. Zilio Grandi, F., and Del Turco, A., “Proceedings of an International Seminar on Practical Applica- tions of Moclclling Techniques to Air Pollution Prediction and Control,” Chemical Engineering lristitute of University of Genoa, Gcnoa, Italy, 1979, p. 775. posium, Technicon Italiana, Rome, 1969, p. 295. Zilio Grandi, I?., and Del Turco, A., Inquzvzavne?zto, 1981, 11, in the press.24 ZILIO GRANDI, DEL TURCO AND RAVASIO 11. “Instructions Manual for Sulphur Dioxide Monitors PW 9700/00,” N.V. Philips Gloeilampen- 12. Brief, R. S., “Basic Industrial Hygiene, Training Manual of the American Industrial Hygiene 13. Bamford, C . H., and Tiffer, C. F. H., “Chemical Kinetics,” Volume 9, Elsevier, Amsterdam, 1973, Received April 13th, 1981 Accepted August 3rd, 1981 fabrieken, Eindhoven, 1973. Association, Section 10, Calibration of Air Sampling Instruments,” 1975, p. 104. p. 32.

ISSN:0003-2654    

DOI:10.1039/AN9820700017

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, Janztary, 1982, Vol. 107, pp. 25-29 25 Comparative Study of the Determination of Nitrates in Calcareous Soils by the Ion-selective Electrode, Chromotropic Acid and Phenoldisulphonic Acid Methods D. G. Hadjidemetriou Agricultural Research Institute, Nicosia, Cyprus A comparative study of the determination of nitrates in calcareous soils by the chromotropic acid, plienoldisalplionic acid and nitrate ion-selective electrode metliods \\-as in\-estijiatcd, using 0.03 s copper(I1) sulphate solution as estractant, which, in addition to being a preservative for nitrates, helps filtration and eliminates intcrfercnce by hydrogen carbonates in the ion- selective electrode method. SilT-cr sulpliate, which had no effect on either procedure, can be added to thc estractant for the precipitation of chlorides.Sitrate in soil suspensions was determinccl by the ion-selective electrode method. Sitrites if present were eliminated by acidifying the extract with 1 N sulphuric acid containing sulphamic acid. To avoid reaction between soluble organic matter and sulphuric acid in the chromotropic acid method, reagents were added in two steps with con- tinuous cooling. Sitrites were eliminated as in the ion-selective electrode method. Both methods were faster than the phenoldisulphonic acid method and gave identical results; the correlation coefficient was 0.999 8. Keywovds : Nitvate determination ; ion-selective electrode ; chromotropic acid method ; phenoldisulplzonic acid method ; calcareous soils The various methods1-18 for the determination of nitrates in soil can be divided into three groups: ~pectrophotometric,~-3 distillation2 and ion-selective e l e ~ t r o d e .~ - ~ ~ ~ l j ~ l ~ ~ ~ * More rapid and accurate methods are needed to replace the lengthy classical phencldisulphonic acid1*2 and distillation2 methods which suffer from interferences from other ions.l12 A rapid spectrophotometric method is that with chromotropic acid used originally for water1* and later for soil a n a l y ~ i s . ~ The nitrate ion-selective electrode has been extensively uscd, even though there are interferences from other ions.5-7~11~13~15~16 The rapidity and the good accuracy achieved using this electrode4-l3J59l8 have made it suitable for use in routine analysis and in soil agrochemical r e s e a r ~ h . ~ ~ ~ ~ , ~ ~ T'arious ~ o r k e r s ~ - ~ ~ ~ l ~ J s have used different extraction solutions in the ion-selective elec- trode method, depending on the soil being analysed. The most important are watcr,5-s~10~18 potassium sulphate,12 aluminium ~ u l p h a t e , ~ copper(I1) sulphate,6 calcium hydroxide5 and copper sulpliatc(I1) with aluminium and silver resins.13 The main purpose of this work was to investigate the application of the nitrate ion-selective electrode to calcareous soil suspensions.For comparison, the chromotropic acid mcthod was modified for the analysis of soils containing calcium carbonate. These methods lvcre com- pared with each other and with the phenoldisulphonic acid method. In addition, ways of eliminating interfering ions were studied. Experimental Apparatus An Orion, I\.Iodel 93-07, nitrate ion-selecti1.e electrode with a 1 x 2 sensing module con- struction, and an Orion, hiodel 90-02, double-junction reference electrode fitted on a Radio- meter, Model PH1162, pH meter were used.The outer chamber was filled with 0.04 M ammonium sulphate solution and the inner chamber with Orion 90-00-02 solution. A Bauscli and Lomb Spectronic 70 spectrophotometer was used in the phenoldisulphonic and chromotropic acid methods.26 HAD JIDEMETRIOU : DETERMINATION OF NITRATES Analyst, VoZ. 207 Reagents All reagents were of analytical-reagent grade, except for phenoldisulphonic and sulphamic acids, which were of laboratory-reagent quality. Phe?aoldisulphonic acid method The solutions were those used by Jacks0n.l Chrornotropic acid method Dissolve 2.5 g of CuS0,.5H20 in water and dilute to 1 1.Dissolve 0.184 g of chromotropic acid, sodium salt, in 100 ml of concentrated sulphuric acid. This solution was stable for 2 weeks when kept in the dark. Dissolve 7.2184 g of potassium nitrate, dried for 2 h at 105 O C , in 1 1 of water. Dilute the stock standard solution wit11 0.02 N copper(I1) sulphate solution. Standard solutions were prepared in the range 1 to 100 mg 1-1 of NO,-N in 0.02 N copper(I1) sulphate solution for the method using tlie ion-selective electrode. Coppcv(II) sztlplznte solzttion, 0.02 N. Clzvomoti~opic acid, 0.1%. Stock stnridnrd izitrnte SoIzitioii, 1000 mg 1-1 of NO,-N. l170~~king stnudnrd nitffntc solzttioizs, 0-3.5 mg 1-1 of NO,-N. Procedures Phe~~oldisztI$lzoii ic acid method The phenoldisulphonic acid method was used as directed by Jacks0n.l Clzifomotropic acid method \iTeigli 10 g of air-dried soil, add 50 ml of 0.02 N copper(I1) sulphate solution, shake the mixture for 15 niin and filter it through a double Wliatman No.42 filter-paper. Transfer a 3-nil volume into a 25-ml calibrated flask and cool in ice. Add 1 ml of O.l(”’ chromotropic acid solution drop by drop, ccol for 3-5 min and then swirl the mixture. Add 6 ml of concentrated acid, cool for 2 min, mix and lePLve at room temperature for 45 min for the colour to dm-elop. Standard solutions and blanks were suhjectcci to tlie same treatment. If filter- papers give coloured solutions with copper( 11) sulphate solution, they must be washed with distilled water and dried before use.Read at 430 nni in a 1-cm cuvette. The colour produced is stable for at least 2 11. Iou-sclcctiTlc elcctrodc method 15 nrin and leal-e the suspension for 30 niin at room temperature to reach equilibrium. after continuous stirring for 1 niin, read on a Radiometer pFI meter. solutions were measured in the same way. Iiour bccause of drift. smic~ or temperature differenceslG must be taken into consideration. \ITeigli 20 g of air-dried soil, add 50 ml of 0.02 x copper(I1) sulphate solution, shake for Then, Standard nitrate Standard solutions have to be re-read every Tlie temperature of the soil suspensions and standards must be the Results and Discussion Tlit soils used were representative of the major soil type of Cyprus. Table I shows that they liaci pH lxlues of more than 7 .5 , with low chloride values and generally high calcium carbonate and hydrogen carbonate values. In the chromotropic acid method, copper( 11) sulpliate lielpcd filtration and gave clear, colourless solutions. In the ion-selective electrode iiietliod, it precipitated carbonates and hydrogen carbonates and eliminated interferences by tliese icns in soil suspensions.6,8111,16 Siins arid Jackson3 used other estractants in the cliromotropic acid method but, n-itli calcareous soils, filtration was slow and some soils gave coloured extracts. Copper(I1) sulpliatc. showed no absorption at 430 nm, gal-e colourless extracts and standards prepared with this solution gave results identical with those for aqueous solutions. As shown in Table I , the soils used were lo\v in organic matter, but it caused some interference.ToJanuary, 1982 I N CALCAREOUS SOILS TABLE I SOIL TEXTURE AND CHEMICAL COMPOSITION OF THEIR AQUEOUS EXTRACTS Soil: water ratio = 1 : 10. 27 Soil No. 1 2 3 4 5 9 10 11 12 1 3 14 Soil type Sandy clay loam Sandy loam Clay loam Silty clay Clay Sandy clay loam Clay -t Loam Clay loam Clay Clay loam Clay Sandy clay loam PH (1: 2.5 H,O) 7.40 8.10 8.15 8.15 8.15 7.50 7.90 8.15 8.10 7.90 7.90 7.65 7.40 7.50 Organic matter, yo mjm* 1.76 0.42 1.70 1.10 0.99 1.34 0.77 0.71 0.53 0.67 1.44 2.08 0.39 1.34 CaCO,, 96 mlm 5.36 9.06 70.4 22.5 18.3 0 2.76 5.86 29.9 27.8 19.2 21:s 18.5 2.47 Composition/mg kg-' r A > Hydrogen Magnesium Calcium Sodium Potassium carbonate Sulphate Chloride 58 39 19 47 54 44 49 54 19 19 24 32 34 46 344 195 176 253 444 30 260 118 304 155 44 34 272 238 120 136 176 111 208 43 200 90 284 101 156 298 44 48 188 20 88 50 80 25 240 6 34 35 54 165 36 26 1135 77 0 1281 48 0 1403 67 0 1281 29 47 1391 29 128 464 19 0 1269 58 78 0 71 805 903 38 21 805 29 21 964 29 25 927 403 36 549 336 163 488 19 0 * Determined by the Walkley and Black method.l3 t Insufficient amount of soil for determination of type.eliminate any undesirable reaction with soluble organic matter, sulphuric acid and chromo- tropic acid, it was found necessary to add the reagents to the aliquot in a different sequence than that used by Sims and Jackson3 and to cool the solution. Addition of hydrochloric acid to the stock solution increased the sensitivity of the chromotropic acid, but the absorbance did not comply with Beer's law.Without hydrochloric acid the calibration graph was linear and followed Beer's law. Although the soils used were low in chlorides, the addition of 50 ml of extraction solution containing 0.06% m/V of silver sulphate eliminated the inter- ference caused by 650 mg kg-l of chlorides and clarified the solutions. This is shown in Table I1 for the addition of 500 mg 1-1 of chlorides to soils. Chlorides interfere in the determination of nitrate^.^^^^ TABLE I1 EFFECT OF SILVER SULPHATE ON CHLORIDE INTERFERENCE I N THE CHROMOTROPIC ACID METHOD Addition of chlorides/mg 1-l I 1 0 500 500 Xtrate-nitrogenlmg kg-l Soil No. 1 2 3 4 5 6 7 8 9 10 11 12 14 , I 0.02 N CUSO, f 0.02 N CuSO, 0.02 N CuSO, 0.06% m/'V Ag,SO, 208 205 203 19.9 21.3 20.4 33.4 39.3 33.9 17.2 26.8 17.8 12.0 11.7 12.7 3.5 2.0 3.2 6.3 6.4 6.4 7.8 10.4 8.9 7.2 7.6 8.0 6.1 5.3 6.2 14.5 14.1 15.5 3.6 2.1 3.2 126 122 125 Nitrites interfere strongly in the determination of nitrates by the chromotropic acid method.The addition of 0.1 ml of 0.2% m/V sulphamic acid in 0.1 N sulphuric acid to the 3-ml sample solution could eliminate up to 150 mg kg-l of nitrites in 10 g of soil. The special 1 x 2 construction of the sensing module of the Orion ion-selective electrode enabled us to make direct readings in the soil suspensions without filtering. The reasons for choosing copper(I1) sulphate have been Calcareous soils are sometimes difficult to filter.28 HAD JIDEMETRIOU : DETERMINATION OF NITRATES Analyst, VoZ. 107 discussed above, and it also acted as a preservative preventing biological degradation of nitrates.Other concentrations of copper( 11) sulphate were investigated by 0ien and Selmer- Olsen6 and it was found that concentrations of copper(I1) sulpliate liigher than 0.02 N increased the ionic strength and decreased the nitrate concentration. With calcareous soils, 0.03 N copper(I1) sulpliate solution gave tlie same results as 0.02 K copper(I1) sulphate solutioii, and the addition of silver sulphate (0.09% m/V) eliminated chloride interference. U'ith regard to interference by nitrites several w o r k e r ~ ~ ~ ~ ~ ~ ~ have solved the problem by addkg sulpliamic acid to the extractant. In this work sulphaniic acid, n-hicli reacts only in weakly acidic solutions, could not be used with the alkaline extracts.If the soil nitrite is high the suspension is filtered and 3-4 drops of 1 x sulpliuric acid containing the necessary concentration of sulphamic acid are added. In order to detect trace amounts of nitrites semi-quantitatively, hlerckoquant nitrate test papers can be used. Tlie influence of interfering ions3~5-7~11~13-16 has been studied on pure nitrate solutions. However, some ions in soil extracts interfered. The chromotropic acid and ion-selective electrode procedures were compared with the plienoldisulphonic acid method without taking into consideration the presence of low concentrations of anions. Table I11 shows the results of the analysis of soils by tlie three methods. Tables I and I11 sliow that both methods under investigation could be used to analyse soils with low and high concentrations of nitrates and with a wide range of calcium carbonate contents.The ion- selective electrode method was the fastest, followed by the chromotropic acid method ; the phenoldisulphonic acid method took much longer. TABLE I11 SOIL NITRATE-NITROGEN DETERMINED BY PHENOLDISULPHONIC ACID, CHROMOTROPIC ACID AND ION-SELECTIVE ELECTRODE hlETHODS Soil No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Nitrate-nitrogen, mg kg-' A I 7 Phenoldisulphonic Chromotropic Ion-selective 210 208 204 acid method acid method electrode method 18.3 19.9 21.0 29.6 33.4 29.4 15.2 17.2 17.1 9.6 12.0 12.4 3.4 3.5 3.8 4.8 6.3 6.4 6.5 7.8 7.7 6.4 7.2 7.5 4.8 6.1 6.1 11.8 14.5 13.7 57.1 60.3 54.2 3.0 3.6 3.3 126 126 122 The relationships between the three methods are shown in Table IV.There is a very close relationship between the methods ; the correlation coefficients are almost unity, indicating that the phenoldisulphonic acid method could be replaced with the ion-selective electrode or the chromotropic acid method. TABLE IV REGRESSION EQUATIONS AND CORRELATION COEFFICIENTS BETWEEN THE THREE METHODS FOR NITRATE-NITROGEN DETERMINATION Regression equation Correlation coefficient Chromotropic method = 1.92 + 0.99 (phenoldisulphonic acid method) . . . . 0.9998 Iori-sclcctive electrode method = 1.58 4- 0.96 (phenoldisulphonic acid method) 0.999 8 Chromotropic nicthod = 0.31 + 1.03 (ion-selective electrode method) . . . . 0.9996 . . The author thanks hlrs. Chrysoulla Kalli and Mrs. Stella Kourtellari for their assistance.January, 1982 I N CALCAREOUS SOILS References 29 1.2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. Jackson, M. L., Editor, “Soil Chemical Analysis,” Constable, London, 1962, p. 197. Bremner, J . M., i n Black, C. A., Editor, “Methods of Soil Analysis,” American Society of Agronomy, Sims, J . R., and Jackson, G. D., Soil Sci. SOC. Am. Proc., 1971, 35, 603. Myers, R. J . K., and Paul, E. A., Can. J . Soil Sci., 1968, 48, 369. Mahendrappa, M. K., Soil Sci., 1969, 108, 132. (aien, A., and Selmer-Olsen, A. R., Analyst, 1969, 94, 888. Fiskell, J. G. A., and Breland, H. L., Soil Crop Sci. SOC. Flu. Proc., 1969, 29, 63. Milham, P. J., Awad, A. S., Paull, R. E., and Bull, J. H., Analyst, 1970, 95, 751. Smith, G. R., Anal. Lett., 1975, 8, 503. Krupsky, N. K., Alexandrova, A. M., Gubareva, D. N., and Varenik, V. A., Agrokhimiya, 1978, Raveh, A., Soil Sci., 1973, 116, 388. Tchagina, E. G., Dubinina, R. I., Golovin, V. A., Materova, E. A., and Grekovitch, A. A., Agro- Houba, V. J. G., and van Schouwenburg, J . Ch., Editors, “Soil Analysis, 11, Methods of Analysis West, P. W., and Lyles, G. L., Anal. China. A d a , 1060, 23, 227. Bound, G. F., J . Sci. Food Agric., 1977, 28, 501. “Instruction Manual for Nitrate Ion Electrode, Model 93-07,” Orion Research, Cambridge, Mass., Nosko, B. S., Alexandrova, A. M., Gubareva, D. N., and Razdaybeda, V. G., Agrokhimiya, 1980, Goodman, D., Analyst, 1976, 101, 943. Madison, Wisc., 1965, Part 2, p. 1191. No. 10, 133. khii~zi?la, 1980, No. 5, 134. for Soils,” Agricultural University, Wageningen, The Netherlands, 1971, pp. 4.3 and 8.3. 1978. No. 4, 131. Received June lst, 1981 Accepted August 3rd, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700025

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

30 Analyst, January, 1982, Vol. 107, pp. 30-34 Determination of Formaldehyde Vapour in the Atmospheres of Clinical Laboratories Using Chromotropic Acid C. W. Lee, Y. S. Fung and K. W. Fung Dcpavtiizeizt of Chemistry, University of Hong Kong, Pokfularn Road, Hong Kong Ethanol and xylene vapours were found to interfere with the determination of formaldehyde vapour by tlie chroniotropic acid method, whether water or a rnisture of chromotropic and sulpliuric acids were used as the scrubbing solution. The interference can be removed by a porous polymer sorbent, Tenas GC. Thus, \\Tit11 a Tenas GC pre-trap, the chroniotropic acid niethod can be applied to the determination of formaldehyde vapour in clinical laboratories where alcohol and xylciie vapours are often encountered. Tlic advantages of selecting clirornotropic - sulphuric acid instead of water as the scrubbing solution are discussed.Keywovds ; Foviiznldehyde deterininntion ; chroinotvopic a c i d ; spectvoplaoto- iizetvy ; ethanol and xylene interfevence ; Tenax-GC $Ye-trap In clinical laboratories, formalin is commonly used for preserving specimens. Formaldehyde is highly odorous, lachrymatory and pliysiologically active and its threshold limit value is 2 p.p.m. V/V.1,2 I t is therefore necessary to monitor the formaldehyde vapour in these work places. Among tlie various existing methods for determining formaldehyde in air 3-5 spectropliotometry with chromotropic acid (1,8-diliydroxynaplitlialene-3,6-disulphonic acid) has received most attention because of its simplicity, sensitivity and reproducibility.6-* However, tliis method cannot be used directly for determining formaldehyde vapour in tlie atmospheres of clinical laboratories because of negative interferences from ethanol and xylene vapo~rs.7-~ Etlianol and xylene are used in these laboratories as a disinfectant and a microscope-slide cleaner, respectively, and their concentrations may be much higher than tliat of formaldeliyde.Rietliods have been suggested for removing tliese interferents before the spectropliotometric measurement. Evaporation of the scrubbing solution to dryncss to expel anj‘ retained interferents had been suggested by Bricker and Aubrey,lO but the pro- cedure is cumbersome and time consuming. Water has been used as an absorbing medium to reduce tlie extent of the int~rference.~ However, tlie trapping efficiency of water for fornialdellyde \rapour is low compared with that of 0.1% m/V chromotropic acid in concen- tratvd sulpliuric acid (cliromotropic - sulphuric acid).’ Also, when wateI is used as tlie scrubbing solution, a large volume of concentrated sulphuric acid has to be added to bring tlie sulpliuric acid strength to above 86% V/V for optimum colour development.ll This means tliat tliere is a greater dilution (often 3-4 fold) and concomitantly a loss in sensitivity, compared v.itli tlie use of chromotropic - sulpliuric acid as the scrubbing s o l ~ t i o n .~ Tlie possibility of using a porous polymer sorbent in a pre-trap to remove ethanol and xylene \rapours so tliat cliromotropic - sulphuric acid can still be used as the scrubbing solution Lvas tlierefore studied.Tlie investigation showed that Tcnax-GC is a suitable sorbent ;is it can cope wit11 more than the maximum levels of interferents commonly en- countered in clinical laboratories. Experimental Reagents and Apparatus Tlie ivater used for preparing tlie reagents and for dilutions was purified by an ion-exchange purification system (Ikirnstead Dli94). I;ol.11znlil(?li~1dr stoclz stniidnrd solirtioii, 1000 p.p.m. m/V. Prepared by appropriate dilution with Iixter of tlic commcrcially available analytical-reagent grade fornialdelijde solution (E. hlerck), standardiscd by a mctliod publislied The standard formaldehydeLEE, FUNG AND FUNG 31 solutions (0-1.6 p.p-rn.) used for constructing the calibration graph were freshly prepared by dilution of the stock solution with water.Prepared by dissolving 1.0 g of sodium 1,8-dihydroxynaphthalene-3,6-disulphonate (BDH Chemicals Ltd .) in 100 ml of concentrated sulphuric acid (98% m/m analytical-reagent grade) and filtering through a sintered-glass crucible (porosity No. 3). A 0.1% wz/V chromotropic acid solution was then obtained by diluting the 1 yo solution with concentrated sulphuric acid. Tenax-GC (Applied Science Laboratory Institute), Porapak Q (Waters Associates Inc.), Chromosorb G (Varian Aerograph), Chromosorb P (Varian Aerograph) and Chromosorb P (AW, DMCS treated, Pierce) were all 80-100 mesh size; 300 mg of each of these porous sorbents were placed in a 1-cm bore x 6-cm long glass tube, which was then attached to the inlet of the scrubber train. Modified from the one described elsewhere,13 by adding a Teflon stopcock (3-mm bore) at the bottom of tlie flask.This small modification was useful as it facilitated the quantitative transfer of solution with minimum rinsing. Spcctro~lzoto?izeter. Absorbance of the purple mono-cationic dibenzoxanthylium dye formed by tlie reaction of formaldeliyde with cliromotropic acid in concentrated sulpliuric acid medium was measurcd at 5i0 nni in a 1-cm silica cell with a Beckman Acta I11 spectro- photometer . Chromotropic acid in sulpphuric acid, 1% m/V. Porous sorbents. Scrubber. Procedure Constant known concentrations of xylene or ethanol vapour were generated continuously by purging the diffusion tube containing the corresponding compound with a constant flow (200 ml min-I) of purified n i t r ~ g e n .~ " l ~ The vapour concentration was calculated from the flow-rate and the mass loss of the diffusion tube.l49l5 A constant concentration of formalde- liyde vapour was similarly produced from tlie formalin solution (37-44% m/V, analytical- reagent grade, E. llerck) and its concentration n-as determined by measuring the absorbance of the resulting scrubbing solution after a known period of time.' The vapour generating system n-as kept at a teniperature of 25.0 3 0.1 "C in a thermostatically controlled water- bath. The amount of vapour produced could be varied easily by altering tlie dimensions of tlie diffusion path. Chromotropic - sulpliuric acid was used as the scrubbing solution. The vapour of formaldeliyde or formaldehyde plus interferent (s) in purified nitrogen was bubbled through about 15 ml of tlie scrubbing solution, with or without a pre-trap, for a known period of time.Tlie resulting solution was then transferred into a 25-ml calibrated flask and the scrubber container was rinsed with 3.5 ml of water. The rinsing water was then added to the calibrated flask and the colour of the solution was developed under the liberated heat of dilution. The solution was made up to tlie mark with concentrated sulphuric acid after cooling to the ambient temperature. The final acid concentration in the solution was about 86% V / V . Tlie calibration graph was constructed by plotting tlie concentration of formaldehyde against the absorbance of solutions prepared by mixing 21.50 ml of 0.1% chromotropic - sulpliuric acid solution with 3.50 ml of various formaldehyde standard solutions.Parallel analyses were also performed with water as the scrubbing solution in order to assess its efficiency to reduce the interfercnces from xylene and ethanol vapours. The procedure for colour development and calibration when water was used as scrubbing solution has been described e l s e ~ l i e r e . ~ ~ Tlie final sulpliuric acid concentration in tlie resulting solution was about 60% VjV. Results and Discussion Table I shows one advantage of chromotropic - sulphuric acid ovcr water as a scrubber in the absence of interferents. The collection efficiency by a single scrubber was found to be practically lOOq,, for cliromotropic - sulpliuric acid and varied betn.een 76% and 8276 for water in tlie concentration range 0.1-1.5 p.p.m.of formaldcliyde. These results support tliose obtained 11:. Altsliuller rt Tlie tliird and fourth columns of Table I1 compare the applicabilitjr of cJiromotropic acid and water in the presence of ethanol and xylcne. Tlie ncgatii-e interfering effect caused by xylene (\\.hen tlie ratio of the concentration of xylene32 LEE et at?. : FORMALDEHYDE VAPOUR IN THE ATMOSPHERES Analyst, Yd. I07 TABLE I EFFICIENCY OF TRAPPING OF FORMALDEHYDE BY CHROMOTROPIC - SULPHURIC ACID AND WATER Trapping efficiency, yo * r 7 Chromotropic - sulphuric acid Water A Formaldchydc vapour ,-*-----, 1 generated, p.p.m.*t First scrubber: First scrubber Second scrubber Overall 0.12 (0.01) 100.0 (8.9) 82.2 (8.9) 15.6 (4.5) 97.8 (9.9) 0.37 (0.02) 100.0 (4.7) 77.8 ( 5 .2 ) 13.3 (2.2) 91.1 (5.6) 1.53 (0.07) 100.0 (3.1) 75.7 (3.0) 13.0 (1.4) 86.7 (3.3) * The \ ~ ~ l u c s in parentheses are the 2a values of triplicate detcrminations. i';tlcul;~tetl from the concentration of formaldehytlc trapped i n chromotropic - sulphuric acid. S o absorhnce was detected in the solution of the sccond scrubber. to tlic concentration of formaldeliyde was about 70) was reduced to about half if chromo- tropic - sulpliuric acid was replaced by water as tlie scrubbing solution. However, both methods ivcre found to give a negative deviation, to about the same extent, in the presence of etlianol. Thus, even disregarding its 1on.cr sensitivity and collection efficiency, water cannot be used directly as tlic scrubbing solution for measuring formaldehyde vapour in a clinical laboratory.The use of clironiotropic - sulphuric acid as tlie scrubbing solution also lias another advantage hitherto unreported, namely, the required sampling time could easily be judged by noting tlic intensity of the colour developed during sampling (clianging from being straw ye110\17 to being tinged witli pink). Based on tlie above results, tlie use of clironiotropic - sulpliuric acid n ~ s preferred to water in formaldeliyde determination and it was used in the subscqucm t studies. Tlic calibration grapli was found to be linear up to a concentration equi\dent to 3.7 p.p.m. l*/J/- of fornialdc~liyde in a 5-1 air sample with a sensitivity of 0.032 absorbance unit per part p- million of fornialdeli\-de per litre of air.The molar absorpti\.ity of the complex in tile resulting solution was found to be 1.9 x lo1 1 mol-I crn-', which is slightly liiglier tlian tliat reported 11~- Sawicki et ~ 1 . ~ In liistopatliology and cytology laboratories, we found sylene, metlianol ethanol and chloroform \-apoui-s to lie present in the air using a gas-chromatographic technique. JItt1i:inol and chloroform were found not to interfere in the chromotropic - sulpliuric acid nictliocl c \ ~ n at concentrations of about 1000 and 100 times Iiiglier, respectively, than tlie conccmtration of fornialdeliyde \-apour. However, sylene and ethanol were found to cause TABLE I1 EFFECTS O F ISTERFERESTS ON THE CHRONOTROPIC ACID METHOD AKD USE OF I'RE-TR.4PS FOR REDUCISG THEIR EFFECTS 'T:ic c~rperinicntnl coiitlitioiis used for all determinations were identical, i.e., total flow of nitrogen carricr pins formaltleh\-tlc ~cneratetl, 200 in1 inin-' ; and duration of each experiinmt, 1 h..Absorbance*$ Porapak 0.056 Q Chrornosorb Chromosorb Chromosorb G \v P , Chromosorb (A\V, DMCS) GC P Tcnax- 0.090 0.085 0 . 2 6 i 0.283 (0.019) (0.015) (1 1.01 ' 7 ) (0.(105) O . ? i G 0.?PG (I.(J;:\ Il.(l,<5 (0.008) (0.005) I).I);s 0.095 ( l l . l l l 2 ) ( O , ~ l ( i $ ) ( I 1.1 N l j ) (I ).I I(!<) U.'Lb-l u.os.1 ((1.1 I( I>) ((J .I IIJS )January, 1982 OF CLINICAL LABORATORIES USING CHROMATROPIC ACID 33 a negative interference as mentioned previously. Therefore, they must be removed before the determination of formaldehyde when using this method.A pre-trap of polymer sorbent was used for this purpose. The prerequisite for the sorbent is that it must allow quantitative breakthrough of formaldehyde. Se\-eral porous polymers were chosen for this study because of their known low retention indices for forrnaldeli~-de.16 The results of the breakthrough cliaracteristics of formaldehjde vapour in tliese sorhents are given in Table IT. The amount of formaldcliyde breaktlirougli can be taken as tlie ratio of the absorbance of the scrubbed solutions with and without using the pre-trap. Tlius, it was found that although Porapak Q had been reported to be successful in rcmoi-ing some organic interferents in tlie determination of formaldehyde,17 it was in fact unsatisfactory as it rctaincd a significant proportion of tlic formaldcliyde vapour especially at low concentrations.Like\i.ise, tlie breakthrough characteristics of formaldehyde in Chromosorb P were also found to bc unsatisfactory. The efficiencies of the sorbents in the removal of interferents are indicated by the ratio of the absorbance of the solution scrubbed with pure formaldehyde to that scrubbed with formaldehyde plus interferents, both in tlie presence of the pre-trap. Thus, from Chromosorb G, Chromosorb \IT, Chromosorb P (AW, DMCS) and Tenax-GC, which allow quantitative breakthrough of formaldehyde vapour, Chromosorb G and Chromosorb W were found to be unsatisfactory in removing ethanol and xylene vapours as shown in Table 11. Although Chromosorb P (A\\.’, DYICS) could be used to remove xylene effectively, it was only partially successful for removing ethanol.Tenax-GC was found to be effective in removing both xylene and ethanol. A detailed study of the dynamic capacity of a 300-mg Tenax-GC pre-trap was carried out. The results are shown in Table 111. No breakthrough was observed up to 3 4 h with con- centrations of interferents higher than those commonly encountered in clinical laboratory atmospheres. As the normal sampling time for the determination is expected to be less than 1 h, Tenax-GC should be an effective sorbent for the required purpose. Tenax-GC can also be used to remove many other neutral and basic organic compound^.^^^^^ Thus, this modified method will also be useful for the determination of formaldehyde vapour in industrial plant environments where high levels of organic pollutants are expected.TABLE I11 DYNAMIC CAPACITIES OF TENAX-GC TO TRAP XYLENE AND ETHANOL Absorbance with Interferent Time/h 300 mg of Tenax-GC* Xylene, 105 p.p.m. . . . . 1 0.085 (0.005) 2 0.088 (0.003) 3 0.088 (0.006) 4 0.076 (0.006) Ethanol, 551 p.p.m. . . . . 1 0.085 (0.004) 2 0.087 (0.006) 3 0.084 (0.004) 4 0.084 (0.008) * Absorbance of the solution that scrubbed 0.22 p.p.m. V / V of formaldehyde in the absence of the interferent and the pre-trap of porous polymer was 0.088 (0.006). The values in parentheses are the 20 values of triplicate determinations. This work was supported by the Research Grant Committee, University of Hong Kong. References 1. 2. Finklca, J . R., “XIOSH Criteria for a Recommended Standard: Occupational Exposure to Formaldehyde,” US Department of Health Education and Welfare, NIOSI-I publication, NO. 77-126, Cincinnati, 1976.Loomis, T. A., Arch. Pathol. Lab. Med., 1979, 103, 321. 3. 4. \\.’alltcr, J . F., “Formaldehyde,” American Chemical Society Monograph, Reinhold Publishing Knight, H., and Tennant, R. W. G., Lab. Pract., 1973, 169. Corp., New York, 1964, p. 159.34 LEE, FUNG AND FUNG Sawicki, E., Hauser, T. R., and McPherson, S., Anal. Chern., 1962, 34, 1460. West, P. W., and Sen, S., Zh. Anal. Khiwz., 1956, 153, 177. Altshuller, A. P., Miller, D. L., and Sleva, S. F., Anal. Chew., 1961, 33, 621. Ekherg, D. R., and Silver, E. C., Anal. Cliem., 1966, 38, 1421. Still, R. H., Wilson, K., and Lynch, B. W. J., Analyst, 1068, 93, 805. Brickcr, C. E., and Aubrey, \V. V., Anal. Chein., 1930, 22, 720. Olansky, A. D. S., and Deming, S. N., Anal. Chinz. Acta, 1976, 83, 241. I<atz, M., Edztov, “Methods of Air Sampling and Analysis,” Second Edition, -Upha Intcrsociety ASTM, D1607-76, in “Annual Book of ASTAM Standards, Part 26,” American Socicty for Testing Altshullcr, A. P., and Cohen, I. R., Anal. Chem., 1860, 32, 802. “Calibration Standard Notebook, ” Analytical Instrument Development I nc., Avondale, Pa., Chromopack Publication No. 9, Chromopack Co. Ltd., Kederland, B.V., Middelburg, The Nether- Frankel, L. D., Madsen, P. R., Siebent, R. R., and Wellisch, I<. L., A?zal. Chem., 1972, 44, 2401. Misure, J. P., and Dietrick, M. W., J . Chvovnatogr. Sci., 1973, 11, 559. Bertsch, W., Chang, R. C., and Zlatkis, A., J . Chromatop. Sci., 1974, 12, 175. Committce, Ryrd Pre-Press, Inc., Springfield, Va., 1076, p. 300. and Materials, Philadelphia, 1076, pp. 487-492. 1979. lands, 1977. Received June 22nd, 1981 Accepted August 14th, 1981 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

ISSN:0003-2654    

DOI:10.1039/AN9820700030

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, January, 1982, VoZ. 107, pp. 35-40 35 Spectrop hotometric Determination of Cobalt in Paints and Environmental Paint Samples F. Garcia Sanchez, A. Navas and J. J. Laserna Department of Analytical Chemistry, Faculty of Sciences, University of Malaga, Malaga, Spain and A. Arbaizar Hygiene and Work Safety Centre, Malaga, Spain A method for the spectrophotometric determination of cobalt, based on the formation of a coloured chelate (Amax. 480 nm, emax, 2.83 x lo4 1 mol-l cm-l and colour contrast 134 nm) with benzyl 2-pyridyl ketone 2-pyridylhydrazone, is described. Under the optimum conditions, colour development is instan- taneous and the colour is stable for several weeks. Under appropriate working conditions, the method is applicable to the determination of cobalt in manufactured paints and environmental samples of paint.The fundamental solution chemistry of the reagcnt and a. brief description of its more interesting colour reactions are also reported. Keywords ; Benzyl 2 - p y ~ i d y l ketone 2-pyridyllaydvazone reagent ; cobalt deter- mination ; paint analysis ; spectvophotonzetry Many methods have been developed for the spectrophotometric determination of cohalt,l but these 1iaL.c not been applicable to paints. Cobalt compounds are reported to be t o x i ~ , ~ - ~ cau4ng health disorders through ingestion, inhalation and skin contact ; some threshold limit \dues (TLJ7s) for environmental cohaltj are 0.1 mg m-3 (USA), 0.5 mg m-3 (Germany) and 0 . 1 mg ni-3 (Sweden). As the prescnce of paint mists in industrial workroom environments often causes thcse TLI’s to be exceedcd, cobalt levels must be checked periodically; therefore, ~netliods for its determination in paint are needed. Benzyl 2-pyridyl ketone 2-pyridylliydrazone (BPKPH) has been reported recently as being a highly scnsitil-e reagent for the fluorinietric determination of galliumj6 zinc7 and cadmium.* In this paper the fundamental solution cliernistry of the BPKPH and a superficial description of its more interesting colour reactions, reported in detail el~ewliere,~ are given.The results sliow that BPKPH is a promising reagent for the spectrophotometric determination of metal ions. A rapid, simple and selective spectrophotometric method for the determination of cobalt in paints and environmental samples of paint is described here. The method demon- strates the reliability and versatility of the reagent for such purposes and presents an alterna- tive to atomic-absorption spectrometry,1° having a similar sensitivity.Experimental Reagents The synthesis of BPKPH has been reported previously.6 Solutions (1 x 10-3 11) were prepared weekly by dissolving 0.028 8 g of BPKPH in 100 ml of absolute ethanol (Merck, analytical-reagent grade). A stock solution containing 6.904 g 1-1 of cobalt, complexometrically contrasted, was prepared from cobalt (I I) nitrate hexahydrate (UCB, analytical-reagent grade). \Vorking solutions were prepared by appropriate dilution of the stock solution. Bemy1 2-pyridyl ketone 2-pyridyllzydrazone. CobnZt solzition. Acetate b u f e r solzition, p H 4.0, 0.5 31.Jlnskiirg solzition. A solution 1 M in sodium thiosulphate, 0.13 31 in sodium citrate and 1.5 JI in ammonium fluoride was prepared in de-ionised, distilled water from analytical- reagent grade reagents. Apparatus The following apparatus was used: a Beckman Acta I11 spectrophotometer; a Beckman, ~ I o d c l IIRGT, spectrophotometer ; a Beckman, Model SS, Expandomatic pH meter; a Perkin- Elmer, Model 306, atomic-absorption spectrophotometer with a deuterium background36 GARC~A SANCHEZ et al. : SPECTROPHOTOMETRIC DETERMINATION Analyst, Vol. 107 corrector and a Perkin-Elmer, Model 056, recorder; an MSA, Model G, personal sampler with 37-mm three-body cassettes ; and a mixed cellulose ester membrane filter (0.8-pm size and 37 mm diameter) mounted on a cellulose support pad.Procedure Transfer a suitable aliquot (up to 3.75 ml) containing 6.25-75 pg of cobalt into a 25-ml calibrated flask. Add, with mixing, 2 ml of masking solution, 5 ml of 1 x M BPKPH solution, 4 ml of buffer solution and 10 ml of absolute ethanol and dilute to volume with de-ionised water. Measure the absorbance at 480 nm against a reagent blank. Sampling and Treatment of the Paint Weigh and digest with 10 ml of nitric acid and then 10 ml of perchloric acid. Evaporate to dryness and dissolve the residue by heating with 10 ml of 10% V/V nitric acid. Remove the binder by filtration and, finally, dilute the filtrate to 10 ml with 0.2 M sodium hydroxide solution. Environmental samples are collected from paint mists in workroom environments, accord- ing to the NIOSH manua1,ll using cellulose ester membrane filters that are mounted on cellulose support pads connected to vacuum personal samplers, operating at a flow-rate of 1 1 niin-l.Dry the sample for 3 h in a current of air at 105 "C. The paint is then treated on the filter according to the above procedure. Results and Discussion Spectral Characteristics and Acid - Base Properties of the Reagent IYhen measured at different pHs, the absorption spectra of BPKPH solutions (2.5 x 10-5 RI) at different ethanol concentrations showed tlie normal transitions from the doubly protonated form (U) to the singly protonated form (11) and from the singly protonated form to the neutral molecule (N). Fig. 1 shows the spectra obtained when using a 40%) VjV concentration of ethanol.The more strongly basic character of the reagent in the excited state resulted in a bathochromic shift in the absorption bands at the longest wavelengths upon successive protonations of the pyridine nitrogen atoms (Table I). The dissociation constants for the tliree acid - base forms are summarised in Table 11. As expected, an increase in tlie ethanol content of the medium also increases the corresponding pK values. 0 250 300 350 400 Wavelength nm Fig. 1 . Absorption spectra of 2.5 x RI solutions of BPKPH at several p H values arid an cthanol con- centration of 40°& I-/[' shon-ing t h e three acid - base forms deri\-ed from the compound. pH values: (1) 1.2G; ( 2 ) 2.41; (3) 2.56; (4) 4.20; (5) 4.70; ( 6 ) i . G l ; arid ( 7 ) 8.08.January, 1982 OF COBALT IN PAINTS AND ENVIRONMENTAL PAINT SAMPLES TABLE I 37 SPECTROSCOPIC CHARACTERISTICS OF THE ACID - BASE FORMS DERIVED FROM BPKPH I N ETHANOL - WATER MIXTURES Amax./nm Emax-/l mol-l cm-l x Ethanol f--*-, ,--*-------, content, % D M N D M N 2 370 350 325 3.70 2.90 3.60 40 365 350 322 2.20 1.92 2.44 90 345 335 320 2.15 2.05 2.80 TABLE I1 DISSOCIATION CONSTANTS FOR THE DICATION (pK,) AND MONOCATION (pK,) DERIVED FROM BPKPH AT DIFFERENT ETHANOL CONCEKTRATIONS The values given in parentheses are the wavelengths (in nanometres) a t which the spectrophotoinetric titrations were plotted.Ethanol content, yo PKl PK, 2 2.73 (370) 5.55 (325) 40 2.96 (365) 5.85 (322) 90 3.14 (350) 6.20 (320) Main Colour Reactions of BPKPH The reagent behaves in a manner broadly similar to other N-heterocyclic hydrazones, acting as a tridentate chelating agent.It forms intensely coloured complexes with a number of metal ions and in most of tlie chelate systems complexation is accompanied by elimination of the imine proton of the ligand. Under basic conditions, most of metal ions form com- plexes with the anionic form of the ligand that are sparingly soluble in water but solubility is achieved when the medium contains sufficient ethanol. The characteristics of the indi- vidual complexes are summarised in Table 111. The data were obtained from the appropriate spectra, which were measured in the presence of a five-fold molar excess of the reagent a t pH values at which tlie different complexes form. As shown, the high molar absorptivities and the colour contrast values of the reactions make the reagent suitable for practical purposes.TABLE I11 ASALYTICAL CHARACTERISTICS OF THE MAIN COLOUR REACTIONS OF BPKPH All values were measured in ethanol - water mixtures containing 20% V / V of ethanol except those of cobalt for which the ethanol - water mixture contained 60y0 V / V of ethanol. Mn . . Fe(I1) Fe (111) c o . . Xi . . Pd . . Zn . . Cd . . Cu(I1) Ion PH .. . . 8.7 . . . . 11.7 . . . . 11.7 . . . . 4.0 . . . . 10.0 . . . . 11.6 . . . . 11.4 . . . . 1 1 . 1 . . . . 10.0 . . . . 10.5 . . . . 8.0 hmitx./nm 510 407 395 480 490 535 505 485 497 485 455 Emax./l mol-l cm-I Ah*/nm x 10-4 187 84 72 134 167 212 182 162 174 162 132 3.85 3.27 1.32 2.83 4.32 1.45 3.93 5.10 4.61 4.21 0.22 * Colour contrast (Ax) = Amax.(complex) - Amax. (reagent) under identical conditions of measurement.38 BPKPH - Cobalt System Spectral behaviour The complexation of BPKPH with cobalt was studied over a wide pH range. The yellow - orange complex showed maximum absorbance at 480 nm over the pH range 2-13.5, and no substantial changes in absorbance were found over this range. The work reported here was carried out a t pH 4.0 (in the acetate buffer solution). GARCIIA SANCHEZ et a,?. : SPECTROPHOTOMETRIC DETERMINATION Analyst, Vol. 107 Reagent concentration and stability of the complex For complete complexation, a five-fold molar excess of BPKPH is sufficient. Colour development is instantaneous and the colour remains stable for several weeks. Slight variations in absorbance can be found by modifying the concentration of ethanol in the medium.Optimum results are obtained in a 60% V/V ethanol - water mixture. No changes in absorbance were observed when the order of addition of the reagents was modified. Characteristics of the complex range of 0.25-2.5 p.p.m., as determined by a Ringbom plot. found to be 0.0021 pg cm-2. relative error of 0.4% and a relative standard deviation of 0.5% were obtained. Beer’s law as obeyed for cobalt levels of up to 3 p.p.rn., with an optimum concentration Sandell’s sensitivity12 was For a series of ten measurements of 1 p.p.m. of cobalt, a Conzposition of the complex ditions by both continuous-variation and molar-ratio methods. cobalt to BPKPH ratio of 1 : 2. The metal to ligand ratio in the complex was studied under the established working con- The results indicated a Inteifereizces a d application to paint analysis Tlie effect of foreign ions on the determination of 1 p.p.m. of cobalt by the proposed pro- cedure was studied and the results are summarised in Table IIT.Tlie tolerance limit for foreign ions was taken as the amount that caused an error in the absorbance value of no greater than 1%. TABLE IV EFFECT OF FOREIGN IONS ON THE DETERMIXATION OF 1 p.p.m. OF COBALT Ion S,O,2-, C,042-, NH4+, C1-, Br-, F-, NO,-, Pb(II), bIn(II), La(III), Ng(II), alkali Cl0,-, citrate .. . . . . metals, APDC* . . . . . . . . Cr(II1) . . . . . . . . . . . . Cd(II), Hg(I1) . . .. . . . . . . Zn(I1) . . . . .. . . . . . . Ni(II), EDTA . . . . . . . . . . Fe(III), Cu(II), CN- . . . . . . . . Amount tolerated/ pg ml-l > 5 000 1 oot 25 10 2 0.1 0.05 * Ammonium tetramethylenedithiocarbamate (ammonium pyrro- t Maximum amount tested.lidinedithiocarbsmate) . In order to apply the method to the analysis of real environmental samples of paint, prior analyses of a number of them were necessary to determine the ratios of the other metal ions present to cobalt. The reason for this is that usually more than one paint is used in any painting operation. In order to obtain representative values, 40 different environmental samples were analysed by atomic-absorption spectrophotometry, and the results are sum- marised in Table V. The maximum ratios of the main metal ions present to cobalt were 27 : 1 for iron, 10T: 1 for zinc, 1.6: 1 for copper, 22: 1 for chromium and 7 8 : 1 for lead, and according to the preceding interference studies, only iron, zinc and copper were above the toleranceJanuary, 1982 OF COBALT IN PAINTS AND ENVIRONMENTAL PAINT SAMPLES TABLE V 39 METAL IONS PRESENT IN ENVIRONMENTAL SAMPLES OF PAINTS AS DETERMINED BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY Sample No.1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Cobalt concentration/ mg m-3 0.03 0.02 0.03 0.03 0.15 0.10 0.18 0.18 0.12 0.09 0.12 0.11 0.04 0.08 0.06 0.04 0.08 0.12 0.12 0.12 0.05 0.07 0.04 0.07 0.88 0.03 0.07 0.26 0.09 0.19 0.16 0.12 0.12 0.08 0.10 0.09 0.09 0.07 0.07 0.07 R* r A > Fe 6.6 8.0 13.6 4.6 27.0 1.5 1.2 1.5 5.1 6.6 4.8 10.5 16.2 16.0 14.9 20.5 18.0 8.8 13.7 10.5 4.6 4.7 6.5 8.7 2.6 2.6 0.9 5.3 3.8 1.8 2.4 5.7 6.2 8.2 6.1 6.0 1.4 1.9 2.4 1.9 Zn 4.3 8.5 6.6 7.0 0.9 0.8 0.2 0.2 63.6 107.0 0.3 1.0 1.7 1.5 5.7 7.8 4.5 3.5 0.3 0.3 4.0 2.6 5.2 3.7 0.3 0.3 0.3 0.9 2.1 0.9 1.0 2.0 2.8 2.8 1.2 1.2 1.2 2.1 1.9 1.6 Pb 3.3 6.0 3.3 6.6 57.0 5.0 1.0 0.8 2.6 1.8 41.3 46.3 28.5 29.5 38.7 57.8 35.2 24.7 5.5 4.7 14.5 15.4 26.3 23.3 3.2 64.6 77.3 3.1 1.8 22.3 25.1 16.2 18.7 20.5 5.3 3.0 4.1 5.6 4.6 4.7 Cr - - - - 0.4 0.4 0.1 0.1 0.1 0.2 9.1 6.3 6.0 6.8 3.8 5.3 4.4 3.8 0.2 0.1 2.4 1.1 4.0 0.9 0.05 21.7 15.7 0.2 0.4 2.4 2.6 2.6 2.7 3.4 0.9 0.5 0.3 0.4 0.4 0.4 c u 1 .o 1.5 1.3 1.6 0.3 0.6 0.4 0.4 - - - - 0.5 0.4 0.5 0.5 0.4 0.2 0.6 0.4 0.4 0.3 0.2 0.3 0.2 0.2 0.1 0.1 - - - - - - 0.7 0.8 - - - - * R is the ratio of the concentration of the metal ions t o cobalt in the environ- mental samples of paint.limit. The interferences from these elements were eliminated individually as follows : iron was niasked \\.it11 1 in1 of 3 11 ammonium fluoride solution; zinc with 0.25 ml of 1 JZ sodium citrate solution; and copper uith 1 nil of 2 31 sodium thiosulphate solution. However, in de17eloping tlie experimental procedure it was found that the addition of 2 ml of a masking solution containing appropriate amounts of the masking agents was satisfactory. Three synthetic samples each containing the maximum ratios of iron, zinc and copper to cobalt were examined using the proposed procedure. I t was found that the optimum results were obtained after a standing time of 30 min, tliis being recommended for the analysis of paints.The results of the analyses of ten replicate samples of an individual paint are presented in Table 1-1. As shown, the results agree with those obtained by atomic-absorption spectro- photometry, showing the reliability of the present method.40 GARC~A SANCHEZ, NAVAS, LASERNA AND ARBAIZAR TABLE VI DETERMINATION OF COBALT IN AN INDIVIDUAL PAINT SAMPLE Cobalt content*/% x lo2 Sample No. 1 2 3 4 5 6 7 8 9 10 Masslg 0.1867 0.2044 0.142 3 0.1354 0.2250 0.1131 0.181 5 0.2149 0.078 9 0.482 0 Atomic-absorption BPKPH method method 5.96 6.32 6.11 6.31 6.15 6.68 6.00 6.42 6.01 6.04 5.99 6.63 5.99 6.17 5.99 6.05 5.83 5.83 5.97 5.83 * Value based on the non-volatile component of the paint (pigment and binder). Table VII shows the determination of cobalt in environmental samples by the spectro- photometric method, the results being expressed in milligrams per cubic metre of cobalt in the environment.TABLE VII ANALYSIS OF ENVIRONMENTAL SAMPLES OF PAINT 1 . 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Sample No. 1 2 3 4 5 6 7 8 9 10 11 12 Volume*/l Absorbance 70 0.156 76 0.123 70 0.133 70 0.122 84 0.174 84 0.154 65 0.146 65 0.102 53 0.144 53 0.178 45 0.077 68 0.129 * Volume of air sampled. References Snell, F. D., “Photometric and Fluorimetric Methods 1978. DD. 931-1003. Cobalt content/mg m-3 0.20 0.16 0.19 0.17 0.20 0.18 0.23 0.16 0.27 0.34 0.14 0.19 of Analysis,” Wiley-lnterscience, New Yorlc, Plunkett,LE. R., “Handbook of Industrial Toxicology,” Chemical Publishing Co., New Yorlr, 1976. “Hygienic Guide Series. Sax, PI;. I., “Dangerous Properties of Industrial Materials,” Van Nostrand Reinhold, New Yorli, “Cobalto,” Ficha Tecnica No. 9, S. S. Higiene y Seguridad del Trabajo, Madrid, 1978. Laserna, J . J., Savas, h., and Garcia Sknchez, F., Aqtal. Chiin. .4cta, 1980, 121, 295. Garcia SBnchez, F., Navas, A., and Laserna, J . J., Talnizta, in the press. Laserna, J . J., Navas, A., and Garcia SAnchez, F., A izal. I A t . , 1981, 14A, 833. Lascrna, J . J., Thesis, University of Malaga, 1980. Sachclcv, S. Id., Robinson, J . I V . , and IVest, P. \V., .-111nl. Cliiiiz. --icfn, 1967, 38, 499. Sational Institute of Safety and Hygiene, “JIanual of Sampling Data Sheets,” US Department of Sandell, E. B., “Colorimetric Determination of Traces of RIctals,” Third Edition, Interscience, New Cobalt,” American Industrial Hygiene -\ssociatioIi, Alrron, 1966. 1968, p. 577. Health, Education and JVelfare, Cincinnati, 1976. ~ o r i c , 1959, p. 83. Received June 4t12, 1981 Accepted July 15th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700035

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

A n a l y s t , January, 1982, Vol. 107, $@. 41-46 Spectrophotometric Determination of 41 Trace Amounts of Molybdenum with 1,4- D i hyd roxyp ht ha I i m ide D i t hiosem ica r bazone M. Ternero Rodriguez Department of Chemistry, E.T.S.I.I., Univenity of Seville, Seville-12, Spain 1,4-Dihydroxyphthalirnide dithiosemicarbazone reacts with molybdenum(V1) t o produce a yellow 1 : 1 complex in acidic dimetliylformamide - water solution. The yellow complex can be extracted into isopentyl alcohol (A,,,. = 435 nm; E = 9.4 x lo3 1 mol-l cm-l) and used for the spectro- photometric determination of trace amounts of molybdenum in the range 10-93 pg. The interferences of many metallic ions have been examined and a sensitive and selective method for the determination of molybdenum is proposed.Keywolds Molybdenum cleternzinntio?a ; spectvoplzotoiiaetry ; 1,4-dihydroxy- phthaliiizide dit~~ioseinicnvbnzone 1,4-DiIiydroxyphtlialiniide ditliiosemicarbazone (OH-PDT) has been previously described and applied as an indicator reagent in kinetic analysis1$2 and catalytic tit ration^,^^^ based on its autoxidation reaction, catalysed by manganese( 11), in alkaline medium. In this work, OH-PDT has been used as a spectropliotometric reagent and applied to determination of trace amounts of molybdenum. The spectrophotometric characteristics and the development of two sensitive and selective methods (in a homogeneous medium and by extraction with isopentyl alcohol) for the determination of molybdenum(V1) are reported. Of these, thiocyanatej and dithio16 are generally used.More recently, some hydroxylated thiosemicarbazones have been utilised as photometric ligands for m~lybdenum,~-l~ but all of these methods require the presence of tin(I1) chloride in mineral acid medium. However, OH-PDT does not react with molybdenum(V1) under similar conditions. In the proposed method, a coloured chelate is formed in moderately acidic medium, the presence of a reducing agent not being necessary. Moreover, the extraction of the chelate gives a high sensitivity and selectivity, similar to those in other methods for the spectrophotometric determination of molybdenum. The number of cliromogenic reagents available for molybdenum is relatively small. Experimental Reagents All reagents and solvents were of analytical-reagent grade. 1,4-Dilzydro,ryphtlzalinzide dithiosenzicarbazo??e.Prepare a 0.03% m/V solution in dimethyl- formamide. The reagent was synthesised from 1,4-diliydroxyphthalimide dioxime and t liiosemicarbazide. Dissolve 1.500 g of molybdenum(V1) oxide in a few millilitres of 0.1 11 sodium hydroxide solution, dilute with water, make slightly acidic with hydrochloric acid and dilute to 1 1 with water. Dilute this stock solution (1000 g 1-l) just before use. Add 65 ml of 0.2 M sodium Stniidnrd molybdemnz( V I ) solution. Cliloroncctic acid - sodium hydroxide buffer solzition, fiH 2.6. hydroxide solution to 300 rnj of 0.2 distilled water. Apparatus Sped roph o t o 11 z e t e T . used. Dz'gitd PH itzctcr. for pH measurements. A Perk in- E lme r An Orion 701A i JGhloroacetic. k i d solution and dilute to 1 1 with 124 spectrophotometer with 1 .O-cm silica cells was nstrument, with glass - calomel electrodes, was used42 TERNERO RODRIGUEZ : SPECTROPHOTOMETRIC DETERMINATION Analyst, VoZ.107 Procedure Determination of molybdenum( V I ) without extraction To a solution containing 40-225pg of molybdenum(V1) in a 25-ml calibrated flask, add 5 ml of 0.03% rn/V OH-PDT solution and 5 ml of pH 2.6 buffer solution and dilute to the mark with distilled water. Measure the absorbance at 425 nm against a reagent blank. Determination of molybdenum( V I ) with extraction To 1-15 ml of sample solution containing 10-95 pg of molybdenum(VI), in a separating funnel, add 2 ml of 0.03% m/V OH-PDT solution and 5 ml of pH 2.6 buffer solution and dilute to 25 ml with distilled water.After mixing, add exactly 8 ml of isopentyl alcohol and shake vigorously for 1 min. Allow the phases to separate and draw off the aqueous layer. Transfer the organic phase into a 10-ml flask containing anhydrous sodium sulphate. Measure the absorbance at 435 nm against a reagent blank similarly prepared. Prepare a calibration graph by using standard solutions of molybdenum(V1) treated in the same way. Results and Discussion Study of the Complex in Aqueous Dimethylformamide Solution molybdenum(V1) ions produces a yellow complex. shown in Fig. l(A). acid buffer). The addition of a 0.03% solution of OH-PDT in dimethylformamide to a solution of Absorption spectra of the complex are The complex remains stable for at least 3 11 at pH - 3 (chloroacetic Injueme of pH (1, OH-PDT is unstable and the molybdenum complex is destroyed.zleysibs pH curve [Fig. 2(A)] shows a useful working pH range of 2.5-3.5. consisting of chloroacetic acid - sodium hydroxide was adopted for all further work. The influence of pH was studied using a series of solutions in the pH range 1-6. At pH The absorbance A buffer solution 0.7 0.6 0.5 a m 0.4 e 0, 0.3 0.2 0.1 Wavelength 'nm Fig. 1. Absorption spectra of molybdenum complexes with OH- PDT : .%, molybdenum(\TI) complex in aqueous dimethylformamide medium a t pH 3.0; B, molybdenum(Y1) complex extracted into isopentyl alcohol a t pH 3.0; C, molybdenuin(Y) complex in aqueous dimethylformamide medium a t pH 3.0; D and E, reagent blanks of X and B, respectively. Concentration of molybdenum, 5 p.p.m.; and concentration of OH-PDT, 1.84 x n ~ .January, 1982 OF TRACE AMOUNTS OF MOLYBDENUM 43 I I I I I I 1 I I 0 1 2 3 4 5 6 7 PH Fig. 2. Influence of pH on the formation of molyb- denum(V1) - OH-PDT complex: A, in aqueous dirnethylformamide solution a t 425 nm ( C M ~ = 5 p.p.m.) ; and B, extracted into isopentyl alcohol at 435 nm ( C M ~ = 5 p.p.m.). E f e c t of reagent concentration The absorbance of the complex was studied as a function of the molar ratio of OH-PDT to molybdenum(V1). A 2-fold molar excess of the reagent over molybdenum was necessary in order to obtain maximum absorbance at 425 nm. A 5-ml volume of 0.03% m/V solution is recommended as a suitable amount of reagent. When the molar ratio of reagent to metal is higher than 7 and the concentration of molyb- denum exceeds 3 p.p.m., the absorbance decreases and a new complex appears in the solution.The study of the molybdenum(V1) - OH-PDT system under those conditions will be the subject of a future investigation. E f x t of iouic stvength and o d e r of additioiz of reagents The ionic strength of the solution does not affect the absorbance of the molybdenum(V1) - OH-PDT system. The same constant absorbance values were obtained when 1-5ml of 0.5 31 potassium nitrate solution or 1-5 ml of 0.5 M potassium chloride solution were added. The sequence metal ion, reagent and buffer solution was adhered to during the preparation of all measured solutions. Tlie order of addition of reagents was found not to be important. hTatwe of the conqblex The metal to reagent ratio in the molybdenum(V1) complex and the stability constant were determined by the molar ratioll and continuous12 methods (Fig.3). The stoicheiometry was found to be 1: 1 and the stability constant was 2.1 x lo5. A study of the retention of this chelate on an anion-exchange resin showed that under these experimental conditions the complex was anionic. From experimental evidence it was concluded that the reagent forms the yellow complex previously mentioned with molybdenum(V1). The presence of hydroxylamine in the solution before adding the OH-PDT reagent changed the absorbance peak from 425 to 410 nm. When molybdenum('l') was used a green complex was formed [Fig. l(C)]. The presence of potassium persulphate altered the absorption peak because oxidising agents destroy the reagent.Solvent Extraction Study The complex is quantitatively extracted with isopentyl alcohol under the conditions described. Tlie absorption spectrum is very similar to that obtained in aqueous medium and a small batliocliromic shift is observed, from 425 to 435 nm [Fig. l(B)]. The complex is stable for at lcast 4 11. \Yhen isobutyl methyl ketone is used, the stability is higher (24 h) but the44 TERNERO RODRIGUEZ : SPECTROPHOTOMETRIC DETERMINATION Analyst, VoZ. I07 I 0.9} 0.8 - 0.7 - $ 0.6 - 0.5 t (0 - 0, a 0.4 - a 0.3 0.2 - - L a I 0.1 0.3 0.5 0.7 0.9 [Mol [Mo] + [OH-PDT] Fig. 3. Stoicheiometry of molybdenum(V1) - OH-PDT complex in aqueous dimethyl- formamide medium, a t pH 3.0 (continuous variation method) : A, a t 425 nm; and B, a t 440 nm.The initial concentration of molybdenum was 1.54 x M. extraction is not quantitative (approximately 8074,). In both instances, the presence of dimethylformamide in the medium is necessary in order to increase the stability of the extract. The extraction of molybdenum(V1) complex is pH dependent [Fig. 2(B)]. The most favourable pH range is 2 4 . The effect of reagent concentration is similar to that reported in homogeneous medium. When a volume of 1.5-3.5 ml of 0.03% OH-PDT solution was utilised, the absorbance remained constant. The extraction of the molybdenum(V1) complex is slightly decreased if the phase ratio (aqueous to organic phase) is increased because of the appreciable solubility of isopentyl alcohol and dimethylformamide in water. When a phase ratio between 1 : 1 and 1:4 was utilised, the absorbance remained constant.I t is concluded that the volume of the aqueous phase should be held approximately constant. A 25-ml volume is utilised in the recommended procedure. Salts such as sodium sulphate, potassium perchlorate, potassium chloride and potassium nitrate do not affect the colour intensity, even at a concentration of 2%. The extraction is quantitative from solutions of the complex in chloroacetic acid buffer, a salting-out reagent being unnecessary. A 2-ml volume of solution is utilised in the recommended procedure. The influence of the phase ratio was studied. The ionic strength does not affect the extraction of the complex. Spectrophotometric Determination of Molybdenum Based on the experimental work, two methods are reported for the determination of trace amounts of molybdenum involving the formation of the yellow complex with OH-PDT and its extraction into isopentyl alcohol.Spectrofihotonzetric characteristics Direct method. In an aqueous dimethylformamide medium, the molybdenum(V1) - The molar OH-PDT system obeys Beer’s law from 1.0 to 10.0 p.p.m. of molybdenum.January, 1982 OF TRACE AMOUNTS OF MOLYBDENUM 45 absorptivity is 9.0 x lo3 1 mol-l cm-l. A Ringbom plot showed that the range for mini- mum error in the determination is 1.6-9.0 p.p.m. of molybdenum. The relative error (P = 0.05) of the method is 51.1% and Sandell’s sensitivity13 is 0.011 pg cm-2 of molyb- denum. Extraction method. Beer’s law is obeyed between 5 and 100 pg of molybdenum under the conditions described.The molar absorptivity is 9.4 x lo3 1 mol-l cm-l and the range fc\r minimum error in the determination is 10-95 pg of molybdenum. The relative error (P = 0.05) of the method is &1.3% and Sandell’s sensitivity is 0.010 pg cm-2. Interference study The recommended procedure was used to analyse standard molybdenum solutions in the presence of potential interfering ions. The results for the determination of 50.0 pg of molybdenum are shown in Table I. Determinations by the direct method were carried out a t the limiting value of the concentration of foreign ion that caused interference in the extraction method. TABLE I RECOVERY OF MOLYBDENUM IN THE PRESENCE OF METAL IONS Ion added Al(II1) . . . . .. Ni(I1) . . . . . . Mn(I1) .. . . .. Zn(I1) . . . . .. Cd(I1) . . . . .. Ti(IV) . . . . .. Cr(II1) . . .. .. Bi(II1) . . . . .. La(II1) . . . . .. W(V1) . . . . . . CO(I1) . . . . .. Cu(I1) . . . . . . Fe(I1) . . . . . . Fe(II1) . . . . . . Pb(I1) . . . . . . Hg(I1) . . . . . . I’(V) . . . . .. Alkali and alkaline earth metals . . . . . . Mass excess relative to Mo 15 15 15 15 15 15 15 15 15 10 2 1 1 1 1 1 1 100 Amount of molvbdenum recovered/pg - Extraction method 49.5 49.5 50.0 50.1 50.0 49.7 50.1 50 0 50.2 49.5 48.0 22.0 26.0 49.5 50.0 60.0 82.0 50.0 Direct method 42.0 85.0 55.0 59.0 57.5 42.0 43.0 72.0 55.0 55.5 > 100 > 100 > 100 > 100 39.0 99.5 100 50.0 From the results, it is concluded that the tolerance limits for the extractive determination of molybdenum are greater than those obtained by the direct method.The major inter- ferences were caused by elements that are known to form OH-PDT complexes [Cu(II), Co(II), Fe(II), Fe(III), Hg(1I) and V(V)]. Other metal ions, such as Al(III), Ni(II), Mn(II), Zn(II), Cd(II), Ti(IV), Cr(III), Bi(II1) and La(III), can be tolerated at up to a 15-fold excess by mass. From the data reported above (spectrophotometric characteristics and interference study), it is concluded that the extraction of the chelate provides the greater sensitivity and selec- tivity, and the extraction method is therefore proposed. In order to assess possible analytical applications, it appears that a prior separation of molybdenum before the determination may be necessary in order to improve the specificity in the presence of interfering ions.Extraction with dithizone in carbon tetrachloride is of value in separating a number of heavy metals from molybdenum(VI), if these are not present in large amounts. A 10-fold excess of Cu(II), Co(II), Fe(II1) and Hg(I1) can be tolerated with prior extraction with 0.1% dithizone solution from aqueous solutions at pH - 3. Rlolybdenum can be also separated from iron, copper, vanadium and lead in the presence of citrate ions.14 When used in conjuction with an ion-exchange procedure, the proposed method may be useful in deter- mining the molybdenum content of some steel alloys. Tungstate(V1) does not interfere a t up to a 10-fold excess. Ion-exchange separation of molybdenum may also be appli~able.l5-~*46 TERNERO RODRIGUEZ Conclusion The suitability of the molybdenum(V1) - OH-PDT system for the development of a rapid, simple, accurate and sensitive method for determining small amounts of molybdenum has been demonstrated.The proposed reagent compares well with previously described thio- semicarbazones (Table 11). It is evident that OH-PDT is the most sensitive thiosemi- carbazone reported for the determination of molybdenum. TABLE I1 CHARACTERISTICS OF MOLYBDENUM - THIOSEMICARBAZONE COMPLEXES Compound Salicylaldehyde thiosemicarbazone . . 2-Hydroxybenzophenone thiosemi- carbazone . . .. . . .. 2,2’-Dihydroxybenzophenone thio- semicarbazone . . .. .. .. 2-Hydroxy-4-methoxy-5-sulpho- benzophenone thiosemicarbazone . . 1,4-Dihydroxyphthalimide dithio- semicarbazone . . . . . . .. Toluene-3,4-dithiol* .. .. Condition SnCl,, pH <2 SnCl,, pH 1.3 SnCl,, pH 1.3 SnCl,, pH 1.3 pH 2-4, extract with isopentyl alcohol H,SO, 8-12 N, extract with benzene 1. 2. 3. 4. 5. 6. 7 . 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. * Included for the sake of comparison. References Sandell’s Amax./nm sensitivity Reference 550 0.020 7 505 0.033 10 505 0.030 8 505 0.020 9 435 0.010 - 680 0.004 6 Pkrez-Bendito, D., Valcarcel, M., Ternero, M., and Pino, F., Anal. Chim. Acta, 1977, 94, 405. Ternero, M., Pino, F., Pkrez-Bendito, D., and Valcarcel, M., Microchem. J . , 1980, 25, 102. Ternero, M., Pino, F., P&ez-Bendito, D., and Valcarcel, M., Anal. Chim. Acta, 1978, 109, 401. Ternero, M., Pgrez-Bendito, D., and Valcarcel, M., Micvochem. J . , 1981, 26, 61. Sandell, E . B., “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience, New Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience, New Pgrez-Bendito, D., and Pino, F., Mikrochim. Acta, 1976, I, 613. Lopez Fernandez, J . M., Pkrez-Bendito, D., and Valcarcel, M., Analyst, 1978, 103, 1210. Toribio, F., Lopez Fernandez, J. M., and Valcarcel, M., An. Quim., 1980, 76, 436. Lopez Fernandez, J. M., Toribio, F., Pkrez-Bendito, D., and Valcarcel, M., Quim. Anal., in the press. Yof, J . H., and Jones, A. L., Ind. Eng. Chem., Anal. Ed., 1944, 16, 211. Job, P., Ann. Chim. (Paris), 1928, 9, 113. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience, New Klement, R., 2. Anal. Chem., 1952, 136, 17. van Erkelens, P. C., Anal. Chim. Acta, 1961, 25, 42. Strelow, F. W. E., J . S . A f r . Chem. Inst., 1961, 14, 51. Nelson, F., Murase, F., and Kraus, K. A., J . Chromatogr., 1964, 13, 503. Riley, J . P., and Taylor, D., Anal. Chim. Acta, 1968, 41, 175. York, 1976, p. 644. York, 1976, p. 650. York, 1976, p. 83. Received July loth, 1981 Accepted August 12th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700041

出版商:RSC    

年代:1982

数据来源: RSC

摘要:

Analyst, January, 1982, Vol. 107,pfi. 47-52 47 Determination of Iron( 11) in Silicates by Gravimetric Titration T. Denis Rice New South Wales Department of Mineral Resources, Chemical Laboratory Branch, P.O. Box 76, Lidcombe, N . S . W. 2141, Australia A4 rapid method, suitable for batch working, is described for determining iron(I1) in silicates. Hot 1 + 1 hydrofluoric - sulphuric acid decomposes the subsample in a polypropylene bottle previously filled with carbon dioxide. After dilution with a boric - phosphoric - sulphuric acid mixture, iron(I1) is titrated gravimetrically with potassium dichromate solution, in the bottle used for sample decomposition. A sample-preparation procedure that mini- mises oxidation of iron(I1) is described. Results obtained for iron(I1) in various international reference rocks by the proposed method agree well with recommended values.Keywovds : Ivon(II) determination ; silicates ; gyavimetvic titvation French and Adamsl have published a procedure for determining iron(I1) oxide in rocks that involves decomposition of the sample by treatment with a hot 1 + 1 mixture of concentrated hydrofluoric and sulphuric acids. A modification of this procedure is used by the CSIRO Division of Alineralogy, North Ryde, New South Wales.2 The latter procedure differs from the former procedure mainly in that the sub-sample is decomposed in a narrow-mouthed 125-ml polypropylene bottle previously filled with carbon dioxide, rather than in a 60-ml wide-mouthed polypropylene bottle, and the redox titration is performed with potassium dicliromate solution and sodium diplienylamine sulplionate, rather than with cerium( IV) sulpliate solution and N-phenylanthranilic acid as titrant and indicator, respectively.Tlie procedure described in this paper is simpler than the two procedures referred to above for the following reasons. ( i ) (ii) The bottle used for decomposing the sub-sample is also the titration vessel. A convenient gravimetric titration system, consisting of the titrant in a polyethylene dispensing bottle3 and an electronic top-loading balance, replaces the traditional burette. Tlie experience of several operators in this laboratory has shown that it is easier to use the above titration system rather than a burette. As the fine tip on the dispensing bottle can deliver individual drops weighing between 0.008 and 0.010 g , the end-point can be located as precisely as with a microburette.The only disadvantage of this gravimetric approach is that an electronic top-loading balance is much more expensive than a burette. However, sucli a balance has many other uses in the laboratory, so this disadvantage is more apparent than real. Experimental Sample-preparation Equipment This consists of the following items installed in a sample-preparation room with dust extraction facilities for operator safety. Hydrniilic lnboirntory cruslzer - breakel. with tuizgsten carbide working faces. Obtainable from Rocklabs, P.O. Box 28-362, Auckland 5, h'ew Zealand. J n ~ l c r i r s l w uiith nzaj?gajiese haideized steel plates. As an alternative to the hydraulic laboratory crusher - breaker.Sicbtccliizik tungsten carbide - cobalt riizg griizder, 300-ml. M'eighing about 11.6 kg and obtainable from Siebtechnik, lIiihlheim, Federal Republic of Germany. TT-cdrrg Znborntoyv Liibmtiizg cup mill, Type ML\T 954/2. Obtainable from KHD Industrie- anlagen SiccZ ndnptoi/ base. This allows the Siebtechnik grinder to be used with the Il'edag vibrating cup mill. Humbold t IVedag, Federal Republic of Germany. Il'eighing about 2.7 kg.48 RICE : DETERMINATION OF IRON(II) IN Analyst, Vol. 107 Apparatus These are of 125-ml capacity and have screw-caps that give a gas-tight seal. In this laboratory, Nalgene, Catalogue No. 2006-0004, bottles (obtainable from Nalge Company, Division of Sybon Corporation, Rochester, N.Y .) have been found suitable; they can be re-used almost indefinitely.The bottles must be dry and filled with carbon dioxide within 6 h before use. Boiling water in a 2000-ml beaker is used to heat four poly- propylene bottles floating upright. This has a capacity of 500-600 ml and a fine tip on its dispensing tube capable of delivering drops weighing between 0.008 and Narrouvnouthed polypropylene bottles. Glass beakers, 2000-ml. PoLyetlayle?ze disfiensi.lzg bottle for use as a weight burette. 0.01 g.3 Mag fietic stirrer and PTFE-coated stirrer bars. Cnlibvnted electronic top-loading balance. This must be capable of weighing up to about 1200 g to the nearest 0.01 g and have a precision of &0.005-0.01 g. Reagents All reagents should be of the appropriate analytical-reagent grade.should be used in the preparation of reagents and throughout the analysis. De-ionised water Caution -Wear safety glasses and gloves when handling the acids listed below. Handle hydrofluoric Handle potassium dichromate with care because it is acid solutions in a well ventilated fume cupboard. toxic and carcinogenic. HjldroJEzioric acid, sp. gr. 1.13. Sirlfiliztvic acid, sp. gr. 1.84. Szdfilzitric acid, 1 + 1. Boric acid, solid ciystals. PlZOs$l1oric acid, sp. gr. 1.75. Hot mixtiwe of Jydrojziovic nizd szilfilzwic acids, 1 + 1. Cautiously add 500 ml of sulpliuric acid (sp. gr. 1.84) with stirring to 500 ml of water, cool and dilute to 1000 ml. Slowly add 50 ml of sulphuric acid (sp. gr. 1.84) to 50 ml of hydrofluoric acid in a 125-ml polypropylene bottle in a fume cupboard.(Heat and hydrogen fluoride vapour are liberated in this process.) To prevent tlie acid mixture from cooling before use, stand the sealed bottle in a 500-ml beaker containing water at 60 "C. Dissolve 75 g of boric acid in 1875 ml of water. Transfer to a 2.5-1 polyethylene bottle fitted with a handle for ease of dispensing, add 250 ml of 1 + 1 sulphuric acid and 125 ml of phosphoric acid and mix. No. 32 Analoids from Ridsdale and Co. Ltd., Middlesbrougli, England, have been found to be suitable. IYeigh 1.2283 g of potassium dichromate into a tared 1100-in1 polypropylene beaker. II'ith the aid of a calibrated top-loading balance and a polyethylene dispensing b ~ t t l e , ~ add 998.78 g of water. Cover and stir magnetically until dissolution is complete.Transfer about half of this solution to the weight burette (see under Apparatus) ; store the remainder in a 500-ml polyethylene bottle. The concentration of this solution is such that 1.000 g is equivalent to 1.800 mg of iron(I1) oxide. Dilirte solzitiorz of boiic, plzosphoric and siilfihzii4c ncids. Sodii1112 d i ~ l i c q ~ l n m i i z e sziZfilzo.rzate, fxllets. Yotnssiirirz dicliromnte solzrtioiz. Procedure Il'itli the aid of a hydraulic crusher - beaker or a jaw crusher with hardened steel plates reduce the rock sample to a grain diameter of about 3 mm or less5 Place a representative 30-g portion of this material in a 300-ml Siebteclinik tungsten carbide - cobalt ring grinder. Secure the loaded ring grinder in the Il'edag mill with the aid of the steel adaptor base; mill for 5 min.Approximately 98y0 of the resulting powder is less than 45 p i in particle diameter. Transfer a 0.25-g subsample of powdered rock, weighed to tlie nearest 0.0002 g, into a 125-nil polypropylene bottle previously filled with carbon dioxide. Add (with a poly- prop>-lene measuring cylinder) 10 ml of a 1 + 1 mixture of liydrofluoric and sulpliuric acids heated to GO "C. Place the bottle upright in a beaker of boiling water and boil for 10 min. (The level of boiling water in the beaker should Iminediately tighten the cap of the bottle.January, 1982 SILICATES BY GRAVIMETRIC TITRATION 49 be high enough to ensure that the cap of the bottle is higher than the top of the beaker. This will prevent the cap from being heated above about 50 “C; the cap may loosen when heated above this temperature.) Add 90ml of the dilute solution of boric, phosphoric and sulphuric acids.Immediately seal the bottle, shake and stand it in a cooling bath in readiness for titration. Any unattacked particles, which may contain iron(I1) (especially if they are dark in colour), can probably be seen settling in the bottle at this stage. If a sample does have unattacked particles, as distinct from precipitated fluorides that dissolve after stirring for a few minutes in the presence of boric acid, repeat the procedure, extending the heating period in boiling water to 30 min. Stir the contents of the bottle magnetically and titrate gravimetrically with the potassium dichromate solution contained in the polyethylene dispensing bottle3 until the solution turns from bright green to grey - green.Place the cap on the bottle and shake to include any droplets on the underside of the cap in the titration. Continue to titrate dropwise until the end-point is reached, when the solution turns purple and remains so for at least 15 s. Cool the bottle in cold water. Add one pellet of sodium diphenylamine sulphonate and a PTFE-coated stirrer bar. Calculations (i) Dichromate titre = Mass of dispensing bottle before titration -mass of dispensing bottle after titration Dichromate titre (g) x 0.18 Mass of subsample (ii) Iron(I1) oxide (%) = Results and Discussion Sample Preparation French and Adamsl found that grinding a rock powder with an agate pestle and mortar for 10 min while continuously moistening with acetone produced “a sufficiently fine grain size and no detectable oxidation.” As acetone, by rapid evaporation, prevents heating of the sample, this finding is in accord with the statement of Hillebrand et aZ.6 that the cause of oxidation of iron(I1) when rocks are ground in air is probably local heating.In this laboratory, a Siebtechnik tungsten carbide - cobalt ring grinder is preferred to an agate pestle and mortar for pulverising silicate rocks for summation analysis because it is more rapid and introduces less risk of contamination. Local heating of the sample when being pulverised in this ring grinder is probably negligible if the ring grinder’s internal parts do not feel warm immediately after grinding. This situation occurs when the Wedag mill is “weighed down” with a 2.7-kg adaptor base so that one can use the Siebtechnik ring grinder with i t ; the internal parts of this grinder are still cool to the touch even immediately after 7 min of grinding.As Table I indicates, the adaptor base causes the effective weight of the Siebteclinik ring grinder to be about 21% higher than the Wedag tungsten carbide - cobalt ring grinder. Ilihen used on tlie \;l’edag mill, the latter ring grinder feels distinctly warm even after grinding for only 1 min. Table I shows that although the Siebtechnik ring grinder plus adaptor base on the Wedag mill requires about twice the time to achieve a given sample fineness, it yields results for iron(I1) oxide in two typical igneous rocks significantly higher at the 95% confidence level than the “all \\‘edag” system.Use of Carbon Dioxide-filled 125-ml Bottles Frencli and Adamsl used 60-ml wide-mouthed polypropylene bottles as decomposition vessels and found that the copious hydrogen fluoride vapour produced by the violent reaction of the hot acid mixture and the sample was sufficient to expel all the atmospheric oxygen from tlie bottle and thus to eliminate tlie negative error due to oxidation of iron(I1) during sample deconiposition. For titration to be performed in the bottle, its capacity has to be about 125 nil rather than 60 nil. To ascertain n-lietlier the 1-apour referred to above was sufficient to expel all the atmos- pheric o s j y p i from 125-nil bottles, the samples listed in Table I1 were analysed in quad- ruplicate u-itli carbon dioxide-filled and air-filled bottles.The caps of the former bottles50 RICE : DETERMINATION OF IRON (11) IN Analyst, Vol. 107 were secured immediately after hot acid was added; however, with the latter bottles the French and Adams procedure1 of waiting for vapour tc appear in the neck of the bottle and then securing the cap was followed. From the results for three of the four samples in Table 11, one can conclude with 95% confidence that 125-ml narrow-mouthed polypropylene bottles have to be pre-filled with carbon dioxide in order to avoid oxidation of iron(I1) during sample decomposition. TABLE I EFFECT OF MASS OF RING GRINDER ON IRON(II) OXIDE RESULT AND GRINDING EFFICIENCY Ring grinder used on Wedag mill Milling time*/min . . .. . . FeO (dry basis),% *i .. . . Sample NSI.TIDMR G791607t- Percentage of sample having particle size (45 pm§ Sample NSWDMR G80/8477- . . FeO (dry basis),% $ . . . . Percentage of sample having particle size <45 pm§ . . * For a 30-g portion. 300-ml Siebtechnik tungsten carbide - cobalt + adaptor base, total mass = 14.38 kg 300-ml Wedag tungsten carbide - cobalt, total mass = 11.87 kg I -l 3 5 7 1 2 3 5 9.45 9.43 9.42 9.35 9.31 9.26 - f0.03 f0.03 A0.03 h0.03 A0.05 1 0 . 0 3 97.8 - 93.9 97.9 99.1 95.1 96.9 4.50 - - 4.28 - 4.08 ,t0.0411 & 0.04 f 0.05 - - 97.911 - - 96.5 - 94.4** Very hard, basic, intrusive rock. Each result is the mean of triplicate determinations & the 957; confidence limit of the mean, obtained by multiplying the range by 1.3.7 5 Particle size was determined by wet sieving. 7 Porphyritic diorite containing hornblende, quartz and plagioclase./I TVhen a separate 80-g portion of this sample was wetted with 15 ml of acetone in the Siebtechnik ring grinder and ground for 5 min, the result for FeO (dry basis) was 4.57 & 0.030,/, and 98.176 of the sample had particle size (45 pm. ** Lower than that obtained after 2 min of grinding because agglomeration occurred during the excessively long grinding time. In discussing the procedure cf French and Adams,l Jefferyg has stated that althougli the use of a mixture of hot sulphuric and liydrofluoric acids may be ideal for some rocks, with others it produces a reaction that is too violent, and so should be used with caution. Loss of sample because the reaction is too violent is much less likely in the present procedure because of the use of 125-ml narrow-mouthed bottles instead of 60-ml wide-mouthed bottles.Blank Determination If the de-ionised water or the reagents contain appreciable dissolved organic matter, a blank determination is required. As iron(I1) has to be present initially for an end-point to occur, a suitable procedure for blank determination is to include in a batch of determina- tions a rock known to be low in iron(I1). The blank is obtained from the difference between the found and expected iron(I1) oxide. Blanks thus determined i n this laboratory with 30-mg subsamples of the syenite NIM-S have been very low, ranging froin -0.01 to $0.03% iron(I1) oxide over a 2-year period. Interferences In common with other wet-chemical methods for determining iron(I1) oxide in silicates, the niethod described herein suffers from a positive interference by sulphide minerals (except pyrite) and organic matter.The spectrophotometric method of Begheijn'O is free froin inter- ference by organic matter but is less rapid than the method described herein.January, 1982 SILICATES BY GRAVIMETRIC TITRATION 51 Although graphite is not attacked by hot hydrofluoric - sulphuric acids and does not react with potassium dichromate, its presence as a dark suspension in a titration solution renders the end-point indiscernible. This problem is readily overcome by weighing the bottle plus initial solution (approximately 100 ml), centrifuging, transferring most of the supernatant liquid to a fresh, tared bottle, weighing this “gravimetric aliquot” and then titrating it in the usual way.The amount of sample titrated can be readily calculated from the weighings referred to in the previous sentence, provided that the mass of the empty bottle used for decomposing the sample is known. TABLE I1 COMPARISON OF RESULTS OBTAINED USING AIR-FILLED AND CARBON DIOXIDE-FILLED 125-ml NARROW-MOUTHED POLYPROPYLENE BOTTLES FeO (dry basis), % r 1 Sample Recommended C0,-filled bottle* Air-filled bottle* Anorthosjte AN-G . . . . 2.247 2.29 f 0.03 2.18 & 0.06 Granite MA-N . . . . . . 0.311 0.30 i 0.03 0.21 0.04 Basalt BE-N . . . . . . 6.777 6.89 & 0.07 6.70 & 0.14 Ammonium iron(I1) sulphate . . 18 23 0.09; 18.3 f 0.2 17.7 & 0.2 * Results are the means of quadruplicate determinations f the 95% confidence limits of the means obtained by multiplying the range of each set of quadruplicate determinations by 0.72.’ t Reported by Govindaraju.E 100-mg subsamples of ammonium iron(I1) sulphate were analysed.The recommended FeO content of the ammonium iron(1I) sulphate (May and Baker, Pronalys analytical reagent) was calculated from its molecular formula, assuming that it was 99.5 & 0.5% pure, as implied by its label The above results for ammonium iron(I1) sulphate are reported on an “as received” basis. Refractory Iron(I1) Minerals The work of French and Adamsl showed that results for iron(I1) oxide determined by the procedure described herein will have a negative error if the sample contains iron( 11) minerals, such as garnet and staurolite, that are resistant to acid attack.RIaxwellll also includes tourmaline, ilmenite and magnetite in this category. French and Adamsl showed that staurolite \IXS only slightly decomposed even after 90 min of digestion; however, gainet was almost completely decomposed when the heating period in boiling water was increased from 10 to 30 min. As mentioned under Procedure, visual examination of the titration solution will probably reveal unattaclied minerals. Increasing the heating period to 30 min will sometimes result in such samples being completely decomposed, especially if they are finely ground as described in this paper. Reference Rocks These results were obtained by including a 0.25-g subsample of a given reference rock in a batch of determinations of iron(I1) oxide in rocks submitted for summation analysis. Table I11 also gives the number of determinations carried out on a particular reference rock and the 95% confidence limits of the mean results.Each determination on a given reference rock was carried out on a separate day. Table I11 gives the mean results for several reference rocks. Throughput of Pre-pulverised Samples tions, i.e., 24 determinations, in a 7-h working day. Using the method described, an analyst can readily perform two batches of 12 determina-52 RICE Conclusion The above method is convenient and rapid, and is used in this laboratory for the determina- tion of iron(I1) oxide in silicate rocks. Accurate results are obtained provided that the rocks do not contain any of the following: iron(I1) minerals incompletely attacked by hot 1 + 1 hydrofluoric - sulphuric acids; organic matter; and sulphide minerals attacked by the above acid mixture.TABLE I11 DETERMINATION OF IRON (11) OXIDE IN SEVERAL REFERENCE ROCKS FeO (dry basis), % A I No. of determinations Rock Recommended This procedure* (this procedure) Anorthosite AN-G . . 2.24t 2.29 f 0.03 7 Granite MA-N . . . . 0.31t 0.30 & 0.01 7 Basalt BE-N . . . . 6.77t 6.89 f 0.04 7 Basalt BCR-1 . . . . 8.963 9.05 f 0.02 2 Norite NIM-N . . . . 7.475 7.48 3 0.06 4 Granite NIM-G . . . . 1.305 1.28 3 0.04 4 Syenite NIM-S . . . . 0 305 0.31 f 0.02 4 Basalt G75/1323T[ . . 8.42 f 0.02 20 - * Results are the means f the 95% confidence limits of the means. t Reported by Govi~idaraju.~ $ Reported by Abbey.12 5 Reported by Steele and Hansen.13 7 Basalt G75/1323 is an in-house reference rock used in this laboratory as a control sample.Permission to publish this paper was given by the Under Secretary, New South Wales The author thanks R. Taylor for assistance with sample Department of Mineral Resources. preparation, and A. Clifton and J. Lyck for assistance with some of the analyses reported. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References French, W. J., and Adams, S. J., Analyst, 1972, 97, 828. Corbett, J . A., G I R O Division of Mineralogy, North Ryde, New South Wales, 2113, Australia, Rice, T. D., Aqaal. Chim. Ada, 1978, 97, 213. Sax, N. I., Editov, “Dangerous Properties of Industrial Materials,” Fifth Edition, Van Nostrand Maxwell, J. A., “Rock and Mineral Analysis,” Interscience, New York, 1968, pp. 50-51. Hillebrand, JV. F., Lundell, G. E. F., Bright, H. A., and Hoffman, J . I., “Applied Inorganic Analysis,” Second Edition, John Wiley, New York, 1953, p. 907. Dean, R. B., and Dixon, JV. J., A7zal. Chem., 1951, 23, 636. Govindaraju, I<., Geostand. N e d . , 1980, 4, 49. Jeffery, P. G., “Chemical Methods of Rock Analysis,” Second Edition, Pergamon Press, Oxford, Begheijn, L. Th., ATzzalyst, 1979, 104, 1055. Rfaxwell, J . A., “Rock and Mineral Analysis,” Interscience, New York, 1968, p. 208. Abbey, S., Geol. Surv. Can. Pap., NO. 80-14, 1980. Steele, T. W., and Hansen, R. G., Nut. Inst. Metall. Repztb. S . Afr. Rep., No. 2016, 1979. personal communication, 1976. Reinhold, New York, 1979, pp. 504-505 and 925. 1975, pp. 280-281. Received June Sth, 1981 Accepted August 13th, 1981

ISSN:0003-2654    

DOI:10.1039/AN9820700047

出版商:RSC    

年代:1982

数据来源: RSC