ISSN:0003-2654    

DOI:10.1039/AN96186FX005

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186BX007

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186FP025

出版商:RSC    

年代:1961

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96186BP037

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

FEBRUARY, 1961 THE ANALYST Vol. 86, No. 1019 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY ORDINARY MEETING AN Ordinary Meeting of the Society, organised by the Physical Methods Group, was held at 7 p.m. on Wednesday, February lst, 1961, in the meeting room of the Chemical Society, Burlington House, London, W.l. The Chair was taken by the President, Mr. R. C. Chirnside, F.R.1 .C. The subject of the meeting was “X-ray Fluorescence” and the following papers were presented and discussed : “X-ray Fluorescence Spectroscopy,” by K. M. Bills ; “Some Analytical Applications of X-ray Fluorescence Spectrometry,’’ by F. Brown, BSc., Ph.D., A.R.I.C. NEW MEMBERS ORDINARY MEMBERS Edmund Hollis Amstein, B.Sc., A.R.C.S., Ph.D. (Lond.) ; Roy Maurice Dagnall, BSc. (Birm.) ; Patrick Paul Donovan, M.Sc., Ph.D.(N.U.I.), M.I.C.I.; Michael Freegarde, B.Sc. (Sheff.), A.R.I.C.; William Edward Hearn; John Charles Henry Jones, B.Sc. (Lond.) ; Ronald Mead Pearson, A.R.I.C. ; Albert George Roach, Ph.D. (Manc.), B.Sc. (Wales), A.R.I.C. ; Ernest Walter Summers, B.Sc. (Lond.); Victor Daniel Tyrrell, B.Sc., M.A. (T.C.D.), M.Sc. (N.U.I.). JUNIOR MEMBERS Peter Edmund Arnold, A.R.I.C. ; James Lewis Tomlinson. DEATH Hubert Thomas Stanley Britton. WE record with regret the death of SCOTTISH SECTION A JOINT Meeting of the Scottish Section of the Society with the Chemical Society, the Society of Chemical Industry and the Royal Institute of Chemistry was held at 7.15 p.m. on Friday, December 2nd, 1960, in the Royal College of Science and Technology, George Street, Glas- gow, C.l.The Chair was taken by the Chairman of the Scottish Section, Mr. A. N. Harrow, The following paper was presented and discussed : “Ramsay, Chemistry and the Electrical A.H.-W.C., F.R.I.C. Industry,” by R. C. Chirnside, F.R.I.C. MIDLANDS SECTION AN Ordinary Meeting of the Section was held at 7 p.m. on Tuesday, January loth, 1961, at the Wolverhampton and Staff ordshire College of Technology, Wulfruna Street, Wolver- hampton. The Chair was taken by the Chairman of the Section, Dr. S. H. Jenkins, F.R.I.C., F.1nst.S.P. The following paper was presented and discussed : “Trace Analysis Using the Polarographic Technique,” by Mrs. B. Lamb, B.Sc., F.R.I.C. 8182 PROCEEIIINGS [Vol. 86 BIOLOGICAL METHODS GROUP THE sixteenth Annual General Meeting of the Group was held at 6.30 p.m. on Thursday, December 8th, 1960, in “The Feathers,” Tudor Street, London, E.C.4. The Chair was taken by the Chairman of the Group, Dr. J. I. M. Jones, F.R.I.C. The following Officers and Committee Members were elected for the forthcoming year :-Chairman-Mr. J. S. Simpson. Vice-Chairman-Mr. W. A. Broom. Hon. Secretary and Treasurer-Mr. K. L. Smith, Standards Department, Boots Pure Drug Co. Ltd., Nottingham. Members of Com- mittee-Mr. P. A. Andrews, Mrs. J. Gammon, Miss A. M. Jones, Dr. J. I. M. Jones, Dr. M. W. Parkes and Dr. G. F. Somers. Mr. D. M. Freeland and Dr. J. H. Hamence were re-appointed on. Auditors. The Annual General Meeting was followed by a Discussion Meeting on “Problems in the Control of Neomycin Quality,” which was opened by J. W. Lightbown M.Sc., Dip. Bact. The Chair at this meeting was taken by the new Chairman, Mr. J. S. Simpson, F.I.M.L.T.

ISSN:0003-2654    

DOI:10.1039/AN9618600081

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

82 PROCEEIIINGS [Vol. 86 Obituary ARNOLD LEES ARNOLD LEES, F.R.I.C., who passed away suddenly at his home in Southport on December 22nd, 1960, in his 71st year, was born in Ramsbottom, Lancashire, and was educated at Cockburn High School, Leeds. His first introduction to public health chemistry was with the York Water Board. In the year 1914 he became an assistant in the laboratory of J. A. Foster, who was Public Analyst for the East Riding of Yorkshire, and in 1916 he joined an Ordnance Factory at Leeds. In 1918 he was appointed an assistant analyst on the staff of the Lancashire County Laboratory under W. Collingwood Williams and spent 37 years in that laboratory, both in Liverpool and later when it was removed to Preston, retiring in January, 1955. He was made Chief Assistant County Analyst and an Additional Public Analyst in 1938 and was appointed Deputy County Analyst in 1946.His work as an analyst was always characterised by its neatness, and he was meticulous (to use his own word) in his attention to the detail of any investigation. In collaboration with G. D. Elsdon and with J. R. Stubbs he contributed papers to The Analyst. Arnold Lees was a staunch supporter of the North of England Section of the Society and never missed a meeting over a period of many years. In 1940 he became Honorary Secretary of the Section, which office he held until 1955. It was largely due to b s efforts that the meetings, and particularly the Summer Meetings, were such an unqualified success. On his resignation a presentation was made to him by the then President, Dr.D. W. Kent-Jones, in recognition of his services to the Section. He was passionately fond of sport, particularly cricket and football. He was a member of Lancashire County Cricket Club and he made a point every year of attending the Scar- brough Cricket Festival, where he enjoyed the company of his many friends. During his last ten years he was President of an amateur football club in his home town, He was also a Committee member of the Northern Section of the Lancashire Amateur League, a member of Council of the Lancashire Amateur League and Area Representative of the Lancashire Football Association, Amateur Committee. In his retirement, therefore, he gave his services to sport as wholeheartedly as he had given them to the North of England Section of the Society during his professional career. He was an exemplary father dearly loved by all his family. He leaves a widow and two sons; the latter have both taken up careers in chemistry. Arnold Lees had many friends in his profession and in the world of sport. GEO. H. WALKER

ISSN:0003-2654    

DOI:10.1039/AN9618600082

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

February, 19611 SMYTHE AND WHITTEM 83 Analytical Chemistry of Beryllium A Review* BY L. E. SMYTHE AND R. N. WHITTEM (A4 ustrailian Atomic Energy Commission Research Establishment, Sydney, New South Wnles, Azsstmlia) SUMMARY OF CONTENTS Introduction Materials requiring chemical analysis General chemical and radiochemical methods Gravimetric Volumetric Colorimetric and fluorimetric Ion exchange Solvent extraction Chromatographic Polarographic Radiochemical Determination of impurities in beryllium Spectrometric methods Emission X-ray Mass spectrometry BEFORE 1939, the analytical chemistry of beryllium dealt mainly with the analysis of beryilium ores, a limited number of beryllium compounds and a few useful alloys with such metals as aluminium (introduced in 1918) and copper (introduced in 1926).During the past 20 years new uses have been found for beryllium in the electrical, chemical and metal industries and, more recently, in the field of atomic energy. The increasing use of beryllium and its com- pounds led to the recognition, in the early 1940’s, of beryllium disease causing respiratory illness and certain kinds of skin reactions. Investigations of this disease and the control of occupational levels of beryllium required the development of methods for detecting small traces of beryllium in a variety of materials (see Table I). Accurate analytical methods were also required for the whole range of macro to micro amounts of beryllium. Although a Russian1 review was published in 1957 and a specialised review2 in 1958, many papers have appeared in the literature during the past 3 years.This review is presented to assist analytical chemists with a coverage of the more recent methods. MATERIALS REQUIRING CHEMICAL ANALYSIS Materials that may require examination are listed in Table I. GENERAL CHEMICAL AND RADIOCHEMICAL METHODS General chemical and radiochemical methods for determining beryllium may be con- veniently grouped under nine headings. This grouping, although convenient for review purposes, is not rigid, since an analytical procedure might involve a combination of methods. GRAVIMETRIC- Gravimetric procedures for determining beryllium are based on the formation of com- pounds insoluble under certain conditions ; some of the more “insoluble” beryllium compounds are listed in Table 11.It should be noted that the exact compositions of many of the compounds listed in Table I1 are unknown and that this is a limitation to their use in accurate gravimetric methods. The literature of beryllium is overburdened with compounds that have been assigned formulae simply from the chemical analysis of the solid phases-mixed crystals, residues from evaporation or indefinite gummy precipitates obtained under various con- dition~.~ For example, the halides of beryllium are hydrolysed by water, and, by careful * Reprints of this paper will be available shortly. For details, please see p. 142.84 SMYTHE AND WHITTEM: ANALYTICAL CHEMISTRY TABLE: I BERYLLIUM-CONTAINING MATERIALS [Vol. 86 Material Beryllium ores (generally 3BeO.A1,0,.6SiO,) Beryllium metal (commercial or high- purity powder, flake or fabricated) Beryllium oxide and hydroxide Beryllium alloys Beryllium compounds (carbide, sulphate, Air Water (or effluent) Filter-paper smears or swabs Biological materials Soils Miscellaneous (e.g., filter elements) halides, organo-metallic, etc.) r Constituents determined Be, Al, $3, K, Na, Fe, Ni, Li, Mg, Cr, Mn, Ca, etc.MetaZs--.Al, Fe, Si, Ca, Mg, Cu, Mn, Cr, Ni, Cd, B, lanthan- Non-mettzZs---O, H, N, C, halogens, He, 3H, etc. As for beryllium metal As for beryllium metal, but larger amounts of alloying Mainly major constituents, e.g., carbide, sulphate, halogen, ides, etc. elements such as Th, U, Pu, etc. etc. Traces of Be or Be compounds NoTE-Eeryllium metal, alloys or compounds Will Contain active constituents after irradiation in high-flux materials-testing reactors.Radiochemical analysis is necessary in such instances. manipulation of the evaporation residues, products of almost any degree of basicity can be prepared. It follows that many compounds of beryllium cannot be prepared by methods involving the use of aqueous solutions. Owing to lack of many insoluble beryllium compounds of definite composition, the most common and reliable method for determining beryllium gravimetrically involves precipitation of beryllium hydroxide and subsequent ignition to the o ~ i d e . ~ s ~ ~ ~ ~ ~ The method is com- paratively straightforward, although appropriate health precautions should be taken with ignited beryllium oxide.' Because of the colloidal nature of beryllium hydroxide, the results of gravimetric determinations are likely to be high, owing to adsorption and occlusion of impurity elements and compounds.Special precipitation procedures are used involving precipitation of beryllium hydroxide from near-neutral solution. Appropriate separation procedures are required in the presence of aluminium, silica, the hydrogen sulphide group, iron, titanium, zirconium, lanthanides, chromium, tungsten, vanadium and thorium. TABLE :[I INSOLUBLE BERYLLIUM COMPOUNDS Compound Notes Beryllium hydroxide, Be(OH),* Beryllium oxide, hydrate, BeO.xH,O* Beryllium oxide, Be0 Beryllium sulphate, BeSO, Sodium fluoroberyllate, Na,BeF,* Potassium fluoroberyllate, K,BeF,* Barium fluoroberyllate, BaBeF,* Beryllium carbonate, basic, [BeCO, + Be(O:H),j * Beryllium ammonium phosphate, BeNH,PO,* Beryllium - cobalt complex, [ (H,O),Be,(CO,),(OH),] Beryllium acetate, Be(C,H,O,),* Beryllium acetylacetonate, Be(C5H,0,), Beryllium - tannin complex* Beryllium - quinaldine complex Beryllium - oxine complex Beryllium - naphthaldehyde complex Beryllium - mercaptobenzothiazole complex Beryllium stearate, Be(C,,H,,O,), Amorphous powder or gel Decomposes on heating giving Be0 Stable to above 2000" C Hydrate (.4H,O) is soluble No normal carbonate [Co(NH,),] .3H,O* * Exact composition unknown.Other gravimetric methods based on the isolation of insoluble beryllium compounds of rather indefinite composition (see Table 11) require close control and careful assessment of interfering elements. These methods include isolation of beryllium as the beryllium - cobalt- ammine complex,8 ,9 barium fluoroberyllate,1° ,l1 beryllium mercaptobenzothiazole,l2 berylliumFebruary, 19611 OF BERYLLIUM.A REVIEW 85 naphthaldehyde,13 ammonium beryllium phosphate,14 the beryllium - tannin complex4 or as its complexes with benzylamine, triethanolamine or oxine.15 Gravimetric methods are suitable for determining milligram to gram amounts of beryl- lium. With careful control the precision and accuracy are to approximately 0.1 to 1 per cent. VOLUMETRIC- Unfortunately there are few specific reactions that can be used for the volumetric deter- mination of beryllium. A reliable volumetric procedure suitable for microgram to milligram amounts of beryllium would be invaluable. One volumetric method for b e r y l l i ~ m ~ ~ ~ 6 ~ ~ 7 is based on formation of the extremely stable complex BeF42-.If an excess of alkali fluoride (e.g., potassium or sodium fluoride) is added to a suspension of beryllium hydroxide, the reaction proceeds almost quantitatively in accordance with the equation- Be(OH), + 4F- -+ BeF42- + 20H- The free base liberated is proportional to the beryllium present and can be titrated with standard hydrochloric acid. However, it should be noted that, because the reaction is not strictly stoicheiometric, it is necessary to use an empirically determined titre for beryllium for the standard hydrochloric acid. Another volumetric method is that in which quinalizarin is used to detect the end- point.4 Iron interferes in this method and is removed by reducing the mixture with hydrogen and selectively dissolving the iron in hydrochloric acid.The quinalizarin reagent is used in a colour-comparison test in the presence of sodium hydroxide. Several modifications of this method have been described, and it is said to be suitable for amounts of beryllium in the microgram to milligram range. Other volumetric methods for beryllium include the salicylate - fluoride method,l* 8-hydroxyquinaldine with a volumetric finish,lg indirect complexometric titration of beryllium with ethylenediaminetetra-acetic acid (EDTA),20y21 potassium iodate - sodium thiosulphate hydrolysis method,22 titration of beryllium oxine with potassium b r ~ m a t e ~ ~ and bismuth oxychloride t i t r a t i ~ n . ~ ~ Volumetric methods for determining beryllium are possibly the least satisfactory of the chemical methods and their reliability under various conditions cannot be stated with any certainty .COLORIMETRIC AND FLUORIMETRIC- Several reliable colorimetric and fluorimetric methods are available for beryllium. Most of these are suited to determining microgram to low milligram amounts of beryllium. For example, a fluorimetric method for determining beryllium in which morin (2‘,4’,3,5,7- pentahydroxyflavone) reagent is used is said to have a detection limit of 0.004 pg and is precise to 0.8 per cent. on 0.2 pg at the 95 per cent. confidence The procedure and precautions to be observed to achieve this precision ( e g . , control of temperature, concentra- tions of morin, sodium hydroxide, salts, etc.) are rather time-consuming. However, with simple fl uorimetric equipment and without any special precautions, measurements can easily be made at levels down to 0.1 pg.Although good colorimetric methods for beryllium are also available, fluorimetric methods appear to have greater sensitivity and are more suitable for the microgram range. The colorimetric methods involve use of adsorption indicators or formation of suit able stoicheiometric complexes. Fluorimetric methods reported in the literature include morin preceded by selective electrolysis26 or selective e ~ t r a c t i o n , ~ ~ 8-hydroxyquinaldine and successive extraction with final fluorimetric determinati0n,~8,~~ quinizarin (1,4-dihydroxyanthraquinone) ,30 l-amino-4- hydroxyanthraq~inone~~ and a rapid routine method for determining sub-microgram and microgram amounts of beryllium in filter-~aper.~~ Colorimetric methods involve use of chrome azurol S,32 s33 8-hydroxyq~inaldine,~~ Erio- chrome cyanine R,34 aluminon (ammonium aurintricarboxylate) ,36 936 chrome blue K (also known as mordant blue 31 or 4-sulpho-2-hydroxyphenylazo-l,8-dihydroxynaphthalene-3,6- disulphonic acid) ,37 gossypin (a glycoside of the flavanol gossypetin) ,38 neothorin (arsenazo),39 thoron [ l-(o-arsonophenylazo)-2-naphthol-3,6-d.isulphonic acid] ,40 s 4 1 alberon (Solochrome brilliant blue B) ,a miscellaneous hydroxyq~inones,~~ salicylic acid,43 beryllon I1 [S-hydroxy- naphthalene-3’, 6‘-disulphonic acid-( l-azo-2’)-1’, S’-dihydroxynaphthalene-3‘, 6’-disulphonic acid, di- or tetra-sodium salt] ,14944,46 quinalizarin (1,2,5,8--tetrahydroxyanthraquinone) ,4*%86 SMYTHE AND WHITTEM : ANALYTICAL CHEMISTRY [Vol.86 hapthazarin (5,6-dihydroxy- 1,4-naphthaquinone) ,47 4- (9-nitrophenylazo) -0rcinol,~8 zenia (9-nitrobenzeneazo-orcinol) ,49 curcumin (diferuloylmethane) ,4 y50 molybdophosphoric acid,51 Naphthochrome green G,52 Naphthochrome azurine 2B,53 quini~arin,~~ 954 5-sulphosalicylic acid,55 l-amino-4-hydroxyanthraquinone30 and acetyla~etone.~~ Generally, the choice of reagent is governed by the material to be analysed and the interfering elements. Masking agents, solvent extraction, ion exchange, chromatography and electrochemical separation have all been used in conjunction with a colorimetric finish. Currently available colorimetric methods are not generally applicable in the sub-microgram range, and the lower limit for normal spectrophotometric precision is approximately 10 pg, with an upper limit in the milligram range.A wide choice of colorimetric reagents for beryllium is therefore available. ION EXCHANGE- Ion exchange has not been extensively used in the determination of microgram amounts of beryllium. Some promising methods have recently been reported, and it is likely that ion-exchange methods will receive more attention. The new ion-exchange chelating resins containing such groups as iminodiacetate, e.g., Dowex A1 chelating resin (The Dow Chemical Co., Midland, Michigan, U.S.A.), or the sodium. diallgl phosphate complexing resins57 may prove promising as the basis of new analytical procedures. The ion-exchange separation of beryllium with salicylate analogues has been studied by Schubert, Lindenbaum and Westfal158 and might also form the basis of an analytical procedure.It was shown that beryllium can be selectively eluted from a cation-exchange resin with sulphosalicylic acid (0.02 to 0.1 M) at pH 3.5 to 4.5. The ions Cu2+, UO,2+ and Ca2+ are not removed under these conditions; however, at pH 4.5 to 4.7 the ion U022+ is eluted. At pH values above 6 in the presence of sulphosalicylic acid, beryllium is strongly adsorbed by an anion-exchange resin. The separa- tion of milligram amounts of beryllium can be carried out on a cross-linked polystyrene cation- exchange resin, such as Amberlite IR-112 or Zeo-Karb 225, in the presence of a complexing agent. 59 When a buffer solution (pH 3-5 to 5.0) containing beryllium, aluminium, chromium, titanium and a slight excess of EDTA is passed thxough the resin in the ammonium form, only beryllium is adsorbed.The beryllium can be eluted with ammonium chloride solution and determined by a suitable method. Belyavskaya. and Fadeeva developed a method60 for the quantitative separation of beryllium from copper and nickel; they used SBS cation-exchange resin in the ammonium form. The separation was carried out in amrnoniacal medium (con- taining ammonium carbonate) at pH 8.5 to 9.0 atnd subsequent elution was with ammonium carbonate solution; full details were not available to us. Another method for determining milligram amounts of beryllium in beryl was described by Nadkarni, Varde and AthavaleGI; beryllium was separated from iron, aluminium and titanium by ion exchange.After fusion with sodium fluoride, digestion with sulphuric acid and addition of EDTA, the solution at pH 3-5 was passed through a cation-exchange resin (Amberlite IRA-120 or Zeo-Karb 225) in the sodium form. Beryllium was retained and the iron and aluminium complexes passed through. The beryllium was then eluted and determined gravimetrically, as BeO. Ion exchange has been used for the concentration of beryllium from sea water.56 Only 10ml of Dowex 50-X8 resin (200 to 400 mesh) converted to the ferric form and hydrolysed with ammonia were required for 50 litres of sea water. The standard deviation of results by ion-exchange methods for beryllium in the milligram range appears to be about 5 per cent., and these methods cannot as yet be recommended for sub-microgram amounts of beryllium.SOLVENT EXTRACTION- Only a limited number of different solvent-extraction systems have been used in the determination of beryllium. Extraction systems based on chelate formation appear to be the most promising for analytical purposes. Six-membered ring systems including beta-diketones and hydroxycarbonyls have proved valuable. Little work, however, has been carried out on analytical ion-association extraction systems in which beryllium may be contained in the cationic or anionic member of the ion pair. Analytical methods based on the solvent extraction of beryllium as acetylacetonate have received the most attention.62 to 68 The method has been applied to metallurgical analysis,65 to radiochemical analysis6* and in conjunction with masking agents.es ss9 MostFebruary, 19611 OF BERYLLIUM.A REVIEW 87 finishes after this extraction are colorimetric. Other solvent-extraction methods for deter- mining beryllium include a fluorimetric method involving semi-micro purification of the beryllium solution by extraction with ethyl acetate and diethyldithi~carbamate~~; sodium hydroxide and sodium ~ u l p h i d e ~ ~ ; 8-hydroxyquinaldine and chloroform70 ; separation as basic acetate and extraction with chl~roform~~ ; extraction with thenoyltrifl~oroacetone.~~~~~ Solvent-extraction methods for beryllium are applicable from microgram to macro amounts. Under carefully controlled conditions recoveries of 95 per cent. and higher can be achieved.In the microgram range the standard deviation is 5 to 10 per cent., with a lower extraction limit of about 1.0 pg. CHROMATOGRAPHIC- Chromatographic methods for determining beryllium have not yet gained wide accept- ance. The disadvantages of such methods for beryllium include (a) lengthy procedures, ( b ) close control is required, (c) results are mostly semi-quantitative and (d) the range of concentrations is restricted. An electrochromatographic method for beryllium has been described by Majumdar and Si11gh.7~ The migration sequences of beryllium and seven other elements were determined in various electrolytes; separations were possible with some mixtures containing up to four constituents. A semi-quantitative determination of beryllium, lithium and boron in minerals by partition chromatography has been de~cribed,’~ but cannot be recommended for routine use.Amounts of beryllium between 10 and 90 pg have been separated from uranium and titanium by paper chromatography, a mixture of isopropyl alcohol, acetylacetone and hydro- chloric acid being used as solvent.76 After separation, the ions were detected with an ethanolic solution of quercetin and a solution of potassium ferrocyanide, and the planimetric method was used for quantitative evaluation of the separated spots. Values of RF for beryl- lium, uranium and titanium were also reported. The average error in the analysis of a (1 + 1) mixture of beryllium and uranium did not exceed 10 per cent. The determination of beryllium after separation by paper chromatography has been described by Elbeih and Abou-Elnaga, 77 who used visual comparison of the oxine complex under ultra-violet illumina- tion.POLAROGMPHIC- Recently, Venkataratnain and Raghava Rao stated79 that, in an 0.5 M lithium chloride supporting electrolyte at pH 3.4, the diffusion current at the first stage of reduction remained proportional to the concentration of beryllium in solution up to 8 x 1 0 - 3 ~ . Earlier workers, however, have failed to detect any “steps” for the reduction of beryllium from aqueous solutions of its salts.*O Proximity of the hydrogen steps resulting from hydrolysis of Be2+ ions plus the need for accurate pH control is likely to cause difficulties in measuring step height. A more promising field of investigation would be polarographic methods based on the reduction of beryllium - organo complexes of the hgdroxyazo type or amperometric methods.s1 $82 RADIOCHEMICAL- Several interesting radiometric-titration, tracer,56 photo-neutron and activation methods for beryllium have been described.Neutron-activation methods for impurities in beryllium are referred to in the next section. A promising method, based on earlier work by Gaudin and PannelP and by Aidarkin and his co-workerss4 has been further developed by Milner and Edwardsg5 The method is based on measuring the photo-neutron flux produced when samples containing beryllium are irradiated with photons from an antimony-124 source. The method is rapid, interferences are small (with the exceptions of boron, cadmium, samarium and gadolinium) and, under the best conditions, the lower limit of detection is less than 0.002 per cent., as BeO.Radio- metric-titration procedures involving use of phosphorus-3ZS6 ss7 and ir0n-59~~ have been described. The phosphorus-32 method is said to be suitable for determining milligram amounts of beryllium in alloys and concentrates. Beryllium, as sulphate, is added to a buffer solution at pH 5 to 5.5 and is titrated with 0.1 M diammonium hydrogen orthophos- phate (containing phosphorus-32) of activity 20,OOO to 30,000 counts per minute per ml. At intervals, 0 6 m l aliquots are withdrawn, spun in a centrifuge and counted, and the equivalence point is found graphically. The precision of current radiometric procedures is Published work on the polarographic determination of beryllium is limited.7888 SMYTHE AND WHITTEM: ANALYTICAL CHEMISTRY [Vol.86 not as good as those of colorimetric and fluorimetric methods. Beryllium has been deter- mined88 in beryl by using radioactive iron-59. A preliminary oxalate precipitation is in- volved, and the procedure in its present form does not appear to be suitable for microgram amounts of beryllium. A beryllium-hazard detector in which polonium-210 alpha particles are used has been described.8s A 2.6-curie polonium-210 source provides a flux of alpha particles, to which samples of beryllium dust on filter-paper are exposed. The gamma-ray yield of the reaction sBe(a, n, y)12C is used to measure the beryllium content. The method is applicable in the range 0.2 to 100 pg of beryllium and a determination can be completed in 5 to 10 minutes.The method, however, necessitates a large capital cost for the source and detection equipment and cannot be recommended for routine use. DETERMINATION OF IMPURITIES IN BERYLLIUM-- Considerable attention has been devoted to the determination of impurities in beryllium, and nearly all the analytical methods described in this review, with the addition of methods such as vacuum fusion extraction, have been applied. Despite the demand for improved analytical methods, few analyses have been carried out on beryllium with very low levels of impurities such as oxygen, hydrogen and nitrogen. In fact, the solid solubility limits for oxygen, hydrogen and nitrogen have apparently never been deter~nined.~~ It is doubtful if ultra-high-purity beryllium has ever been prepared and submitted to comprehensive analysis.A comprehensive review dealing with methods for the detection of twenty-one compon- ents in beryllium has been published,2 but no such review covers the literature of the past 2 years. Conventional methods have been used for detecting common anions, such as chloride and sulphate, in beryllium compounds. Spectrographic methods have been extensively used for determining metallic impurities in beryllium and beryllium compounds and are detailed later. Vacuum fusion extraction procedures have been extensively used for determining hydrogen, oxygen and other gases in beryllium. A wide variety of chemical and radio- chemical methods has also been used. Some of the more important papers are summarised in Table 111.EMISSION- The major advantages of emission spectrornetry are specificity, high sensitivity, rapid processing of large numbers of samples and the relatively small samples. Against this, limitations are imposed by the empirical nature of the method when quantitative results are required. For every form and composition of sample, it is necessary either to have a range of standard samples or to carry out more or less extensive sample preparation to ensure that standards and samples can be directly compared. Even when this is done, spectrographic analyses for major elements show a considerably higher standard deviation than do most other methods. Thus, although many spectrographic techniques have been described for determining beryllium in bery1,134,136J38 it is our opinion that this determination is prefer- ably carried out by conventional “wet” chemical methods. This section will therefore deal with methods for determining traces of beryllium and its compounds and with methods for determining traces of impurities in beryllium and beryllium oxide.Determination of very small amounts of beryllium-Several techniques have been described for determining traces of beryllium in various materials, for example, plutonium,137 thoria,138 bismuth - uranium alloys139 and ~ 1 a n t s . l ~ ~ The major effort, however, appears to have been directed to meet the requirements of speed and sensitivity limits set for environmental moni- toring by smears, air-sampling, etc., reviewed recently by Br00ks.l~~ A wide variety of techniques has been used, the highest sensitivity claimed being obtained by the cathode- layer technique.In this method,136s142,143 the sample, as a powder or a solution dried on the graphite cathode, is arced at 10 to 15 amps, and only light originating very near the cathode is allowed to enter the spectrograph. g of beryllium has been ~1airned.l~~ Conventional d.c. arc techniques, in which the sample is placed on the anode and light from all the gap is admitted to the spectrograph have also been de~cribed.1~~~1~~ Thallium was used as internal standard in both these methods, and Landis and Coons145 claimed that, by adding barium chloride as carrier, higher sensitivity and reproducibility were obtained over the range 0-002 to 0.1 pg of beryllium. Another d.c.arc technique is the “iron-flux” method used by Garton, Webb and Sayer,146 who covered the range 0.006 to 4 pg of beryllium with a single exposure. SPECTROMETRIC METHODS A limiting sensitivity of 2 xFebruary, 19611 OF BERYLLIUM. A REVIEW 89 Fred, Nachtrieb and Tomkinsl47 were able to detect 0.002 pg of beryllium with a copper spark. Davis, Parker and Webb148 used a condensed spark between graphite electrodes, on one of which the sample, collected on a paper pad, was glued; with a direct-reading spectrograph, these workers were able to detect 0.003 pg of beryllium. Aluminium was used as internal standard, aluminium sulphate solution being placed on the pad (by pipette) before sparking. Spark techniques have also been described. TABLE I11 SUMMARY OF METHODS FOR DETERMINING IMPURITIES IN BERYLLIUM AND ITS COMPOUNDS Impurity determined Oxygen (as BeO) in Be .. Oxygen in Be . . .. .. Helium and tritium in Be . . Free carbon in Be . . . . Combined carbon in Be . . Fluorine in Be . . .. . . Oxygen in Be0 and Be . . Chlorine in Be0 and Be . . Combined nitrogen in Be . . Beryllium oxide in BeF, . . Fluorine in Be compounds . . Lanthanides in Be . . . . Various metals in Be . . . . Silicon in Be . . . . .. Iron in Be . . . . . . Chromium in Be . . . . Nickel in Be . . . . .. Cadmium in Be .. .. Uranium in Be.. .. . . Silicon in Be compounds . . Free Be and carbide-C in Be Various impurities in Be Copper in Be . . . . . . Manganese in Be . . .. Tungsten in Be . . . . Iron in Be compounds . . . . - Solution methods have .. . . . . . . .. . . . . . . .. .. . . . . . . .. .. .. . . .. . . . . . . . . . . . . . . .. been Literature reference* 91 92 93 ‘1 94, 95 96, 97 98 99 100 101 102 94, 95 105 106 107 108, 109, 110 111, 112, 113 112, 114 115, 116, 117 118, 119 120 121, 122, 123, 124 125 126 127, 128 129 130, 131 132 133 2 i { 2 Remarks Chemical method (methanol - Br,) ; error -&O.Ol per cent., as Be0 Chemical method (HCl) Activation method (lSN measured) ; relative error &5 per cent. Micro vacuum fusion; coefficient of variation 15 to 20 per cent. a t 0.01 per cent. level Fusion and extraction Solution, ignition and gasometric finish Solution and gasometric finish Absorptiometric method Chemical method (CuSO,) Chemical method Vacuum fusion method Chemical method (Kjeldahl) Chemical method (diffusion) Chemical method Chemical method Chemical method (oxalate) Spectrographic methods Absorptiometric methods Absorptiometric methods Absorptiometric methods Absorptiometric methods Absorptiometric method Absorptiometric and volumetric methods Absorptiometric method X-ray fluorescence spectroscopy Absorptiometric method Chemical methods Chemical methods Activation analysis Review of various methods - * See reference list, p.91. described. Feldman149 and Owen and his c o - w ~ r k e r s ~ ~ ~ used a porous-cup- spark technique in which the sample was fed into the discharge by percolation through the thin base of a hollow graphite electrode. A closely allied technique is the rotating-disc method in which the lower electrode is a rotating disc of graphite partly immersed in the sample solution.This has been used by Smith and his co-workersl61 and by the U.K. Atomic Energy A~th0rity.l~~ The latter workers used scandium as internal standard and, by choosing various beryllium and scandium lines, covered the range 0.03 to 50 pg of beryllium. Since air and smear samples are usually collected on filter-paper, most of the techniques mentioned above involve considerable chemical pre-treatment, such as wet ashing and separa- tion of interfering elements. Since these determinations are most usually carried out for Health Physics purposes, it seems to us that speed is more important than accuracy, especially when “spills” give rise to suspected contamination. Accordingly, in these laboratories, smear- and air-sample analyses are carried out by a modification of the method described by Davis, Parker and Webb,l** in which the filter-paper samples are glued to graphite electrodes 10 mm in diameter.A pulsed-arc discharge is used, with the sample as cathode and a blunt graphite anode 6 mm in diameter; the gap is 4 mm. Results are reported in the range (0.01 to >2 pg of beryllium by making visual comparison with standards prepared by90 SMYTHE AND WHITTEM 1 ANALYTICAL CHEMISTRY pol. 86 drying solutions of beryllium sulphate on similar filter-papers. If a stock of machinedelec- trodes is maintained, a batch of twenty samples can be processed by one operator in about 90 minutes. The main disadvantage of the method is that results may vary appreciably with different physical forms and chemical compositions of the beryllium contamination.How- ever, as the biological effects vary far more with the particle size and chemical state of the beryllium, the method is regarded as acceptable for Health Physics control. The speed of the method is invaluable in assessing the spread of contamination after a “spill.” Churchill and Gilliesonlm introduced a technique for the continuous monitoring of air for beryllium by drawing a constant flow of air across a spark gap. The light from the spark was dispersed by a grating monochromator set for the beryllium doublet at 3130 A and was continuously recorded by means of a photomultiplier, d.c. amplifier and circular-chart recorder. An improved version of this monitor has been described1” in which integration over a short time was used rather than continuous recording.Other refinements were higher dispersion to improve sensitivity, backgroundL correction and periodic standardisation (involving a discharge between beryllium - copper electrodes to generate particles of beryllium oxide). It was claimed that, by re-designing the spark gap and using a pulsed-arc discharge axially in the air stream, rather than a condensed spark across it, results were independent of the particle size and chemical composition of the beryllium aerosol. A similar instrument is being constructed in these laboratories. Here, very high dispersion is used, since it has been found that titanium interferes if the monochromaator “window” is set to admit both lines of the beryllium doublet at 3130 A.Our monochroinator has a band pass of 0.1 A, sufficient to resolve one beryllium line from titanium. A number of the refinements made by Webb, Webb and Wildyl” have been omitted, since we require specificity and sensitivity to changes in concentration, rather than absolute accuracy. Recently, commercially constructed auto- matic monitors have become available. One of these1& is based on the rotating-disc tech- nique161 and will automatically process up to sixty samples at the rate of six per hour. The otherlE6 involves a filter tape that is directly excited in a spark, providing a result every minute. Determination of traces of impurities in beryllium and its compounds-Over the last two decades, the requirements for determining impurities in beryllium and its compounds have varied as pure forms of beryllium have becorne available.Arc methods are used almost exclusively; in general, all samples are converted to beryllium oxide before arcing. A typical early example is the method described by Lee Srnith and Fa~se1,l~~ who used a 16-amp arc and a barium hydroxide - graphite mixture as a “spectroscopic buffer’’ for determining alu- minium, calcium, chromium, iron, magnesium, manganese and silicon, mostly in the range above 100 p.p.m. The carrier-distillation method originally [developed for determining impurities in uranium168 has been used for determining silver, cadmium, molybdenum, lead, zinc, lithium and calcium,2 but the limits of detection were not stated. With lanthanum as carrier, Zaidel and his co-worker~l~~ detected gadolinium, europium and samarium in beryllium down to 0.1 p.p.m.Zaidel and othersls8 have also described an interesting method for determining very small amounts of boron in beryllium oxide. A 30-mg sample of oxide is heated in vacuo and volatile impurities are collected on a copper electrode, which is then sparked. Recently, Karabash and his co-workers7l have described a chemico-spectrographic method for determining twenty-five elements as impurities in beryllium and beryllium oxide. The sample (2 g) is converted to basic beryllium acetate and is then extracted with chloroform until about 5 per cent. of the beryllium remains in the aqueous layer, which is then separated, evaporated to dryness and converted to oxide. The oxide is then arced in a graphite cup at 12 amps.Detection limits are given as- 5 p.p.m. for Zn, 3 p.p.m. for Ca and Al, 2 p.p.m. for Ba, Ti, Fe, Sb, Te, In and T1, 1 p.p.m. for Mg, Mo, Co, Ni, Sn, Pb and Na, 0-5 p.p.m. for V, Cr, Bi and Ga, 0.3 p.p.m. for Cu, 0.2 p.p.m. for Ag, 0.1 p.p.m. for Mn and 0.05 p.p.m. for Cd.February, 19611 OF BERYLLIUM. A REVIEW 91 X-RAY METHODS- When X-ray techniques are considered for analytical problems in the beryllium field, the most notable factor is the very low absorption and scattering of X-rays by beryllium atoms. This means that the sensitivity of detection of beryllium metal is poor. On the other hand, the sensitivity of detection of impurities in beryllium metal is fairly high, particularly if thick specimens can be used. An example, now being examined in these laboratories, is the determination of beryllium oxide in beryllium, which is usually present in the range 0.3 to 1 per cent.Since the solu- bilities of oxygen and beryllium oxide in beryllium are likely to be very low, it is expected that this technique will give results for total oxygen in beryllium if the metallurgical history of the sample is such as to favour reasonable crystallite size for the beryllium oxide. X-ray fluorescence methods for many impurities are also feasible and are very sensitive. Qualitatively, chromium, copper, iron, lead, manganese, nickel, vanadium, zinc and zirconium have been detectedlZ8 in MTR shim rod; uranium was estimated at the 500 p.p.m. level. At present, we are examining a method for determining iron in beryllium and in beryllium oxide.It is planned to extend this to other common impurities, e.g., chromium, copper, manganese, nickel, zinc and uranium. MASS SPECTROMETRY- Mass spectrometry is not generally applicable to the detemination of beryllium, since beryllium and most of its compounds are insufficiently volatile for application of the usual gas-phase techniques. The use of stable-isotope dilution methods involving solid sources has not been reported, perhaps because of the difficulty of obtaining supplies of beryllium-10. However, mass spectrometry in conjunction with vacuum fusion has been used in Canadag6 and in England9’ for determining gases formed in beryllium by (n, 212) reactions; the gases found were 4He, 3He, ‘H, and 3H,. The British workers found a large and variable content of lH,, which may be attributable to the reduction, during vacuum fusion, of water adsorbed on the specimens or to radiolysis of water occluded in the specimens.1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. REFERENCES Goryushina, V. G., Zavod. Lab., 1967,23, 1300. Melick, E. S., US. Atomic Energy Commission Report TID-7555 (Part l), Oak Ridge, Tennessee, 1958, p. 158. Mellor, J. W., “A Comprehensive Treatise on Inorganic and Theoretical Chemistry,” Longmans, Green & Co. Ltd., London, 1953, Volume IV, p. 215. Scott, W. W., and Furman, N. H., “Standard Methods of Chemical Analysis,” Fifth Edition, D. Van Nostrand Co. Inc., New York, and The Technical Press Ltd., London, 1939, Volume I, p.139. Tananaev, I. V., and Shcheglova, E. I?., Zh. Anal. Khim., 1957, 12, 671. Rogers, N. E., and Prather, D. W., Anal. Chem., 1959, 31, 1081. Chadwick, V. R., U.K. Atomic Energy Authority Industrial Group Report, Series 37 (Rd/R), Pirtea, T. I., and Mihail, G., Z. anal. Chem., 1958, 159, 205. Pirtea, T. I., and Constantinescu, V., Ilbid., 1959, 165, 183. Gupta, A. K. S., J . Indian Chem. SOG., 1957, 34, 725; Sci. & Culture, 1956, 21, 687. Dutta, R. K., and Gupta, A. K. S., J . Indian Chem. SOC., 1956,33, 146. Spacu, G., and Pirtea, T. I., Rev. Chim. Romania, 1956, 1, 5. Gusev, S. I., Kumov, V. I., and Sokolova, E. V., Zh. Anal. Khim., 1957, 12, 55. Tsyvina, B. S., and Davidovich, N. K., Zavod. Lab., 1957, 23, 280. Sastri, C. L., Sriramulu, G., and Raghava Rao, B.S. V., J . Sci. Ind. Res., India, 1955, 14 , 171. McClure, J. H., and Banks, C. V., U.S. Atomic Energy Commission Reports AECU-812, 1950, and TID-7015, Oak Ridge, Tennessee, 1958, Section I, No. 211070. Wallace, C. G., U.K. Atomic Energy Research Establishment Reports AM29 and AM30, Harwell, 1959. Zolotukhin, V. K., Nauk. Zap. L’vivs’k Univ., 1955, 34, 115. Motojima, K., Bull. Chem. SOC. Japan, 1956, 29, 6. Misumi, S., and Taketatsu, T., Jbid., 1959,32, 6. Whitehead, T. S., U.S. Atomic Energy Commission Report TID-7568 (Part l), Oak Ridge, Ten- Suseela, B., Zh. Anal. Khim., 1955, 10, 286. Motojima, K., Nippon Kagaku Zasshi, 1956, 77, 95. Das, M. S., and Athavale, V. T., Anal. Chirn. Ada, 1955, 12, 6. Sill, C. W., and Willis, C. P., Anal.Chem., 1959, 31, 598. Fedorchuk, C. N., Trudy Inst. Mineral. Geokhim. Kristallokhim. Redk. Element. Akad. Nauk. Vanossi, R., An. Asoc. Qulm. Argentina, 1957, 45, 215. Ishibashi, M., Shigematsu, T., and Nishikawa, Y., J . Chem. SOC. Japan, Pure Chem. Sect., 1957, Springfields, Lancs., 1959. nessee, 1958, p. 96. SSSR. 1967, 178; Anal. Abstr., 1959, 6, 853. 78, 1139.92 29. 30. 31. 32. 33. 34. 15. 3 6 . 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. SMYTHE AND WHITTEM : ANALYTICAL CHEMISTRY [Vol. 86 Motojima, K., Bull. Chem. Soc. Japan, 1956, 29, 75; Proceedings of the Second International Conference on the Peaceful Uses of Atomic Energy, United Nations, Geneva, 1958, 28, 667.Carlson, A. B. Neuman, W. F., and Underwood, A. L., U.S. Atomic Energy Commission Report MDDC-941, Oak Ridge, Tennessee, 1947. Florence, T. M., Anal. Chim. Acta, 1959, 20, 472. Silverman, L., and Shideler, M. E., Anal. Chem., 1959, 31, 152. Mustafin, I. S., and Matveev, L. O., Zavod. Lab., 1958,24, 259. Hill, U. T., Anal. Chem., 1958, 30, 521. Mukherji, A. K., and Dey, A. K., Chim. Anal., 1958, 40, 299. Luke, C. L., and Campbell, M. E., Anal. Chem., 1952, 24, 1056. Kalinchenko, L. P., Strakhov, N. P., and Kalinchenko, I. I., Zavod. Lab., 1958, 24, 22. Bose, J. C., Srinivasulu, K., and Raghava Rao, B. S. V., 2. anal. Chem., 1958, 162, 93. Shibata, S., Takeuchi, F., and Matsumac, T., Bull.Chem. Soc. Japan, 1958,31, 888. Adamovich, L. P., and Yutsis, B. V., Ukr. Khim. Zh., 1957, 23, 784. Adamovich, L. P., and Mirnaya, A. P., Symposium report, “Sovrem. Metody Anal. Metall. M , Przheval’skii, E. S., Belyavskaya, T. A., and Golovina, A. P., Vestn. Moskov. Univ., 1956, 191. Adamovich, L. P., and Kravchenko, T. U., Zavod. Lab., 1957,23, 416. Lukin, A. M., and Zavarikhina, G. B., Zh. Anal. Khim., 1956, 11, 393. Budanova, L. M., and Zhurkova, B. A., Zavod. Lab., 1959,25, 411. Vasil’ev, P. V., Sb. Nauch. Tekhn. I n f . Ministerstvo Geol. i Okhrany Near., 1955, 131. Karanovich, G. G., Trudy Vses. Nauch. Inst. Khim. Reaktivov, 1956, 21, 43. Covington, L. C., and Miles, M. J., Anal. Chem., 1956, 28, 1728. Stross, W., and Osborn, G. H., J . SOC.Chem. Iizd., 1944, 63, 249. Kolthoff, I. M., J . Amer. Chem. Soc., 1928,50, 393. Sunderasan, M., and Das, M. S.. Analyst, 1955, 80, 697. Aldridge, W. N., and Liddell, H. F., Ibid., 1948’, 73, 607. Cucci, M. W., Neuman, W. F. and Mulryan, B. J., U.S. Atomic Energy Commission Report AECD- Fletcher, M. H., and White, C. E., Amer. Mineralogist, 1946, 31, 82. Meek, H. V., and Banks, C. V., Anal. Chem., 1!350,22, 1512. Merril, J. R., Honda, M., and Arnold, J. R., Proceedings of the Second International Conference on the Peaceful Uses of Atomic Energy, United Nations, Geneva, 1958, A/CONF./15P/412. Kennedy, J., and Wheeler, V. J., Anal. Chim. .4cta, 1959, 20, 412; U.K. Atomic Energy Research Establishment Report C/M 359, Harwell, 1958. Schubert, J., Lindenbaum, A., and Westfall, W., Chimia, 1957,11,50; J .Phys. Chem., 1958,62,390. Taketatsu, T., J . Chem. Sac. Japan, Pure Chem. Sect., 1958,79, 586 and 590. Belyavskaya, T. A., and Fadeeva, V. I., Vestn. Moskov. Univ., 1956, 6, 73. Nadkarni, M. N., Varde, M. S., and Athavale, V. T., Anal. Chim. Acta, 1957, 16, 421. Adam, J. A., Booth, E., and Strickland, J. D. H., Ibid., 1952, 6, 462. Tabushi, M., Bull. Inst. Chem. Res. Kyoto Univ., 1958, 36, 156. Buchanan, J. D., J . Inorg. Nuclear Chem., 1958, 7, 140. Kett, M., Hutn. Listy, 1958, 13, 250. Alimarin, I. P., and Gibalo, I. M., Zh. Anal. Khim., 1956, 11, 389. Cool, R. D., Chemist Analyst, 1959, 48, 7. Shigematsu, T., and Tabushi, M., J . Chem. SOC Japan, Pure Chem. Sect., 1959, 80, 159 and 162. Krishen, A., Dissert. Abstr., 1957, 17, 970.Motojima, K., Bull. Chem. SOC. Japan, 1956, 29, 71. Karabash, A. G., Peizulaev, Sh. I., Slyusareva, R. L., and Lipatova, V. M., Zh. Anal. Khim., Toribara, T. Y., and Sherman, R. E., Anal. Chmn., 1953,25, 1594. Bolomey, R. A., and Wish, L., J . Amer. Chem. SOC., 1950, 72, 4483 and 4486. Majumdar, A. K., and Singh, B. R., Anal. Chim. Acta, 1958, 18, 224. Agrinier, H., Bull. SOC. FranF. Mine’. et Crist., 1957, 80, 181; Bull. d’lnf. Sci. et Tech., 1958, 22, 10. Michal, J., Chem. Listy, 1956, 50, 542. Elbeih, I. I. M., and Abou-Elnaga, M. A., Chemist Analyst, 1958,47, 59. Milner, G. W. C., “The Principles and Applications of Polarography and other Electroanalyticai Processes,” Longmans, Green & Co. Ltd., London, 1957, p. 203. Venkataratnam, G., and Raghava Rao, B.S. Tr., J . Sci. I n d . Res., India, 1958, 17B, 360. Heyrovskg, J., and Berezicky, S., Coll. Czech. Chem. Comm., 1929, 1, 19. Willard, H. H., and Dean, J. A., Anal. Chem., 1950, 20, 1264. Kostromin, A. I., Uch. Zap. Kazansk. Univ., 1855, 115, 65. Gaudin, A. M., and Pannell, J. H., Anal. Chem., 1951,23, 1261. Aidarkin, B. S., Gorshkov, G. V., Grammakov, A. G., Zhadin, V. S., and Kolchina, A. G., Trudy Milner, G. W. C., and Edwards, J. W., U.K. Atomic Energy Research Establishment Report Alimarin, I. P., and Gibalo, I. M., Zavod. Lab., 1957, 23, 412. Alirnarin, I. P., Gibalo, I. M., and Sirotina, I. A., I n t . J . Appl. Radiation Isotopes, 1957, 2, 117. Bleshinskii, S. V., and Abramova, V. F., Izv. Akad. Nauk. SSSR, 1955, 37. Gold, R., Nucleonics, 1957, 15, 114.Lane, J. R., J . Metals, 1958, 738. Eberle, A. R., and Lerner, M. W., Metallurgia, 1959, 59, 49. Metallurgizdat,” Moscow, 1955, 172. 1990, Rochester University, New York, 194 8. 1959, 14, 94. Radiev. Inst. Akad. Nauk, SSSR, 1957, 5, 891. C/R 2965, Hanvell, 1959.February, 19611 OF BERYLLIUM. A REVIEW 93 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 218. 119. 120. 121. 122. 123. 124. 135. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. Tighe, J. J., Gerds, A. F., Center, E. J., and Mallett, M. W., U.S. Atomic Energy Commission Coleman, R. F., and Perkin, J. L., Analyst, 1959, 84, 233. Booth, E., and Parker, A., Ibid., 1959, 84, 546; 1958, 83, 241.Booth, E., Bryant, F. J., and Parker, A., Ibid., 1957,82, 50; U.K. Atomic Energy Research Estab- Ells, C. E., and Perryman, E. C. W., Atomic Energy of Canada Ltd. Reports 776, 1959, and 719, Rich, J. B., Redding, G. B., and Barnes, R. S., U.K. Atomic Energy Research Establishment U.K. Atomic Energy Authority Industrial Group Report AM/S 150, Springfields, Lancs., 1959. U.K. Atomic Energy Authority Industrial Group Report AM/S 151, Springfields, Lancs., 1959. U. K. Atomic Energy Authority Industrial Group Report AM/S 152, Springfields, Lancs., 1959. Wallace, C. G., U.K. Atomic Energy Research Establishment Report A/M 16, Harwell, 1959. Walkden, J ., U.K. Atomic Energy Research Establishment Report A/M 42, Harwell, 1959.- , U.K. Atomic Energy Research Establishment Report A/M 43, Hanvell, 1959. Conway, E. J., “Microdiffusion Analysis and Volumetric Error,” Fourth Edition, Crosby Lock- Walkden, J., U. K. Atomic Energy Research Establishment Report A/M 47, Hanvell, 1959. Huguet, J. L., Proceedings of the Second International Conference on the Peaceful Uses of Atomic Lerner, M. W., and Pinto, L. J., Anal. Chem., 1959, 31, 549. Killeen, 0. P., U.S. Atomic Energy Commission Report Y-1259, Oak Ridge, Tennessee, 1959. Polyakov, P. M., Rusanov, A. K., and Blokh, I . M., Zavod. Lab., 1957,23, 1320. Lee Smith, A., and Fassel, V. A., Anal. Chem., 1949, 21, 1095. U. K. Atomic Energy Authority Industrial Group Report AM/S 130, Springfields, Lancs., 1959. Walkden, J., U.K. Atomic Energy Research Establishment Report A/M 21, Harwell, 1959.-, U.K. Atomic Energy Research Establishment Report A/M 53, Harwell, 1959. U.K. Atomic Energy Authority Industrial Group Report AM/S 131, Springfields, Lancs., 1959. Bowell, E., U.K. Atomic Energy Research Establishment Report A/M 33, Harwell, 1959. U. K. Atomic Energy Authority Industrial Group Report AM/S 132, Springfields, Lancs., 1959. Walkden, J., U.K. Atomic Energy Research Establishment Report A/M 52, Harwell, 1959. -, U.K. Atomic Energy Research Establishment Report A/M 17, Harwell, 1959. U.K. Atomic Energy Research Establishment Report AM/S 133, Harwell, 1959. Walkden, J., U. K. Atomic Energy Research Establishment Report A/M 6, Harwell, 1959. U. K. Atomic Energy Authority Industrial Group Report AM/S 134, Springfields, Lancs., 1959.U. K. Atomic Energy Authority Industrial Group Report AM/S 135, Springfields, Lancs., 1959. Walkden, J., U. K. Atomic Energy Research Establishment Report A/M 22, Hanvell, 1959. -, U.K. Atomic Energy Research Establishment Report A/M 54, Harwell, 1959. U. K. Atomic Energy Authority Industrial Group Report AM/S 190, Springfields, Lancs., 1959. Booth, E., and Parker, A., U.K. Atomic Energy Research Establishment Report A/M 10, Harwell, Walkden, J., U. K. Atomic Energy Research Establishment Report A/M 45, Harwell, 1959. Mueller, J. I., and Simens, H. G., U.S. Atomic Energy Commission Report TID-7568 (Part l), Bowell, E., U.K. Atomic Energy Research Establishment Report A/M 39, Hanvell, 1959. Weber, W. P., Quirk, J. F. Lemmon, A.W., jun., and Filbert, R. B. jun., US. Atomic Energy Walkden, J., U.K. Atomic Energy Research Establishment Report A/M 44, Harwell, 1959. -, U.K. Atomic Energy Research Establishment Report A/M 49, Harwell, 1959. Yakovlev, Yu. V., Ref. Zh. Khim., 1956, Abstr. No. 36,101. Greitz, E. B., U.S. Bureau of Mines Report No. 5407, Washington, D.C., 1958. Kehres, P. W., and Poehlman, W. J., Appl. Spectroscopy, 1954, 8, 36. Alelrseeva, V. M., and Rusanov, A. K., Zh. Anal. Khim., 1957, 12, 23. Zotov, G., and Fowler, C. A., Atomic Energy of Canada Ltd. Report No. 298, Chalk River, Ontario, Zaidel, A. N., Kaliteevskii, N. I., Lipis, L. V., and Chaika, M. P., Zh. Anal. Khim., 1957, 12, 17. Owens, M. J., and Webb, M S. W. U.K. Atomic Energy Research Establishment Report C/R 2115, Wells, N., N.Z. J . Sci. Tech., 1956, 37B, 473. Brooks, R. 0. R., Nuclear Power, 1958, 3, 549. Cholak, J., and Hubbard, D. M., Anal. Chem., 1948, 20, 970. Birks, F. T., Spectrochim. Acta, 1955, 7. 231. Cholak, J., and Hubbard, D. M., Anal. Chem., 1948, 20, 73. Landis, F. P., and Coons, M. C., Appl. Spectroscopy, 1954, 8, 71. Garton, F. W. J , , Webb, M. S. W., and Sayer, J. A., U.K. Ministry of Supply CI/R42, Woolwich, Fred, M., Nachtrieb, N. H., and Tomkins, F. S., J . Opt. Soc. Amer., 1947, 37, 279. Davis, H. M., Parker, A., and Webb, R. J., U.K. Ministry of Supply CI/R82, Woolwich, 1963. Feldman, C., Anal. Chem., 1949, 21, 1041. Owen, L. E., Delaney, J. C., and Neff, C. M., Amer. Ind. Hyg. Ass. Quart., 1951, 12, 112. Smith, R. G., Boyle, A. J., Frederick, W. G., and Zak, B., Anal. Chem., 1952, 24, 406. U.K. Atomic Energy Authority Industrial Group Report AM/S 127, Springfields, Lancs., 1958. Report BMI-799, Battelle Memorial Institute, Columbus, Ohio, 1952. lishment Report C/R 2759, Harwell, 1959. 1958, Chalk River, Ontario; J . Nuclear Mat., 1959, 1, 73. Report, M/R 2685, Harwell, 1958. wood & Son Ltd., London, 1957. Energy, United Nations, Geneva, 1958, A/CONF./15P/1582. 1959. Oak Ridge, Tennessee, 1958, p. 265. Commission Report BMI-1165, Battelle Memorial Institute, Columbus, Ohio, 1957. 1955. Harwell, 1956. 1952.94. 153. 154. 155. 156. 157. 168. 159. ROBERTS AND SMITH : SPECTROPHOTOMETRIC MEASUREMENTS [Vol. 86 Churchill, W. L., and Gillieson, A. H. C. P., Sjkmtrochim. Ada, 1962,5, 238. Webb, R. J., Webb, M. S. W., and Wildy, P. C., U.K. Atomic Energy Research Establishment Report R/2868, Harwell, 1959. Anon., Instrum. Prucfice, 1959, 13, 1261. Anon., Chem. Eng. News, 1959, 74. Lee Smith, A., and Fassel, V. A., U.S. Atomic Energy Commission Report AECD-2100, Iowa State Scribner, B. F., and Mullin, H. R., J. Res. Nut. Bur. Stand., 1946, 37, 379. Zaidel, A. N., Kaliteevskii, N. I., Lipovskii, A. A., Razumovskii, A. N., and Yakimova, P. P., Received May 23rd, 1960 College, decl. 1948. Vestn. Leningrad Univ., 11, No. 22; Ser. Fis. i Khim., 1956, 18.

ISSN:0003-2654    

DOI:10.1039/AN9618600083

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

94. ROBERTS AND SMITH : SPECTROPHOTOMETRIC MEASUREMENTS [Vol. 86 Spectrophotometric Measurements of Theaf lavins and Thearubigins in Black Tea Liquors in Assessments of Quality in Teas* BY E. A. H. ROBERTS (Indian Tea Association, Butlw’s Wharf, London, S . E. 1) AND R. F. SMITH (The Laboratories, J . Lyons 6. Co. L.td., Kensington, London, W.14) The theaflavins and thearubigins formed as a result of enzymic oxidations of polyphenols during the fermentation process in tea manufacture determine the colour of a tea liquor and are associated with some of the other liquor characters recognised by tea tasters. A spectrophotometric method for determining theaflavin is described. This method also yields an approxima- tion of the total thearubigin content, and the ratio of optical-density values for thearubigins at 380 and 460 mp provides a further parameter giving an indication of whether lightly or deeply coloured thearubigins predominate.-4s a result of the fermentation process in tea manufacture the liquor of black tea develops certain characters not present in green or unfermented teas. Among the terms used by tasters to describe these characters are colour, strength, quality and briskness. It has been shown that these characters, measured organoleptically, are influenced by the extent to which fermentation has taken place, and, further, that colour and strength are correlated with the polyphenol oxidase activity and total polyphenol content of the plucked shoots from which the tea was manufactured.lS2 It follows that some, at least, of the liquor charac- ters of a black tea are due to the presence of polyphenolic enzymic oxidation products.The main polyphenolic oxidation products found in black tea extracts are the thear~bigins.~~~ These are a complex mixture of substances derived largely, if not entirely, from two parent substances, l-epigallocatechin and its gallic acid ester. These substances are not polymers, as previously thought, in fact they are probably mainly dimeric. They have fairly strongly acidic properties and in tea liquors considerable proportions of them are present as potassium and calcium salts. Although no pure substances have been isolated from the thearubigin mixture, some separation into fractions has been achieved. S I and S I1 fractions differ in their solubility relationships and in their chromatographic behaviour.3 The absorption spectrum of the S I1 fraction is similar to that of the S I fraction; it never- theless shows significantly greater absorption in the visible region, but not in the ultra- violet r e g i ~ n .~ The other oxidation products detected in black tea include theaflavin and theaflavin gallate, a bis-flavanol and its mono- and di-galloyl esters and several trace substance^.^ to 8 The theaflavin and theaflavin gallate have been isolated as pure substances and rather tentative structures have been suggested. The bis-flavanols show a single absorption band (Amax. = 280 m ~ ) ~ and contribute nothing towards the colour. Contributions by the trace substances may be disregarded, so that colour :in tea liquors may be considered as due to * Presented at the meeting of the Society on Tuesday, September 20th, 1960.February, 19611 OF THEAFLAVINS AND THEARUBIGINS IN BLACK TEA LIQUORS 95 theaflavins and thearubigins alone.Determination of these two groups of substances should therefore give a precise method of measuring the colour of a tea liquor, and such measurements might also be expected to be correlated with some of the other liquor characters. The method developed for determining theaflavins and thearubigins depends on the fact that the theaflavins are almost quantitatively extracted from a tea liquor by one extrac- tion with either ethyl acetate or isobutyl methyl ketone. These solvents do not extract thearubigins of the S I1 type, but there is a partial extraction of the free-acid forms of the S I type thearubigins.Potassium and calcium salts are not extracted. The thearubigins extracted by ethyl acetate or isobutyl methyl ketone are soluble in aqueous sodium hydrogen carbonate] whereas the theaflavins are insoluble. Complete separation of theaflavins and thearubigins is therefore effected by shaking the ethyl acetate or isobutyl methyl ketone extract with aqueous sodium hydrogen carbonate. Theaflavin and its gallate have well defined absorption maxima at 380 and 460mp.6*8 Either of these wavelengths is suitable for direct spectrophotometry in the extract washed with sodium hydrogen carbonate, as no other substances are present that absorb at these wavelengths. The fall in optical density, which results from the washing with sodium hydrogen carbonate, affords a method of determining the extractable thearubigins.Direct spectrophotometric determination of residual thearubigins in the aqueous layer after extraction with ethyl acetate or isobutyl methyl ketone is not possible, as a high pro- portion of the theambigin molecules is present as anions, which are more deeply coloured than the free acids. Addition of excess of aqueous oxalic acid reduces the colour intensity to that of the free acids, and spectrophotometry is possible after this acidification. METHOD PREPARATION OF THE TEA EXTRACT- In the traditional technique of tea tasting a weighed sample of tea is infused with boiling water for from 5 to 10 minutes in a porcelain pot fitted with a lid. The resulting liquor is decanted into a porcelain cup.This method was not sufficiently reproducible for analytical purposes and it was necessary to modify it; the modified method is described below. Weigh 9 g of tea into a 500-ml Erlenmeyer flask placed on a boiling-water bath. Bring 375 ml of distilled water to the boil, and pour it on to the tea. Allow extraction to continue for 10 minutes, without letting the temperature fall below 85°C. Filter through a plug of cotton-wool, and allow to cool just to room temperature. PROCEDURES FOR PARTITION AND SPECTROPHOTOMETRY- Shake 50 ml of the cooled, well shaken, filtered extract with 50 ml of isobutyl methyl ketone, taking care to avoid the formation of an emulsion. Allow the layers to separate, and dilute a 4-ml portion of the isobutyl methyl ketone layer to 25ml with methanol (solution A).Dilute a 2-ml portion of the aqueous layer to 10ml with water and then to 25ml with methanol (solution B). Shake 25 ml of the isobutyl methyl ketone layer vigorously for 30 seconds with 25 ml of a 2.5 per cent. aqueous solution of sodium hydrogen carbonate. Allow the layers to separate, and discard the aqueous layer. Dilute 4 ml of the washed isobutyl methyl ketone layer to 25ml with methanol (solution C). Add 2ml of a saturated aqueous solution of oxalic acid and 6ml of water to a 2-ml portion of the aqueous layer left from the first extraction with isobutyl methyl ketone, and dilute to 25 ml with methanol (solution D). Measure the optical densities] E,, E,, E, and ED, of solutions A, B, C and D, respectively, in 1-cm cells at 380 and 460 mp with a Unicam SP500 or similar spectrophotometer.NOTES-The two layers obtained from a 1 + 1 mixture of isobutyl methyl ketone and water are nearly equal in volume. With isobutyl methyl ketone there is no danger of the volumes of the layers being affected by partial saponification of the organic solvent, as might occur with ethyl acetate, particularly if recovered solvents are used. In the preparation of solutions B and D from the aqueous phase from the first partition, it is essential that the methanol content should not exceed 60 per cent. otherwise a cloudiness may develop, owing to precipitation of pectins. If the sodium hydrogen carbonate contains more than a small amount of sodium carbonate, theaflavins may be Iost by alkaline autoxidation during washing of the isobutyl methyl ketone extract.Analytical-reagent grade sodium hydrogen carbonate should therefore be used, and the solution should be freshly prepared. To minimise the possibility of losses by autoxidation the period of shaking96 ROBERTS AND SMITH SPECTROPHOTOMETRIC MEASUREMENTS [Vol. 86 must be as short as possible, and one layer should be completely removed from the other immediately after they have separated. Under normal conditions solution C is stable, but the measurement of its optical density should not be delayed. How- ever, the &/ED ratio has a particular significance, which will be referred to in subsequent publications. EVALUATION OF THEAFLAVIN CONTENT- There is always considerably more theaflavin gallate than theaflavin in a tea, so that little error will be introduced by expressing total theaflavins as theaflavin gallate.The optical densities, E,, at 380 and 460 mp are converted into a percentage of anhydrous theaflavin gallate by multiplying by the factors 2-25 and 6.69, respectively. These factors are applicable only when the conditions of extraction, partition and spectrophotometry are as described above, and are calculated from the optical-density values for theaflavin gallate shown in Table I. It is assumed that the theaflavin gallate used for standardisation purposes is the dihydrate of molecular weight 892-7.s In routine determinations of theaflavin and thearubigins no use is made of the E, values. NOTE-The ratio of optical densities at 380 and 460mp should be 2.98 to 1.If the ratio is appreciably greater than this, incomplete removal of thearubigins by sodium hydrogen carbonate is indicated. A lower value would suggest that the ratio of theaflavin to theaflavin gallate was greater than usual (see Table I). EVALUATION OF THEARUBIGIN CONTENT- Any factor for converting optical density into a percentage of thearubigin must be somewhat arbitrary, as we are dealing with a variable and rather complex mixture of sub- stances, none of which has been isolated in a pure state. A further complication is introduced by the uncertainty as to the degree of hydration of the thearubigins. It is apparent from Table I that the optical densities for the S I and S I1 fractions differ appreciably at 460 mp, but do not differ much from each other at 380mp.The average value of EY*:Zh at 380 mp for the two fractions is 0.733. Optical densities at 380 mp obtained for these two fractions with four other teas were- Sample No. .. .. .. 1 2 3 4 Optical density of S I fraction . . 0.712 0.655 0.658 0.739 Optical density of S I1 fraction . . 0.734 0.796 0.720 0.810 The acceptance of 0-733 as an average value of EP':Eh at 380mp for total thearubigins is therefore unlikely to lead to any considerable error, and, if this average value is assumed, the percentage of extractable thearubigins in a tea is, for the conditions of extraction, partition and spectrophotometry described above, approximately 7.06 (2E, + E, - Ec). The results must be accepted with certain reservations, but they are of the expected order of magnitude.THE ESIO/EIIO RATIO FOR THEARUBIGINS- The ratio of the values for 2E, + E, - E, at 380 and 460 mp in different teas has varied from 3.6 to 8.2. This indicates that the mixture of thearubigins in teas is by no means constant. If it is assumed that thearubigins have approximately the same optical densities at 380 mp, the variation in this ratio gives some indication of the average intensity of colour (at 460mp) of the thearubigins. A high ratio implies relatively light colour, and a low ratio a correspondingly deeper colour. As will be apparent from Table I, the more deeply coloured S I1 fraction has a lower ratio than the corresponding S I fraction. TABLE I E",Eh VALUES FOR THEAFLAVIN, THEAFLAVIN GALLATE AND THEARUBIGIN FRACTIONS The absorption spectra of the theaflavins and the thearubigin fractions were plotted over the range 220 to 600 mp.The resulting absorption curves and optical densities at certain wavelengths have been published previously5 ~8 E P ' ~ ~ ~ value at- Sample r - - - A - , Ratio 380 mp 460 mP E380/E460 Theaflavin . . .. .. . . 3400 1.235 2.75 Theaflavin gallate . . . . . . 2.225 0.747 2.98 Thearubigins (fraction S I) . . 0-717 0-138 5.20 Thearubigins (fraction S 11) . . 0.750 0.233 3.20February, 19611 OF THEAFLAVINS AND THEARUBIGINS IN BLACK TEA LIQUORS 97 This ratio is a useful extra parameter in describing the thearubigins of a tea, for thearubigin content is not necessarily proportional to depth of colour, particularly if the E380/E@) ratio is high. A rough estimate of the percentage of thearubigin in such teas is given by 4 x 7-06 (2E, + E, - Ea) when optical densities are measured at 460 mp.In the earlier stages of this investigation measurements were carried out at 460mp only, but, so long as the teas originated from Assam, the optical densities could be used to give an approximation of their thearubigin contents. With nearly all Assam teas this ratio falls between 3.6 and 4.4 (average 4.0). RESULTS Many commercial and experimentally manufactured teas have been analysed by the proposed procedure and detailed results will be reported elsewhere. Typical results are considered below. Those in Table I1 give some idea of the range of values obtained. TABLE I1 REPRESENTATIVE ANALYSES OF COMMERCIAL TEAS Thearubigin contents in brackets were calculated from optical densities at 460 mp Theaflavin Thearubigin Sample content, content, Ratio Taster’s remarks % % E,*O/E4liO N.E.India- - Golden colour, strong - Bright - Coloury, dull, common Good quality (C) . . .. 0.78 (8.9) Medium quality (C) .. 0.68 (7.6) Poor quality (C) .. 0.36 (7.1) Duars O.F. grade (C) . . 0-58 7.6 4.38 - 4-24 - Assam O.F. grade (Cj . . 0-70 13.1 Assam B.P. grade (CTC) . . 1.45 16.7 3.64 Very bright, golden Duars (Legg cut) . . .. 1.08 14.7 4.57 Hard, bright colour, with P.F. grade (C) . . .. 0-92 14.6 8.25 Bright, golden B.O.P.F. grade (C) . . . . 0.81 17.1 7.33 Rich, very coloury B.O.P. grade (C) . . .. 0.42 13-6 6.00 Dull B.P. grade (C) . . .. 0.34 8-5 - Thin, grey P.F. grade (C) . . . . 0.23 15-0 4.53 Very thick, dull, muddy rich colour Ceylon- Nyasaland- Kenya- Argentine- C = Conventional manufacture.CTC = C.T.C. manufacture. The thearubigin content of a tea is always considerably higher than the theaflavin content, but, as theaflavins are much more intensely coloured (see Table I), their contribution to total colour is a decidedly significant one. In assessing tone of colour, the taster is influenced more by the theaflavin content than the thearubigin content, and in conventionally manufactured teas the preferred liquor colours are associated with theaflavin contents of 0.75 per cent. or more. Conventional methods of manufacture, particularly in N.E. India, have to some extent been replaced by other methods (C.T.C. and Legg cut) in which the leaf receives a more thorough bruising, resulting in a quicker and more extensive fermentation.Analyses of such teas show increased theaflavin and thearubigin contents and an increase in the ratio of theaflavin to thearubigin. This accounts for the increased “brightness” and greater depth of colour normally associated with these methods of manufacture. Table I1 also shows the different E380/&60 ratios obtained for thearubigins. These are lowest in teas originating from N.E. India. Theaflavin and thearubigin contents are also affected by fermentation conditions, as indicated in Table 111. It will be noted that the theaflavin content reaches a maximum level comparatively early in the fermentation process, and that, in the later stages of fer- mentation, thearubigin contents increase slowly a t the expense of the theaflavins.As the liquor characters of strength, briskness and quality are affected by variations in the duration of fennentation,l it is considered probable that these particular characters are deterrnined, to some extent at least, by the theaflavin and thearubigin contents.98 SHANKARANARAYANA AND PATEL : THE VOLUMETRIC v o l . 86 TABLE I11 EFFECT OF DURATION OF FERMENTATION ON THEAFLAVIN AND THEARUBIGIN CONTENTS OF C.T.C. MANUFACTURED TEAS Each result is an average from six separate experimental manufactures Fermentation time, hours . . .. 1 2 3 4 5 Theaflavin content, % . . . . 1.61 1.46 1.34 1.27 1-17 Thearubigin content, % . . . . 13.0 16.2 16.6 16.7 17.1 CONCLUSIONS We are of the opinion that the measurement of theaflavins and thearubigins, coupled with the E,,,,/E,,, ratio for thearubigins, represents the best available method for determining colour in tea liquors.In view of the probable association of these variables with other liquor characters, it might be thought that these methods would serve as a basis for a chemical evaluation of the market value of a tea. Quality in tea is not determined by polyphenolic oxidation products alone, for there are other im- portant factors, among which caffeine and the volatile substances responsible for aroma may be mentioned. Market valuations are also affected by such factors as appearance and keeping properties on storage. The replacement of traditional methods of tea tasting by chemical analysis is not yet therefore in sight, but it is claimed that the methods described above should prove extremely useful in supplementing a taster’s report. Analytical results, unlike a taster’s evaluation, are not affected by market fluctuations, and are much more suitable when it is desirable to maintain records of the properties of teas manufactured. The analytical method may also be expected to be developed as a means of control during manufacture. We thank the Indian Tea Association (London) and Messrs. J. Lyons & Co. Ltd. for permission to publish this paper. This, however, is too optimistic a view. REFERENCES 1. 2. 3. 4. 5. 6. 7. -,- , Ibid., 1959, 10, 172. Roberts, E. A. H., J. Sci. Food Agric., 1958, 9, 381. Roberts, E. A. H., in preparation. Roberts, E. A. H., Cartwright, R. A., and Oldschool, M., J. Sci. Food Agric., 1957, 8, 72. Roberts, E. A. H., Ibid., 1958, 9, 212. Roberts, E. A. H., and Williams, D. M., Ibid., 1958, 9, 217. Roberts, E. A. H., and Myers, M., Ibid., 1959, 10, 167. 0. -, -, Ibid., 1959, lo, 176. Received October SEith, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600094

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

98 SHANKARANARAYANA AND PATEL : THE VOLUMETRIC v o l . 86 The Volumetric Determination of Dixanthogen BY M. L. SHANKARANARAYANA AND C. C. PATEL (Department of Inorganic and Physical Chemistry, Indian Institute of Science, Bangalore 12, India) A convenient method for determining dixanthogen has been developed ; it is based on the quantitative reaction of dixanthogen with potassium cyanide to form xanthogen monosulphide and potassium thiocyanate. Xanthogen monosulphide and the excess of cyanide are removed, and the thiocyanate is determined iodimetrically after conversion to cyanogen bromide. The method is applicable to the determination of dixanthogen in mixtures also containing xanthate. DURING work on the kinetics of the oxidation of xanthate to dixanthogen, it was necessary to determine the dixanthogen formed by the reaction.A search of the literature for a suitable method was unsuccessful, and attempts 'were, therefore, made to determine dixan- thogen by utilising the reaction reported by Whitby and Greenbergl According to theseFebruary, 19Sl-J DETERMINATION OF DIXANTHOGEN 99 workers, potassium cyanide reacts with dixanthogen to form xanthogen monosulphide and alkali thiocyanate, the reaction being represented by the equation- p S S 11 II B0-C-S-S-C-OR + KCN _~_f KSCN + RO- 4-C-OR A similar reaction has recently been used by Chatterjee, Banerjee and Sircar2 to determine thiuramdisulphide, but their method of determination is complicated. We have simplified this determination by adopting Schulek's method3 for determining thiocyanate (after the removal of xanthogen monosulphide and the excess of cyanide). In this method, thio- cyanate is allowed to react quantitatively with bromine to form cyanogen bromide in accordance with the equation- SCN- + 4Br, + 4H,O ___+ CNBr + 7Br- + SO,2- + 8H+ The cyanogen bromide is then allowed to react with potassium iodide to liberate iodine, the reaction being represented by the equation- CNBr + 21- - CN- + Br- + I, and this iodine is titrated with standard thiosulphate solution ; the amount of dixanthogen originally present is then calculated.METHOD REAGENTS- Unless otherwise stated, all reagents were of recognised analytical grade. Potassium cyanide solution, 2 per cent. w/v. Ammonium nitrate solution, 10 per cent. wlv. Bromine water, 5 per cent.w/v. Sodium thiosulphate, 0.05 N . Orthophosphoric acid, 17 per cent.-Dilute 1 volunie of 88 per cent. orthophosphoric acid with 4 volumes of water. Xanthates-Prepare potassium isopropyl-, n-butyl- and isopentylxan t hat e by Foster's method,4 and purify as described by Dewitt and Roper.5 The purity of the xanthates so prepared is greater than 99 per cent. as determined by the iodine method.6 Di-n-butyl dixanthogen solution-Oxidise a concentrated aqueous solution of n-butyl- xanthate by adding a concentrated solution of iodine, as described by Whitby and Green- berg.1 Extract the dixanthogen formed with light petroleum (boiling range 40" to 60" C), and evaporate the extract at room temperature by means of a current of air. Di-n-butyl dixanthogen is obtained as a yellow oil [sulphur found (Carius method), 42-39 per cent.; C,,H,,O?S, requires 42-98 per cent. of sulphur].Prepare a 1 to 1.5 per cent. w/v solution of the dxanthogen in acetone. PROCEDURE- Place 10 ml of di-n-butyl dixanthogen solution in a conical flask, and add 5 ml of 10 per cent. ammonium nitrate solution, 10 ml of 2 per cent. potassium cyanide solution and 10 ml of acetone. Heat the mixture at 40" to 50" C for about 25 minutes in a fume cupboard to ensure that the reaction between dixanthogen and cyanide is complete. Cool, transfer to a separating funnel, and shake well with 15 ml of benzene to extract the xanthogen mono- sulphide formed; repeat the extraction with a further 15 ml of benzene to ensure complete removal of the monosulphide (this is judged by the disappearance of the yellow colour of the solution).Transfer the aqueous solution to the original conical flask, add 10ml of 17 per cent. orthophosphoric acid, and heat on a bath of boiling water in a fume cupboard for about 30 minutes to remove hydrocyanic acid. Cool the solution, and add bromine water dropwise until a yellow colour stable for at least 5 minutes is obtained. Remove the excess of bromine by adding small amounts of ferrous sulphate, with stirring, until the yellow colour disappears. Add 3 g of potassium iodide and then a little sodium hydrogen carbonate, insert a stopper in the neck of the flask, shake well, and set aside in the dark for 10 minutes. Titrate the liberated iodine with100 SHANKARANARAYANA AND F’ATEL : THE VOLUMETRIC [Vol.86 0.05 N sodium thiosulphate; use starch as indicator. Calculate the amount of dixanthogen present in the original solution from the titre (after correction for the blank value). 1 ml of N sodium thiosulphate = 0.14925 g of di-n-butyl dixanthogen. DISCUSSION OF THE METHOD Solutions containing di-n-butyl dixanthogen were analysed by the proposed method; the results, which show that good accuracy is attainable, were- Dixanthogen present, g . . 0.1397 0.1397 0.1397 0.1397 0.1185 0.1185 0.1011 0.1011 Dixanthogen found, g . . 0.1409 0.1402 0.1402 0.1395 0.1188 0.1180 0.1010 0.1008 Error, g . . .. . . 0.0012 0*0005 0.0005 -0.0002 0.0003 -0.0005 -0*0001 -0.0003 In this work, the use of orthophosphoric acid has many advantages. It assists in con- verting potassium cyanide to hydrocyanic acid, which can be easily removed by boiling.Further, since orthophosphoric acid is a weak acid, the cyanogen bromide formed is stable in the solution. Also, orthophosphoric acid combines with the yellow ferric ions, thereby rendering the solution colourless. Removal of dixanthogen monosulphide from the solution is essential or the titration of the liberated iodine will be unreliable. The use of ammonium nitrate solution enhances the extraction of dixanthogen monosulphide by the non-aqueous solvent and so improves the accuracy of the method. DETERMINATION OF DIXANTHOGEN IN MIXTURES ALSO CONTAINING XANTHATE- It is well known that a xanthate is oxidised to the corresponding dixanthogen when exposed to the atmosphere. Further, an old or exposed sample of xanthate is invariably associated with the corresponding dixanthogen.It was, therefore, thought desirable to determine a dixanthogen in a mixture also containing xanthate; such mixtures were accordingly prepared. The xanthate was deteimined by Hirschkind’s method7 and the dixanthogen by the proposed method. In Hirschkind’s method, the xanthate is quantita- tively converted (by a mineral acid) to an unstable xanthic acid, which immediately dis- sociates into carbon disulphide and the corresponding alcohol. The amount of xanthate present can be calculated from the amount of mineral acid consumed. The procedure used is described below. Mix 10ml of an ethanolic solution of isopropyl- or isopentylxanthate with 10ml of dixanthogen solution (in acetone) in a conical flask, and determine the xanthate by Hirschkind’s method.’ (Use 0.05 N sulphuric acid, and determine the excess of acid by titration with 0.05 N alkali; calculate the amount of xanthate present from the acid consumed.) Add 15 to 20ml of acetone to the contents of the flask to obtain a clear solution, and then determine the dixanthogen by the proposed method. TABLE I AMOUNTS OF XANTHATE AND DIXANTHOGEN FOUND IN PREPARED MIXTURES Xanthate Dixanthogen Xanthate Dixanthogen Components of mixture present, present, found, found, g g g g Potassium isopropylxanthate and 0.0447 0.1146 0.0446, 0.0445 0.1151, 0.1151 di-n-butyl dixanthogen .. Potassium isopentylxanthate and di-n-butyl dixanthogen . . 0.0637 0.1027 0.0635, 0.0635 0.1029, 0.1036 Table I shows the results found when this method was applied to two prepared mixtures; We thank Professor M. R. A. Rao for his keen interest in the work. it can be seen that both xanthate and dixanthogen were detennined with good accuracy. REFERENC:ES 1. 2. 3. Whitby, G. S., and Greenberg, H., Trans. Roy. SOC. Canada, 1929, 23, 21. Chatterjee, P. K., Banerjee, D., and Sircar, A. K., J. Sci. Ind. Res., India, 1960, 1 9 ~ , 118. Schulek, E., 2. anal. Chem., 1923, 62, 337.February, 19611 DETERMINATION OF DIXANTHOGEN 101 4. Foster, L. S., illinera1 and Metallurgical Research Technical Paper No. 2, University of Utah, 1927, quoted by Gaudin, A. M., “Flotation,” McGraw-Hill Book Co. Inc., New York, 1957, p. 224. Dewitt, C. C . , and Roper, E. E., J . Amer. Chem. SOC., 1932, 54, 445. Matuszak, M. P., Ind. Eng. Chem., Anal. Ed., 1932, 4, 98. Hirschkind, W., Eng. Mining J., 1925, 119, 968. 5. 6. 7. Received August 17th, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600098

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

February, 19611 DETERMINATION OF DIXANTHOGEN 101 The Semi=micro Determination of Fluorine in Organic Compounds BY C. A. JOHNSON AND M. A. LEONARD (Analytical Development Group, Standards Department, Boots Pure Drug Co. Ltd., Station Street, Nottingham) A method is described for determining fluorine in organic compounds; i t is based on colorimetric determination of fluoride ions resulting from combustion by the oxygen flask method. The determination depends on the formation of a blue complex between fluoride ions and the cerium111 chelate of alizarin complexan. The influence of the type of glass of which the combustion flask is constructed has been quantitatively examined. A SPOT test involving use of the cerium111 complex of alizarin complexan (1,2-dihydroxy- anthraquinon-3-ylmethylamine-NN-diacetic acid) for detecting fluoride has been described by Belcher, Leonard and West .1 Subsequently, these workers devised a sub-micro method for determining fluorine in organic compounds, based on the same reaction.2 More recently, the theoretical basis of the method and the nature of the chelates involved have been des- ~ r i b e d , ~ and the purpose of this paper is to describe the application of the principle to the semi-micro determination of fluorine in organic compounds.EXPERIMENTAL The method investigated consists in decomposing the sample by the flask combustion m e t h ~ d , ~ a flask constructed of glass essentially free from boron and aluminium being used (see “Results and Discussion of the Method,” p. 103), and then colorimetrically determining fluoride in the resulting solution.In the colorimetric reaction used, fluoride does not exert its usual bleaching effect on the dye - metal complex, but itself enters the structure. The blue colour of the fluoride-containing complex (Lax. = 565 mp) is completely distinguishable from either the yellow of the free dye (Amax. = 423 mp) or the red of its cerium111 chelate (Amax. = 495 m ~ ) . ~ The wavelength chosen for the colorimetric determination (610 mp) corresponds to the maximum difference between the absorption spectra of the fluoride-con- taining complex and the cerium111 chelate. METHOD APPARATUS- The combustion apparatus consists of a 500-ml Erlenmeyer flask constructed of suitable glass. Into the stopper is fused one end of a length of platinum wire, 1 mm in diameter, to the free end of which is attached a piece of 36-mesh platinum gauze, 1.5 cm x 2 cm, to act as sample holder.Optical densities were measured in 4-cm cells with a battery-operated Unicam SP600 visual-range spec t ropho t ometer . REAGENTS- Alizarin comjdexan, 0.0005 MTransfer 0-385 g of alizarin complexan to a Z-litre cali- brated flask by means of 20 ml of recently prepared 0.5 N sodium hydroxide, and set aside for 5 minutes, with occasional swirling, to ensure complete solution. Dilute to about 1500 ml with water, add 0-2 g of hydrated sodium acetate, and adjust the pH to about 5 (thin layer of solution pink) by careful addition of 1 N hydrochloric acid. Dilute to the mark, and filter into a brown-glass bottle. This solution is stable for at least 4 months.102 JOHNSON AND LEONARD : THE SIIMI-MICRO DETERMINATION [Vol. 86 CerozGs nitrate, 0.0005 M-st andardise approximately 0.02 M cerous nitrate by titration against standard ethylenediaminetetra-acetic acid solution a t pH 6 with xylenol orange as indicator.To a suitable volume of this solution (about 50 ml) add 0.2 ml of concentrated nitric acid, 0.1 g of hydroxylamine hydrochloride and sufficient water to produce 2 litres, and filter. Acetate bufer solution, pH 4.6-Dissolve 150g of hydrated sodium acetate in about 600 ml of water, add 75 ml of glacial acetic acid, dilute to 1 litre with water, and filter. Standard jhoride solution, 5 pg $er ml-Dissolve about 22 mg (accurately weighed) of dried analytical-reagent grade sodium fluoride in water, and adjust the volume to 2 litres.Store in a polythene container. PREPARATION OF CALIBRATION GRAPH- In each of a series of 100-ml calibrated flasks place 50 ml of distilled water, an accurately measured volume between 2 and 8 ml of standard fluoride solution, 10 ml of alizarin complexan solution and 3ml of acetate buffer solution. Mix each solution thoroughly, add 10ml of 0.0005 M cerous nitrate, dilute to the mark with distilled water, and set aside, protected from direct light, for 1 hour. At the same time, prepare a blank solution in similar fashion by omitting the standard fluoride solution. Measure the optical densities of the test solutions against the blank in 4-cm cells at 610 mp, and plot a graph of optical density against amount of fluoride present.PROCEDURE- Accurately weigh an appropriate amount of the sample (5 to 25 mg) on a strip of filter- paper approximately 3 cm x 4 crn (Whatman No. 1 grade is suitable) that has been folded into three along its length. Enclose the sample in the paper, insert a narrow strip of filter- paper to act as a fuse, and fix it in the platinum-gauze sample holder. Place 20 ml of water in the combustion flask, fill the flask with oxygen, ignite the fuse, and immediately insert the stopper. Carefully tilt the flask, and, when combustion is complete, set it aside for 10 minutes, with intermittent shaking. Quanlitatively transfer the liquid to a 250-ml calibrated flask, dilute to the mark, and treat an aliquot expected to contain about 25 pg of fluoride by the procedure described for colour development under “Preparation of Cali- bration Graph.’’ At the same time, prepare a standard colour from 5 ml of standard fluoride solution to serve as a check on the calibration graph.Liquid samples can be satisfactorily decomposed by burning in a small gelatin or, preferably, methylcellulose capsule containing approximately 30 mg of powdered cellulose. For solutions derived from the combustion of sulphur-containing compounds, boil gently for about 10 seconds with 1 ml of 100-volume hydrogen peroxide, neutralise to phenolphthalein with 1 N sodium hydroxide, and then add 1 ml in excess; boil to destroy excess of peroxide, cool, and adjust the pH to about 4 with 1 N hyclrochloric acid. EFFECT OF FOREIGN IONS A qualitative survey of ions likely to interfere with the reaction has already been carried 0ut.l The results of a quantitative evaluation of some of the more common of these ions are shown in Table I, from which it can be seen that aluminium and iron cause serious inter- ference.Ions likely to be present as a result of the oxygen flask method of combustion of organic compounds exert a negligible effect, with the exception of phosphate, interference from which becomes significant if a large excess is present. TABLE I INFLUENCE OF SELECTED IONS ON :DETERMINATION OF FLUORIDE Amount of ion taming 10 per cent. decrease in colour produced by 26 pg of fluoride, pg Ion Aluminium .. .. 7 Ferric iron .. .. 8* Citrate . . .. .. 93 Phosphate .. .. 1000 Sulphate . . .. .. 16,500 Chloride . . .. ..315,000 Mole ratio of interfering ion to fluoride 0.19 0.105 0-36 7.7 123 6490 * Ferric iron forms a violet chelate that absorbs strongly a t 610mp and hence causes an increase in colour.February, 19611 OF FLUORINE IN ORGANIC COMPOUNDS 103 RESULTS AND DISCUSSION OF THE METHOD Calibration graphs prepared as described above are linear over the range 10 to 30% of fluoride, but above this range the sensitivity decreases, owing to the smaller excess of reagent. The slope of the calibration graph is positive, in contrast to those obtained by conventional bleaching methods, and the sensitivity is such that 1 pg of fluorine causes a change in optical density of about 0-014. This sensitivity is inversely affected by the concentration of acetate in the buffer solution.However, a decrease in the concentration of the buffer, although it increases the sensitivity, impairs the reproducibility. For this reason, test solutions should contain about 25 pg of fluoride. TABLE I1 FLUORINE CONTENTS FOUND BY PROPOSED METHOD AFTER COMBUSTION IN A SILICA FLASK Standard deviation Standard Number Mean of a single deviation as of Theoretical fluorine deter- percentage of Sample NO. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Compound 9-Fluorobenzoic acid, F.C,H,COOH Trifluoroacetanilide, C,H,.NHCOCF, (micro-analytical standard) . . m-Trifluoromethylbenzoic acid, CF,.C,H,.COOH (micro-analytical standard) . . .. . . . . (M.A.R. grade) . . .. .. . . Triamcinolone, C,,H,,O,F . . , . 9-Fluorohydrocortisone, C,,H,,O,F Hydroflumethiazide, C8H8N,04S,F, Bendrofluazide, C,,H,,04N3S2F3 .. Polytetrafluoroethylene, (CF,) . . Research compound, CBF,H,NO . . Research compound, C,,F,,SO,Na . . p-Fluorophenyl-fi-chlorobenzyl sulphone, Research compound, C8F,, (perfluoro- Research compound, Research compound, l-Fluoro-2,4-dinitrobenzene, C,H,O,N,F . . C,,H,,O,SClF . . .. . . . . cyclohexylethane) .. . . . . [CF,.C,H,P : (C&,),]&I - - . . .. [CFSC,H,P: (C2H,),]Cd12 . . .. .. Fluorobenzene, C,H,F . . .. .. Research compound, CF2C1CHC1, . . .. .. .. .. .. .. .. . . . . .. deter- fluorine minations content, (n) s/o 14 13.56 16 30-14 15 29-98 31 4-81 5 4.98 10 17-20 9 13.58 7 76-0 14 56-12 7 64-2 4 6-68 9 76-0 9 16.22 6 9.50 4 10.22 8 19.80 6 22.45 content mination* mean fluorine found (y), % 13-55 30.10 30.00 4.81 4.89 17-27 13.48 75.8 55.30 62.4 6.69 72.42 15-63 9-15 10-39 19.69 21.96 (S), content found % 0-167 1-16 0.382 1.27 0.248 0.83 0.059 1-22 0.035 0-72 0.058 0-34 0.142 1.05 0.80 1.05 0.85 1.53 0.36 0.58 0.041 0.62 0.79 1.1 0.23 1.47 0.056 0.61 0.089 0.86 0.192 0.98 0.127 0.58 * Calculated from the expression S = dZ.(x - y)z/(n - l), in which x is the result of anindividual determination. obtained (with the use of a silica flask for combustion) are shown in Table 11. results were satisfactory, but certain of the research compounds gave low results. The proposed method has been applied to a wide range of compounds, ar.d the results Most of these Sample TABLE I11 COMPARISON BETWEEN RESULTS FOR FLUORINE AFTER COMBUSTION IN FLASKS OF DIFFERENT MATERIALS Theoretical fluorine Compound content, % Triamcinolone .. .. .. 4-81 p-Fluorobenzoic acid . . . . 13-56 m-Trifluoromethylbenzoic acid 29.98 Polytetrafluoroethylene . . 76.0 Hydroflumethiazide . . . . 17.20 Fluorine content found after combustion in- silica boron-free-glass* borosilicate-glass flask, flask, flask, 4.8 1 4.79 4-68 13-55 13.47 13-03 30.00 29.95 28.81 75.8 76.9 72.1 17.27 - 16.71 ( L 3 % % % * This glass contained 3 per cent. of alumina.104 DREWES: THE ANALYSIS OF LEAD TANNATE AND [Vol. 86 No. 10 was reported to be difficult to decompose by fusion with an alkali metal, but the pre- cision of the results obtained by the flask method points to complete combustion. Samples Nos. 12 and 17 were volatile and non-inflammable. It was noted in the initial stages of this work that the use of borosilicate-glass combustion flasks gave consistently low results. Combustions were subsequently carried out in silica and in boron-free-glass flasks, and some comparative results are shown in Table 111.It is considered that boron-free-glass flasks are suitable for routine determinations, although the results obtained are slightly lower than those found when silica flasks are used. This is probably due to the presence of a small amount: of aluminium and is consistent with Rogers and Yasuda’s observation^.^ Our work has confirmed Rogers and Yasuda’s contention that decomposition of the -CF, group by flask combustion is complete; we have not found it necessary to use oxidation aids, such as sodium peroxide,6 potassium chlorate2 or potassium nitrate. We thank Dr. A. M. G. Macdonald of Birmingham University for supplying a number of the research compounds listed in Table 11. REFERENCES 1. 2. 3. 4. 5. 6. Belcher, R., Leonard, M. A., and West, T. S., Taluntu, 1959, 2, 92. , J. Chem. SOC., 1959, 3577. Leoiard, M. A., and West, T. S., Ibid., 1960, 4477. Schoniger, W., Mikrochim. Ada, 1956, 869. Rogers, R. N., and Yasuda, S. K., Anal. Chem., 1959, 31, 616. Steyermark, A., Kaup, R. R., Petras, D. A., and Bass, E. A., Mzcrochem. J., 1959, 3, 523. Received November 2nd, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600101

出版商:RSC    

年代:1961

数据来源: RSC