摘要:

SEPTEMBER, 1963 THE ANALYST Vol. 88, No. I050 Report of the Analytical Methods Committee I96 1 I196 2 THIS seventh Report of the Analytical Methods Committee of The Society for Analytical Chemistry reviews the progress of work during the two years 1961 and 1962, covering the period January lst, 1961, to December 31st, 1962. ANALYTICAL METHODS COMMITTEE Chairman: D. C. Ganatt, Ph.D., IISc., Hon.M.P.S., F.R.I.C. Boots Pure Drug Go. Ltd. A. J. Amos, O.B.E., BSc., Ph.D., F.R.I.C. R. Belches, Ph.D., D.Sc., F.Inst.F., F.R.I.C.* E. Bishop, BSc., A.R.C.S.T., F.R.I.C. L. Brealey, B.Sc., F.R.I.C. S . G. Burgess, BSc., Ph.D., F.Inst.Pet., J . H. Hamence, M.Sc., Ph.D., F.II.1.C. W. C. Johnson, M.B.E., F.R.I.C. A. G. Jones, BSc., F.R.I.C.7 D. Mi’. Kent-Jones, B.Sc., Ph.D., F.R.I.C. E.Q. Laws, B.Sc., F.R.I.C. D. T. Lewis, C.B., BSc., Ph.D., D.Sc., F.R.I.C. R. F. Milton, B.Sc., Ph.D., M.I,Biol., F.R.T.C. R. M. Pearson, A.R.I.C.? A. A. Smales, O.B.E., B.Sc., F.R.I.C. R. E. Stuckey, Ph.D., D.Sc., F.P.S., F.R.I.C. C . Whalley, B.Sc., F.R.I.C. K. A. Williams, B.Sc., Ph.D., A.Inst.P., H. N. Wilson, F.R.1.C.t F.InstS.P., F.R.I.C+ M.Inst.Pet., F.R.I.C. Analytical and Consulting Chemist (President of Univevsity of Birmingham (Professor of AnaZyticaZ University of Exeter (Department of Chemistry) International Combustion Ltd. Scienti3c Advisey, London County Council Public Analyst, Uficiat Agricuituraal Analyst and Hopkin & Williams Ltd. Inz$erial Chemical Industries Ltd. (Plastics Analytical and Consulting Chemist D.S.I.R., Laboratovy of the Governnzant Chemist D.S.I.R., Labovatory of the Government Chemist Analytical and Consdting Biochemist Imperial Chemical I-rzdustries L i d .(Billingham A tomic Energy Research Establishment, Harwelt Honorary Secretary of the Society Laporte Chemicals Lid. Analytical and Consulting Chemist Inq5evial Chemical Industvies Ltd. (Biliinghawz the Society) Chemistry) Consulting Chemist Division) (Honorary Treasurer of the Society) Division) Division) Secretary: Miss C. H. Tinker, B.Sc., Ph.D., A.R.I.C. Assistant Secretary: Miss A. M. Parry, BSc. {Resigned November 1961) Miss V. Lewis (Appointed November, 1962) * Temporarily resigned May, 1961. 7 Appointed May, 1962. $ Resigned July, 1962. GENERAL REVIEW During 1961, the Committee was sorry to lose Dr. J. Haslam, who resigned on his retirement from Imperial Chemical Industries Ltd.; he had been a valuable member of the Committee for many years.The Committee was also sorry to lose, through resignation to take up another appoint- ment, the services of Miss A. 111. Parry in November, 1961. Miss Parry had been Assistant Secretary to the Committee since January, 1956 ; she was succeeded, in November, 1962, by Miss V. Lewis. PROGRESS OF WORK- Some projects have been completed and the Sub-committees and Panels, having prepared their Reports, which will be published during the coming year, have been disbanded. Two new committees were appointed during 1961. One of these is a Sub-committee of the Analytical Methods Committee for investigating methods for Particle Size Analysis. This Sub-committee had 659 The Committee can once again report steady progress in its work.660 l<EI’ORT OF THE AN.iLYTIC.4L hi ETHODS COMMITTEE 1961/196;! AndJSt, VOl.88 its first meeting i n March, 1961, and, after nearly 2 years of vigorous work, it has prepared a comprchensi1.e classification of methods that, it is hoped, will prove useful to chemists and physicists new to thc subject and who require guidance on how to sct about their task of finding suitahle methods for their own particular types of materials and fields of application. This classification is bcing published in thc March, 1963, issue of The Analyst. Thc other com- mittee to be appointed is a ncw Panel of the Joint Committee of the Society (as represented by the Analytical Methods Committee) and The Pharmaceutical Society on Methods of Assay of Crude Drugs.This Panel, the seventli to be appointed by the Joint Committee, is inlmti- gating chemical methods for assaying the biologically active constituents of thyroid. An exploratory meeting was lield on the subject in Ilccember, 1961, and thc Panel had its first meeting in Februarj., 1962. Unlike most sub-committees and panels of the A.M.C., which are commissioned to examine published or known methods of assay with a view to ensuring their applicability and reproducibility for reference purposes, this Panel has the more difficult task of devising a method practically de m v o ; for this purpose, its inklestigations arc more in the nature of research un a collaborative basis. The other existing Sub-Committees of the h.3I.C.and Panels of the Joint Committee have, on the whole, been actively at work during the period under review. One of the Joint Committee’s Panels--that dealing with the investigation of chemical methods for anthra- quinone drugs (senna, cascara, etc.)--has resumed work again having been in suspension for nearly 2 years pending a parallel investigation on biological methods by an associated Panel (Panel .?A). Panel 3.4 has now unified a biological techniquc for senna fruit that can be used as a form of “yardstick” for correlating chemical methods with purgative activity in mice. Another Panel of the Joint Committee-that on methods for lonchocarpus and derris (Panel 5)-cornpleted its programme of work at the end of 1961 and its second Report, on the Colorimetric Dctermination of Rotenone, was published in The A tta2yst in lu’ovember of that year.The IDGO Report of the A.M.C. presented rather a gloomy picture with regard to the difficulties encountered by the Additives in Animal Feeding Stuffs Sub-committee in its endeavour to apply known methods of anallsis to anirnal and poultry feeds. I t says much for all the members of the five Panels of that Sub-Committee that during the past two >‘cars these difficulties have been largely resolved, with the result that four of these Panels have completed their programmes and h a w heen disbanded, and their Reports are to be published shortly. They include methods of assay for antibiotics (penicillin, chlortetracycline and oxytetracycline) , for synthetic hormones (stilboestrol and liexoestrol) and for a number of vitamins (vitamin -4 and p-carotene, of the fat-soluble types, and vitamin HI,, nicotinic acid, pantothenic acid and riboflavin, of the water-soluble group of %vitamins).Only the methods for two of the 13-vitamins-choline and pyridoxin -have proved to be somewhat intractable, and it became evident that these would require lengthy and specialised investi- gation; some of the problcms encountered are described later (see p. 665). The fifth Panel o f this Sub-Committee---that on prophylactics-has completed its work on one item, nitrofuranone, and has prepared its Report, which will he published with those on the other additives. In view of this YaneI’s lengthy programme, it has been decided that it should be re-organised as a Sub-Corninittee in its own right directly under the aegis of the A.M.C.This arrangement has enabled the Additives Sub-committee to wind up its commitments, and it has been disbanded. Associated with the Additives Sub-Committee is another Sub-Committee that dealt, independentl~., with mcthods for determining trace elements in fertilisers and feeding stuffs, This latter committee started its work in 1957- about a year and a half before that on additives -and it has now completed its IengtEiy programme of work; tlic collected methods for about fourteen elements are to be published in l!)M by the Society as n. separate booklet, since it is envisaged that they will be capable of application to products other than fertilisers and feeding stuffs. Another Sub-Committee to complete part of its programme is that on Metallic Impurities in Organic Matter.In 1960, it published two Keports, on Small Amounts of Arsenic and on Methods for the I>cstruction of Organic Matter; during 1961, the work on a method for copper was completed and the Report is being published in the April, 1963, issue of The Analyst. Meanwhile, the Sub-Committee has continued with its programme and has been con- This Panel has now becn disbanded.September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE 196111962 661 sidering methods for zinc and mercury; after examining its specificity in the presence of certain other metals that react with dithizone, the Sub-committee can recommend the method for zinc adopted by the Trace Elements in Fertilisers and Feeding Stuffs Sub-committee.The deter- mination of very small amounts of mercuryin organic matter has presented a problem to analysts €or a long time, since the removal of the organic matter can frequently lead to a significant loss of the metal. It is possible that radiochemical procedures (particularly isotope dilution), which are rapidly increasing in popularity and becoming more available to smaller labora- tories, might be used as the basis of a reference method for mercury, and the Sub-committee has been considering this possibility. However, because some laboratories may not have the necessary facilities for some time, the Sub-Committee thinks that a chemical method should be investigated in the first place, and collaborative work is proceeding on a method that appears to be promising and that is sensitive for measuring amounts down to half a microgram.During this collaborative work, use is being made of radiochemical techniques for monitoring the various stages during the wet-combustion procedure to check that there is no loss of mercury. Two Reports were published in the September, 1961, issue of The ArtaZyst by the Meat Products Sub-Committee. The first Report, on Nitrogen Factors for Pork, presents a vast amount of evidence from results collected from laboratories in the United Kingdom and in Europe to show that the hitherto-accepted factor of 3.6 for the conversion of meat nitrogen into meat content was too high for pork. This had meant that in the past manufacturers of pork products often had to put in more meat than necessary to ensure that analysis of their products would give results within the statutory limits; the factor of 3.45, as recommended in the Report, appears to be the best compromise on the evidence of the results presented, and it has been adopted for use by the Association of Public Analysts and by the meat trade.The second Report is a short one dealing with the Nitrogen Content of Rusk Filler. Since rusk filler is used in the manufacture of sausages and some other meat products, it was considered necessary to revise the value for its nitrogen content in view of changes that have occurred in milling practice since the second World War. The Sub-committee is continuing its programme of work by collecting values for the nitrogen contents of beef and of chicken meat, in a similar manner to that carried out for pork.Its Report on the Nitrogen Factors for Beef has been completed and will be published in 1963, and its Report on chicken meat is being prepared. The Chlorine in Organic Compounds Sub-committee completed its programme of work during 1962, and its Report on the application of the oxygen-flask combustion method to the determination of pesticides containing organically-bound chlorine will be published in 1963. It was reported in 1960 that two more Panels (set up jointly by The Society for Analytical Chemistry , the Association of British Manufacturers of Agricultural Chemicals, and the Scientific Sub-Commit tee of the Int erdepart ment a1 Advisory Commit tee on Poisonous Substances used in Agriculture and Food Storage) had been set up to study methods for determining malathion and organo-mercury residues.The Report on malathion was published in The Analyst at the end of 1960, and that on organo-mercury appeared in September, 1961. Another Panel set up in 1960 was that on demeton-methyl residues in fruits and vegetables; its Report was published in The A.nalyst in June, 1962. The Joint Committee of The Pharmaceutical Society and The Society for Analytical Chemistry (represented by the Analytical Methods Committee) has continued its work during the past 2 years and, as mentioned earlier, the Panel on Lonchocarpus and Derris has been disbanded on completion and publication of its second Report. Also, as mentioned earlier, the new Panel on Thyroid (Panel 7) has now been at work for nearly a year.The Panel engaged on methods for the chemical assay of the Capsaicin Content of Capsicum and its Preparation has completed its programme of work and is preparing its second Report, the first being published in 1959. The first Report included recommended methods for the assay of the drug and for three official preparations (oleoresin, tincture and ointment). Since then, experience of the methods has revealed the need for some revisions and these are included in the second Report, together with a simpler method for ointment; a more practicable colorimetric method, involving use of Gibb’s phenol reagent, has been devised €or end- determination (the spectrophotometric difference method remaining the method of choice) and is also included.662 REPORT OF THE ANALYTICAL METHODS COMMITTEE 1061/1962 [AfldySt, VOl.88 Other Sub-Committces of the A.M.C. and Panels of the Joint Committee have continued Details of individual committees-their personnel and with their programmes of work. reports of work-will be found below. S.A.C. SCHOLARSHIP- L)r. J. H. Stevenson was appointed in October, 1960, to carry out investigation into bio-assay methods for determining pesticide residues, a t Rothamsted Experimental Station. He was unable to continue his work for a second )rear, and his results arc now awaiting publication. ASALYTICAL METHODS TRUST- The second of the three-year periods for the receipt of promised subscriptions to the Trust Fund ended in 1960 and, as in 1958, an Appeal to Industry to continuc its financial aid of the work of the Analytical Methods Committee was made in June, 1061.The response to this appeal was extremely gratifying; 76 organisations promised donations, some of which were by Deed of Covenant for 7 years, others for 3 years and some others for 1961 only. This reprcscnted an increase of 51 over the number in 1960 and showed the wider response t o the appeal. The total amount received for 1961 was E5,G65, and firm promises totalling #,64(i were made for each of the years 1982 and 1963. ASK u AL ACCO U NTS- The audited statements of accounts for the two financial years ending October 31st, 1961, and October 31st, 1962, are shown in Appendixes I and 11, respectivcly. REPORTS OF SUB-COMMITTEES OF THE ANALYTICAL METHODS COMMIlTEE PUBLICATIONS SUB-COMMITTEE COKSTITUTIOY- L>.C . Garratt, Ph.D., T>.Sc., Hon.M.P.S., F.IC1.C. J. B. httrill, M.X., F.R.I.C. J . H. Hamence, M.Sc., Ph.L)., F.H.I.C. D. W. Kent-Jones, B.Sc., Ph.D., V.H.I.C. H. X. IYilson, F.K.Z.C.’ S. C . Jolly, RSc., B.Pharm., M.Y.S., A.R.I.C. Chairman, Analytical Mellrods Committee Ediiov, The Analyst Membev, A izalytical .Methods Committee Meinber, A nalylical Methods Commitlee Menaber, Analytical Methods Committee Editor, Scientijc Publzcalions, The I’havnzaceuti- (Chairmaw) (Secvetavy and Editor) cal Society of Gveal Ilvitazn * Resigned July, 1962. PROGRESS OF WORK- Much progress has been made in the collection and editing of all those recommended methods of analysis published since 1936 by the Analytical Methods Committee in its various Reports.These collected methods were published in March, 1063, under the title “Official, Standardised and Recornmended Methods of Analysis,” and included a revised and expanded version of the Bibliography of Standard and Recommended Methods originally published in 1951. ADDITIVES IK ANIMAL FEEDISG-STUFFS SUB-COMMITTEE CONSTITUTION- D. C. Garratt, Ph.D., Il.Sc., 14on.BI.l’.S., V.H.I.C. A. 3 . Amos, O.B.E., B.Sc., I’h.D., F.R.T.C. W. T. Barker* j. H. Hamcnce, M.Sc., Ph.D., F.II.1.C. R. F. l’hipcrs, R.Sc., Ph.Tl. S. A. Price, BSc., F.1C.I.C. C . J. liegan, RSc., F.R.I.C. Formerly Chemist-in-Chief, London Caunty W. I,. Sheppard, F.R.I.C. Boots Puve Drug Co. Lid. (Formerly with C’tai- R. I?. Stuckcy, Ph.D., I>.Sc., F.P.S., F.K.I.C. British Drug Houses Lld.F. R. Williams7 Ministry of Agviculture, Fishevies and Food Miss C. H. ‘Tinkcr (Secretary) Boots Pure Drug Go. Ltd. AnaLyyticaZ and Consulting Chemist Ministry of Agriculture, Fishevies ond I:ood Public Analyst, tlflcial Agriculttrral Analyst and Coopev Technical Bureau Vitamins Ltd. (Chaivman) Consulting Chemist Council levev I-td.) Appointed May, 1962. 7 Resigned May, 1962.September, 19631 REPORT O F THE ANALYTICAL METHODS COMMITTEE 198111962 663 TERMS OF REFERENCE-"TO investigate and prepare methods for determining the amounts of additives (nutrients, stimulants and prophylactics) in animal and poultry feeding stuffs. " PROGRESS OF WORK- Although the task was expected to prove difficult when originally undertaken by the Sub-committee, the efforts of its Panels have led to a resolution of most of the problems. The Antibiotics, Hormones and the Vitamins (fat-soluble) Panels have completed their programmes; the Vitamins (water-soluble) Panel has completed its work on all but two of the vitamins listed in its programme.The various Reports by both these groups, giving recommended methods of assay, are being submitted for publication. The Prophylactics Panel has completed its Report on Nitrofurazone. Since the remaining projects require a considerable amount of further investigation, either collaboratively or by individual research, it was decided to disband the Sub-Committee, which had acted purely in a steering and advistory capacity for all the Panels, This has now been done, and the opportunity is taken here to thank the members, not only of the Sub-committee itself, but of all the Panels, for the vast amount of hard work that has been so willingly undertaken.Only the Prophylactics Panel survives to carry on its programme of work, and, to allow of its re-organisation and broadening of scope, it has been decided to reconstitute it as a Sub-committee of the Analytical Methods Committee in its own right. During the work of the Additives Sub-committee it has become apparent that, in some instances, wide tolerances on the assay must be allowed; an unrealistic amount of extra work would be necessary to increase the accuracy. Throughout the work, the problem of sampling has been apparent and the Sub-committee emphasises that the stated accuracy of a recom- mended method can be attained only if the analyst's sample is properly prepared; problems involved in sampling the bulk material are, however, not within the Sub-Committee's province.ANTIBIOTICS CONSTITUTION- S. A. Price, B.Sc., F.R.I.C. A. J. CavelI, M.Sc., A.R.C.S., D.I,C., F.R.I.C. Mrs. J. Gammon, B.Sc. 0. Hughes W. P, Jones, F.P.S., F.R.I.C. G. Sykes, MSc., F.R.I.C. J. H. Taylor, Ph.D., M.R.C,V.S. S. Varsanyi, A.I.S.T. Miss A. M. Parry (Secretary) (Chairman) PANEL Vitamins Ltd. Ministry of Agriculture, Fisheries and Food, National Agricultural Advisory Service Formerly Cyanamid of Great Britain Ltd., Lederle Laboratories Pjizer Ltd. Cyanamid of Great Britain Ltd., Ledevle Labora- tories Boots Pure Drug Co. Ltd. Cyanamid of Great Britain Ltd., Agricultural Glaxo Laboratwies Ltd.Division PROGRESS O F WORK- The Panel has completed its programme of work and its Report will be published in 1963. Methods for the three permitted antibiotics (penicillin, chlortetracycline and oxy- tetracycline) can now be recommended for the assay of feeding stuffs, with the limitation that the identity of the antiobiotic is known and also that only one antibiotic is present. Because of the higher levels of antibiotics in the former (a few grams per pound), chemical methods of assay are recommended; for the much lower levels in feeds (a few grams per ton), microbiological methods are required. CONSTITUTION- Both supplements and supplemented feeding stuffs are covered. HORMONES PANEL R. E. Stuckey, Ph.D., D.Sc., F.P.S., F.R.I.C. (Chairman) J. Allen, A.R.I.C. L.Brealey, B.Sc., F.R.I.C. J . A. Potter, A. R.I.C. W. L. Sheppard, F.R.I.C. Miss A. M. Parry (Secretary) British Drug Houses Ltd. British Drug Houses Ltd. Formerly Boots Pure Drug Co. Ltd. Analytical and Consulting Chemist Formerly Unilever Ltd.664 REPORT OF THE AXALYTICAL METHODS COMMITTEE 1961 I1962 :Analyst, Vol. 88 The Panel has completed its programme of work and its Report will be published in 1963. Only the two synthetic hormones, stilboestrol and hexoestrol, have been investigated and physico-chemical methods are recommended. PROGRESS OF WORK- CONSTITUTIOS- I<. I;. Phipcrs, U.Sc., I”h.L>. C . Ff’. Batlard, B.Sc., F.P.S., F.K.I.C. N. C. Brown, M A . , B.Sc., ;L’lI.I.C. H. C;. Dickcnson, Iz.Sc., 1’h.D. C;. Drewery, BSc., F.R.I.C.? A.If‘. Hsrtley, I:.R.I.C. A. Holbruok, F.R.I.C. (Chairman) PROPHYLACTICS PANEL D. 11. Mitchell: S. G. E. Stevens, B.Sc., F.R.I.C. J . A. Stuhbles, L1.Sc. Miss A. M. Parry (Secvetariesl Iliss C. H . Tinker } * Corresponding member. Cooper Technical Uuveau May G. Daker Ltd. Coopcr Technical Uureazc Ward, Rlenkinsop 6 Co. Lid. .Merck Shai.9 6 Dohme Ltd. I wipevial Chemical Industvies Ltd. (Phanntrceldi- Wellcome Chemical Wovks Smith Klim (4, 1;rench I,abouatovies Lid. May G. Ilaker I-td. Sfiillers I-td. cals Division) t Appointed January, 1962. Appointed April, 1 W I . PROGRESS OF WORK- The Panel has a long programme of work, complicated by tlic fact that the fashion in prophylactic chemicals is constantly changing, so that new substances are frequently appearing and others are likely to go out of fashion before there has been time to devise a suitable method for determining them in feeds.A method for nitrofurazone can now be recommended, and the Panel’s Report on this will be published in 1963. The work on methods for acinitrazole, sulphaquinosalinc and llmprolium is virtually cornpletcd, and ICeports ;ire to be preparcd for publication. As mentioned above, the future work of rhe Panel is to be continued by the “Yrophylac- tics in Animal Feeds Sub-committee,” which. is being formed from the Pancl members with additional representation from other organisations interested in the subject ; Dr. Pliipers will continue as Chairman of the ncw Sub-committee. VITAMISS (FAT-SOLUBLE) P-WEJ, COSSTlTVTIOX- JY. L. Sheppard, F.R.1 .C.C . It. Louden, n.Sc., F.R.I.C. H. Pritchard, >I.%., F.IZ.1.C. S. A. Reed, B.Sc., X.R.I.C. C;. \Vhalley, B.Sc., F.R.I.C. 3. M’illiams, B.Sc., Ph.ll., l:.l<. I .C. Miss C. H. Tinker (Secretary) (Chair Mian) PROGRESS OF WORK- B(~0t.q Puve Drug Co. I A . (Formerly with L-nz- R. Silcock G. Sons I M . Analytical and Consultiiig Chemist Uvitislz Cod L i w Ozls (Hull and C;rzrnsl+~) L f d . L-ndever L f d . Spillevs L!d. lever Ltd.) As reportcd 9 years ago, tht: Panel has completed its work on the spectrophotomctric method for both \vitamin A and for p-carotene, and thc Report has been approved for publication. In addition, a colorimetric method has sine-o been tested collaborativcly and is to be included in the Keport as a quick routine method; however, it is not sufficiently reliable to recommend it as a reference method.‘The Panel has now been disbanded,September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE 1961/1962 665 VITAMINS (WATER-SOLUBLE) PANEL CONSTITUTION- A. J. Amos, O.B.E., B.Sc., Ph.D., F.R.I.C. J . E. Ford, B.Sc., Ph.D. B. M. Gibbs, B.Sc., A.R.I.C. F. W. Norris, Ph.D., D.Sc., A.R.C.S., D.I.C., S. A. Price, BSc., F.R.I.C. H. Pritchard, M.Sc., F.R.I.C. F. Clermont Scott, BSc., F.R.I.C. S. Varsanyi, A.I.S.T. J. Williams, B.Sc., Ph.D., F.R.I.C. Miss C . H. Tinker (Secretary) AnaEytical and GonsztEting Chemist National Institute f o r Research in Dairying Unilever Ltd. University of Birmingham (Department of Bio- F. R.1 .C. chemistry) Vitamins Ltd. Analytical and Consulting Chemist Vitamins Ltd. Glaxo Laboratories Ltd, Spillers Lid.(Chairman) PROGRESS OF WORK- The Panel has studied and tested collaboratively on animal feeding stuffs selected methods of assaying riboflavin, nicotinic acid, vitamin B,, pantothenic acid, pyridoxin and choline. Because of their specificity and sensitivity, microbiological procedures have been the methods of first choice. It has been possible to determine microbiologically nicotinic acid, vitamin BIZ, pantothenic acid and riboflavin in unsupplemented and supplemented animal feeding stuffs with an acceptable “between-laboratories” variance, and the Panel’s two reports on recommended methods for these four vitamins has been approved for publication. During the Panel’s earlier investigation of the determination of riboflavin by the microbiological method recommended by the Analytical Methods Committee for foods (Amlyst, 1946, 71, 397), it was suggested that the unacceptable divergence between the results obtained in different laboratories was attributable to the extraction procedure.As a result of a study of the effect of variations in the method of extraction, it has been possible to introduce modifications that allow the riboflavin activity of animal feeding stuffs to be determined with an accuracy commensurate with those of the assay methods recommended for the other three vitamins. Collaborative studies of a microbiological method involving S. zymogenes have revealed that, despite its greater sensitivity, the method has no advantage over the recommended method, which relies on L. heZvei!iczts as the test organism.Continued studies of the remaining two vitamins on the Panel’s original list-namely, pyridoxin and choline-have failed to produce methods of assay that can be recommended for use with animal feeding stuffs. Several microbiological methods of assaying pyridoxin have been subjected to collaborative trials, but none has given results with an acceptable “between-laboratories” variance. Collaborative studies of microbiological and chemical methods of assaying choline have been no more successful. ANALYTICAL STANDARDS SUB-COMMITTEE CONSTITUTION- E. Bishop, B.Sc., A.R.C.S.T., F.R.I.C. (Chaivmm) S . Andrus, A.R.I.C. P. R. W. Baker, M.Sc., A.R.I.C. A. G. Hill, F.R.I.C. R. M. Pearson, F.R.I.C. Imperial Chemical Industries Lid. (Billingham J .M. Skinner, B,Sc., Ph.D., E.R.T.C. Imperial Chemical Industries Ltd. (Billingham IV, I . Stephen, B.Sc., Ph.D., A.R.I.C. University of Birmingham (Department of K. E. Topp, B.Sc., Ph.D., A.K.I.C.* J. T. Yardley, B.Sc., F.R.I.C. Miss C. H. Tinker } University of Exeter (Department of Chernisfry) Lupovte Chemicals Ltd. Wellcome Research Laboratories British Drztg Houses Ltd. DiUiSiOPZ) D i V i S i O f i ) Chemistry) National Chemical Laboratovy Hopkin G. Williams Ltd. Miss A, M- Parry (Secretaries) * Appointed April, 1961.666 REPORT OF THE -4NALYTICAL METHODS COMMITTEE 1961/1962 [Analyst, VOl. 88 TEims OF REFEREKCE-“TO examine existing analytical standards and to select suitable substances. ” PROGRESS OF WORK- Continuing the critical assessment of standards for acid - base titrimetry with a view to recommending a suitable standard to IUPAC, the Sub-committee has set up a carefully defined scale of standards. After a detailed appraisal of the material submitted for con- sideration, sodium carbonate was selected for initial experimental examination in view of new work on the preparation of this compound.A statistical design was set up for a collab- orative assay involving four laboratories, four samples of sodium carbonate prepared by two different methods and two reference standards-atomic-weight silver and zone-refined benzoic acid, Experimental proccdurcs were devised to give the required precision of *O-l per cent., and have been tested. The first collaborative assay broke down over the difficulty of storing and transporting molar hydrochloric acid without change of concen- tration; this effect was later discovered to be due mainly to leakage through the stoppers of polythene bottles.The second attempt failed becausc of the difficulty of removing carbon dioxide from solution without simultaneous loss of traces of benzoic acid. The third assay, in which the sodium carbonate samples were related to a uniform sample of atomic-weight silver through individual hydrochloric acid samples, met with considerable success, but revealed an interesting laboratory bias of 50.02 per cent. Although the result of this assay fell just within the prescribed limits, a fourth and final assay on fresh samples was made, and confirmed that the samples were assaying at about 99.96 per cent. The varieties of sodium carbonate comprising the test samples did not therefore meet the requirements for a primary standard.Nevertheless, confirmation was received that the methods of assay developed for this work were giving adequate replication and that certain modifications and alternatives also reached this standard. Other sources of sodium carbonate were examined, and a fifth collaborative assay showed that samples assaying at 99.096 per cent. were available. Meanwhilc, attempts were made to trace the source of the 0.04 per cent. deficiency in the earlier samples, but were met with only partial success. The latest assays, showing a scatter of 18 parts per million over 35 analyses, were regarded as satisfactory evidence that the pro- posed methods were suitable and that sodium carbonate was an appropriate primary standard for the calibration of solutions of strong acids.Finally, a readily accessible source of sodium carbonate, on which preliminary assays are promising, is being investigated. -4 final report, embodying the recommendation of sodium carbonate, together with the method of assay, is being prepared and will be made available to the IWPAC Conference in July, 1963. A detailed examination of losses of benzoic acid has also been made, with the conclusion that such losses come within the experimental error and can readily be determined. The scatter of the assay results, however, of 200 parts per million indicate the unsuitability of benzoic acid for applications to titrimetry, although the Sub-committee entertains no doubts as to its purity and suitability in other applications.The possibilities of using the coulomb as a universal reference standard are being explored, and a proposal has been made to the Analytical Methods Committee that specialised research work on this project would be well worthwhile. CHLOKI KE IS ORGANIC, COMPOC‘SDS SUB-COMMITTEE CONSTITUTION- li. Belcher, l’h.I>., I).%., F.Inst.F., F.R.I.C. J . H. Dunn, RSc., A.II.1.C. I<. Gardner, R S c . , F.H+I.C. R. Goulden, F.1I.I.C. C. A. Johnson, n.Sc., B.Pharm., F.P.S., F.Ii.1.C. , I r i s A. M. G. Jiacdonald, >I.%., Ph.L)., .\.li.I.C, bliss C . 1.1. Tinker (Secrefary) (Ckairman) Linitlerszty of L3ivntinghum (Professor o f A nalyticul Plant Protection Ltd. Fisons Pest Contvol Lfd. “Skell” Research I-td.t3ooi‘s Pure Drug Co. Ltd. Unicersity of Birmilzgham (Deparlmeizt o j Chemistry) Chemistry) TERMS OF REFE:HEE;CE--(‘TO prepare methods for the determination of organically-bound chlorine, having special reference to commercial preparations such as pesticides.”September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE X961/1962 667 PROGRESS OF WORK- The work of the Sub-Committee has now been completed, and its Report will be pubfished in 1963. The method recommended is based on the oxygen-flask combustion technique, and it has been tested collaboratively on (a) chlorobenzoic acid, (b) dieldrin 50 per cent. water- dispersible powder and (c) a miscible oil containing pentachlorophenol, by both the members of the Sub-committee and other analysts who had not had experience of the technique. Results were excellent and the method can be accepted with confidence as a standard method for all but low-concentrate pesticide dusts, in which the high proportion of inorganic filler prevents complete combustion; for this latter type of product the Stepanow method should be used.ESSENTIAL OILS SUB-COMMITTEE CONSTITUTION- G. W. Ferguson, BSc,, Ph.D., F.R.I.C. A. J. M. Bailey, B.Sc., F.P.S., F.R.I.C. H. E. Brookes, BSc., F.R.I.C. K. Field, M.Sc., Ph.D.* D. Holness, B.A. H. T. Islip, BSc., F.R.I.C. P. McGregor, B.Sc., AX.-W.C., F.R.1.C.t T. L. Parkinson, B.Sc., Ph.D., F.R.I.C. Miss €3. M. Ferry, M.Sc., F.R.I.C. G. B. Pickering, M.A., B.Sc., Ph.D., A.R.I.C. J. H. Seager, M.Sc., F.R.I.C. S. G. E. Stevens, B.Sc., F.R.I.C. B. D.Sully, BSc., Ph.D., A.R.C.S., F.R.I.C. Miss C. H. Tinker (Secretary) (Chairman) * Appointed July, 1962. Analytical and Consulting Chemist W . J . Bush & Co. Ltd. Boots Pure Drug Co. Ltd. D.S.I.R., Laboratory of the GovernrPze%t Chemist Proprietary Perfumes Ltd. Formerly Tropical P~oducts Institute D.S.I.R., Labovatory of the Government Chemist Beecham Foods Ltd. Staford Allen G. Sons Ltd. D.S.I.R., TropicaE Products Institute Yardley Q Co. Ltd. Smith Kline G- French Laboratories Ltd. A . Boake, Roberts & Co. Ltd. Resigned July, 1962. PROGRESS OF WORK- Studies of methods for the evaluation of specific groups of substances in essential oils have been continued. Completion of Mr. Holness’s study of the reactions of citronellol and geraniol, on formylation, and the publication of his Report (Analyst, 1961, 86, 231) provided a major contribution to the proceedings of the I.S.O.meeting on Essential Oils at The Hague in May, 1961. The van 0 s and Elema method for determining geraniol and citronellol when these occur together has also been studied, as have been alternative methods of acylation of alcohols. However, the Sub-committee is now much in favour of Dr. Sully’s stearoylation method (Andyst, 1962, 87, 940) for the determination of hydroxyl groups, and a series of collaborative tests is at present being carried out. Investigations into the determination of carbonyl groups have continued, and a pre- liminary study has been made of spectrophotometric methods that have been published for the determination of citral as a, p-unsaturated aldehyde.Work has continued on the deter- mination of phenols and some collaborative tests have been made on Dernetrius and Sin- sheimer’s method; it is hoped that a short Report on these investigations will be published during the coming year. MEAT PRODUCTS SUB-COMMITTEE CONSTITUTION- S. M. Herschdoerfer, Ph.D., F.R.I.C. S. Back, RSc., F.R.I.C. P. 0. Dennis, B.Sc., F.R.I.C. J. R. Fraser, BSc., A.C.G.F.C., F.R.I.C. H. C. Hornsey, F.R.I.C. -4. J. Kidney, B.Sc., Ph.D., A.R.C.S., A.R.I.C.* R. A. Lawrie, B.Sc., Ph.D., F.R.I.C. T. McLachlan, D.C.M., A.C.G.F.C., M.I.Biol., A. McM. Taylor, B.Sc., Ph.D., F.R.I.C. E. F. Williams, M.A., F.R.I.C. Miss C. H. Tinker (Secvetary) T . Wall & Sons (Ice Cream) Lid. Crosse G. Blackwell Ltd. 0 x 0 Ltd. D.S.I.R., Labovatory of the Government Chemist J .Sainsbury Ltd. T . Wall G. Sons (Meat and Handy Foods) Ltd. A .R.C., Low Temperature Research Station Public Analyst British Food Manufacturing Industries Research J . Sainsbwry Ltd. (Chairman) F.R.1.C.t Association * Appointed August, 1961. t Appointed June, 1962.668 [A.rtalyst, Vol. 88 TERMS OF HEFEHESCE-"(~) the determination of the meat content of products containing meat; (b) the determination of the constituents of meat and meat products. REPORT OF THE AK'ALYTICAL METHODS COMMITTEE 1!361/1962 Xo-rE-The term 'meat products' to include hydrolyscd protein and, if found necessary, fish pastes." PROGRESS OF WORK- The Sub-committee's Reports on Xitrogcn Factors for Pork and on Nitrogen Content of Rusk Filler were published in The Analyst in September, 1Ml.Work has continued on the collection, on similar lines, of values for the nitrogen contents of beef and chicken meat. The Sub-committee's Reports on beef and chicken factors will be published in 1963. COXSTITUTION- JfETALLIC IMPURITIES IN OKGAXIC MATTER SUB-COMMITTEE I\'. C. Johnson, M.B.E., F.1C.I.C. J . C. (;age, T3.Sc., Ph.U., F.K.1.C. Intpevial Chemical Industries L f d . (Indtdriui T. T. Gorsuch, B.Sc., Ph.l>., A.1LI.C. L'.I<. A toinic Etzrvgy Autliority, ?'he Hadiocitemi- Miss E. $1. Johnson, AI.Sc., -1.R.I.C. ljpitish Food ;Ma?zujactitr*i-itg Industries Ileseavcli H. F. hlilton, ILSc., I'h.I>., bI.l.l3iol., F.K.I.C. -1 iiaiyfical and COnSdtlNg Uiochemisl 13. J . ru'ewtnan, RSc., F.1i.I.C. Hopkin & Williams 1,td.IY. G. Sharples, A.R.I.C. Imperial Chemical Industries Lid. (IJyestufls G. B. Thackray, B.Sc,, I;.R.l.C.* E. I. Johnson, 31,Sc., I;.H.l.C.t J . 1;. C. 'l*j.Ier, n.Sc., Ph. I)., A.K.l.C,; hliss A. hl. I'arry (Secveluuies) Miss c'. H. 'Tinker } Hopkin 15 U'illianis Lid. (Chairman) Hygiene La4Joralovies) c ( I 1 Cent re Association Dzais i o ~ ) Sla,t;/ovdsliire Couiity Council L > . S . I . R., Luboralovy of llre (;oi!errtnienl Cheinisl D..S.l. I ( . , Laborafovy of the Goveriiiirent Ci1e.nzi.d * .4ppointed May, 1 N i I . i ;\ppointcd May, I9fid. l<esignecl hhy, 1962. TERNS OF I~:ERESCE-'"~O investigate the determination of small quantities of metals in organic matter. " PROGRESS OF WORK- The Sub-committee cornplctcd its work on the determination of trace amounts of copper in organic matter, and its Report will be, published in 19(i3. The method for zinc adopted by the Trace Elerncnts in Fertilisers and Fceding Stuffs Sub-Committee has been accepted as being applicable to organic matter generally, mid the specificity of the method for zinc in the presence of certain other metals that react with dithizone has been confirmed.The rather special problems attending the dctcrmination of mercury in organic matter have been considered at some length, and a programme of individual and collaborativc work is in hand; use is being made of radiochemical techniques to c1iet:k m y losses of mercury during t h : various stages o f the procedure. D 111 ECT >I I C RO- DE TE I< hi I S A TI 0 S 0 F OX Y G E 5 I: N 0 H Cr .A N I C M A'ITE R S U H -C 0 M 31 I TTE E COXSTIT~XIOS- L).lyilson, M.Sc., F.R.1 .C. 1.'. It. \\'. Thker, X S c . , :\.I<. I.C. JIiss 13. H. Hauininger, l'h.l)., A . I . . f < . I . , li. K.1.C. i\-. '1'. (:liarnbers, B.Sc., l'h. I)., -4. It. 1.C:. S i r .JoJm Cuss Colicge (Depwtiitmt I ) / Clrenzisfvy) 1.1 'rllco tiie 11'esm rcli l. ah o m to vies I)u?iIup Jtesrurch C'cnfvc l l r i f i s h It'ubbev Yimlzrccrs' Kcsearch .-I ssocintioia (CIf r r i r n m ) F. Colson, 13.Sc., Z'h.l>., F.1I.I.C. 1 J i i p P 7 Z U / Cliemical 1 nilustries I-td. ( A Ikali Di?:isi~iz) (Drparlrrtent of Chemistry) Miss bl Comer, R.Sc., F.R.1 .C. * BIiss J . C'uckncy irripryial College oj. Sciencc aiid T ~ c l ~ o i ~ g y 1:. l<llington, 13.Sc., A.K.C.S., F.1C.l.C. .Yutioiid Coal Board, Ctml Research Establish~iient 1:.J , McMurray 11 'dicoine ChPmicui IVorlis 31. 1'. m11(1~)z:L, n s c . , A.R.C.S. Ijriiisli Coal L-tilizniion Research .+lssocialioiz H. J . iVarlow 11..5. I . R . , Tropic01 l'rodi4cfs lwstitlrtr c i\I1dlcy, B.SC., 1:.1<.1.c. I.ciportc Ckcwicn1.s Ltd. 3Iiss C. I I . 'l'inkcr (Sccvctury) II..S.l.R., .Yatiuttal C k e n i i c d Laboratory 1:. H. Oliver I W { ~ . r107.i.q co. * T)ccoaserl So\wnber, 1962.September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE 1961/1962 669 TERMS OF REFERENCE-~~TO investigate the Unterzaucher method, and its modifications, for the micro-determination of oxygen.” PROGRESS OF WORK- since 1960. No progress can be reported since this Sub-committee has been in enforced abeyance PARTICLE SIZE ANALYSIS SUB-COMMITTEE CONSTITUTION- E.Q. Laws, B.Sc., F.R.I.C. (Chairman) R. de E. Ashworth, M.Sc., Ph.D., F.R.I.C. D. G. Beech, Ph.D. C. G. L, Furmidge, B.Sc., Ph.D., A.R.I.C. H. Heywood, Ph.D. H. W. Hibbott J. F. Hinsley, F.I.M.* R. Howes, B.Sc. R. Jackson, Ph.D. B. €3. Kaye, BSc., Ph.D. Miss C. H. Tinker (Secretary) B.S.T.R., Laboradory of the Government Chew.& Ministry of Agriculture, Fishevies and Food, Plant Pathology Laboratory British Ceramic Research A ssociation “Shell” Research Ltd. Director, WooEwich Polytechnic D. R. Collins Ltd. Edgar Allen & Co. Ltd. Chesterford Park Research Station British Coal Utilisation Research Association Research Council of the British Whiting Federalion {Formerly with Nottingharn & Disfrict Tech- nical Colkge) * Appointed March, 1962.TERMS OF REPERENCE-“TO study methods of particle size analysis, to survey available instruments and to evaluate them with regard to their principles of operation and fields of application. ’ PROGRESS OF WORK- The Sub-Committee held its first meeting in March, 1961, and its initial task has been to list and classify existing methods for determining particle size in the sub-sieve range (i.e., below 76 p ) . After 2 years’ hard work this classification, comprising 74 methods, has been completed and is being published as a Review article in The Analyst in March, 1963. The principle of each method is briefly described and accompanied by adequate literature references. The second, and more difficult, stage of the Sub-Committee’s work will consist of some form of appraisement or evaluation of methods or apparatus, and its is envisaged that, since this will entail collaborative tests of a wide variety of apparatus, the constitution of the Sub-committee will be supplemented by co-opted members, as and when necessary for special tests.TRACE ELEMENTS IN FERTILISERS AND CONSTITUTION- C. J. Regan, B.Sc., A.R.I.C. S. M. Boden, B.Sc., F.R.I.C. (Chairman) S . G. Burgess, B.Sc., Ph.D., F.Inst.Pet., J. H. Hamence, M.Sc., Ph.D., F.R.I.C. E. I. Johnson, M.Sc., F.R.I.C. R. F. Milton, B.Sc., Ph.D., M.I.Biol., F.R.1.C K. L. Mitchell, B.Sc., Ph.D., F.R.I.C., F.R.S.E. J. R. E. Patterson, M.Sc., F.R.I.C. W. L. Sheppard, F.R.I.C. j.,TVilliams, B.Sc., Ph.D., F.R.I.C. icliss C. H . Tinker (Secretary) M.Inst.S.P., F.R.I.C.FEEDING STUFFS SUB-COMMITTEE Formerly Chemist-in-Chief, London County Council Ministry of Agriculture, Fisheries and Food, National Agricultural Advisory Service Scienti3c Adviser, London County Council Public Analyst, UBcial AgriculturaE AnaEysd and D.S.I.R., L ~ O Y ~ Q Y Y of the Government Chemist Analytical and Consulting Biochemist Macaulay Institute for Soil Research Ministry of Agyiculture, Fisheries and Food, National Agricultuml Advisory Service Boo€s Pure Drug Co. Ltd. (Formerly with Unilever Ltd.) SpiElers Ltd. Consulting Chemist670 REPORT OF THE ANALYTICAL METHODS COMMITTEE 1961 11962 [ A ndyd, VOl. 88 T E r i h i s OF IIEFERENCE-‘‘TO devise appropriate methods of analysis (to be recommended for inclusion in the Kegulations under the Fertilisers and Feeding Stuffs ,4ct, 1926) for the dt:termination of boron, cobalt, copper, fluorine, iodine, iron, magnesium, molybdenum, selenium and zinc, which can be expected to be present in fertilisers and feeding stuffs.” UJ- the end of 1961, the first draft of the final Kcport had been prepared; this comprised :in introductory section, a section on general considerations and then detailed procedures for t h t preparation of the sample and for the determination of the eleven elements listed above.In addition, methods for chromium and nickel were added to cater for fertilisers con- taining sewage sludge; methods for calcium and chloride ion were also included as a result of a request from thc Additives in Animal Feeding Stuffs Sub-Committee, of which this Sub-commit tee acted as the Minerals Panel.For several of the elements, alternative methods arc given for application in various circumstances. Publication of these collected methods is to he in the form of a separate booklet, under the titIe “Determination of Trace Elements, with Special Reference to Fertilisers and Feeding Stuffs,” since it is envisaged that the methods will find a wider application than that for which the>- were originall!- intended; the book is being published early in 1963. PROGRESS OF WORK--- A1 2: C 0 31 M I TTE I: C o N s r I T u *r I o N - Ti. K. Capper, l’h.l>., RPharm., F.l’.S., 11.1 .C.. J . ..Illen, .A.Ii.I .C.* ‘r. C. Ilenston, 13.Pharm.z J . IY. Fairbairn, B.Sc., Ph.l)., F.P,S., I;.l-.S., .-I. J . l=euell, B.Sc., Ph.l)., A.R.Z.C.I-). C. Chrratt, I’h,D., I).Sc., FIon.M.P.S., F,lI.l+C. Phnmrnceitlicul Society of G i ~ u t Hvitain lrrilish l k u g Houses Lad. llvitish Pharmacoporiu Contmission C:n ive Y d y of I- n n do n { .Pvcflcsso r of I’h arm acog - U.S.I.H., Tropical Products Insiitute Chairman, Analytical Meihods Conrnoittee of the (Chaivmalz) 17. It. E .c. “ O S Y ) S, A .C. I<. Higson, I;.P.S. C . A. Johnson, n.Sc., B.l-’harm., F.P.S., F:,R.I.C, 14. C . Macfarlanc, A.R.T.C.S., F.K.1.C.S IV. Mitchell, 13.Sc., Ph.l)., F.R.I.C. R . I;. l’hipers, RSc., 1’h.D. J . 31. I h w s o n , JI.Sc., Yh.D., l:.l’,S,, I:.I,.S. 1). IYatt, F.P.S. Yiss C. H. Tinker (Secretary) 3liss :I. 31. I’arry (Assistant Secvetavy)§ i ~ i s t v y of MeaEfh (Supplies Uvamh) Huots Puve Dvug Co. Ltd. .4 nalytical and Comdting Chemid Stafioovd Allen (5 S o n s IAd.Cooper Technical Uureazr Ilradford Institufe of Techmlngy (Departmnt of T . & If. SmiM Ltd. Pliarnzncy) * Appointed February, 1962. Rcsigncd October, 1962. t Appointed February, 1962. $ licsigncd Sovember, 1961. ‘rERMS OF HEFEIwxcE---‘‘To prepare standard met hods of assay of crude drugs and kindred materials.” PROGIIESS OF N’ORK- The Main Committee, in its steering capacity, is able to report continued progrcss in most of its Panels. Since its establishment in 1!)5ti, eight Panels have hem set up to examine methocis of assay of different drugs that arc widely used but have no offcia1 or rccogniscd ;issay procedure. During the past 2 years, one panel has been disbanded on completion of its programme of work and one new Panel has been sct up; two other Panels have bccn in abeyance pending the outcome of associated work clscwhere.Panel 5 (Lonchocarpus and Derris) was disbanded during 1961 on completion of its second Keport-The Colorimetric Determination of Rotenone-which u.as pul)lishc!d in 7’he Analyst in Xovember, 1961. Mention has becn ma&: carlier of the work of tlic new Panel on chemical methods for Thj,roid (Panel 7 ) , and further details of its progress and problems are given below.September, 19631 67 1 Panel 1 (Digitalis Purpurea) was suspended in 1959 because no further progress could be made on a collaborative basis until considerable individual research had been undertaken. Since the problems encountered by the Panel have not yet been resolved elsewhere, it has has been decided to disband it; it is hoped, however, that it will be reconstituted at some later date.The other Panel that has been in suspension since 1960 is that dealing with chemical methods for anthraquinone drugs (Panel 3); as mentioned earlier this Panel began work again in December, 1962. The three remaining Panels have continued with their programmes of work, and one of them-Panel2 (Capsicum-Capsakin Content)--is preparing its second, and find, Report. REPORT OF THE ANALYTICAL METHODS COMMITTEE 1961/1962 PANEL 2 : CAPSICUM-CAPSAICIN CONTENT CONSTITUTION- H. B. Heath, M.B.E., B.Pharm. C. F. G. Fost, M.P.S. C. A. Macdonald, B.Sc., F.R.I.C. E. A. Elsbury, F.R.I.C.7 A. J . Middleton, B.Pharm., M.P.S., A.R,I.C,* G. R. A. Short, F.P.S., F.L.S.G. I. Srnales, B.Sc., F.R.I.C. Miss G. M. Wells, B.Sc., APJ. A. J. Woodgate, BSc. Miss C. H. Tinker } (Chairman) Miss A. M. Parry (Secretaries) Staflord AlEen G. Sons Ltd. W . J . Bush G. Co. Ltd. Evans Medical Research Labovatovies Parke, Davis & Co. Pavke, Davis G. Co. W . J . Bush G. Co. Ltd. Pavke, Davis G. Co. Reecham Research Labovatovies Ltd. Staffoord Allen & Sons Ltd. * ,Appointed May, 1962. t Resigned March, 1962. TERMS OF REFERENCE-“TO investigate methods of assay of capsicum and capsicum products with particular reference to the determinatian of the capsaicin content.” PROGRESS OF WORK- Owing to the potential explosion hazards associated with the preparation of dry diazo- compounds, the Panel’s work has been concentrated on perfecting an alternative colorimetric method of assay for the determination of capsaicin.This is based on the reaction of phenols with dichloroquinone-chloroimide (Gibb’s reagent) to give a blue colour that can be evaluated spectrophotometrically. The Panel has also developed an assay procedure for determining capsaicin in Capsicum Wool B.P.C. A report embodying these, together with certain amendments that have been found desirable to the originally published methods (Analyst, 1959, 84, 8031, is in preparation. With the completion of this report, the Panel’s programme of work will come to an end. PdXEL 3: ANTHRAQUINONE DRUGS CONSTZTUTION- W. Mitchell, B.Sc., Ph.D., F.R.I.C. J. W. Fairbairn, B.Sc., Ph.D., F.P.S., F.L.S., C . A. Johnson, B.Sc., B.Pharm., F.P.S., F.R.I.C.S. C. Jolly, B.Sc., B.Pharm., M.P.S., A.R.I.C. Miss H. M. Perry, M.Sc., F.R.I.C. J. M. Rowson, M.Sc., Ph.D., F.P.S., F.L.S. H. A. Ryan, B.Sc., F.R,I.C. W. Smith, B.Sc., F.R.I.C,* R. V. Swann, B.Sc., F.R.I.C. Miss A. M. Parry 1 Miss C . H. Tinker ISecre’avies) (Chairman) F.R,I,C. Staflord Allen 6 Sons Ltd. URiversity of London (Professor of Pharnzacog- Boots Pure Drug Go. Ltd. Pharmaceutical Society of Great Bf itain Stafford Allen G. Sons Ltd. Bradford Institute of Technology (Department of Westminster Laboratories L t d . Allen G. Hanbu~ys Ltd. Allen G- HaPzburys Ltd. nosy) Phurnzacy ) * Resigned September, 1962. TERMS OF REFERENCE--I ‘TO investigate methods for estimating the purgative activity of drugs and preparations of drugs containing anthraquinone derivatives with a view to recom- mending standard methods of assay.”67 2 PROGRESS OF iVORK- REPORT OF THE .AS.\LYTIC.\L METHODS COMMITTEE ltt61/196? ! F! ?lalJJSt, 1:01.s8 Since this Panel had becn in suspension since SIaj’, 1960, no progress or work can be reported. As mentioned earlier in this Report, it resumed work in December, 1962, d t c r its associated Panel (3.4) on bioassa!, methods had submittcd its report on a unified biological procutlure for senna fruit that can bc used as a form o f “>-ardstick” for correlating chemical met hods with purgative activity in micc. L)r. Rowson, the Panel’s original Chairrnm, resigned from that office in July, 1957, on his appointment to a post abroad, and Ih. Mitchell succeeded hirn. Now, on Dr. Rowson’s rcturn to this country, I)r.Mitchell has askcd t o be rclicwd of this duty, which he nssurnccl on a temporary basis, and I h . Ronwn i5 oncv more the Chairman of tlit I’anvl. rlhc Joint Committee thanks Dr. JIitchell for so ably stepping into the Imacli. * Resigned 1Iarc-h. 19ti2. PltOC,R1:SS OF WORK- The reproducibility of results between Ialioratories for the biological assaj- mcthocl when inice and the same solutions and thc same conditions were used has been examined. \\‘ith this information, re-esamination of two samples of senna pod (pre\-iouslj* examined by Panel 3 b!r a chemical method) against a common sennoside standard has been made. The dis- agreement betnywn laboratories beyond the limits of esperiniental error lias been removed. jI full report of these investigations ivas submitted to Pancl 3, n.110 made the original request for the \vork to be done.TERMS OF ~FERESCT.:-‘‘~‘O investigate nit tho& of ma!’ of derris, Ionchocarpus and thc4r preparations, with particular reference to the dvtcrniination of the rotcnonc content.” PROGRESS OF 1VORK- The Panel continud its work on a colorimetric method for dctcrmining low conccntIa- tioris of rotenonu, as n routine control proccdur c. Optinium conditions ha\-c hcen establisl~ecl and details of the method are publishcd in the Panel’s second Report (iltzal~lst, 1061,86, 748), but it is not intended to be applied as a stand;ird method. l‘he Pancl has now completed its programme of work and has hcen disbanded.September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE 1961/1963 PANEL 6 : PYRETHRUM: CONSTITUTION- 673 W.Mitchell, B.Sc., Ph.D., F.R.I.C. E . A. Baum, D.Sc., 1.C.K.” H. E. Coomber, B.Sc. L. Donegan M. Elliott, PhD. J. R. Furlong, O.B.E., Ph.D.7 A. D. Harford S. C. Jolly, B.Sc., B.I’harrn., A L P S . , X.R.I.C. W. S. Manson, BSc., -1.R.I.C. R. A. Kabnott F. H. Tresadern T. F. West, Ph.D., L).Sc., -I.M.I.Chem.E., F.R.I.C. (Chairman) Staflord ,411en G. Sons Ltd. Pyrethrum Bureau ,Witchell Colts G. Co. Ltd. D.S.I.R., Tropical Products I’nstitute Rothamsted Experiwaen fa2 Station Pyvethruwt Bureau British Petvoleurn Research Centre Pharmaceutical Society of Eveat Britain Cooper Technical Bureau ,4 nalytical and Consulting Chemist Stafford Allen 6 Sons Ltd. Society of Chemical Industry Miss A. &‘I. Parry ‘1 Miss C.H. Tinker J (Secreta”ies) * .kppointed May, 1982. t Resigned ,2pril, 1962. TERMS OF REFERENCE-~~TO investigate methods of assay of pyrethrum flowers and pyrethrum extract with a view to recommending a standard chemical or physical method of assay.” PROGRESS OF WORK- The variant of the method that has generally been used in this country was revised, and has given excellent inter-laboratory agreement of results in the hands of members of the Panel, this applying both to pyrethrum extracts and to pyrethrum flowers. Unfortunately, use of the revised version of the method by three commercial laboratories (each represented on the Panel) has not secured improved agreement between their routine results. Independent work carried out in the laboratories of two companies represented on the Panel has shown that the Panel’s variant of the mercury-reduction method records as “pyrethrin I” a significant amount of extraneous material.Several alternative procedures that appear partly to eliminate this interference have been proposed. Pending a fuller study of these, and other, possibilities, it has been decided to study the A.O.A.C. version of the mercury-reduction method, which appears t o include less extraneous material. A modified procedure ( J . Sci. Food Agric., 1956, 6, 465) for the determination of “pyrethrin 11” has been studied. Concurrently, it has been approved by the A.O.A.C. for inclusion in their published method. The Panel has continued its study of the mercury-reduction method. PANEL 7 : THYROID CONSTITUTION- C. A. Johnson, BSc., B.Pharm., F.P.S., F.R.I.C.R. E. A. Drey, BSc., F.R.I.C. Miss S. J. Patterson, B.Sc., A.R.I.C. R. L. Clements, BSc.* N. A. Terry, B.Pharm., F.P.S. C. Vickers, BSc., A.R.I.C. Miss C. H. Tinker (Secretary) Boots Pure Dyug Co. Ltd. Wellcome Chemical Wovks D.S.I.R., Laboratovy of the Govevnment Chemist D.S.I,R., Laboratory of the Govevnnzexzf Chemist British Dvug Houses Ltd. Boots Pure Drug Co. Ltd. (Chairman) * ,Appointed December, 1962. TERMS OF REFERENCE-~ITO investigate the possibility of devising a chemical method for determining the pharmacologically active constituents of thyroid.” PROGRESS OF WORK- During the first year of its work, the Panel has concentrated on a general investigation of the problems involved in order that it might assess the possibility of success.Work has been carried out on each of the three main sub-divisions of a possible assay, namely (i) hydro- lysis of the thyroid, (ii) separation of the active iodinated amino acids by paper or thin-layer chromatography and (iii) determination of the separated fractions.67.1 ; A nal~lst, Vol. SK Stago (i), which is likclj. to prove the most difficult for quantitative work, has so fiir recei\*ed only a minimum of attention. 130th alkaline and enzjmatic conditions of hydrolysis are being considered. Chromatographic separation of synthetic mixtures of iodinated amino acids, Stage (ii), has been successfully carried out by a number of diffcrcnt systems, and this work is now being (Lxtcntled to the qualitative examination of thyroid hydrolysates.For ttic determination of the \w-y small amounts of iodine deriving from such separations, Stagc (iii), several methods have been investigated. The ccric sulpliate - arsenious oxide catalytic method is considered unsuitable for routine application to occasional samples, such as u.oulc1 be necessary for the commercial examination of thyroid. ’The use of ultraviolet absorption measurements (at 288 mp) and the formation oi a starch - iodine complex both shon- considcrable promise, arid these arc hcing furt1ir.r in\-c.stignted. H w o w OF T H I ~ ~I?;.~LYTICAI. METIIODS COMMITTEE 196lj1962 42 t t ! Kent, Light, licat and Salaries . . . . . . 2969 I I Omcc Equipment . . 63 l’rinting and Stationcry 358 Trav~lliiig Expenscs . . 33 I7xpcnscs of hlcetings . . 29 .Iudit Fec arid :\ccmun- Teltiphonc .. . . 272 i tam}. . . 9 . . . 21 pmscs . . . . . . 123 Postagt: and ‘Petty Ex- ~ 3868 ! Scholarship Grant and I Award for Hcscnrch . . 1132 , C‘ontributioii to Fittings I and Dewrations of C’oiiricil Room - Excess of Iricornr over i 15spt:nditurc for tht: year endcd 31st Octo- I lwr, 1961 traiisfcrred ! I t o .Ac:curnrilatcd Fund 1873 I I W O L i .i d k ‘ l Suh.ripticins froin Industry 6 1 0 Incornc Tax re(-ovcrccl on Covenanted Subscriptions for the !.cars 1954i55 t o 955 1958 -59 5.116 - - . Intercst from Investments 10 (gross) 8 53 I lhnk I.)eposiit Intcrrst , . 650 Profit on Sales of “Kecon- metidctd bIctIiods for tht: .\nalysis of ‘I‘ratlc T3fflu- tmCs” rcceivt:d from thc Society for :\nnlytical 3 :i 6 (‘hcriiistrq* . . * .. . 105 ! 3!) I I I I _ L 17,504 15,X3 I - - . . .- . L 13alance a t Octobvr 31~1, I960 Excess of Income over Expendi- ture for the y t w ended 31st Octobrr, 1961 . , .. . . 1,673 Increase in \‘due on Redemption of L l O O Governrncnt o f Ccylon . . 15,831 34:;, Stock l!)Ei9 . . . .675 A1 8,559 i18,568 --- I00 A100 3Q% War Stock (Market Value at 31.10.61, L105) 100 18.3 - 183 36 Sundry Debtors . , 3‘7 1 I, 000 Deposit Account 11,000 Cash: At Banks on- 7340 Current Account i348 18,340 -- 18,348 k18,559 L18,568 llll__ =- 1961 L L 272 2969 63 358 33 29 21 123 3868 - 1132 I673 1961 ACCOUNTS FOR THE YEAR ENDED OCTOBER 31ST, 1962 Income and Experzditure Account for the Year Ended Octobev 31st, 1962 i 1961 L i Rent, Light, Heat and Salaries . , . . .. 2371 Office Equipment .. 13 Printing and Stationery 83 Travelling Expenses . . 30 Expenses of Meetings . . 41 Audit Fee and Accoun- Postage and Petty Ex- Scholarship Grant and Award for Research - Excess of Income over Expenditure for the year ended 31st Octo- ber, 1962 transferred to Accumulated Fund 1589 Telephone . . . * 334 tancy . . . . .. 21 penses , , * . . . 97 2990 _- ,L 5910 Subscriptions from Industrl- . . Interest from Investments 8 Gross . . . . .. . . - Taxed (net) . . . I . . 650 Bank Deposit Interest . . - . Profit on Sales of “Recom- mended Methods for the Analysis of Trade Effluents” received from the Society for 105 Analytical Chemistry 1 . A ccumulated Fund I 1961 & Balance a t 31st October, 1962 f 7,504 carried to Balance Sheet. . 1.9,093 k19,093 P .L L 15,831 Balance at October 31st 1961, . .17,504 Excess of Income over Expendi- ture for the year ended 31st 1673 October, 1962 .. . . . . 1589 A1 7,504 ,t;19,093676 REPORT OF THE .w.iLyrIc.iL METHODS COMMITTEE 1961/1962 :-Analj'st, Vol. 88 Balaizce Sheet at 31sf Ocfobcr, 1962 196 I L , i ,i L ' Sundry Creditors I Thc Society for Ana- I 7043 lytical Chemistry 2924 ! lO(i.1 -- _- 0'345 I 17,304 A cc 11 rn dated F u nd 10,093 I ? I -4ccountsncg , . 21 83 100 _ i l S , S S S 3 7 I I , 000 ,i Irsveslnienls (at Cost) : L100 34 ?(, Convcr- sion Stock . . 83 LlOO 34% War Stock 100 350 Ikbenhams Ikd. 10s. Ordy Shares $47 AS50 Itnpcrial Chem- cal lndiistries I,td. Ordy Stock . . -736 600 Philip H i l l Invostmcnt Trust L,td. 5s. Ordy Shares . . . . 70s i300 lienold Chains I.td.Ordy Stock. . 685 300 Royal Insurance (h. Ltd. 6s. Ordy Shares . . 535 400 IVIiarf i.Ioldings 1,td. Ordy Shares 706 (31,arket Value at Cash at Banks: I:nitt:d Dominions Trust TAd. on Deposit Account 11,000 narclays nank Timi- tcd on Current Accoii nt . . 6512 .. - 4500 31.10.62, i4225). . Sundry Debtors . , 26 1 7 3 1 3 Sigjied on b~lratf of ihc .4~iaZyticaZ -1Icthods Trust Fm7d G. H. T J . 0 1 ' 1 ) - JACOB, Chairman, J. HUBERT H..\MLIEXCE, Hot?orary Z'veaszrrev. Rcporf of ihs A uditors lo the Tritstees of The Society for A ~zalyttca2 Chmzistry .4 nalyiicd Methods Trust F m d We have examined the above Balance Sheet which in our opinion gi\w a true and fair view of the state o f affairs of thc Trust Fund a t 31st October, 1962 and is in accordancc with thc Books kept hy thc Trustees. JVe have verified thc Investments and found them to bc in order. 10 Xcw Court, Lincoln's Inn, LObiL)OS, \ir.c.2. 28th June, 1063. (Signed) Rll.)IXY, IIEST.OP & S:lINEK C h a rfe red -1 cc o u n fa 12 t s , -1 zrdiiovs.September, 19631 REPORT OF THE ANALYTICAL METHODS COMMITTEE 196111962 677 APPENDIX I11 SUESCRIBERS TO THE TRUST hlbright & Wilson Ltd. Alginate Industries Ltd. Ashburton Chemical Works Ltd. The Associated Octel Co. Ltd. Associated Chemical Companies Ltd. Associated Portland Cement Manufacturers Boots Pure Drug Co. Ltd. The British Aluminium Co. Ltd. The British Arkady Co. Ltd. British Cod Liver Oils (Hull and Grimsby) The British Drug Houses Ltd. British Oxygen Research & Development Cadbury Brothers Ltd. Central Electricity Generating Board. Cerebos Ltd. Cooper, McDougall & Robertson Ltd. Courtaulds Ltd. Crosse & Blackwell (Holdings) Ltd. Cyanamid of Great Britain Ltd. The Distillers Co. Ltd, Dunlop Rubber Co. Ltd. Norman Evans & Rais Ltd. Fisons Ltd. A. Gallenkamp & Co. Ltd. Glaxo Laboratories Ltd. Goulding Fertilisers Ltd. &\rthur Guinness, Son & Co. (Park Royal) Ltd. T. Hedley & Co. Ltd. H. K. Heinz Co. Ltd. E. M. Hobbs, Esq. Hopkin & Williams Ltd. Horlicks Ltd. Hovis McDougall. Huntley & Palmers Ltd. Imperial Chemical Industries Ltd. International Flavors & Fragrances (Great Johnson Matthey & Co. Ltd. Ltd. Ltd. Ltd. Britain) Lid. FUND DURING 1961 AND 1962 Kellogg Co. of Great Britain Etd. Kodak Ltd. Laporte Chemicals Ltd. Laporte Industries Ltd. London County Council. 5. Lyons & Co. Ltd. May & Baker Ltd. Merck Sharp & Dohme Ltd. The Metal Box Co. Ltd. The Millers’ Mutual Association Monsanto Chemicals Ltd. National Coal Board Nest14 Co. Ltd. Novadel Ltd. Ox0 Ltd. Parke Davis & Co. Pfizer Ltd. The Pharmaceutical Society of Great Britain. Procea Products Ltd. The Pyrethrum Bureau Quaker Oats Ltd. Roche Products Ltd. Rowntree & Co. Ltd. J. Sainsbury Ltd. Schweppes (Home) Ltd. “Shell’ ’ Research Ltd. Smith Kline & French Laboratories Ltd. Stafford Allen & Sons Ltd. John & E. Sturge Ltd. Tate & Lyle Ltd. Unilever Lt d. Vitamins Ltd. Wallace & Tiernan Ltd. Washington Chemical Co. Ltd, The Wellcome Foundation Ltd. Weston Research Laboratories Ltd. West Norfolk Farmers Manure & Chemical William R. Warner & Co. Ltd. John Wyeth & Brother Ltd. Co-operative Co. Ltd.

ISSN:0003-2654    

DOI:10.1039/AN9638800659

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

678 LOTHIAS: BEER’S LhW ASD ITS USE IN .WALYSIS [Analyst, 1’01. $8 Beer’s Law and its Use in Analysis A Review BY G. I;. LOTHIAX ( L‘w ivevsily o j Exefer, Deparlrnenf of Physics, The Washington Singev Lnbovafories, Prigice o j Walcs Road, Exetsr) SUMMARY OF COSTESTS Introduction : dcduction of Beer’s law Terms and symbols The validity of Beer’s law: The use of monochromatic radiation: finite waveband and stray radiation l~arallelism of beam Xon-homogeneous specimens : chemical analysis by turbidirnetric mcasurernents, determination of particle size, alrsorptinn curves of granulated materials Mdecular interactions Determination of absorbing materials IF a parallel beam of homogeneous (ie., one wavelength) radiation of intensity I falls 011 a sufliciently thin layer (thickness dl) of absorbing material, a small fraction will bc absorbed.If the thickness of the layer is doubled, the fraction absorbed will be doubled; this is obvious if successive layers behave independently of one another (no multiple scattering) and if each thin layer is so sparsely populated with absorbing moleculcs that they do not hide behind one another in the path of the beam. Thus for thin layers the fraction absorbed is propor- tional to the layer thickness, i.e,- where p is a constant known as the absorption coefficicnt. an increase in path means a decrease in intensity). where I,, is the incident intensity at I = 0. . . . . * (1) dI/I -= -pdl . . . . (The minus sign occiirs because Integration of this expression gives log, I,!I r= pz . , . . . . . . .. (2) ‘I‘his equation niay be rewritten in various ways- log,, Io/I = 04343 log, I,/I = 0.4343 pz = d .. .. .. . . * . * . ( 3 ) .. . . - - (4) = KZ where K is a new constant, or- where t is the fraction transmitted. Any of these forms expresses the law first formulated about 1730 by Lambert and by Bouguer . If tlie concentration, c, of absorbing molecules is doubled and the path-length halved so that the total number remains the same, the absorption (d or t) will remain the same, pro- vided that the molecules do not come so close together that they come within each otlm-’s influence, modifying the energj. levels or even causing chemical changes. That is, tIic absorption will be a function of the totaI nlimber of molecules, i.e., of tlie product cl, and equation (3) may be rewritten- E, E are new constants, the latter symbol being used when concentration c is in gram-molccules per litre.This last equation is Beer’s law, first statcd in 1852’; it includes thc earlicr Lambert’s law. In words it may be surnmarised 11~- saj-ing that thc absorption of radiation by molecules depends only on their total number. Its original formulation has been recently discussed by Pfciffer and Licbhafsky.2 t = I/I, = e--Pl or 1 0 - K ’ . . d = log,, I,/I = Ed or ECZ . . . . . . , . (5)September, 19631 LOTHLIS: BEER’S I-~IW ASD ITS USE IN ASSLYSIS 679 For a mixture of several different absorbing substances the abovc argument may he repeated-provided again that the mixture is sufficiently dilute-when it mag- be shown that the optical densities of the several components are added, 2.c.- This is the generaliscd form of Bcer’s law used in analysis of mixtures.d =- (E,c, + E,c, - - - - ) I . . .. . . - - (6) *. 1 EKMS ASL) SY3lHOI.S d, K, E and .E (equations 3 and 5) are fundamental in the use of Beer’s law. Their usefulness is obvious, since the value of d is proportional to concentration and path length, and the value of K is proportional to concentration. The names of these quantities arc summarised, together with the definitions, in Table 1. In 1942 a report in The A d p P recommended the use of names shown in column I and of the symbols shown in column 4. A t later dates some altcrnati1.e names shown in columns 2 and 3 were rec~mrnended~-~.e in the I‘.S.A. and are now commonly iiscd there. The absorption coefficient, p, is in common use for absorption of x radiation, but not for optical radiations. Spcific ex tint tion coefficient coc ffic ien t .\l)sorpti vity 3lolcc:ular c:xtinction 3lolnr absorp t i \.i t y Sationsl 13rirr:au of Standardss Symbol 1)cfinition .\ bsorhanc-c d t hg,,,(I0 1) logarithm of reciprocal t rmsmission .\bsorI);ince indrs I< d /-optical dcnsity (cltc.) per unit 1)ath length .lbsorl~;lncy indcs 1: d 1-1 q t i c a l dcmsity for unit path It*nglli and concentration JIolsr absol-bmc y E d LI-Lsprcific vxtinction cocfficimt (ctc.) for conccntration o f 1 gram- mole pvr litrv indcx * Sow The Society for .\nalJ.tical L’hcniistry.t I), 13 or E soinctinws uscd. THE VALIDITY OF REEK’S LAW If a measurement of optical density, d , is made, with known path length 2, then the concentration, c, can be calculated from equation 5 if the specific extinction coefficient (E or c) is known.Hut before this equation can be used it is necessarj- to test that the equation holds under the particular conditions of a measurement. If it is found not t o hold, it is possible to establish an empirical relation (calibration curve) between optical density and concen- tration. But it is dangerous to do this blindly; unless one knows why 13cer’s law is not hoId- ing, one may not know within what limits the empirical relation may be expected to apply. I t is therefore most important to consider in detail reasons for departure from R e d s law. I t will be apprcciated from the introductor?. paragraphs that Beer’s law may be expected to apply for- (I) perfectly monochromatic radiation (2) travelling in an optically homogeneous medium (no scattering of radiation) (3) as a strictly paralltl beam.(4) I t is further ncccssary that the absorbing molecules are never close enough to one another or to other (impurity) moleculcs that the molecule structure and hence thc energy levels arc affected. These four conditions are ideals to whicli it is possible only to approximate. The degrees of approximation necessary and the effects that appcar when the limits are exceedcd wiI1 110w bc discussed for each of thcse conditions in turn.680 um-ii.i?;: HEER~S L.IW ASU ITS I:SE IS .ANALYSIS ;,4?lalJ,st, 1701. 88 TIIl-: I’SE OF MOX;OCHHO!tIATIC RADl.4TIOK- Thc constant E (or E ) varies with wavelength so that, if tlic incident radiation contains ;L numhr of wavdcngths, the wavelength distribution of the radiation rcacliing the deeper layers of an absorbing specimen will be modified and the conditions assumed in equation (1) , that all elementary laj’ors absorb the same fraction of radiation, is no longer true.In practice, radiation can never be perfectly monochromatic and will ticpart from this condition in two respects- I t will iricludc a continuous band of radiation extending ovvr a finite waveband centred about the required wavelcngtli. Radiation of widely different wavelength from the desired wavelength arising from scatter, unwanted reflections, etc., in a monochromator will also iorm part of the measuring beam. Thcsc two factors will tie considercd in turn.Finite waveband-Fig. I shows spectrophotometric curves for two concentrations (ratio 4 to 1) of the same substance. If, in endeavouring to measure the optical density at tlw narrow long-wave peak, the “monochromatic” radiation used covers the waveband cd, it is obvious that the observed optical density will be less than thc true \diie for the peak. (2) Finite wavebands. (ii) Scattered radiation. a W av el e ng t h Fig. 1. Typical absorption curvc with broad and narrow peaks \ \ I 1 Fig. 3. Relation bctrveen optical density and concentration : (a) Beer’s law; (b) 1;ailurc of IJeer’s law owing to finite waveband ; (c) Failure o f 13err’s law owing to stray radiation. Drawn t o scale for 1 per cent. of stray radiation that is unatisorbcd by tlic spcci- n1cn The difference between the observed and true maximum valw will lit: proportionately greater for the higher concentration from which it follows that the relation between optical density and concentration will no longer be linear, but curved as in b of Fig.2. Calculations b > r Brodersen7 have shown that even if measured optical densities arc found to be proportional to concentration (curve a of Fig. 2) the observed values may be less than the “true” values for perfectly monochromatic radiation situated at the wavelength of the absorption maximum. (The difference amounted to 8 per cent. in a case calculated by him.) I t follows from all this that a spectrophotometric curve obtained with a monochromator passing a finite waveband will have its maxima depressed and its minima raised.Thct “true” curve can be estimated approximately from an obscrvcd ciirvc by a graphical method first given by I’asclien in 1897 and more recentIjv explained bj- Slatera and also given in various text-books on spectrophotometry. Of course no trcatmtmt of observations can rcvoxl detailed structure within the waveband uscd for tht: rncasiirements. The maximum waveband permissible, if Beer’s law is to hold, thus dcpcnds on the measixre- Incnts being made. Conditions arc less stringent at a broad peak (a of Fig. 1) than a t tlicSeptember, 19631 68 1 narrow peak b of this figure; the former is obviously more satisfactory to use for analytical purposes. In atomic-absorption spectrophotometry, the width of an absorption line may be about 0-02 A, and the source of radiation has to be a hollow-cathode discharge giving emission linewidths appreciably less than this.For measurements made at the peak of the cc band (A = 5770 a) of oxyhaemoglobin, a bandwidth of 2 to 3 A would be desirable-this is not easy to attain with a monochromator and a source giving a continuous spectrum, and it is preferable in determining haemoglobin to avoid this narrow absorption band, But with measurements on potassium chromate at the broad maximum at X = 3720 A, a bandwidth of 50 A will leave Beer’s law valid within about 1 per cent. It is obvious from Fig. 1 that the higher the concentration the sharper a maximum becomes, and efforts to increase precision by working at high optical densities (differential spectrophotornetry) must mean a narrower waveband if Beer’s law is to be maintained.Stray radiation-In general a monochromator passes, in addition to an intensity I, of radiation of the required wavelength, an intensity S of stray radiation of other wavelengths. If a specimen whose optical density is being measured transmits this stray radiation without any absorption, the apparent or measured value will be LOTHIAN: BEER’S LAW -4ND ITS USE I N ANALYSIS in place of the value given by equation (5). As the concentration approaches infinity, when according to Beer’s law d should approach infinity, the apparent optical density value will now approach a maximum value of I, -k s dappt = 10&0 7 and a curve of optical density against concentration will now be as c in Fig. 2. If the stray radiation is 1 per cent.of I,, the maximum value of apparent optical density will be 2.0. Departure from Beer’s law from this cause can be a frequent source of error. It is dangerous to use empirical calibration curves such as c of Fig. 2. This is partly because the curve may change from day to day with changing instrumental conditions (changing values of S due to dust, tarnished surfaces, etc.). But, in addition, the measurement will be affected by the presence of impurities; to take an extreme instance, if a specimen being measured contains, unknown t o the operator, impurities completely transparent at the working wave- length, but which completely absorb the stray radiation of other wavelengths, the calibration curve c of Fig. 2 will be replaced by the straight line a-and the operator will not know this! Tests for the effective presence of stray radiation may be made in one of several ways- (a) By making measurements to construct a calibration curve as in Figa 2 under condi- tions where one knows that Beer’s law ought to be applicable.Some of the first workers to describe the use of this method were Hogness, Zscheile and SidwelP who found, using their own instrument at h = 3700 A with a hydrogen-discharge lamp, a value for S/I,.equal t o 0-004. A modified form of this method is to insert in the beam a specimen having a narrow absorption band at such a thickness that one would expect zero transmission, and then to measure the resulting intensity of radiation or the optical density of the specimen. (6) An alternative method, complementary to the first, is to endeavour to reduce the stray radiation by inserting a filter having high transmission at the required wave- length and low transmission at the wavelengths at which stray radiation might be expected.With a double-beam instrument the filter should, of course, be inserted into both beams. If the use of such a filter then brings about a change in the measured value of optical density, the presence of stray radiation is to be inferred, and it would be desirable to retain the added filter for measurements under these conditions. PritchardlO has described how to determine the stray radiation of various wavelengths for various wavelength settings, by passing pure radiation from a double mono- chromator into the spectrophotometer.The effective amount of stray radiation is likely to be great (S/I,-, large) when the effective This will be most likely to occur when working at wavelengths near the (c) value of 1, is small.682 LOTHI-~S: BEER’S L . ~ W .\XLI I-rs USE IS ;\S.ILYSIS IAiiuLj~st, \‘ol. 88 limits of usefulness of the radiation source, tlic opticat systcm or the detector. €;or example, a tiingsten-filament lamp can be used at X = 3500 A, but the energy here is so small compared with the emission at longer wavelengths that there is grave risk of measurements of optical density being grossly in error. Thus a syccirnen whose absorption is small tliroughotit the \visible spectrum and continuously increasing in going to shorter wavelengths in the ultraviolct, ma!’ well show a completell- spurious absorption maximum a t , say, A = :3700 a smaller apparent optical density at A = 3500 i! bein? measured because the energy here ma).bc. mostl5- visible radiation. Similar difficulties occur in the far infrared, where t tit> emission of one of the usual incandescent sourccs is ver!- small compared with the emission in the region h = 1 to 3 p. I,ikewise, measurements with a quartz optical system near its useful limit of X = 2000 A, or a photomultiplier near its long wave limit in the red, need careful consideration before being passed as valid. This is not the placc to discuss how to reduce the amount of stray radiation; but the examples mentioned are intended to emphasise tlw need to test for Becr’s law iinder the exact conditions used in subsequent measurements-in particular: nature of specimen, soiirce of radiation, slit width, dispersing sJ-stem, radiation detector and alignment of components.Goldring, Hawes, Hare, Reckman and Stickneyll have published some curves showing the effects of stra>+ radiation on measuremcnts of potass- ium chromate; in their paper they discuss this arid other sources of error in some detail, and include a useful bibliography of 19 references. I’-4KXLLELISM OF BE-431- T n the introduction we have prcsupposed a parallel beam --an ideal that can nc\w he obtained. A perfectly parallel beam must come from a point source and can carry onl?. an infinitesimal amount of energy. In practice a beam of finite angular size (see Fig. 3) must IF used. For a beam passing through a specimen at an anglc 9 to thc axis the p a t h lcngth and hence the optical dcnsitj- will be increased b>- a factor l/cos 8.If the tkxtrcmc value of 9 tlirough the specimen is 9’, the optical tlrmit!. will be increased by 1 pcr cent. (l/cos 9‘ = 1.01). If the refractive index of thc specimen is 4/3, the corresponding angle outside tlie specimen wiIl be 0 = 12”; this is greater tlian the semi-angle of beam in most commercial spectrophotorneters, so that the finite angle will not in general be a source of error, 13ut if efforts arc made to increase precision by working at high optical densities (sa?. d = 3) to a precision of 04K&, as is sometimes done in differcntial spectrophotometr!., the finite angular size of beam may well cause departure from Beer’s law.The effect is analogous to that arising from a finite waveband, but is complementat-!. to the latter-tlie ccntral axial beam giving a minimum optical density. As will be seen in the nest section, the angular size of beam is more important wlicn scattering specimens arc bcing measured. s OS-HOMOGE?; EOU S SPECIMES S-SCAT’I’EIII SC IS THE Sl’E<’IMEX- A specimen may be fortuitously turbid; i.e., it ma!- include some suspendcd or colloidal material as an impurity in what would idcally be a homogeneous solution. Provided the turbiditjv is small, the optical densities due to absorption and scattering arc. additive, and it is often possihlc to determine the latter scparatel!. b!r extrapolation of measurements at adjacent wavclcngths where the absorption is zero.Much more must be said about specimens that inhercntl?. scatter radiation---colloidal solutions and suspensions, where it is the light scattering material in which one is interested. Such measurements may be made for \,arious purposes- (I) for chemical analysis-as in the determination of sulpliate as barium sulphate, (2) as a means of determining particle size, (3) to determine the actual absorption of tliu material forming (part of) the particles --e.g., the absorption of powdered materials incorporated in a pressed disc of potassium bromide or the absorption of pigment in plant or animal cells. The argument used in the opening paragraphs to deducc Beer’s law ma>- he repeated for a suspension or colloid simply by spcaking of “particles” instead of molecules.One assumes that some or all of the radiation incident on a particle is lost from the transmitted beam-by absorption or by scattering or refraction into another direction. The fraction so lost depends in a complicated way on a number of factors : the matter can onlj. be summarised liere by saying that for a perfectly parallel beam thc obscuring power of 8 single particleSeptember, 19631 683 is not equal to its projected area A, as might be expected from geometrical optics, but to KA, where K (quite distinct from the K in Table I) is often known as the scattering area co- efficient. LOTHLW: BEER’S LAW AND ITS USE IK ..\SXLYSIS The value of K (somewhere between 0 and 5) is a function of- (;) particle size rclativc to wavelength; (ii) particle shape and orientation in the beam; (izz) refractive index of the particle relative to the surrounding medium ; (iv) the true absorption of the particle.In practice the angles 8, and Od of the incident and measured beams must be finite (see Fig. 3) so that some of the scattered radiation wiIl be picked up by the detector, rcducing the effective opacity; i.e., reducing the coefficient K. Thus K is also a function of- ( v ) thc h a m angles Os and Ba. may be expected to hold provided- If all these five conditions remain constant so that K remains constant, Beer’s law (1) The beam angles Bs and Od of Fig. 3 are so small that the path lengths are not significantly different for the various beams through the specimen--as alreadv discussed for homogeneous specimens. (2) Secondary scatter may be neglected.Radiation scattered out of the observed beam may suffer secondary scatter back into the observed beam. This secondarj. scatter will be more pronounced at high optical densities, rcducing their value and giving an optical-density - concentration curve as in b of Fig. 2. The larger the angles 0, and Od the lower the optical density at which such failure of Beer’s law will set in. The effects of these considerations on measurements made for the three different purposes listed on p. 682 are discussed below in turn. C’hemical arzalysis by tzirbidimetric me.mzwement.s-In the determination of barium suIphate, Treon and Crutchfic1d,l1 for example, found Beer’s law to hold for the largest optical densities they used (up to 1.0) at h = 5700 -4.Rut this conclusion would be useful only under the conditions of their work-type of spectrophotometer and its adjustment and the exact technique they used in preparing their suspensions. I t is not really possible to be sure of repeating the technique exactly from day to day, and it is essential that all measurements be comparative-an unknown concentration being measured by comparison with known con- centrations not very different from the unknown, the two suspensions being prepared side by side at the same time. FVorking in this way, one does not rely on the applicability of Beer’s law except for interpolation over a small range. Ihtermination of fiarticle size-For this sort of work it is desirable to have the angles 8, and Od of the incident and emergent beams as small as possibic in order to approximate to the conditions (8, -= @d == 0) under which the scattering properties have been c a l ~ u l a t e d .~ ~ Fig. 3. Illiistration of the finitc angular sizes o f the beam incidcnt on a s~)~cimcn-semi-angle O,--and of the beam recei\red by the detector --semi-angle684 LOTHIAN: BEER’S LAW AN11 ITS CSE 13 .lS.\I.E’SIS ‘+4 nalyst, Lvol. SS GouIden14 lias described modifications of comniercial instruments with this end in vicw and has made measurements to determine the size of the fat globules in milk. As explained above, such small angles help to maintain Beer’s law to higher values of optical density. I t is of course important to check by measurements at several concentrations or cell lengths that Beer’s law is holding under ttie conditions of measurement.If secondary scatter is affecting the mtwurements they cannot , without correction, be used to calculate particle size. Detcrmiizatioti of ubsorption curves of granulated maferials-More or less successful attempts have been made to deduce true absorption curves from transmission measurements on granulated materials; for example, haemoglobin in wholc blood cells15-16 and chlorophyll in cells of the alga, Ch10rella.l~ But the theory developed so far involves some approxi- mations, and at present it must be said that in general it is not safe to assume that absorption curves obtained from such mcasurements give the true curve that would be obtained from a homogeneous sohtion of the absorbing compound. If tlie measurements are to be a function of the absorption only and not of the scattering propertics, it is necessary- to include all the scattered radiation in the measurement.Ideallj- this may be done by surrounding the specimen with an integrating sphere. But the theor>- and practice of this arrangement are complicated, and the method is not in general use. A good approximation is to use a translucent diffusing surface at thc exit face of the spccimcn cell to direct a representative sample of the forward scattered radiation into the detector. IVith these conditions of observation Beer’s law may be expected to hold, but only in the limited sense that optical density will be proportional to the number density (number per unit area} of a given type of particle provided this is small enough for secondary scatter not to come into play.The optical densities will not be proportional to the values that would be obtained for the same amount of material uniformly dispersed in ii homogeneous solution. Some o€ the beam of radiation passes outside the absorbing particles to reach the detector unabsorbed. This means that for a given number of particles the optical density will not exceed a Limiting valuc, however great the absorption in each particle; hence absorption peaks will be depressed. I t may be shownl8 that this effect disappears as the absorption of a single particle approaches zero. A conse- quencc of this is that powdered materials (as in the potassium bromide pressed-disc mcthotl) should bt ground as fine as possible. In addition, peaks ma37 be slightiy shifted because of refractive index changes near an intense absorption band.And if not enough of the scattered radiation is picked up, an absorption peak may well disappear altogether when a pigment is localised in cells.l6.l9 This is because of ttie “sieve effect.” MOLECULAR IXTERACTIOKS- bYhcn tlie concentration is sufficiently increased an absorbing rnoleculc can no longcr be rcgardcd as being isolated from other similar molecules by empty space (gas samples) or by an infinite ocean of solvent molecules. The energy levels, and hence the absorption spectrum, will become modified when molecules are in siihcient concentration to influence one another and consequently to reduce the influence of surrounding solvent molecules. In the classical language of chemistry this modification is explained as a chemical change that may lie described as a change of polymerisation, isomcrisation, ionisation, lijdrolysis, etc.These effects form too vast a subject to be dealt with in this review and arc appropriate to rt:\hvs on these individual chemical topics. I t milst suffice to say that, if a lack o f proportion- ality of optical density and coriccntration cannot be ascribed to any o f the causes already discussed in some detail, a chemical change must be considered. A short mention should, however, be made of the case of atmrption in a gas unclcr cl-ianging pressure. As the pressure of a gas is incrcascd, an absorption band is in general broatlened with the result that thc specific extinction coefficient at the ccntre o f the hand is deprcsscd, i.e., Beer’s law is no longer valid.But i t is found that wcn wlicn this is soJ the total area under an absorption band (A = JKdv or Jpdv when the integration is carried out over the whole range of frequencies, ‘J, covered by the band) is often approximatcl). proportional to thc total concentration of absorbing molecules. This may be regarded as a modified version of Beer’s law. Some spectrophotometers providc mechanical integrating attachments for determining this integral. *Hie use o f the area under an absorption curve for analytical work Iias heen discussed in several original papers ; see, for example, Oswald,g* Kamsay21 and Mills and Thompson.22 Thest: last workcrs show how, VV(W w h n the spectro-September, 19631 685 meter will not completely resolve a band, a correct value for A can be obtained by extrapola- tion to zero partial pressure by measurements on mixtures with an inert non-absorbing gas.DETERMINATION OF ABSORBING MATERIALS The use of Beer’s law for determining a single absorbing component (equation 5) and of mixtures of several such components (equation 6) is common practice and is well covered in the various text-books on spectrophotometry, and so needs little consideration here. As already explained, it is desirable to make measurements at the peaks of broad bands for several reasons: (a) At a peak so that the sensitivity is a maximum and interfering effects of absorption due to impurities is a minimum, and further so that a small error in wavelength setting will cause minimum change in optical density-measurement on the steep slope of a band (e in Fig.I) is bad practice. (b) As already explained the use of a broad peak reduces troubles due to finite waveband. For measurements on mixtures of pz components, n measurements a t n different wave- lengths are needed to solve n equations of the type of equation (6). The wavelengths need to be chosen for maximum sensitivity-at each of the n wavelengths one component having a large absorption and the others having small absorption. Simplified methods of solution are given in various text-books. The range of optical density in which a measurement is made can be adjusted by suitable choice of cell lengths. Much has been written about the optimum value of optical density to choose. For photographic and visual spectrophotometry a high value is, in general, the best.Various tests show that, with simple assumptions about the limiting sensitivity of a photo- electric or thermo-electric detector, the optimum precision is attainable at an optical density of 0-43. But it has also been shown23 that this conclusion is modified by taking account of the fact that a measurement of optical density involves several operations-including setting a galvanometer or similar scale with no radiation falling on the detector, then with a blank cell (solvent) in the beam and finally with the absorbing specimen in the beam. These considerations may need to be further modified by reason of instability in the radiation source (voltage fluctuations). I t is my opinion that the optimum optical density depends on so many factors-voltage stability and level of photo-electric (or thermo-electric) current- that an optimum value of d cannot be usefully calculated, but must be determined in any particular investigation by trial and error determination of repeatability of observations. LOTHIAN: BEER’S LAW AND ITS USE I N ANALYSIS 1.2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 2 1. 22. 23. REFERENCES Beer, A., Ann. Phys. Chew.., 1852, 163, 78. Pfeiffer, H. G., and Liebhafsky, H. A., J . Chem. Educ., 1951, 28, 123. Cooper, B. S., Gillarn, A. E., Lothian, G. F., and Morton, R. A., Analyst, 1942, 67, 164. Brode, W. R., J . Opt. Soc. Amer., 1949, 39, 1022. Hughes, H. K., Anal. Chem., 1952, 24, 1349. U S . National Bureau of Standards, Circular No. 484, Washington, 1949. Brodersen, S., J . Opt. Soc, Amer., 1954, 44, 22. Slater, J. C., Phys. Rev., 1926, 25, 783. Hogness, T. R., Zscheile, F. P., and Sidwell, A, E,, J . Phys. Chem., 1937, 41, 379. Pritchard, €3. S., J . Opt. SOC. Amer., 1955, 45, 365. Goldring, L. S . , Hawes, R. C., Hare, G. H., Beckman, A. O., and Stickney, M. E., Anal. Chem., Treon, J. F., and Crutchfield, W. E., Ind. Eng. Chem., ,4naZ. Ed., 1942, 14, 119. A number of tables showing values of scattering area coefficient, K, for particular values of particle size (relative t o wavelength), refractive index and extinction coefficient have been published (and are continuing t o appear) in various journals. Rouwkamp, C. J., Rep. Prog. Physics, 3954, 17, 35. (This paper contains an extensive Lowan, A. N., Nut. Bur. Stand., Appl. Math. Series, 1949, No. 4. Penndorf, R. B., J . O f t . Sac. Amer., 1962, 52, 896. 1953, 25, 869. See for example- bibliography). Goulden, J. D. S., Brit. J . AppZ. Phys., 1961, 12, 456. MacRac, R. A., McClure, J. A,, and Latimer, P., J . Opt. SOC. Amer., 1961, 51, 1366. Lothian, G. F., and Lewis, P. C., Nature, 1956, 178, 1342. Latimer, P., Science, 1958, 127, 29. Lothian, G. E., “Absorption Spectrophotornetry,” Hilger & Watts Ltd., London, 1958, p. 38. Adams, G. A,, Bradley, R. C., and MacaIlum, A. R., Biochem. J . , 1934, 28, 482. Oswald, F., Z. Elektrochem., 1954, 58, 345. Ramsay, D. A., J . Amer. Chem. SOL., 1952, 74, 72. Mills, I. M., and Thompson, H. W., PYOG. Roy. Soc., A , 1955, 228, 287. Gridgeman, N. T., Anal. Chem., 1952, 24, 445. Received April 24th, 1963

ISSN:0003-2654    

DOI:10.1039/AN9638800678

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

686 FURMIDGE : APPLICATION OF ELYING-SPOT [Amlyst, Vol, 88 The Application Size Analysis Of in Flying-spot Scanning to Particle the Formulation of Pesticides* BY C. G, L. FURMIDGE (Woodstock Agriculhral Researdz Centre, Sittingbourne, Keni) Pesticides are usually applied as dusts or sprays in which particle size is one factor that controls the impaction on and coverage of the target surface and the amount of drift of toxic material away from the target. Most pesticide sprays are formulated as emuIsions, or suspensions of solid particles in water ; the particulate size of these dispersions is important in determining both their relative stability and the biological toxicity of the deposits they produce on the target surface. Particle size analysis in the pesticide field necessitates the study of a wide variety of particulate matter under many different conditions, and it has been found possible to use the “flying spot” automatic scanning technique to cover many of the problems involved.The advantages and disadvantages of this automatic method of particle sizing are discussed with particular reference to the sampling requirements to obtain the greatest accuracy in the results. ONE of the essential functions of formulation is to make possible the distribution of extremely small amounts of pesticide over large surface areas. The most obvious way of achieving this is to reduce the particle size of the pesticide so that its surface area approaches that of the target surface. It is impossible to do this to the toxicant alone, but, if its bulk can be extended by dilution with biologically inert solids or liquids and this larger amount is applied in the form of extremely fine particles, a satisfactory distribution of pesticide becomes possibIe.Thus, the usual ways in which pesticides are applied are as sprays or as dusts. Aqueous sprays are usually preferred when reasonable amounts of water are available, but few pesticides are soluble in water, and so they have to be formulated as emulsions or as suspensions of solid particles. The formulation and application of pesticides is concerned, therefore, with the dispersion of particulate material in various ways, the most important of which are solid particles in air (dusts), oil droplets in water (emulsions), solid particles in water (“wettable powders”) and sprays of these last two dispersed in air.A control over the particle size in all these types of dispersions may be important in determining the biological efficiency of the pesticide application. SPRAY S- The droplet spectrum formed when a liquid is sprayed into air can vary considerably with the spray equipment both in terms of the actual size of the drops and in the range of drop size found in the spray cloud. Both these factors may influence spray performance in several ways. The effect of droplet size on coverage of the target surface is illustrated in Fig. 1. This shows the area (in acres) covered by 10 and 50 gallons of an ideal spray (when all the droplets are the same size), the coverage being calculated on the assumption that each droplet will impact and spread t o cover a circle whose diameter is 4 times that of the original droplet. A spread factor of this magnitude is an average figure for aqueous sprays containing a reasonable wetting agent on foliage.For many pests, good coverage of the target surface is essential for satisfactory control, but the total leaf area of an acre of crop may be anything from 1 to 30 acres. To provide a spray coverage equivalent to the crop area, either extremely large volumes of spray must be used or the spray must be composed of extremely small droplets. The modern tendency is to apply smaller and smaller volumes of spray per unit area, and between 10 and 50 gallons to the acre is common; a drop size of less than 100 p is therefore desirable. However, the process of spraying is not as simple as this, because, if small droplets are used, drift of the droplets away from the target may become serious.Fig. 2 shows the *Presented at the meeting of the Society on Wednesday, February 6th, 1963.September, 19631 SCANNING TO PARTXCLE SIZE ANALYSIS Diameter of spray droplet, p Area covered by spray, assuming a spread fact5r of 4 x diameter of drop: curve A, 10 gallons; carve B, 50 gallons Fig. 1. Spraying point I 10 rn.p.h. wind ___t 687 D i m e t e r of spray droplet. p Fig. 2. Influence of drop size on spray drift variation of spray drift with droplet size, the spray droplets being assumed to be initially stationary at 6 feet above ground level in a 10 m.p.h. wind. The spray drift is shown by the horizontal distance that the droplets travel before reaching the ground.In practice, 10 m.p.h. is about the maximum wind speed in which spraying would be carried out, but under these circumstances drops as large as 200 p may drift a considerable distance.688 FURMIDGE APPLICATION OF FLYING-SPOT [Analyst, Vol. 88 The droplet size is one of the factors that influence the impaction efficiency of spray drop- lets and the retention of liquid on sprayed surfaces. I t is impossible to generalise on the optimum drop size required for impaction on agricultural spray targets, because the latter vary so considerably in size, surface characteristics, degree of packing and in their movement relative to the spray droplets. The larger the drop size, the more readily will the drops impact on foliage, so that if it is desired that the spray should penetrate through thick foliage or dense undergrowth, then extremely small drop sizes are required.Work on the effects on flying mosquitoes of droplets of lubricating oil containing 8 per cent. of DDT shows that the maximum activity per unit weight of spray is found with a droplet size just over 10 p ; drop sizes either larger or smaller show much reduced actiGty.l A similar drop size - toxicity effect is found with many other flying pests, and this is important in the formulation of the aerosol generators that are now so widely used. Such formulations consist of a liquefied propellent gas, mixed with a solution of the toxicant in a non-volatile oil and held under pressure in a metal container. On releasing the pressure, the propellent gas vaporises and produces extremely fine droplets of the non-volatile oil; the droplet size is controlled by the proportion of propellent to oil and by the dimensions of the nozzle through which it is released.The droplet size also governs the toxicity of a spray to flying insects. DUSTS- These have not, so far, been studied as thoroughly as sprays, but similar considerations apply in respect to drift and coverage of target. Dusts for foliage application would normally have a particle size between 50 and 70 p, and so drift is again a serious problem unless the dusts are applied under still or extremely light wind conditions. One important factor in the application of dusts is the relative size of toxicant particles to filler particles. If con- siderable differences in size exist, toxicant and filler may settle out of the dust cloud in different areas and thus produce a patchy cover of toxicant.The size of the particles in a dust cloud has a considerable effect on their impaction on foliage, their adhesion and on their resistance to weathering. Generally, impaction is improved as the particle size is increased, but resistance to weathering (by wind or rain) is decreased. EMULSIONS AND SUSPENSIONS- Most pesticides are formulated as water-dispersible oils (usually called emulsifiable concentrates) or as water-dispersible powders (wettable powders). The essential requirements in a wettable powder are that it should flow freely, even after prolonged storage, and that it should suspend well in water.The particle size can affect both these properties, and it is also one of the factors that determine the ease with which the toxicant is deposited on to the sprayed surface once the spray has impacted. The biological efficiency of the toxicant deposit is also governed by its particle size; in general, the smaller the particles, the better the results. Similar considerations apply to the size of the oil globules in oil-in-water emulsion sprays. In studying emulsion behaviour, it is valuable to be able to measure the rate of coalescence of emulsion globules in the dilute emulsion systems used in practice. 'I his rate of coalescence is related to the ease of formation of the emulsion, its stability on standing and its behaviour after impaction on a plant surface. In a similar way, it is interesting to measure the rate of aggregation of solid particles in suspensions of pesticides.PARTICLE SIZE ANALYSIS Research into the formulation of pesticides necessitates the study of an extremely diverse range of particulate systems. Several methods are available for determining particle size,2 but most of these methods are limited to one type of particulate system, q., dusts in air or solid particles suspended in water. Many of these methods are used in studying specific formulation problems, e.g., the over-all stability of emulsions and suspensions is usually estimated by fractional decantation methods. These have the advantage that the results can be directly related to the extent to which the dispersed phase will settle-out in the spray tank.Several other sedimentation methods are equally satisfactory for studying this type of system, but such methods are not generally applicable to all the particulate systems involved in the formulation and application of pesticides. For example, the measure- ment of spray droplet size can be satisfactorily carried out only by collecting the dropletsSeptember, 19631 SCANNING TO PARTICLE SIZE ANALYSIS 689 on a suitable surface and sizing them. This method, in which the sample is placed under a microscope and the particles are compared by means of a suitable eye-piece graticule, has been used to a considerable extent, but it is really too tedious and slow to be satisfactory except in those instances when few samples have to be measured or when the particle size is reasonably uniform.Even then the method tends to be inaccurate because of the eye-strain that is in~olved.~,~ These difficulties can be overcome by using an automatic method of counting and sizing the samples, and, if such methods are sufficiently versatile, they may be used for many of the particulate studies besides spray droplet sizing that are important in the formulation of pesticides. The flying-spot system5,6 of scanning the sample was selected as being the most versatile method, with an instrument based on the flying-spot microscope and developed by Messrs. Rank-Cintel Ltd. n t I1 FLY IN G- SPOT PARTI c LE RE s OLVE R- The operation of the flying-spot particle resolver is shown diagrammatically in Fig. 3. A 700-line scanning raster, produced on the face of the scanning tube, is passed into the optical system.For convenience, this usually consists of a standard optical microscope, but, if lower magnifications are desired, it can be replaced by a simple projection lens. The image of the scan raster is focused on the sample to be examined, the scanned area decreasing as the magnification increases. The amount of light passing through the sample, when transparent backgrounds are used, or reflected from the sample if it is opaque, varies according to the optical density and configuration of the objects on the sample; these changes in light intensity are detected by a multiplier photocell, in which they are converted into an electrical signal. This signal is fed to the video amplifier, the output of which modulates the monitor cathode-ray tube, and thus a magnified image of the sample is produced on the monitor screen.Automatic counting of all the particles within the scanned field is carried out by using the extra units shown in Fig. 3. If only one particle is on the sample slide when this is Optical system / c amplifier d display I I; unit Sizing + Counter S Quantizer Monitor tube 1, I f lf Fig. 3. Flying-spot particle resolver scanned by the flying spot, an intercept pulse will be produced which passes to the quantizer. This unit will pass on the pulse only if it is greater than a predetermined voltage level; it thus prevents counts from spurious particles and random noise. From here the pulse passes to a magnetic memory system (the one-line delay), the sum circuit and the pulse-forming unit where it is suitably shaped.After shaping, the pulse passes through an electronic gate, which is normally open, and then operates a dekatron counter tube. Acceptance of a count pulse is shown on the monitor screen by a bright spot appearing on the image of the particle. If this particle is larger than one picture element, i.e., greater than the distance between two adjacent scanning lines, the flying spot will scan the same particle for a second time on its next scan line. A similar sequence of events occurs, but this time the sum circuit receives the new pulse from the quantizer plus the pulse from the previous scanning line, which is Sum One line + Trigger- circuit - delay690 FURMIDGE : APPLICATIOS OF FLYIKG-SPOT [Anazyst, VOl.88 released from the one-line delay unit. This state of coincident inputs produces an inhibiting pulse that shuts the gate and prevents a second count being recorded for the same particle. Thus, in any given field every particle, however large, will be counted only once. I t should be noted that this is an ideal situation and that, should the profile of thc particle present a re-entrant along the line of the scan, two counts will be recorded from the same particle. Also, aggregates or particles that are not scparatcd by more than two picture elements may be counted falsely. Picture elements 1.5 A 2.5 3-5 4.5 Scanning lines One pict element .ure Length of subtraction pulse n n I;ig. 4. Operation of sizing unit : Sizing unit set on 0; no suhtrac- tion pulse is produced and thcrc- fore all four particles are counted.Sizing unit set on 2; the subtrac- tion pulse is equivalmt t o 2 picture clcmcnts and no particle is counted whose size is less than this value. Therefore three parti- cles are counted. Sizing unit set on 245; no particle is countcd whose size is less than 2.83 picturc elements; therefore two particles are counted Sizing unit set on 4; only one particle is counted. Sizing u n i t set on 4 6 2 ; no particles arc counted. The automatic sizing is performed by the sizing unit, which produces a pulse of selectable width that corresponds to a given size of particle on the sample. Any particle whose size is less than that selected will not be recorded during a field count, so that by successive counts with selected pulses of progressively greater width a complete size analysis of the sample in the field can be obtained.The standard pulse widths obtained from the sizing unit are given in terms of picture element size, and increase in a l / 2 progression; the picture element size, of course, varies with the magnification used in the optical system. The operation of the sizing units is illustrated diagrammatically in Fig. 4.September, 19631 SCANNING TO PARTICLE SIZE ANALYSIS 691 When counting, one scan across the field takes 8 seconds, and a further complete scan has to take place to re-set the sizing unit. The total time taken to count each size range is, therefore, 16 seconds, and a complete size analysis of the field may take up to 3 minutes (if all the size ranges have to be explored).Normally, between 10 and 25 fields must be sized to obtain a statistically sound size analysis of the whole sample, the number of fields depending on the particle density of the sample and on its degree of heterogeneity. Prepara- tion of the instrument €or a particular type of sample may take from 15 to 30 minutes; many similar samples can be measured with little further adjustment of the instrument. The total time taken to count and size a sample is usually between 10 and 60 minutes. ACCURACY AND RELIABILITY OF THE METHOD- The accuracy of the instrument in counting and sizing any particular sample will depend on two main possible sources of error. First, errors within the instrument itself, i.e., those inherent in its design or those due to faulty operation.Measurement on the reliability and accuracy of the instrument showed it to be somewhat better than visual measurement on similar samples , and operating errors, with reasonable experience, were negligible.4 The second and by far the more important source of error in automatic counting and sizing lies in the nature of the sample. It has already been shown that aggregates and irregularly shaped particles can produce spurious counts, and it must be remembered that any form of artificial scanning can never be as discriminating as the human eye. An ideal sample for automtic counting with this instrument should have the properties listed below. 1. A reasonable degree of contrast between particle and background; a minimum of LO per cent. optical contrast and preferably more.2. The particles should have clear-cut, well defined edges ; any haziness or reduction in contrast towards the edges will affect the length of the pulse that the particle produces, and this cannot always be overcome by adjustment of the instrument 3. The optical contrast should be reasonably uniform over each individual particle, e.g., emulsion globules that appear under transmitted light as black rings with light centres will be counted as several particles. 4. The particles should be regular in shape and preferably circular (or spherical). Ellipsoidal or irregular shapes can be counted and sized to produce a mean or average figure provided the particles are randomly distributed on the sample. Very irregularly shaped particles may be counted more than once; the counts are registered by a bright spot appearing on the monitor screen so that these false counts can be seen and may be corrected visually.5. The sample must not be overcrowded, because the scan will not differentiate between the individual particles in groups and aggregates. On the other hand, the samples should not be too sparse, or the total count per field would be low and many fields would have to be counted and much time wasted, The errors due to overcrowding may not be very important when a number frequency is to be measured, but when volume or mass parameters are to be calculated (and this is frequently the parameter of greatest interest in the pesticide field) the errors can be extremely serious. To summarise, the speed, reliability and degree of accuracy of the automatic instrument are superior to those of visual counting.The instrument is versatile in both the range of particulate matter it can measure and in the range of magnification that can be used. Particles from 1 p upwards may be measured by transmitted light and from 10 p upwards by reflected light. These ranges may be extended by using an intermediate photographic process. Thus the use of such an instrument removes most of the difficulties involved in the measurement of particle size, but it tends to increase the difficulties involved in obtaining a sample of suitable quality. It is the quality of the sample that is important; normal samp- ling errurs can be reduced because the increase in speed of counting permits more particles to be measured than is usually possible with visual counting.SAMPLING- to settle out in a sedimentation chamber on to suitable collecting surfaces. particles on the sample can be readily controlled to avoid undue overcrowding. The sampling of sprays and dusts can be carried out in the laboratory by allowing them The density of Dust particles69% FTRMIDW : XPPLICATIOS OF mwx-sPo.r [AItalyst, \rol. 88 are rt:latively c'asy because they can lie collected and measured directly on glass slides. Spray drop1c.t sampling is more complicated in that the droplets will tend to spread when they impact on ;t collecting surface, and thc droplets themselves must form a coloured stain that provides sufficient contrast to distinguish them from the liackground. In agricultural sprays, tlie Iattcr requirement is hest covered by adding a dye to the spray liquid; the degrcc of spread is dependent to a large degree on the nature of the sampling surface.Collection on various tj-pes of papcr surfaces, commonly used in droplet sampling, is unsatisfactorjr for automatic counting, because the edges of the stains arc alwaj-s diffuse and the spread Factor is often cxtremel>- largo. Glazed photographic paper gi\,cs stains that are idcal for automatic counting, but thcrc the stain size varies with the impaction velocitv of tlie drop as well as with its size. Up to the present, siliconcd surfaces, as described by Courshee,' have proved to be thc rnost satisfactory in terms of qualitjv of sample and constancy of spread factor. 1,'nder practical spraying conditions in the open, sampling is very much more difficult and it is almost impossible to get a perfect size spectrum of a spray or a dust cloud.13esides the difficultics of sampling under isokinetic conditions, the size of a practical spray or dust cloud makes it virtuall). impossible to get samples representative o f the whole cloud. The automatic scanning technique is particularly .cxluabIe in determining tlie degree of cover obtained on sprayed or dusted surfaces. This is not riecessarily conctbrned with the measurement of particles or microscupic stains because, with sprays particularly, tlic droplets may run togetlicr to gi\re a more continuous film of deposit. Deposit arcas can be measnrrd by ignoring the counting and sizing circuits and electronically timing the total period during one scan that the scanning spot is obscured by the deposit. 'rhe time necessary for one complcte scan with a completely clear field can also be measured, and the ratio of thcse two times gives the proportional area covered by the deposit.T l i c samples for such mcasiircmcnts may be collected on artificial surfaces, but the most \valuable information on coverage can only bc obtained by studying the deposit on tlic surface that is used in practice, i.c., the leaf. The measurement of samples collctcted on leal surfaces presents serious difficul- ties, but these can be overcome hy incorporating a fluorescent material into the spray or dust and then photop-apliing the sprayed or dusted leaves in ultraviolet light.s The areas of deposit show up black on a light background, as shown in Fig.5, and the photographic negative can be directlJ- scanned to measure deposit area. Considerable practical difficriltics arise in svlecting the most appropriate fluorescent tracer to iise in a particular pesticide formulation arid the wa). to incorporate it into the fi~rmulation. It is essential that it will trace the pesticide deposit closely, because the area covered by pesticide deposited from sprays may he less than the total arca wetted by tlie aqueous phase of tlic sprajr. The measurement of oil droplets and solid particles dispersed in water poses some rather different problems. Thc automatic scanning of such particles is reasonably straightforward provided that tlic particles all lie in the same plane and, for emulsions, that the dispersed phase is suitably coloured.The first of these requirements can be met by placing a sample of the emulsion or suspension in a suitable ccll, e.g., a haemacytometer slide, and allowing the particles to rise to the surface or settle to the bottom of the cell before measurement. If the particles are extremely small and subject to Brownian motion, the sample can he photographed and the measurements made from the negative. One of the greatest difficulties in this sampling process lies in extracting the sample from the bulk emulsion or suspension in a way such that it completely reproduces the particle size spectrum at the point of sampling. The samples arc usually required at successive time inter- vals from the moment of preparation and at various depths below the surface of the emulsion or suspension, but the act of removing a sample can cause a considerable disturbance.Also aggregates or large emulsion globules in the extracted sample may he broken down by rapid passage through tubcs or by passage through a narrow aperture. This brief review of the most usual types of particulate matter encountered in pesticide formulation has dealt essentially with aqueous dispersions. However, it is also necessary to consider sprays of oil solutions, suspensions of solid particles in oil, and evcn multiple dis- persions of suspensions of solid particles in oil that is itself emulsified in water. The mcasure- mcnt of particle size in any of these systems remains the same in principle, but different sampling problems may be involved. The measurement of particle size by automatic methods is fairly straightforward, but it is the obtaining of good samples that raises the greatest difficulties.Fig. 6 . Photograph of a fluorescent spray deposit on a leaf surface under ultraviolet light /To face p. 692September, 19631 SCANNING TO PARTICLE SIZE ANALYSIS 693 REFERENCES 1 2. 3. 4. 5 . 6 . 7 . 8. La Mer, V. K., Hochberg, S., Hodges, K., Wilson, I., Fales, J. A,, and Latta, R., J . Colloid Sci., Analytical Methods Committee, Analyst, 1963, 88, 156. Courshee, R. J., Brit. J . AFpE. Phys., 1954, Suppl. No. 3, 5, S.162. Furmidge, C. G. L., Ibid., 1961, 12, S.268. Taylor, W. K., Ibid., 1954, Suppl. No. 3, 5, S.173. Causley, D., and Young, J. Z., Reseawh, 1955, 8, 430. Courshee, R. J., and Byass, J. B., National Institute of Agricultural Engineering, Report No. 32, Staniland, L. N., J . Agric. Eng. Res,, 1959, 4, 110. Received March 21st. 1963 1947, 2, 539. 1953.

ISSN:0003-2654    

DOI:10.1039/AN9638800686

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

694 SHXW AND DUNCOMBE : CONTISUOUS AUTOMhTIC [Analyst, Vol. 88 The Continuous Automatic Microbiological Assay of Antibiotics BY W. H. C. SHAW AIW R. E. DUKCOMBE (Glnxo Lnboralories L t d , , Greenford, Middlesex) The principle of continuous culture of the test organism is applicd to the 4utoAnalyzer niethod for the microbiological assay of antibiotics by measurement of respiratory carbon dioxide. The use of E. coli as n. general purpose tcst organism is described, and the method is extended to the assay of high potency samples of benzylpenicillin, streptomycins, neomycin, iiovobiocin and cephalosporin C. A modification is suggested for samples of lower potency. In comparison with a conventional agar-diffusion method (8 x 8 quasi-latin square), the AutoAtialyzer results sho\ved no bias and, with duplicatc recording o f each sample, were of greater precision.AN AutoAnalyzer instrumental system for the microbiological assay of tetracyclines and polyene antibiotics has been described by Hancy et al.l and Gerke, Haney and Pagano.2 The method is based on measurement of the carbon dioxide resulting from free respiration by the test organism during a fixed incubation time and of the depression of respiration by graded concentrations of the antibiotic to be determined. In preliminary trials with tlic method it became dear that the instrumental system was satisfactory, but wc encountered considerable difficulties in preparing a standard inoculum of a test organism (Escherichia coli) similar to that dcscribed for the assay of tetracj*clines.I t has been specified1,* that the organism be grown by overnight cuIturc in a vigorously shaken medium; the cells are then collected by centrifugation, washed and re-suspended in fresh cold nutrient medium. Alterna- tively, the cdture may be suitably diluted without centrifugatiom3 The inoculum produced in this way is maintained at 0" to 3" C during the working day in order to reduce to a minimum respiration of the cells in the inoculum rescrvoir. The number of cells produced by overnight ciiltiire and their vigoiir depend on the degree of aeration the medium receives, and any variation in aeration necessitates a corre- sponding variation in the concentration of cells required for the working inoculum to maintain constant production of carbon dioxide in the AutoAnalyzer system.Moreover, tlic dcgrec of sensitivity towards a given antiobiotic appeared to vary with the numbers of cells obtained, and this made it necessary to re-check for each inoculum the concentration of antibiotic required to give the working mid-point of the dose - response curve. Considerable timc was often required at the beginning of the working day to establish the inoculum dilution and antibiotic concentration required for both standards and samples. To assay certain antibiotics to which E . coli grown in this way is not sensitive we wished to use Hacillzts subtilis. When grown overnight, cultures of I?. szcblilis contain a large propor- tion of spores, and a suitably diluted inoculrnn showcd a. rapid cleclinc in ability to prodricc carbon dioxidc when maintained at 0" tu 3" C.Much younger cultures (4 hour), predomi- nant]!* in the vegetative form, showed a similar rapid cleclinc in vigour. l'hesc difficulties led us to consider the possibility of continuoils culture of the test organism; if successful, this w ~ u M provide a vigorous and consistent inoculiim for the assay and eliminate the need for maintaining a supply of icc-cold inoculum. l h e size of the reservoir required for continuous growth of the inoculum depends on the generation time of the organism at the incubation temperature and on thc throughput of nutrient medium. With the short generation time of E. coli (about 20 minutes), a reservoir volumc of 50 ml and the continuous addition of 0.8 ml per minute of stcrile medium permits a sixitable concentra- tion of organisms to be maintaincd, and the whole unit can be accommodatcd in the Auto- Analyzer incubation bath.Vigorous aeration serves to mix this volumc of culture and to remove both the respiratory carbon dioxide as it is formed and culturc in excess of that rcquircd for the assay. The sensitivity of the organism grown in this way towards certain antibiotics, notably cephalosporin C and, to a lesser extent, Iienzylpenicillin, differs markedly from that of the same organism grown in overnight cultures.September, 19631 MICROBIOLOGICAL. ASSAY O F ANTIBIOTICS METHOD APPARATUS- 695 AzttoAnalyzer-This consists of a modified sampling unit, a 15-channel proportionating pump, an incubation bath at 37" C, a colorimeter fitted with a 15-rnm tubular flow-cell and 550-m~ filters and a single-pen recorder.Fit the sampler with a reverse sampling crook and with a modified cam lever t o provide a 2g-minute sample aspiration time. The constant-level device (see Fig. 2) should be clamped to the sampler, so that formaldehyde is aspirated for 24 seconds in each three-rninute cycle, Fix t o the lower side of the cover of the bath four support pillars, approximately 7 inches in length, through the four central holes provided in the cover, and attach the lower end of the pillars to a 4-inch diameter disc of non-corrodible metal, suitably drilled. To one inlet of the cover attach 100 feet of polythene tubing, 0-110 inch internal diameter (Portex tubing 54B, Portland Plastics Ltd., Hythe, Kent), coiling it round the supports.Attach a glass T-joint to the other end of the tubing. To another inlet in the cover attach a short length of polythene tubing, a &length mixing coil and another length of polythene tubing to connect with the lower limb of the T-joint. Connect the remaining limb of the T-joint to two mixing coils in series and then to one of the outlets in the cover. For making polythene-to-glass connections, the ends of the tubing may be softened by brief immersion in boiling water, the tubing being then pushed over the glass fitting and secured with thin copper wire. Fit the prepared inoculum reservoir (see below) through the hole in the rear of the incubation bath-top, and clamp in position. Magnetic diaphragm air-jhmp-Obtainable from Chas. Austen Pumps Ltd. , Byfleet, Surrey.Inoczzltxm reservoir and Jittiwgs-See Fig. 1. Rcageutt reservoirs-Pyrex bottles, of 5-litre capacity, fitted with stoppers, each provided with one glass tube reaching to the bottom and an inlet connected to a gas wash-bottle containing sodium hydroxide solution. REAGENTS- Prepare all reagents with carbon dioxide free distilled water. Reagents 1, 3 and 4 are conveniently stored in 5-litre containers protected from atmospheric carbon dioxide. 1. Sztlphuric acid, N-Add 0.1 per cent. v/v of MS Silicone Antifoam Emulsion RD (Hopkin & Williams Ltd., Chadwell Heath, Essex) to N sulphuric acid. 2. Carbonate bufey-Dissolve 56 g of analytical-reagent grade sodium hydrogen car- bonate and 35 g of analytical-reagent grade anhydrous sodium carbonate in sufficient water to produce 1 litre.3. Bz4,fered pIzzenolg5htkalein reagent-Add 7.0 ml of a 1 per cent, w/v solution of phenol- phthalein in methanol to about 900 rnl of water, and mix. For assay with E. coli 39SE as test organism, add 5.5 ml of carbonate buffer, and dilute with water to 1 litre. Add 0.25 ml of capryl alcohol, and shake well. Adjust the strength of this reagent as described on pa 698. 4. Diluent-Add 0.1 per cent. v/v of Tween 20 to distilled water, and mix. 5. Formaldehyde solutiopzs, (a) 1 per cent. w l v , (b) 0.1 par cent. w/v. (a) Dilute 2.5 ml of a 40 per cent. w/v solution of formaldehyde to 100 ml with diluent. (b) Dilute solution (a) (1 + 9) with diluent. Medium 1-Prepare from the materials listed below. NUTRIENT MEDIA- Dehydrated Penassay broth (Difco") .. . . 17*5g Yeast extract (Difco*) . . . . .. . . 5.0g Tryptone (Difco") . . * . . . * . . . 10-og Polypropylene glycol, P2000 . f . . . . 0.02ml Distilled water .. . . . . ,. . . to I litre * Equivalent materials can be substituted, but they may not necessarily give the same yield of carbon dioxide, the same sensitivity to the antibiotic or the same dose - response dope.696 SIIAW ANT) ~CNCOMBE : COXTISCOCS Liu-roxi-rIc [A~talysl, Vol. 88 Sterilise 1-5-litre amounts in 80-oz bottles in an autoclave at 121" C (15 lb. per sq. inch) for 20 minutes. The bottles should be covered with aluminium foil during t tic sterilisation and tlien sealed immediately witti sterile rubber caps. Medium 2-Prepare exactly as described for medium 1 except for the substitution of 2 ml of Tween 20 per litre in place of the polypropylene glycol.Sterlilise 2-litrc amounts in 80-oz bottles. Store, preferably at 4" (1. Store, preferably at 4" C. TEST ORGAXISMS- As a general-purpose test organism, suitable for assaying many antibiotics, E . coli 397E has proved satisfactory, but the principle of continuous culture is of general application to other organisms including spore-formers. For such organisms the volume of the reservoir and the input of medium may be adjusted so that the inoculum used in the assay consists almost entirely of organisms in the vegetative form. Whatever organism is selected it must be capable of rapid growth and respiration and must be of sufficient sensitivity to the anti- biotic being assayed. The sensitivity shouId be assessed under continuous culture conditions, since other methods for measuring the sensitivity to an antibiotic are not necessarily relevant. Maintain test organisms as agar-slope cultures, fresh %&hour slope cultures being pre- pared as required for the assay.~ E P A R A T I O W ASD CONTINUOUS CULTURE OF INOCLLUM- Attach non-absorbent cotton filters to the air-inlet tube of the inoculum reservoir (see Fig. 1) and to thc inlet side of the fittings for the reservoir of medium 1. \:rap tlie fittings, G, of the reservoir assembly in aluminium foil, and cover the tops of tubes C and E with foil. Sterilise the whole assembly in an autoclave at 121" C (15 lb. per sq. inch) for 20 minutes. Fig. 1. Assembly for continuous culture of inoculum \Vhen required for use, aseptically remove the foil film from assembly G, and insert into a bottle of sterile medium 1.Slightly ease out the plug on the inociilum resen-oir, and by means of air-pressure applied to the filter, H, blow over 50 rnl of medium 1 into the inoculum rescrvoir. Aseptically remove 5 ml of medium from the inocuIum reservoir, and add it to an over- night agar-slope culture of the test organism. Suspend the organisms in the medium, andSeptember, 19633 MICROBIOLOGICAL ASSAY OF ANTIBIOTICS 697 transfer the suspension to the inoculum reservoir. Re-secure the plug and fittings, ensuring that the jet of the air-inlet tube B is directed round the circumference of the reservoir and away from the bottom of the inoculum tube E. Immerse the inoculum reservoir in the incubation bath, and clamp in position, so that the level of liquid in the reservoir is below the level of the water in the bath.Insert the Tygon pump tube into one channel of the Auto- Analyzer pump, and start the pump. Connect the air-pump to filter A, and pass a steady stream of air (about 40 litres per hour) into the inoculum reservoir. Allow the system, assembled as described below, to reach equilibrium, as will be shown by the production of a constant amount of respiratory carbon dioxide during the fixed incubation time of the assay. For organisms with a short generation time, 6 hours should be sufficient, but it is convenient to allow equilibration to proceed overnight. The inoculum can be renewed at the end of each working day or allowed to continue for several days by aseptically transferring the assembly G to a fresh bottle of medium 1 each day.Alternatively, a reservoir of medium 1 large enough to last for a week may be attached, provided contamination of the medium and inoculum reservoir can be prevented. ASSEMBLY OF AWTOANALYZER- The assembly shown in Fig. 2 incorporates a preliminary 1 + 9 dilution stage. This is optional but convenient for the assay of solid samples with potencies approaching those of the pure antibiotics, since the need for making many accurate dilutions by hand is thereby avoided. If, however, greater sensitivity is desired, the preliminary dilution stage can be omitted, as shown in Fig. 3, or a smaller dilution can be substituted. Whatever arrangement is adopted, it is essential that the variable amount of air admitted by the alternating operation of the antibiotic and formaldehyde sample tubes be eliminated from the system.Only liquid may be permitted to enter the re-sampling tube. If the preliminary dilution does not exceed about 1 + 9 the whole manifold can be pumped satisfactorily with one pump, but it may be more convenient to use a separate pump for the medium 1 and the preliminary dilution stage, particularly if a larger preliminary dilution is desired. Formal d e h y d e sol u t ion, 3 Waste Re-sample } :dim 2 lnoculum Medium I N Sulphuric acid Buffered phenolphthalein Absorptiometer (15-rnm flow cell; 550 mp) Fig. 2. Flow diagram for automatic microbiological assay of high-potency antibiotic samplesGO8 SHAW AKD DCNCOMRE : COSTISCOCS AC'TOM.ATIC ylzalyst, Vol.88 Fo rma id eh y dc so Iu ti o n, Sample w- Air l i g . 3. Partial flow diagram for incrertscd sciisitivity Apart from the continuous culture assembly the rest of the XiitoAnalyzer arrangement is substantially that suggested by Haney, Gerkc, Madigan, Pangano and Ferrail However, we prcfer to carry out the gas - liquid separation at 37'' C: with the \axiim-jackcttrl separator shown in Fig. 4; this, if mounted immediately on top of the incubation bath, permits visual examination of its performance. The substitution of a 25-rnm tubular flow-cell in place of the 10-mm conventional one gives improved separation of samples and a bctter dose - response calibration curve. The glass cactw-fitting, A (see Fig. l ) , must he of sufficiently wide borc to gcncratc air bubbles large enough to maintain a regular bubble pattern througliout thc incubation coil.l o set up the AutoAnaljmr for the assay, fill the constant-le\.el device with 1 per cent. fomaldchydc solution (or 0-1 per cent. formaldehyde solution if the preliminarjr dilution is omitted, as shown in Fig. 3), and connect tube E of the inoculum reservoir to the inoculum tube of the manifold with 0.045-inch bore polythene tubing. This conncction should be as short as possible, to minimise the continued growth of the organisms bciore joining the medium 2 and diluted sample streams. Connect thc appropriate tulles to the dilircnt, K sulpliuric acid, mcdium 2 and reagent reservoirs, making the connection to the buffered p€tcnolplithalcin reagent with O445-incli bore polj.thene tubing:.Set the sample-plate in oycration at 20 tests PI- lmur, with water in the sample-cups. When peaks begin to be recorded, acljiist the strength of the buffered phcno1,phtlialein reagent with either carbonate buffer or distilled watcr containing 0.7 per cent. ,-. Fig. 4. Vacuum-jackcttcd gas - liquid separator. working capacity, approximatcly 0.7 mlSeptember, 19631 MICROBIOLOGICAL ASSAY OF ANTIBIOTICS 699 v/v of a 1 per cent. w/v solution of phenolphthalein in methanol, so as to obtain peak transmissions of 90 per cent. Determine the concentration of the antibiotic under test required to give the mid-point of the dose - response curve (50 per cent. transmission) by placing a series of suitable dilutions in the sample-cups, each dilution being placed in three successive cups.Approximate mid-point concentrations for some antibiotics (with the preliminary 1 -+ 9 dilution) can be read from Fig. 5. Re-establish the mid-point concentration (medium standard) weekly or when any change is made in the manifold. Once this is established the concentration of the buffered phenolphthalein reagent may be adjusted to give 50 per cent. transmission with the medium standard in the sample cups. Haney et aE.1 recommended that the concentration of the high and low standards, required to establish the slope of the dose-response curve, be 1Q times and Q of the concentration of the medium standards, respectively. For the arrangement shown in Fig. 2 and with continuous inoculum culture, a narrower range of concentration is necessary to retain the responses on the approximately linear portion of the dose - response curves (preferably within the limits of 35 to 65 per cent.transmission). The permissible ratio depends on the slope ; for streptomycin, suitable concentrations are and 4 of that of the medium standard, giving, for example, cup concentrations of 3.0, 2.5 and 2.08 rng per rnl for the three standards. At the end of each working day disconnect the inoculum tube from the inoculum reservoir, and plug the top of tube E. Pass 2 N sodium hydroxide through the inoculum, medium 2 and buffered Phenolphthalein tubes for 5 minutes, and then pump diluent through to clear any deposit from the incubation and mixing coils. Fig. 5. Calibration graphs : curve A, novobiocin; curve B, sodium benzyl- penicillin ; curve C , streptomycin sulphate ; curve D, sodium cephalosporin C; curve E, neomycin sulphate ASSAY DESIGN- In the design proposed by Haney et a2.l each sample-plate is loaded with a series of 4 high, 4 low and 4 medium standards, 5 samples and 2 medium standards, with the last7 0 SHAW Ah'D DUNCOMRE : COSTINCOUS .4CTC)M:ITIC !Analyst, Vol.88 is assayed in duplicate, The responses of the medium standards are connected on the recording by a line (the drift line), and the responses of all samples arc measured as differences from tliis line. The slopes of the dose - response curve above and below the medium standard are determined, and these permit calculation of thc potencies of samples as differences from that of the medium standard.Many differcnt assay designs are possible according to the replication, and hence the precision, desired in relation to throughput of samples. In one design, which permits mathe- niaticaI correction for drift, the slope of the dose - response curve over the narrower range of potencies suitable for the AutoAnalyzer arrangement described above is first established with 4 low and 4 high standards. The remaining sample-plate positions are then filled with two similar 16-position patterns, each of 6 samples in duplicate and 4 mediiim standards. The samples and standards are so arranged that the second eight solutions in each pattern arc: in the reversc order of thc first eight. Thus solutions in the order A to H arc immediately followed by the same solutions arranged H to A.The mean response for each sample and for the four medium standards is then independent of drift, if it is assumed that this is linear over tlie 48 minutes required for recording the 16 responses. Tlic medium standards may occupy any positions in the pattern, but are conveniently placed in positions R and F. Both the designs considered above are one-level (1 x 1) assays, in which the slopes of the dose - response lines for the standard and for every sample are known to be the same or when this may be assumed. Moreover, with E. coli 397E as test organism similar slopes are given by different antibiotics (Fig- 5 ) , anti it is therefore necessary that the qualitative composition of the sample be known. C ALCL- LAT I ON o F RESU LTS- Read off from the recording the percentage transmission (TY;) for the low and liigli standards, ignoring for each the first of the four responses.Calculate the mean response for each standard, determine the difference, 1.1, and calculate tlie slope constant, K, from the K L) equation I( = -, where K is the log of the dose ratio (high to low). The slope constant represents on a log scale the concentration difference for each 1 per cent. transmission over the working range and is positive or negative according to whether the response o f a sample (in TO,/,) is numerically less or more than that of the medium standard. Calculate tlie mean responses for each sample and for the four mediiim standards. Determine the difference in TO; from the medium standard for each sample, and multiply Sample number 1 2 3 4 5 6 7 H 9 10 11 12 l'repared strength of streptomycin sulphate, xng pcr ml 3.00 3.50 3.20 3.10 2-90 2.60 2.70 3.16 3.20 3-52 2.86 2-78 STREPTOMYCIN ASSAYS Amount of streptomycin found by using .- Ail to Analyzer Agar-plate method r-.-.-A.~ 7 ~ ~ 7- - 7 7 <---A - 1,' 2, * 3 , t 1,: 2,: mg per ml mg per ml mg per rnl mg pcr ml ing per rnl 3-01 3.00 2.95 3-46 3-52 3 4 2 3-17 3.20 3-08 3-08 3.09 2-99 2.83 2.93 2.79 2-60 2.54 2.6 1 - 2.64 -- - 3-17 - - 3.34 - - .. 3.60 - _ . 2.90 -_ _ . 2.77 __ 2.85 3-47 3-30 3.07 2.99 2.54 2.76 3.1 1 3.30 3-5 1 3.03 2.93 3.05 3-48 3-28 3.03 2.88 2.43 2-87 3.4 1 3.27 3-51 2-83 2-78 * Calculated by thc proposed method. I ;he results (if an 8 x 8 quasi-latindsquare design with U. s~rbtilis 841 as test organism. alculated bv the method of I-Ianev et al.'September, 19631 MICROBIOLOGICAL ASSAY OF ANTIBIOTICS 701 by K, maintaining the correct sign. Add this to (or subtract it from, as appropriate) the log of the potency of the medium standard. Convert to the antilog to give the concentration of the sample solution, and calculate the result on the original sample. The results shown in Table I were obtained on a series of accurately prepared dilutions (unknown to the operator) of streptomycin sulphate by the two AutoAnalyzer designs and methods of calculation discussed above. In comparison with the known concentration and with results on the same samples assayed at the same time by a conventional two-level agar-diffusion method (8 x 8 quasi-latin square) with a different test organism, the Auto- Analyzer results show no bias and demonstrate that, with the duplicate recording of samples recommended, an appreciably lower error can be expected. We thank all those at the Squibb Institute for Medical Research, New Brunswick, N. J., U.S.A., who generously supplied advance information of their procedures, Mr. W. K. Anslow, Dr. A. Ferrari and Mr. J. P. R. Tootill for helpful discussions and Mr. K. Clover for assistance with the practical work. REFEKENCES 1. Haney, T. A., Gerke, J. R., Madigan, M. E., Pagano, J. F., and Ferrari, A., Ann. N.Y. Acad. 2. Gerke, J . R., Haney, T. A., and Pagano, J. R., Ibid., 1962, 93, 640. 3. Kavanagh, F., Editor, “Analytical Microbiology,” Academic Press Inc., London and New York, Received April 16th, 1963 Sci., 1962, 93, 627. 1963.

ISSN:0003-2654    

DOI:10.1039/AN9638800694

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

7 02 SCHOLES AKD THULBOURXE AUTOKITIC PHOCED GRES FOR THE [A.id)td, VOl. 88 Automatic Procedures for the Colorimetric Analysis of Iron- and Steel-making Slags BY P. H. SCHOLES AND CYKTHJA THULBOCRKE ( T h e Byitish Iron and Steel Itesenrcli ,4ssoczatzo??, Hoyk .ShwL, Sht!frcld 3) Thc AutoAnalyzer system of automatic colorimetric analysis has been successfully applicd t o the determination of total iron and the oxidcs of inanganesc, phosphorus and aluminiu~n in iron- and steel-making slags. The sample is decomposed in acid, silicon removed as metasilicic acid and the solution &luted to a fixed volume. From this stage analysis is com- pletely automatic, the operator bcing required only to read optical-density values from ;I chart recorder and convert these in to percentage conteri t by reference to a calibration factor.l‘he a u tomatccl procedures arc primarily intended for analysing at least tcn sample solutions a t a. time. Each slag constituent is determined separately by using a specially designed flow system that can be assembled and calibrated in about 40 miniilcs. The speed o f operation is either 20 or 40 samplcs per hour dcpcndent on thc type of chemicai reaction involved. THE use of the Technicon AntoA4nalyzer is well established in clinical and industrial analj.sis. Typical applications of this cquipmen t in the Unitcd Kingdom have included the determina- tion of streptomycin in fermentation broths1 and the determination of zinc, lead, molybdenum and nickel in soil extracts.2 The sample in solution form is aspirated, pumped through the apparatus and mixed with appropriate reagents to produce a colouretl solution.This colourcd solution is passed through a colorimcter; the optical density is recorded on a chart and related to percentage content from a calihation graph. A fuller description of the mechanics and princip1t.s of operation has been provided by F*errari, Kusso-Alesi and Kelly.3 As an initial study of thc instrument’s application to steelworks’ matcrials, four spectro- photometric metlwds for analysing iron- and sttiel-making slags have been automated. These mctliods form part of an analytical scheme used in our lat)oratorjr,4 in which the sarnpIe of slag is decomposed in acid, silica rcmovcd and the soliition diluted to a fixed volume, From this solution, separate portions are taken for the determination of total iron, manganese oxide, phosphorus pentoxide and alumina by spectrophotometry, and calcium and magnesium oxides by complexometric titration.This papcr described the adaptation of the spectro- photometric procedures for use with thc ‘Technicon AutoAnalyzer. PREPAlCATION OF SAMPLE SOLUTION A powdered slag sample weighing 0-5 g is partly decomposed in hydrochloric acid; the solution is then oxidised, perchloric acid is added, and the solution is evaporated until fumes of perchloric acid are evolvcd. Fuming is continued under reflux for 10 to 15 minutes to separate silicon as metasilicic acid. After removal by filtration, the precipitatc is ignited and weighed as silica, Finally, the silica is volatiliscd as the fluoride, the rcsidusl oxides bcing fused and added to the sample solution.This solution is eXTapor-atcd t o fumes for a second time to remove hydrochloric acid and to oxidise chromium to the scxavalcnt state. The volume of perchloric acid remaining should be about 10 to 15 ml, giving an approximately 2.5 per cent. v/v acid soliition on dilution t o 5 0 ml. Full details o f the preparation of the sample solution arc given in a British Iron and Steel Research Association Publication.4 DETE RM I SATIOX Oh* ALU 11 I KA In the manual method, iron and manganese are separated from aluminium by preciyita- tion with sodium hydroxide solution in the presence of hydrogen peroxide. Interference tluc to titanium and vanadium is prevented by a further addition of hydrogen peroside.After tlit: solution has been carefully neutralised and adjusted to pH 6-0, Eriocliromc cyaninc is added to produce the orange-red aluminium - dye complex; colour dcvelopmcnt is completc in 20 to 30 minutes.September, 19631 COLORIMETRIC ANALYSIS OF IKOS- AND STEEL-hlAKINC SLAGS 703 Initial tests indicated that the colour reaction could be automated, provided that two tirne-delay coils were placed in the flow system to allow sufficient time for reproducible coIour developmcnt. Thc colour reaction is extremely sensitive, and it was difficult to dcsign a flow system t o cover thc dcsircd percentage range (up to 16 per cent. of alumina, equi\ralcnt t o 8 mg of aluminiiim per 100 ml of sample solution). Considerable dilution was necessary, and this was achieved by a method of “opposed dilution” introtlucccl by Technicon Tnstru- merits Ltd.In the proposed system, a tube pumping diluent at 0.32 ml per minute in opposi- tion to a sample line pumping a t 0.42 ml per minute gives an cffcctive sample intrike of 0.1 ml per minute. (A tube of flow rate 0.32 ml per minute was the smallest available to us when these procedures were developed.) With a high over-all flow rate of 18 ml per minute, it was now possible to cover at least half tlie desired percentage range. For concentrations greater than about 7.5 per cent. of alumina, a diluted sample solution must be iisctI. Considerablc difficulties were caiiscd by excessively noisy records. This was traced to the formation of a blue deposit in the flow-through cell of the colorimetcr and was probably due to the decomposition of thc aluminium - dyc complcx.This dificulty WZLS overcome by adding a small amount of acetone to ttic dyc solution. Contamination between successive samples is not a serious problttm in thc AutoAnalyzer system provided that a properly designed manifold of tiibes is used. Thc problem is, however, accentuated when such strongly coloured dye solntions as Erioclirome cyanine arc used. For this rcason, it is preferable to IISC alttirnate plastic cups filled with dilritc pcrchloric acid to provide a thorough washing after each sample has been analysed. An additional advantage of this technique is tlic impro\vcmc.nt in resolution between successive peaks on the chart rccorder. In the proposed flow system, sampler-plate operating a t 40 tcsts per hour with alternate acid-filled cups gives an effective speed of 20 samples per hour.IN TE 13 FE H I S G ELF: ME STS- In an attempt to automate the procedure for removing iron, etc., in the manual method, the sample stream was mixed with dilute solutions of sodium hyciroxidtb and 113-drogen yer- oxide, and dialyserl through a Celloplmnc membrane. Only partial siicccss was attained since there was frequent rupturing of the membrane. In subsequent work, the use of ascorbic acid for complexing iron made it possible to dispense with the dialyser unit. Ascorbic acid was first proposed by Hill5 as a suitable complexing agent for iron and certain othcr elements in the aluminium - Eriochrornc cyaninc reaction. Tn our experience, the presence of this reagent during colour development in tlie manual procedurc results in the formation of a coloured product that is not suf‘ficicntly stable with respect to time to permit optical-density measurements to bc made.This latter consideration is of less im- portance in the AutoQlnalyzer system, as all measiircments are made under identical condi- tions, and, provided that tlie rate of colour developmcnt is reproduriblc, the production of TABLE r THE EFFECT OF V.i\RIOCS EL13hlEKTS interfcring c1t:mcnt Titanium . . .. Mangariesc . . . . Chrorniuni>‘I . , Vanadium . . * . L’Imspliorus . . .. .. .. .. .. Amount o f interfcring clemrnt added, mg pcr 100 ml 1.0 2-5 20.0 1.0 2-5 10-0 20.0 1 . 0 2-5 10.0 20.0 20.0 50.0 05 TIIE DI3TERMISATIOK OF ALUMINA rlmount of alumi~iium found when- - r-.-- 0.2 mg of A1 was added, mg per 100 ml 0-25 0.25 0-20 - ._ _- _.- - 0-20704 SCHOLES AND TIIULROURKE : AUTOMATIC PROCEDURES FOR THE [Analyst, T.'Ol. 88 a time-stable complex is not essential. Tests confirmed that in a flow system operating under closely controlled conditions (see Fig. 1) iron in amounts up to 25 rng per 100 ml of sample solution could be successfully complexed with ascorbic acid. There was, however, a slight positive interference, equivalent to about 0-05 rng of aluminium, when the amount of iron added was increased to 50mg per 100rnl. Previous experience in determining aluminium by using Eriochrorne cyanine suggested that, of the other elements present in slags, titanium, vanadium, manganese, chromium and phosphorus might interfere in the colour reaction.The effect of thew elements was deter- mined at three concentration levels of aluminium, and the rcsults are summarised in Table 1. Interference was considered to be significant when the amount of aluminium found lay outside arbitrary limits of 5 0.03 mg at the 1-0 mg and 3.0 mg per 100 rnl levels, and f 0-05 mg at the 0.2 mg per 100 ml level when the analytical precision is not great owing to the sigmoid character of the calibration graph. Titanium and vanadium in amounts of more than 1 mg and 2.5 mg per 100 ml, respec- tivdy, interfcre seriously giving high resuits. I t was not possible to mask interference of these elements with hydrogen peroxide as in the manual procedure, presumably because of the influence of the ascorbic acid present in the flow system.Chromium in the sexsvalent statc can be tolerated in amounts up to 2-5 mg per 100 ml, but with chromium contents above this level low results for aluminium will be obtained. Manganese up to 20 rng and phos- phorus (as phosphate) up to 50 mg per 100 ml are without effect. In terms of the percentage oxide content, the limits are 2 per cent. of titania, 4 per cent. of chromium trioxide and 4 per cent. of vanadium pentoxide. BI ETH 0 D REAGESTS- of Teepol, and dilute to 1 litre. Ascorbic acid, 0-5 per cent. w/v-l)issolve 5 g o€ ascorbic acid in water, add 10 drops Perchloric acid, diluted (1 + 3) and (1 + 39)-Prepare from perchloric acid, sp.gr. 1.54. Dye soZul~o~--l)issolve 0-25 g of Merck Eriochrome cyanine in water, add 1 0 ml of Rufer solzttion, p H 6.4-Dissolve 400 g of analytical-reagent grade hydrated sodium The pH of this solution Standard aluminium solution-1)issolve 0.6293 g of high purity aluminium metal in 20 ml Cool, add water, Dilute to 1 litre in a calibrated flask, and acetone, and dilute to 1 litre in a caIibrated flask. acetate in water, add 10 ml of glacial acetic acid, and dilute to 1 litre. should be bctween 6.35 and 6.45.of hydrochloric acid, add 15 ml of perchloric acid, and evaporate to fumes. and warm, if necessary, to redissolve the salts. store in a stoppered polythcne bottle. PROCETIIJRE- Set the sampler-plate at the rate of 40 samples pcr hour, and fill alternate sample-cups with dilute perchloric acid (1 + 39). Prepare a series of calibration solutions by adding up to 10 ml of standard aluminium solution to 10 ml of diluted perchloric acid (1 + 3).Dilute each solution to 100 rnl in a calibrated flask, and store in a tightly stoppered polythene bottle. Prepare a calibration curve by running the calibration solutions in duplicate. Draw the base line on the chart recorder by joining together the small peaks due to the intermediate acid cups, and record the peak height of each calibration solution in terms of optical dcnsity. The base line drifts slightly with time owing to fading of the dye solution, but the drift should not exceed 0,005 optical- density units per hour. Deduct the optical density of the reagent blank solution in each determination. The graph relating aluminium content to optical density is s i p o i d in character (see Fig.2) ; it is, however, rectilinear in the range 1.5 to 7-5 mg of alumina per 100 ml, and the curve between thesc two points can be expressed as y = mx + c. For contents greater than 7-5 mg of alumina per 100 ml, the graph is curved and not reproducible. In routine use, a graph should be plotted with each batch of tests, but if the alumina content of the samples is known to exceed 2 per cent., i.e., falling on the linear part of the graph, it is only necessary Prepare a fresh solution daily. Store in a stoppered bottle, 1 ml -= 1 mg of alumina. Assemble the AutoAnalyzer as shown in Fig. 1.September, 19631 COLORIMETRIC ANALYSIS OF IRON- AND STEEL-MAKING SLAGS 705 to run, say, three calibration solutions each day to establish its slope.At least one calibration solution should be measured with each sampler-plate of test solutions as a check on calibration drift. Run the test samples through the apparatus, and calculate the percentage of alumina by reference to the calibration graph. Deduct the apparent percentage of alumina found in the reagent blank solution. For samples that give optical-density values on the upper curved part of the graph, repeat the tests on a diluted sample solution containing sufficient added dilute perchloric acid to maintain the acid concentration at (I + 391, e.g., for a (1 + 1) dilution, take 50 rnl of sample solution, add 5 ml of diluted perchloric (1 + 3), and dilute to 100 ml in a calibrated pump 1 I 0.32 ml/min 0.42 ml/rnin flask. Proportioning -1 0.7 r I / Dilute perchloric acid (I + 39) Dye solution Buffer solution, pH 6.4 0 Time de tay delay coil I 0 5 10 15 1 Concentration, yo Colorimeter (535-mp Fig.2. Calibration graph for the analysis of filters; 10-mm flow cell) steel-making slags with the AutoAnalyzer : curve A, Fig. 1. AutoAnalyzer flow total iron; curve B, manga- system for determining alumina nese oxide; curve C, phos- in steel-making slags phorus pentoxide ; curve D, alumina -7 DETERMINATION OF PHOSPHORUS PENTOXIDE 0.6 - 0-5 - x u .- 2 0.4- -0 m I ._ $ 0.3- 0-2 - 0.1 - In the manual method, a solution containing ammonium vanadate, ammonium molybdate and dilute nitric acid is added to a portion of the sample solution to form the yellow molybdo- vanadophosphoric acid complex. This simple procedure proved easy to automate, but it was necessary to give some consideration to the choice uf light filter and to possible inter- ference from iron present in the sample solution.Initial tests indicated a slight positive interference from iron when the 440-mp filter (corresponding to the wavelength used in the manual method) was used. At this wavelength the optical-density readings were rather low, e.g., 0-21 optical-density units for solutions containing 5 mg of phosphorus per 100 ml, and for this reason the more sensitive 420-mp filter was preferred (0.35 optical-density units for 5 rng of phosphorus per 100 ml). Unfortunately, interference from iron is more pro- nounced as the ultraviolet part of the spectrum is approached. A series of tests in which iron was added to solutions containing different amounts of phosphorus showed that inter- ference was roughly proportional to the amount of iron present, up to phosphorus contents of 8 rng per 100 ml of test solution, Thus it is possible to make a small deduction from the apparent phosphorus content to allow for interference from iron ; the deduction is 0.002 rng of phosphorus per mg of iron present.706 SCHOLES AND THULBOGRNE : AUTOMATIC YIIOCEDURES FOR THE [AHdySt, VOl.88 METHOD REAG E m s - i'Molybdovanadate soldion-Transfer 0.31 g of ammonium metavanadate to a beaker containing 50 ml of water and slowly add 40 ml of nitric acid, sp.gr. 1-42, whilc swirling the solution. Dissolve 12.5 g of ammonium molybdate in 100ml of water, warming if necessary. Cool both solutions thoroughly, mix, and dilute to 1 litre.Store in a stoppered polythene bottle; the solution is stable for about 3 days. Slandard fihosphorus solution-Dissolve 1.9 17 g of potassium dihydrogcn orthophosphate in water, dilute to I litrc in a calibrated flask, and store in a stoppered polythcnc bottle. Heat below the boiIing-point until dissolution is complete. 1 ml -= 1 rng of phosphorus pentoxide. P I ~ O C E ~ U R E - Assemble the AutoAnalyzcr as shown in Fig. 3, and set thc sampler-plate at the rate of 40 samples per hour. Prepare a standard calibration graph from a series of solutions contain- ing 10 ml of diluted perchloric acid (1 + 3) and up t o 20 ml of standard phosphorus solution, diluted to 100ml to cover a range of contents up to 20 per cent. of phosphorus pentoxide, Proportioning pump 080 ml min Sample Air Water Mol ybdovanadate solution Mixing coil 3.40 rnl min -- - 1.60 mi min Colorimeter (420-mp filters; 10-mm flow cell) Proportioning pump Sample Mixing coil i h0& Dilute phosphoric acid r- - - I 20 ml m i n t * S01Ution Potassium periodate I L S o d i u r n nitrate solution or air o 80 ml mm- Colorimecer (535-mp filters.I O+mm flow cell) Fig. 3. Autohnalyzer flow systems for determining (a) phos- phorus pentoxide arid ( h ) mangancse oxide in stecl-making slags With the instrument used by us the calibration graph wa.s linear up to 15 pcr cent. of phospliorus pentoxide (see Fig. 2 ) ; a diluted sample solution was used when tlic content exceeded this limit. For routine use plot thrce calibration points in duplicate to establish the slope of the graph.At lcast one calibration solution should be measured with each sampler-plate of test solutions as a chcck on the calibration. For sample solutions containing coloured ions, repeat thc determinations substituting dilute nitric acid (4 + 96) for the molybdovanadate solution. Ileduct these readings to give the net optical density due to phosphorus pentoxide.September, 19631 COLORIMETRIC ANALYSIS OF IRON- AND STEEL-MAKING SLAGS 707 DETERMINATION OF MANGANESE OXIDE Manganese is determined manually by oxidising it to permanganic acid by heating with potassium periodate in a strongly acid solution. In the automated procedure, development of the manganese colour is effected by passing the solution through a 40 f t length of glass tubing in the form of a coil immersed in a heating bath maintained at 95” C.Provision is made for reducing the manganese colour with sodium nitrite in order to determine the optical density of any coloured ions, such as chromium, present in solution. METHOD REAGENTS- Phosphoric acid, diluted (1 + 3). Sodium nitrite solution, 0.5 per cent. w/v, aqzleous. Potassium periodate solzctiout, 2 per cent. w/v-Transfer 20 g of potassium periodate to a 1-litre beaker, add 800 rnl of diluted phosphoric acid (1 + 3), heat almost to the boiling- point, and stir until dissolved. Standard manganese solzdion-Dissolve 1-1 144 g of potassium permanganate in 500 rnl of water, reduce with a slight excess of sulphurous acid, dilute to 1 litre in a calibrated flask, and store in a stoppered polythene bottle.1 ml = 0.5 mg of manganese oxide. C,ool, and dilute to 1 litre with diluted phosphoric acid. PROCEDURE- Assemble the AutoAnalyzer as shown in Fig. 3, and set the sampler-plate at the rate of 40 samples per hour. Prepare a standard calibration graph as described under “Determina- tion of Phosphorus Pentoxide,” by using suitable volumes of standard manganese solution to cover a range of up to 20 per cent. of manganese oxide. With our instrument the calibration graph was linear up to 12.5 per cent. of manganese oxide (see Fig. 2); a diluted sample solution was used when the content exceeded this limit. For a sample solution containing coloured ions repeat the determinations, substituting 0.5 per cent. w/v solution of sodium nitrite for the air-line immediately after the heating bath.Deduct these readings to give the net optical density due to manganese oxide. DETERMINATION OF TOTAL IRON In the manual procedure, iron is reduced to the bivalent state with hydroxyammonium chloride; the solution is buffered to about pH 4 with sodium citrate solution, and 1,lO-phenan- throline hydrate is added to form the orange-red complex. Colour development is complete in about 20 minutes. The two principal difficulties encountered in automating the manual method were Contamination and lack of resolution between samples. The problem of contamination was overcome by introducing intermediate acid-filled cups between successive samples as in the method for alumina. In addition, an extra large-diameter air-line was included in the flow system to scavenge contaminants from the time-delay coil used for colour development.These modifications also considerably improved resolution between successive peaks on the chart recorder. It was established that minor variations in the amount of perchloric acid in the main sample solution would not affect the optical density of the coloured solution provided that the acid content was between 2 per cent. v/v and 3 per cent. v/v; acid contents in excess of this range would lead to low results. The slope of the cahbration graph was found to depend on the concentration of 1,lO-phenanthroline hydrate present in the system, but at all concentration levels the graph was linear up to 15 mg of iron per 100 ml of test solution. Above this value the graph is curved and non-reproducible (see Fig.2).708 REAGEKTS- SCHOLES AKL) THCLROURNE : AGTOMATIC PROCEDCRES FOR THE [Anahst, Vol. 88 METHOD Hydroxyammonium chloride solulion, 0.5 per cent. w/v, aqueous. Sodium citrate solution, 3 par cent. w/v, aqueous. 1,10-Yhenanthroline hydrate solution, 0.2 per cent. w/v, aqueous. Standard iron solutio+-L)issolve exactly 1 g of Matthey iron sponge in 20 ml of hydro- cliloric acid, oxidise with nitric acid, add 10 ml of perchloric acid, and evaporate to fumes. Cool, add water, and warm, if necessary, to dissolve salts. M u t e to 1 litre in a cdibratcd flask, and storc in a stoppered polythene bottle. 1 ml = 1 rng of iron. PROCEDURE- Assemble the AutoAnalyzer as shown in Fig. 4. Set the sampler-plate at 40 samples per hour, and fill alternate cups with dilute perchloric acid (1 -!- 39).Prepare a standard graph as described under “Determination of Phosphorus Pentoxide” ; use suit able volumes of a standard iron solution to cover a range of contents up to 20 per cent. of total iron. With our instrument, the calibration graph was linear up to 15 per cent. total iron (see Fig. 2); a diluted sample solution was used when the content exceeded this limit. Proportioning pump Sample Air Hydroxyammoniu rn chloride solution Air Sodium citrate SOIUI Phenant hrolme hydrate solution :ion Colorimeter (505-mp filters; 6-mm flow cell) Fig. 4. AutoAnalpzcr flow systcm for determining total iron in steel-making slags RESULTS, PRECISIOK AXD ACCURACY The instrumental precision of the AutoAnalyzer systcm was determined by making 12 successive determinations on three solutions prepared from samples selected to give different optical-density values.Table TI indicates that instrumental precision is related to optical density and varics from about 2 per cent. (as coefficient of variation) at the 0.05 level of optical density to about 0.3 per cent. at the 0.5 level. These errors do not include minor calibration errors that might be experienced in day-to-day operation or chemical errors involved in the preparation of the sample solution. To measure the over-all analytical precision, six samples of British Chemical Standard So. 174/1 basic slag were analysed six times over a period of 3 months by the AutoAnalyzer procedures. These same sample solutions were also analysed by the manual methods; the individual results obtained are shown in part of Table IV, and the precision obtained with both manuaI and automatic procedures is compared in Table 111. Over-a11 precision with the AutoAnalyzer is at least as good as that obtained by carefully operated manual processes and in the determination of the alumina it is noticeably better.This is probably due to the simpler procedure developed for automatic operation, which avoids the nccd for chemical separation to remove interfering elements.September, 19631 COLORIMETRIC ANALYSIS OF IRON- AND STEEL-MAKING SLAGS 709 The results obtained for the sample of standard slag (see Table 111) are in close agreement with Certificate values. As an additional check on the accuracy of the automated methods, results obtained on a series of 10 duplicated sample solutions were compared with results obtained by the manual methods.The results are presented in Table IV together with results for the sample of standard slag mentioned above. TABLE I1 PRECISION OF THE INSTRUMENT Mean Percentage Standard Coefficient Sample optical density equivalent, deviation, of variation, % O/ % /Q Afanganese ozide- Slag “10”’ (- 1.75 per cent.) . . 0-0495 1.81 0.015 0.84 B.C.S. 174/1 (5-11 per cent.) . . 0.165 4-96 0.025 0-50 Slag “6” (- 12.5 per cent.) . . 0-380 12-61 0.059 0 4 7 A~UW~~FZLZ- B.C.S. 174/lt (1.82 per cent,) . . 0.035 1.7 1 0.050 2.9 Slag “l0”t (- 5-7 per cent.) . . 0.195 5.68 0.063 1.1 Slag “6” (w 7 per cent.) . . . . 0.255 7.16 0.073 1.0 Total i v m - Slag “6” (- 2.5 per cent.) - ., . 0.112 2.54 0.014 0.56 B.C.S. 174/1 (8.45 per cent.) . . 0.400 8-48 0-023 0.27 Slag “10” (- 15 per cent.) . . . . 0-659 14-91 0-034 0.23 Phosphorus $entoxide- Slag ‘‘6” (- 1 per cent.) . . . . 0.031 1-00 0.020 2.00 B.C.S. 174/1 (12.3 per cent.) . . 0.374 12-20 0.060 0.49 Slag “10” (w 15 per cent.) . . . . 0.457 14-89 0.054 0.36 * Range expansion x 4. t Range expansion x 2. TABLE I11 COMPARISON OF PRECISION OBTAINED WITH AUTOANALYZER AND MANUAL METHODS FOR ANALYSING B.C.S. 174/1 BASIC SLAG Precision Precision obtained with AutoAnalyzer of manual method h A f \ r -7 Certificate Mean Coefficient Mean Coefficient value and of Standard of of Standard of range, results, deviation, variation, resdts, deviation, variation, % % % % % % % Manganese oxide 5.11 4-93 0.025 0.51 4.95 0.045 0.91 Total iron 8*45* 8-55 0.10 1.2 8-41 0.12 1-4 Phosphorus pent- 12.30 12-17 0.10 0-98 12-23 0.10 0-82 Alumina 1-82 1.76 0-07 4.0 1.69 0.12 7.1 (5.00 t o 5-15) oxide (12-16 to 12-50) (1-78 to 1.92)f * Sum of mean results quoted for FeO and Fe,O,.t B.I.S.R.A. Methods of Analysis Committee’s results. To establish whether or not a bias existed between the two procedures, the average discrepancy between the results was calculated, together with standard deviation and 95 per cent. confidence limits. There is evidence of a positive bias in the determination of total iron, which is confirmed by the slightly higher results obtained for B.C.S. 174/1 standard slag (see Table 111). Con- sideration of the manual method suggests that this bias and the high standard deviation of the average discrepancy may be due to errors introduced into the manual method during theTABLE IV COMPARATIVE RESULTS OBTAINED BY THE MANUAL AND AUTOMATIC COLORIMETRIC PROCEDURES Phosphorus pentoxide found B Y manual % Sample number procedure, Blast furnace slags- M.G.S.408 . , 0.20 0-25 M.G.S. 409 . . 0.10 0.05 Basic arc furnace skzgs- M.G.S. 406 . . B.I.S.R.A. 1 .. B.I.S.R.A. 2 . . B.I.S.R.A. 3 , . B.I.S.R.A. 4 . . B.I.S.R.A. 5 .. B.I.S.R.A. 6 , . B.I.S.R.A. 7 . . B.I.S.R.A. 8 . . B.I.S.R.A. 9 . . 1-00 1-05 0.95 1.00 0.35 0.40 0.40 0.45 0.55 0.50 0-60 0.50 0.85 1.00 1.00 1.05 1-15 1.10 1.15 1.10 1.10 1.05 1.20 1.10 1.40 1-35 1-40 1-40 0.15 0.15 0.15 0.15 0.25 0.20 0.25 0.20 0.15 0.15 0.15 0.15 BY Auto- % Analyzer, 0-15 0.15 <o-1 <0*I Dif- ference, % - 0.05 - 0.10 - - + 0.05 + 0.05 + 0.05 + 0.05 - 0.05 - 0.10 -1-0-15 + 0.05 - 0.05 - 0.05 - 0.05 -0.10 - 0.05 0.00 0.00 0-00 - 0.05 - 0.05 0.00 0.00 Total iron found / --7 BY % manual procedure, 0.40 0.40 0.40 0-40 20-8 20-6 7.40 7.40 7.45 7.65 16.8 16.8 20.2 20.1 24.0 24.4 18.7 18-9 1.05 1-05 0.85 0.85 2-35 2.30 BY % Auto- Analyzer, 0-28 0-30 0-30 0.30 20.6 20.6 7-35 7.55 7.60 7.70 17-2 17.2 20.6 20.5 24.2 24.5 19-1 19.3 1.10 1-00 1-10 1.00 2.25 2.15 Dif- ference, % -0.16 -0-10 -0.10 - 0.10 - 0.20 0-00 - 0.05 +0-15 + 0-15 +Om15 + 0.40 + 0.40 + 0-40 + 0.40 + 0.20 +0*10 + 0-40 + 0.40 + 0-05 + 0-25 + 0.15 - 0.10 -0.15 - 0.05 Manganese oxide found 7- ~ - BY % manual procedure, 1-30 1.40 0.95 0.95 9-15 9-00 4-45 4.55 4.80 4.75 10.9 10.8 12.4 12.4 13.1 13.% 12.5 12-4 0.80 0.85 0.95 0.95 0.65 0.65 BY Analyzer, % Auto- 1.35 1.30 0.95 0.95 9.30 9-20 4-35 4-35 4.70 4.65 10.8 10.7 12.1 12.4 12.9 12.9 12.6 12.3 0.60 0.55 1-00 1.05 0-60 0.70 Dif- ferencc, % + 0.05 - 0.10 0.00 0.00 + 0.15 + 0*20 - 0.10 - 0.20 -0.10 -0.10 - 0.10 -0.10 -0.30 0.00 - 0.20 - 0.30 + 0.10 - 0.10 - 0.20 -0'30 + 0.05 + 0.10 - 0-05 + 0.06 Alumina f ouna - w BY BY manual Auto- procedure, Analyzer, % 14.5 14.5 14.8 15.0 5-70 5.75 3.65 3.65 3.45 3.50 2.70 2.65 2.50 2.35 1.65 1.75 2.00 1.85 1-45 1.55 1.55 1-65 4.10 4.25 % 14.2 14.4 14.7 14.7 5-35 5.45 3-85 3-75 3 4 0 3-45 2-80 2-75 2-15 2.05 2.00 2.00 2.10 2-05 1-55 1-75 1-90 1.90 4-40 4.45 -T Dif- f erence, % - 0.3 -0.1 - 0.1 - 0.3 - 0.35 - 0.30 + 0.20 + 0.10 - 0.05 - 0.05 + O .l O + 0.10 - 0.35 - 0.30 + 0.35 + a.25 + O s l o + 0.10 $0.10 + 0.20 + 0.35 + 0.25 + 0.30 + 0-20Phosphorus pentoxide found ' I Dif- BY B Y manual Auto- Sample number procedure, Analyzer, ference, % Basic open hearth fwnace slags- M.G.S. 401 M.G.S. 402 M.G.S. 403 M.G.S. 410 M.G.S. 411 M.G.S. 412 M.G.S. 413 B.C.S. 174/1 . , 17.1 17-1 . . 4-35 4-30 . . 5-90 5.85 . . 9.90 . . 8-75 6.85 . . 4.35 4.40 . . 7.35 7.25 10.0 12.4 12-2 12-1 12.2 12.3 12.2 Average discrepancy and confidence limits . . Standard deviation of average discrepancy . . % % 17.0 17.1 4.60 4.55 5-85 5.75 9.85 9.90 6.80 6-80 4.65 4.70 7.55 7-50 12.2 12.2 12.0 12.2 12.1 12-3 -0.10 0.00 + 0.25 + 0.25 - 0-05 -0.10 - 0.05 -0.10 + 0.05 - 0-05 + 0.30 + 0.30 + 0.20 + 0.25 - 0.02 0.00 -0.10 0.00 - 0.20 + O s l o + O - O l 0.04 0.12 TABLE IV-corntimed BY % manual procedure, 11.7 11-8 24-9 24-9 11-3 11.4 14.4 14.5 12.6 12.6 29.4 29.4 4-50 4.50 8.35 8-40 8-25 8-40 8-60 8-45 Auto- By Dif- Analyzer, ference, % % 11-5 11.8 25.1 25.2 11.0 11-2 14.2 14.3 12.6 12-4 29.1 29.3 4-60 4.55 8.60 8.55 8.45 8.65 8.60 8-40 - 0.20 0.00 + 0.20 + O s l o - 0.30 - 0.20 - 0.20 - 0.20 0.00 - 0-20 - 0.30 - 0.10 + O s l o + 0.05 + 0.25 $0.15 + 0.20 + 0.25 0.00 - 0.05 +0-05 t 0.06 0.20 Manganese oxide found BY manual procedure, % 3-40 3-45 5.05 5.05 5-40 5-40 14-3 14-3 10-5 10-5 2.00 2.00 4.65 4.70 5.00 5.00 4-90 4-90 4.95 4.95 Auto- By Dif- Analyzer, ference, % % 3.40 3-40 5-00 5.00 5.45 5.45 14.2 14.2 10.4 10.4 2.05 2-00 4.70 4.70 4.95 4-95 4-90 4.95 4.90 4-90 0.00 - 0.05 - 0.05 - 0.05 + 0.05 + 0.05 - 0.10 - 0.10 - 0.10 -0.10 + 0-05 0.00 + 0-05 0-00 - 0.05 - 0.05 0.00 + 0.05 - 0.05 - 0-05 -0.05 F 0.03 0.12 Alumina found 7 r--__h_li- manual Auto- BY Dif- BY procedure, Analyzer, ference, % 0.45 0.50 1.70 1.80 2-95 2.95 1.80 1.90 3.15 3.10 0-75 0.80 1.80 1-75 1.80 1-75 1-70 1.55 1.55 1.80 % % 0.50 + 0-05 0.50 0.00 1.70 0.00 1-85 +0-05 2-75 - 0.20 2.75 - 0.20 1-55 - 0.25 1-60 - 0.30 3.15 0.00 3-10 0.00 0-75 0.00 0.80 0.00 1-80 0.00 1.80 +0*05 1-85 f-0.05 2.85 + O s l o 1.70 0-00 1-70 +Om15 1.75 +0-20 1.70 -0.10 0.00 & 0.06 0.19712 SCH OLE3 S AX 1) TH U I, 130 U KN E transfer of small portions (1 or 2 ml) of the sample solution when the iron is present in high concentrations, i.e., more than 10 per cent.Errors involved in volumetric transfer are eliminated in the automated procedure.Results for manganese oxide show a small negative bias, but it is difficult to determine whether or not this is of practical significance. There is no evidence of significant bias in the determination of phosphorus pentoxidc and alumina. h high standard deviation of the average discrepancy was obserLw1 in the comparison of the alumina results, and this may possibly reflect errors introduced into the manual procedure by thc cliemical separation, as mentioned ahovc. ;Analyst, 1’01. 88 co s c 1-1: s I OK s Tlie Xuto:Inalyzer system of automatic colorimetric analysis can hc u s d for determining the total iron and the oxides of phosptmrus, manganese and aluminium in iron- and stcel- making slag. Accuracy and precision are equal to and possibly bctter than those obtained by carefully operated manual proccdiires. .4fter sample preparation and manual determination of silica, the time required for each of the automated proccdures is 1 hour for the analysis of either 20 or 40 samples, depending on the type of chemical reaction involved. To this time must be added a period of about 30 to 40 minutes for assembling the apparatus, the passage of the first test solution through the flow system and the measurement of calibration and reagent blank solutions. Aftcr assembly and calibration of the apparatus, operation is completely automatic apart from the filling of sample-cups. \Vhile the instrument is in automatic operation, the operator can prepare additional sample solutions and convert opt ical-densi ty readings to percentage concentration. Recent developmtmts in tlw -4utoAnalyzer system have now made it possible t o perlorm multiplc analytical determinations. One sampler unit is couplcd to a series of separate flow systems, each with its own colorirneter and chart recorder. ‘llus, if a combined system of this type were used, it would be possiblc to determine the four slag constituents simul- taneousIy with a further saving in time. The helpful co-operation and advice of Dr. J. Marten, Chief Chemist of Technicon Instru- ments Ltd., is gratefully acknowledged. We also thank Mr. P. Haycox for the manual test results reported here. REFEREXCES 1. 2. 3. Fcrrari, A., Russo-Alcsi, F. M., and Kellcy, J. )I., Anal. Chem., 1959, 31, 1710. 4. 5. I€ill, L. T., Anal. Clzenz., 1959, 31, 429. Shaw, IV. H. C , and Fortune, I., Analyst, 1962, 87, 187. Stanton, R. E., and McThnald, A . J., Chew. d> I n d . , 1961, 1406. Scholes, 1’. H., Speight, I<., and I,oxlcy, R*, HTSRA Restricted Heport hiI‘;~I1~246,’63. Keceived Febritwy 181h, 1063

ISSN:0003-2654    

DOI:10.1039/AN9638800702

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

September, 19631 CROMPTON AND REID : ANALYSIS OF ALUMINIUM ALKYLS 713 Analysis of Aluminium Alkyls BY T. R. CROMPTOPU’ AND V. W. REID (Shell Chemical Co. Ltd., CarrixgtoPz Research Laboratories, Urmston, Maxchester) Organo-aluminium compounds, containing alkyl groups between methyl and butyl, are now used extensively as catalysts in processes for the manufac- ture of polyethylene and polypropylene, and are also of interest as intermediates in the manufacture of other chemicals. Trialkylaluminium compounds, dialkylaluminium chlorides or dialkylaluminium alkoxides containing these alkyl groups are all used as polymerisation catalysts. As well as alkyl groups these compounds frequently contain different proportions of aluminium-bound hydride groups produced during manufacture. Procedures are described for determining such hydride and alkyl groups in organo-aluminium compounds.BONITZ~ and Ziegler et aZ.2 have developed gasometric procedures €or the analysis of the lower aluminium alkyls, such as triethyl-, tripropyl- or tributylaluminium. In these proce- dures a known weight of sample reacts at a low temperature with an alcohol of high boiling- point, such as 2-ethylhexanol, under an atmosphere of nitrogen or helium. On alcoholysis each alkyl group liberates one mole of gaseous alkane and each hydride group liberates one mole of hydrogen. ......................... >A1 CnH2.n+l + H 0CHz - CH.CH,.CH, --+ >Al-OCH,-CH.CH,.CH, + CnH2n+2 I ......................... i ! C2H5 C2H5 >A]: H + H / OCH, - CH.CH,.CH, >Al-OCH,-CH.CH,.CH, + H2 .................1 C2H5 CZH, The alkyl and hydride contents of the samples are then calculated from the amount of the gas evolved and its composition. Ziegler et aL2 employed mass spectrometric and other methods of gas analysis for determining the composition of the gas mixture evolved. Ziegler ef aZ,2 state that recoveries of gas by their procedure were lower than expected from the composition of the samples analysed. They attributed low gas yields to a partial dissolution of the evolved paraffin - hydrogen mixture in the 2-ethylhexanol reagent. We analysed several carefully purified samples of triethylaluminium and tri-isobutylaluminiurn and confirmed that recoveries of gas were lower than expected when 2-ethylhexanol or n- hexanol was used as the alcoholysis reagent. From the results obtained in this work we considered it probable, however, that the low recoveries of gas obtained were due to incomplete reaction of alkyl and hydride groups with the alcoholic reagent, rather than dissolution of gas in the reagent.Thus, appreciably higher gas yields were obtained when sample decomposition was effected with a mixture of n-hexanol and monoethylene glycol or a mixture of water and monoethylene glycol than when 2-ethylhexanol was used alone. We therefore studied the reaction of lower alkyl groups (up to butyl) and hydride groups with a range of hydroxylic compounds (alcohols, glycol, water) in order to discover suitable reagents for the quantitative decomposition of each type of organo-aluminium compound. Gas - liquid chromatography offered a simple and rapid method €or analysing the gaseous mixture of alkane, hydrogen and nitrogen withdrawn from the gasornetric apparatus after decomposition of the sample.We used this method of analysis throughout the investigation. Work carried out to permit selection of suitable reagents for the determination of alkyl and hydride groups in trimethylaluminium, triethylaluminium, tri-n-propylaluminium, tri- isobutylaluminium and some of their chloro- and alkoxide derivatives is described below. A description is also given of the apparatus and procedure used in carrying out the analysis.714 CROMPTOS AND REID: XXS’XLYSIS OF -4LUMIYICM ALKYLS [RndySt, VOl, 88 SELECTIOS OF HYDROXYLIC REAGEKT hfETHYLALUMIXIUM ASD ETHYLALUMIKIC‘M COMPOUKDS- I t has been stated already that incomplete decomposition of the aIkyl and hydride groups in triethylaluminium samples appeared to occur on reaction with 2-ethylhexanol or n-hexanol. Alternative reagents that would react quantitativdy and smoothly with tri- ethylaluminium were therefore sought.We found that higher gas yields wcrc obtained when a 4 -1- 1 mixture of n-hexanol and monoethylene glycol was used for decomposing triethylaluminium instead of anhydrous n-hexanol alone. I t was decided, therefore, to see if a still higher yield of gas was obtained when an aqueous reagent was used for decomposing the sample. I t is not practicable to add water directly to triethylaluminium, dicthylaluminium chloride or diethylaluminium ethoxide, as the ensuing reaction is extremely vigorous, even when carried out at -70” C.Also, we showed that an undcsirablc “fissioning” side reaction, which converts alkyl groups to ethylene and hydrogen instead of ethane, occurs to somc extent when aqueous reagents or aqueous monoethylene glycol reagents are added directly to neat organo-aluminium compounds of low molecular weight. AJo m a E reclction- > AZC,H, + H2O -+ C2H6 -+ >AlOH “Fissioning” YeaciioH- > ALC,H, f H,O -+ CzH4 + H, + >AlOH We found it possible to obtain complete reaction without any “fissioning” side reaction by using a combination of n-hexanol and aqueous reagents. The major decomposition of the alkyl is effected by an initial reaction with n-hexanol. An aqueous solution, con- taining 20 per cent. of sulphuric acid, is then added, and complete reaction occurs without “fissioning.” No ethylene was detected in the gases Liberated in these reactions.Results obtained on typical samples of distilled trimcthylaluminium and triethyl- aluminium are shown in Table I ; it can be seen t h a t yields of gas wcrc good. TABLE I ANALYSIS OF FIIESHLY D I s r I L r x u METHYL ASD ETHYL c o ~ ~ o u s ~ ) s Con tent f ou nd , Content found, -ll(CHJ3 . . . . 99-5 99.2 A1(V2€I5)* . . .. 94.4 9 4 0 0) Trimelhylalzcman iuin - ?& W/U’ Tuw f hylalumin ium- /o W‘/ w XI(CH,),H . . . . Sil Kil Al(C,H,},H . . . . 2.2 2.2 -- .ll(C2H&(C4~iD) . . 2.0 2.1 Total . . . . 99-5 99.2 Total . . . . 98.6 98.9 Hydrolysis of chloro- or alkoxide derivatives results in the formation of hydrochloric acid or the corresponding alcohol ; however, these compounds dissolve in the rcagent, and we have shown that they do not interfere in the determination of alkyl groups.PROPYLALGMISIUM COMI’OCKDS- Yropylaluminium compounds are less reactive than are the ethyl compounds, and it was considered possible that thc “fissioning” side reaction might not occur when a rcagcnt containing water was added directly to the propyl compounds to which liexanol had not been added. The first reagent tried was a 3 -1- 7 v/v mixture of monoethylene glycol and water; it was added to a sample of di-n-propylalurninium isopropoxide coolcd to -30” C. Analysis of the gas obtained, however, showed the presence of considerable amounts of hydrogen and propylene, indicating that extensive “fissioning” of the propy1 groups to hydrogen and propylene had occurred under these conditions.>Al-CH2-CHz-CH, -t HZO -+ >AlOH + CH,-CI-.=CH, -t- H,. We next tried a 3 + 7 v/v mixture of monoethylene glycol and 20 per cent. aqueous sulphuric acid. Rather surprisingly, propylene W~S completely absent from the gas generated in these experiments, indicating that “fissioning” did not occur when the aqueous reagent used was acidic. This decomposition procedure was then applied to the analysis of a sample of di-n-propylaluminium isopropoxide and also to a sample of tri-n-propylaluminium. TheSeptember, 19631 CROMPTON AND REID : ANALYSIS OF ALUMINIUM ALKYLS 715 results obtained in the determination of propyl and hydride groups, together with separate determination of aluminium and n-propoxide3 are shown in Table 11.TABLE I1 ANALYSIS OF PROPYLALUMINIUM COMPOUNDS Sum of Departure of n-Pro- Alu- Total subscripts sum from n-Propyl Hydride poxide minium of com- i n 3.00 (i.e., Sample found, found, found, found, ponents, Empirical empirical stoicheiometric Di-n-propylaluminitm isojwopoxide- No. %w/w %w/w %w/w %w/w % formula formula AIR,)J % 1 48.8 (0.01 30.1 14.7 93.6 Al~.OOPr~.o*(OPr)o.~~ 3-02 +1 2 43-3 t o - 0 1 38.1 14.8 96.2 All.,,oPrl.84(OPr)l.18 3.02 + I Tri-n-pyopylal~~ini~~- 3 78.7 0.01 0-7 16.7 96.1 All.00Pr2.y6 4 78-0 0.01 0.6 16-5 95.1 All.o,,Pr2.97 H,. oz(OPr) 0.02 3-00 0 H,. 01 (OW,. 02 3.00 0 The total of the constituents determined was about 95 per cent. As it was known that these samples contained a few per cent.of hydrocarbon solvent, it was not possible to check the recovery of gas directly. Aluminium is always tervalent, however, and a test of the reliability of the analytical results is obtained when the calculated valencies, based on the analytical results, are compared with the value of 3. Values for the valency of aluminium calculated in this way are shown in Table 11; it can be seen that they are within 1 per cent. of the theoretical value of 3. A 3 + 7 v/v mixture of monoethylene glycol and 20 per cent. aqueous sulphuric acid was therefore adopted as a suitable reagent for the direct decomposition of propylaluminiurn compounds. BUTYLALWMINXUM COMPOUNDS- The two types of reagent developed for the ethylaluminium and the propylaluminium compounds were then applied to the analysis of butylaluminium cornpounds.The acidic glycol reagent was found to produce some “fissioning” of the butyl groups, which was over- come when the decomposition was carried out at -65” C, and recoveries of gas were slightly higher than with the n-hexanol- aqueous sulphuric acid reagent. The results obtained on a sample of tri-isobutylaluminium are shown in Table 111. In this sample the isobutoxide group content was determined by the method previously des- cribed.3 It can be seen that the total of the components determined is close to 100 per cent., indicating good recovery of gas in the decomposition procedure. The recoveries of gas were therefore good. TABLE III ANALYSIS OF TRI-ISOBUTYLALUMINIUM Constituent determined % w/w Content found, .. .. ... . 87.6 87.1 .. . . . . 6.6 6-6 Al(C4Hy)2(OC4€lii . . . . . . 6-0 6-0 Total . . .. . . I . 100.2 99.7 AV4HJ3 AWPY),H LOWER ALKYL AND HYDRIDE GROUPS IN HIGHER ALKYLALUMINIUM COMPOUNDS- It is sometimes necessary to determine alkyl groups, up to butyl, and hydride groups in organo-aluminium compounds containing alkyl groups higher than butyl. Higher molecu- lar-weight organo-aluminium compounds can be prepared by the displacement of butyl groups in tri-isobutylalurninium with the appropriate higher olefin. These higher molecular- weight aluminium alkyls may still contain small amounts of hydride and lower alkyl groups.716 ;Analjvst, Jrol. 88 The aqueous sulpliuric acid - monoetliylene glycol reagent, as uscd for the analysis of butylaluminiums, was found to be suitahle for the analysis of these compounds.These materials were oftcn highly \Gscous, and good r(tcoverj(:s of gas were obtained only wlien good mixing during the reaction was maintained b5- means of magnctic stirring. CROWTON AK;D I ~ U : AE;ALYSIS 01; ALUJII?~IUM XLKYLS Jlrrrroo i;cm I)ECOMPC)SING AIAYLS mn UE’rmbiIs ISG - r m GASES EI-OLI~BD A 1’PA K AT U S - The apparatus, wliiclt is similar to that (1cscribt:d by Zicgler 41 U Z . , ~ is sliow-n in Fig. I . I t consists of a reaction vcssi.1 attached by means of a flexible coupling t o a gas manifold 1 ) s2 ‘J Reaction vessel Gas burette in water jacket Nitrogen 1 I Zinc sulphate filled torpedo V-bore Ball and L stopcock SI socket j o i n t c inlet Reaction vessel head - 0 Three-way stopcock S3 D ’E Nitrogen +3+ inlet torpedo tube Fig.I . I I ydrolysis alcoholysis apparatus system, incorporating a gas burette for volumetric mcasimment of thc gas evolved, arid leads to a gas sampling torpedo for transfer of thc gas to the gas - liqiiid chromatograph for suhse- qucnt mal~~sis. Mercury is used as the confining liquid in both the gas burettta and in the sampling torpedo. The manifold system connects to rz supply of piire nitrogen (white spot), which is dricd by passage through a drying tower packed with Linclc 4A molccular sieve. The reaction vessel connects, lia a stopcock, to a supply of aqucous zinc sulphate solution, which is used for discharging the reaction gases into the sampling torpedo. Tile apparatus should be cleaned and dried thoroughly before each determination, and the cones at G and H and stopcocks S,, S, and S, should be lubricated with silicon(: grcasc!.REAGIXTS- Dccomfmition reagent A-Mix 30 ml of 20 per cent. v/v aqueous sulpliuric acid with 70 ml of rnonoetliylenc glycol. To 100 ml of mixed reagent add 1 ml of a non-ionic surface- active agent, e.g., Xonidet P43, a condensation product of (Iioctyl pl~cnol and etli>.lene oxidc obtainable from the British l h g Houses Limited. Decomfiosition reugent B-11-Hexanol. llecompusiiion reagent C-Aqueous sulphuric acid (20 per cent. v/v) containing 0.05 per cent. of water-soluble methyl orange. SAM PI- I: ?; C- Organo-aluminium compounds rcceived for analysis map contain a hydrocarbon dilnent. Hydrocarbon diliicnts that Eloil bdow 180” C can bc removed by vacuum distillation at aSeptember, 19831 CROMPTOPU' AND REID : ANALYSIS OF ALUMINIUM ALKYLS 717 pressure of 0.1 mm of mercury at a maximum temperature of 65" C.Avoid heating above 65" C, as, at this pressure, many organo-aluminium compounds are somewhat volatile or may decompose above this temperature. Transfer the appropriate weight of sample to the weighed reaction vessel by means of a safety pipette. Purge the reaction vessel and sampling pipette with dry nitrogen during transfer of the sample, as described by Cr~mpton.~ The weight of an organo-aluminium compound required for a determination is that from which approximately 70 rnl of gas at S.T.P. will be evolved. Calculate the weight of neat sample required from- 70 x 1000 M 22400 n x - mg Weight of sample = where M = molecular weight of the compound and PZ = number of alkyl groups per molecule of compound.Samples should be free from metallic sediment, such as free aluminium, as reaction of metal with the acid decomposition reagents would occur with the evolution of hydrogen, which would cause high results for aluminium-bound hydride. Metallic sediments can usually be completely removed from the sample by centrifugation. PROCEDURE FOR PURGING THE APFARATUS- At this stage the reagent side-limb, the reaction vessel and sampling torpedo T, are not connected to the apparatus. Open the gas burette to atmosphere, and raise mercury reservoir R, until the burette is filled with mercury to the barrel of stopcock S,. Connect A to B with stopcock S,, and purge with nitrogen through inlet I.Connect B and C, and allow the nitrogen pressure to depress the mercury until about 15 ml of nitrogen have entered the gas burette. Lower reservoir R, until the mercury levels in the burette and reservoir are the s m e . Connect D to F with stopcock S,, and then slowly raise reservoir R, until 5 rnl of nitrogen remain in the burette. Con- tinue purging with nitrogen through inlet I. DECOMPOSITION OF METHYL- AND ETHYLALUMINIUM- By pipette place 1.5 ml of reagent C in the reagent side-limb, ensuring that no drops of this aqueous reagent remain above the liquid level. Then place from a pipette 1-5 ml of immiscible reagent B in the side-limb on top of the aqueous phase. This reagent, n-hexanol, will float on top of the aqueous phase, and no globules of aqueous reagent should be present in the upper hexanol layer.Connect the side-limb to the reaction vessel head, and connect springs across the glass lugs. Nitrogen now leaves via socket H. Purge with nitrogen the interior of a clean oven-dried reaction vessel; use a glass inlet tube. Remove the inlet, and apply a gentle purge to the side-arm of the loosely stoppered reaction vessel. Discontinue the purge with nitrogen, close stopcock S,, and closely stopper the vessel. Weigh the reaction vessel accurately. Purge the reaction vessel gently with nitrogen, remove the stopper, and, by pipette, place the required volume of sample in the reaction vessel. Remove the nitrogen supply line from inlet I, and immediately transfer it t o the side-arm of the reaction vessel.Open stopcock S, on the reaction vessel, and then remove the stopper. Clamp the reaction vessel to the head of the apparatus, and close stopcock S,. Fasten the springs connecting the reaction vessel to the head. Connect D to E with stopcock S,, and rotate stopcock 5, through one complete revolution to bring the internal pressure of the system to atmospheric pressure; connect A to C with stopcock S,. Adjust the height of mercury reservoir R, until the mercury level in both limbs of the U-levelling tube are the same, with stopcock S, open. Measure the volume of nitrogen in the burette, and record the atmospheric pressure and ambient temperature. The temperature of the water-jacket surrounding the gas burette should not differ by more than lo C from room temperature when gas volumes are being read.Immerse the lower bulb of the reaction vessel in a cooling bath maintained at -60" C. Leave for 5 minutes to cool, and level off the mercury in the gas burette and reservoir R, when necessary. Cut off the nitrogen pressure by connecting E to F with stopcock S,. Connect A to B with stopcock S, and then D to E with stopcock S,. Stopper the vessel, close stopcock S,, and re-weigh.7 18 CROMIPTOS ASD REID: AW.-\LYSIS OF . * l L L M I S I ~ M -4LKYLS ‘APl@VSt, l’-Ol, 88 Slowly rotate the reagent side-limb until about half of the n-hexanol layer has flowed into the reaction vessel; ensure that none of the aqueous phase enters the reaction vessel a t this stage. As generation of gas proceeds, equalise the mercur?. levels in the burettc and reservoir R,.iVhen evolution of gas appears to be complete, remove the cooling bath, and allow the reaction vessel to attain room temperature. Immerse the reaction vessel in a cold-water bath, and heat up to 50’ Add the aqueous reagent C by again slowly rotating the side-limb. k-urther evolution of gas occurs when this reagent is added, so continuously equalise the mercury levels in the gas burette and reservoir R,. Heat the water-bath sur- rounding the reaction vessel to the boiling-point, and maintain at the boil for 50 minutes. Remove the water-bath, and again equalise the mercury levels as the gas contracts. Adjust the reservoir K, u n t i l the mercury le\rels in both limbs of the U levelling tube are the same, stopcock S, being open. Measure the volume of gas in the burette, and record thc atmospheric pressure and room temperature.Displace all the gas in the reaction vessel, etc., by first connecting mercury-filled sampling torpedo T,, fitted with a reservoir, R,, to inlet I; on stopcock S,. (:onnect E to I!, and raise the mercury level to the barrel of stopcock S,. Sow connect L) to 1; with stopcock S, and A to B with stopcock S,. Close stopcock S, on the U levelling tube. NOW displace reaction gases by attaching a source of saturated zinc sulphate, supplied from a torpedo, T,, to stopcock S,. Open stopcocks S,, 1, and 11, and apply a gentle pressure of nitrogen at stopcock M. The zinc sulphate solution will now expel the gas contained in the reaction vessel and reagent side-limb. Ml’hen the level of the zinc sulphate soIution reaches A on stopcock S, close stopcock S,, and connect B to C.Raise reservoir R, to displace the gas from the burette into torpedo T2. Keep the mercury in torpedo T, and reservoir R, at the same level during transfer of the gas. Close stopcocks j and K, and disconnect tlie torpedo from the manifold. Shake the torpedo to mix the gas sample, and anaI?.se the gas by gas - liquid chromatograph!-. AIlow the system to come to equilibrium overnight. I fydrogen, methane and C2 hydrocarbons C, and C , hydrocarbons Column . . .. . . . . 3 feet 7 &-inch i.d. glass Column packing . . . . . . I>a\.isoIi silica gel 912, 28- to GO-mesh Column and detector tc>mpcraturt* 3oj (: Sample size . . I . . . 5 ml Ihtector . . .. .. .. lia t haro me t tar Bridge current .. . I .. 150 m.4 Carrier gas .. .. .. Sitrogcn (whitc spot) Carrier gas flow . . . . .. 2 litrcs pcir hour 18 fcct x {-inch stainless stct-1 I2 fect o f dimcth~tsnlp~iolanc -:- 6 fcct of dinoiiyl phthalatr (t)otli 20 pcr cent. on 44- to 60-mcsh Celi t c b ) 30” c 0.25 ml Kat haromc tcc 150 m.1 H y d rogcn 3 litrrs pcr hour F)ECONPOSITIOS OF I’KOPYL- ASL) H ~ T I ’ I . . - 2 L ~ ’ ~ I I S I ~ ~ l - - Carry out the decomposition of propyl- and but>-laluminium compounds in tlie same way as described for the methyl- and ethylaluminiums, with tlie csception of the reagent composition. Add this reagent, as before, by slowly rotating the reagent side-limb to give dropwise addition. During the warm-up period from -60’ C to room temperature, however, the reaction ma!.become vigorous. ShouId this happen, control it by tcmporarily returning the react ion vessel to the cooling bath. After reaction is complete, heat the water-bath surrounding the reaction vessel to the boiling-point, cool, and expel the gas into the mercury-filled torpedo, TP, as previousll- described. Use only one reagent, 2 4 m l of reagent A, in the reagent side-limb.September, 19631 ANALYSIS OF EVOLVED GAS- The gas-chromatographic analysis of the evolved gas is carried out in two stages. In the first stage hydrogen, methane and C, hydrocarbons are determined on a column of silica gel with nitrogen as carrier gas. In the second stage a combination of dimethylsulpholane and dinonyl phthalate columns is used for determining the C, and C, hydrocarbons.Suitable gas - chromatographic conditions are shown in Table IV; any conventional gas Chromatograph adapted for the analysis of gases can be used. Calibrate the apparatus for hydrogen, methane and C, hydrocarbons by analysing accurately prepared mixtures of each gas with nitrogen. Determine the peak areas by elec- trical integration or from the product peak height x width at half peak height x attenuation factor. The preparation of these calibration mixtures can be simplified by using a Wosthoff gas-blending pump (type A18/2a, obtainable from H. Wosthoff, O.H.G., Apparatbau, Bochum, Germany). Calibrate the gas - liquid chromatograph in terms of peak area against volume per cent. for hydrogen, methane and C, hydrocarbons ; check this calibration periodically.For the C, and C, hydrocarbons, use the gas-mixing pump to prepare mixtures of the gases with hydrogen, and determine the response factors for each gas relative to n-butane. This should be done accurately once every 4 months. Carry out a daily calibration check on a mixture of n-butane and hydrogen, and determine the area responses; the relative area response for each gas can then be determined for that day from the relative response factors. Calculate the percentage, by volume, of each gas from the product peak area x relative area response factor. Calculate the total gas composition by normalising the individual percentage volumes so that their sum is 100 per cent. CROMPTON AND REID : ANALYSIS OF ALUMINIUM ALKYLS 719 CALCULATION OF RESULTS CA4LCULATION OF GAS YIELD- The volume of gas (V ml) , corrected to S.T.P., generated during the reaction is given by- where D = CALCULATION volume (in rnillilitres) of dead space in apparatus, i.e., the combined volume of the reaction vessel, side-limb and reaction vessel head. Determine D by weighing the amount of mercury needed to fill these three vessels. atmospheric pressure in mm of mercury when measuring initial and final gas volumes, respectively . ambient temperatures in degrees absolute when measuring initial and final gas volumes, respectively. volume of gas in millilitres in burette before and after evolution of gas, respectively. a small correction term in rnm of mercury allowing for the saturation vapour pressure exerted by the aqueous sulphuric acid - monoethylene glycol Reagent A.This correction term is sufficiently small to be ignored when the two phase n-hexanol- aqueous sulphuric acid reagent is used. The values oi P at different temperatures are- Temperature, "C . . .. 15 20 25 30 P, mm of mercury . . .. 6.9 9.4 12-7 16-9 OF ALKYL AND HYDRIDE CONTENTS- These are given by- a x v x 1.~08 ?/ w/w Hydride in sample = VC' x 22,400 b x V x M Alkyl in sample = w c %w/w720 CROMIYTOS AS;U IZElI); AXALYSIS OF .II.UlrlISIUM -4I.E;YLS 1,4?ZUl?lSf, v@1, 88 where a = percentage by volume of hydrogvn in generated gas. b = percentage by volume of alkane in generated gas. 31 = group weight of alkyl group being determined. T.’ -= volume in miIlilitres of S.T.P. of gas generated in anal!+. I$’ : weight in grams of organo-aluminium samplc C 7 - volume in millilitres per mole (at S.T.P.), rclationship of the alkane gas in\-olved, e . g . , for methane and ethane C = 22,400, for propane C = 21,970 and for butane c‘ = 21,830. cA1,CVLXTION OF (~OMI’OCSI) COMPOSITION OF SAM 1’1-E- The method of calculating the compound cornyosition of an organo-aluminium samplc from its determined hydridc, alkyl and alk0xide3 contents is shown below. A, B, C and 1 ) are the percentage w,/w contents of ethjd, h)dride, ethoside and h t y l groups determined, rcspect ively. Then- H x (mol. wt. of Xl(C H ) H -:= 86.1) *, , o \V/W H r ) , H =. - -2 2 3.2 (atomic wt. of hg’drogen = 1.008) (where group wkight of rtthyl groups = 2946 and molecular wciglit oi \Vc thank the Directors of Shell Chemical Companj- T-td. for permission to publish this triethylaluminium r:~ 114.15). pa per. RE FElZESCE S 1, Honitz, K., Chem. B e y . . 1955, 88, 542. 2. 3. Zicgler, I<., Gcllert, H. G., Martin, H., Sagcl, K., and S(.hntitdcr, J., .-J w)icrlen, 1954, 589, !>I. C‘rompton, T. .K., .-I ira!+ysl, I W l , 86, 652. IXcceived .-?firil kith, 19tj.3

ISSN:0003-2654    

DOI:10.1039/AN9638800713

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

September, 19631 BOWEN AND CAWSE 72 1 The Deterniina tion of Selenium in Biological Material by Radioactivation BY H. J. M. BOWEN AND P. A. CAWSE ( U. K . Atomic Energy Authority, Wantage Reseavch Laboratory, Wantage, Berks.) Neutron-activation analysis has been applied to the determination of selenium in biological material. Samples were activated in a flux of 1012 neutrons per sq. cm per second for 1.8 minutes, and a rapid chemical separation was developed for selenium-81, Under these conditions the practical limit of sensitivity was found to be 5 X g. Decontamination from arsenic, bromine, manganese, sodium and zinc was tested and found to be satisfactory. Selenium contents of fertilisers, tomato tissue and human blood were measured by the procedure. OUR knowledge of the behaviour of selenium in plants and animals has undoubtedly been hampered by the absence of sufficiently sensitive methods for determining the element. L'ntil recently the only microanalytical method was that described by Robinson et aZ.,l which measures the red colour of elementary selenium after a chemical separation.With the discovery that trace amounts of selenium are important and possibly essential to mammaIs233 there has been a revival of interest in other methods for determining the element. I t appears that levels in normal biological tissue range from 0.005 to 0.5 pg per g so that only ultra- micromethods are suitable. Such methods include the combined spectrophotornetric and isotope-dilution method proposed by Kelleher and Johnson4 and Watkinson's fluorimetric m e t h ~ d , ~ but most colorimetric methods1 8 y 7 and X-ray fluorescence methodss are not yet sufficiently sensitive, To avoid errors due to reagent contamination, several workers have used neutron-activation techniques, and one such technique is described below.Table I lists the radionuclides of selenium produced by thermal neutron activation and some of their relevant properties. I t can be seen that only three of the nuclides can be made in high specific activities, and of these the best (selenium-77m) is exceedingly short-lived, with a half-life of only 17.5 seconds. Nevertheless, it has been used by several workersgJoJ1J2 for determining the element, with the use of an activation period of a few seconds and a rnulti- channel analyser focussed on the O.16-MeV gamma-ray peak.This can be a satisfactory method provided the half-life of the peak is measured, since many other short-lived nuclides have a gamma-ray of about the same energy. TABLE I: RADIONUCLIDES PRODUCED BY A THERMAL NEUTRON FLUX OF 1012 neutrons per sq. crn per second ON SELENIUM Activity of selenium after activation for mC per g Radionuclide one half-life, Selenium-'75 . . .. 25 Selenium-77m . , . . 97 Selenium-79m . . ._ 2.9 Selenium-8lm . . ,. 1.5 Selenium-81 . . .. 25 Selenium-83m . . - . 0.46 Selenium-83 . . .. 0.04 Half-lif t 120 days 17-5 seconds 3-9 minutes 57 minutes 18.6 minutes 69 seconds 25 minutes Maximum Gamma beta energy, energies, MeV MeV - 0.1 6 - 0.096 - 0.103 1 *60 none 1.01 and 2-02 3.40 1-70 0.36 and 2.34 0.27 and 0.14 - Most workers have used selenium-75 to determine the element.13 to l9 This nuclide has a convenient gamma-ray for counting, and its half-life is more than adequate to allow complete chemical separation from other activities.Its main disadvantage is the long activation time needed. Most analysts have activated their samples for only 7 to 14 days, which gives them only 5 to 10 per cent. of the specific activity quoted in Table I. This results in an equivalent loss of sensitivity, which is undesirable for biological samples.722 BOWEX AND CAWSE: DETEKMIN-4TION OF SELESIL‘M Analyst, VOl. 88 The use of shorter lived selenium isotopes for activation analysis has also been suggcstcd. The group at the University of Michigan2*J1 have used a rapid chemical separation method coupled with gamma-ray spectrometry for detecting 3.9-rninute selenium-79m, hut the limit of detection is only 10 pg.Yajirna el al.22 have employed 57-minute selenium-81m for deter- mining selenium in tellurium, but the sensitivity is also poor. In this work we have used 18-minute selenium-81, which has the advantage of giving a theoretical sensitivity equal to that obtainable by using selenium-75 and requiring a much shorter activation time. Selenium-81 is virtually a pure beta-emitter, which may explain why it has not previously bccn considered for activation work, though its use is mentioned by Leddicotte.17 ACTIVATION- Samples were collccted under the cleanest possible conditions and scaled into clean polythene ampules. Standards were prepared by first dissolving selenium dioxide in water spotted on to weighed 1-cm x I-cm squares of \l’hatman No.541 filter-paper, which wore then re-weighed. The selenium content of these filter-papers was found to be <@01 pg per square. For each run, four samples and two standards were packed into a plastic “Rabbit” and activated for 18 minutes in a fliix of about 10I2 neutrons per sq. cm pcr second in the Harwell reactor REPO. 31 ETHO D \Then dry, the standards were scaled in small polythcnc bags. 12EAGEXTS- A11 reagents were of recognised analytical grade. Asking m i x t w e - A ( I + 1) mixture, by volume, of 16 s nitric acid and 70 per cent. Nitric acid, 2 s. Hydrochloric acid, 12 S . Hydrobromic acid, 48 per cenl. w / v . Hydrogen peroxide, 30 per cent. 7@/c. Teepol solution, 1 per cent.v/xx. A cehme. Sulphur dioxide, liquefied. Xitrogen, 99.9 per cant. pure. Selenium carrier solution-Prepared by dissolving 2-8 I06 g of sclen iurn dioxide in distilled perchloric acid. water and diluting to 100 ml. 1 ml r 20 mg of selenium. Arsenic carrier solution-Sodium arsenate solution, 10 per cent. w/v. -k€aqyanese carrier solzttion-Manganese nitrate solution, 5O per cent, w/v. Pkosfihorus carrier soZzcfio~+Ammonium dihydrogen orthophosphate solution, 10 per cent. w/v. Tellzirium carrier solzition-Sodium tellurate, 10 per cent. w/v in N h>.drochloric acicl, CHEMICAL SEYAHATIOS OF SELESIL-M- Activated samples containing organic matter were placed in 150-rnl beakers togethcr with 10 ml of ashing mixture, 1 ml of selenium carrier solution and 2 drops each of arsenic, manganese, phosphorus and tellurium carrier solutions.They were boiled until the organic matter was destroyed and all nitric acid had volatilised, so that white fumes of perchloric acid were visible. They were then cooled and transferred to 50-ml round-bottomed flasks containing 5 ml of hydrochloric acid and -5 ml of hydrobromic acid. Ilistillation lleadsa were fitted immediately, and a current of nitrogen was passed through the flasks for approsi- matel>- 4 minutes while they were heated with a small bunsen burner. The distillates were collected in 5O-ml centrifuge tubes containing 5 ml of hydrochloric acid, 10 ml of water, 0-1 ml of Teepol solution and 2 drops of manganese carrier solution. Then sulphur dioxide was immediately passed into the centrifuge tubes through a glass capillary until a dark red precipitate of selenium was seen.This precipitate was separated by centrifugation, and dissdvcd in about 0.25 ml of nitric acid. Hjdrogen peroxide (0.1 ml), 10 ml of hot water, and 5 ml of hydrochloric acid were added, and sulphur dioxide was again passed through the hot solution to precipitate selenium. This selenium was separated by centrifugation, washed with water and acetone and finally transferred to a weighed aluminium counting tray as aSeptember, 19631 IN BIOLOGICAL MATERIAL BY RADIOACTIVATION 723 slurry with acetone. When dry, it was counted with the minimum of delay and was sub- sequently weighed. The mean chemical yield was 70 per cent., and most of the loss was mechanical.In practice it was possible to begin counting two half-lives (36 minutes) after removal of six samples from the reactor, with two analysts performing the chemical manipulations. Table I1 lists the time taken by these operations. TABLE 11 TIME REQUIRED FOR UNIT PROCESSES I N THE CHEMICAL SEPARATION Operation Opening samples . . .. . . * * .. 1 . Xshing . . . . .. . . .. .. .. First precipitation by sulphur dioxide . . . . .. Centrifugation . . .. . . .. .. .. Dissolution of selenium . . , . - . .. .. Second precipitation by sulphur dioxide* .. .. Centriiugation and washing . . . . .. . * Transfer to counting tray . . . . .. .. .. Drying . . .. .. . . .. ,. .. Distillation . . .. . . . . .. . . .. Time required, minutes 1 3 4 1 1 0-5 1 4 0.5 2 * This was the least reproducible step, and sometimes took up to 3 minutes.TREATMENT OF STANDARDS- The filter-paper standards were boiled for 4 minutes with 5ml of hydrochloric acid, 10 ml of water, 0-1 ml of Teepol solution and 2 drops of manganese carrier solution. (Tracer experiments with selenium-75 showed that 99.7 per cent. of the selenium activity was recovered in this step.) The solutions were decanted into 40-rnl centrifuge tubes, and sulphur dioxide was passed through the solutions. The selenium precipitates were washed and plated out as described above for the samples. 2oo* 3 1500 Selenium added, pg Fig. 1. Recovery of known amounts of selenium spotted on to: curve A, filter-paper ; curve B, tomato seeds724 BOWEN .iND CAWSIX 1)ETIHSII?I’.lTIOK OF SELESIUhI , .+! IlalL’St, vol.%$ DETEKMINATIOS oi: KAL>IOACTIVITY-- The beta-activity of the precipitates was counted with a SR2 end-window Geiger counter of approximately 44 per cent. efficiency. Radiochemical purity was cliecked by comparing the decay curves of samples and standards over several half-lives. T h decay curve of neutron-activated selenium is complex, sincc after the selenium-81 has largelj. decayed there are still counts from selenium-83, selenium-81m and selenium-$5, which have longer half-lives (see Table I). Gamma spectrometry was also used for dctecting possihlc impurities in precipitates having high count-rates. DISCUSSIOK OF THE METHOD Fig. 1 shows a recovery curve for known amounts of selenium spotted on to (a) filtcr- paper and ( b ) tomato seeds. Each point represents tht: mean o f four determinations agreeing to 2 5 per cent. It can be seen that thc count rate is dircctly proportional to seleniirrn content in the range 0.5 to 5 p g of selenium.The theoretical sensitivity of thc mcthod is dctemincd both b3- the flus available and bv the background of the counter used. In our experiments, assuming a chemical separation time of 40 minutes and a background of 30 counts per minute, 5 p..g of selenium in the sample would double this background. This amount can hc takcn as the practical limit of scnsitivit>+ when REP0 is used. I’ESTI S (; TH I< KA1)TOCH EM I CAL PROC ED C I< ES- The chemical procedures described ahovc were tested with portions of radiochemically pure arsenic-76, bromine-82, rnanganese-5ti, selenium-75, sodium-24 and zinc-65.The por- centages of these nuclides contaminating the final precipitate and other fractions wore dcter- mined by scintillation counting, and the results shown in Table TI1 were obtained. As might be expected, the volatile clement brominc is largely eliminated during the initial asliing step, whereas the four metals remain in the residue after distillation. The amounts of arsenic, manganese, sodium and zinc found in the distihte must be a measurc of the amount of spray carried over in our distillation procedurv. ,4CCVlCACY OF THE METHOD- Errors may occur in the activation process, in the chemical processes, or during counting. The flux gradient in HEPO is less than 2 per cent. over the Rabbit volume. The high flus o f fast neutrons in BEY0 could give rise to the production of selenium-81 by the reactions- (i) 81Br (TZ,$)~*SC.(ii) K r (%,a) %e. (iii) (n ,f) 81Se. Little is known about reaction ( z ) , though the cross-section for 14-Mel’ neutrons has been calculated to be 0423 barns.= By activating ammonium bromide in BEPO we have shown that 1 g of bromine gives an apparent content of 22 pg of selenium. This is an upper limit since the ammonium bromide may have contained some selenium as an impurity. Since bromine is present in biological tissue in amounts ranging from 1 to 10 pg per g, the ( I Z , ~ ) reaction (i) should not give rise to a significant error. Reaction ( z z ) can be neglected in biological material becausc of the extreme rarity of krypton and its low cross-section.For cxamplo, the normal concentration of krypton In blood is estimated to be only 04002 pg per ml, which is far bclow thc conccntration o f selenium.25 Reaction (izz) can also be neglected since the fission yield of selenium-81 is only 0.13 per cent., and uranium is present in cxtrcmcly small amounts in biological material, e.g., 0.014 pg per rnl of whole As regards the chemical stages, the elements that are likely to distil under the conditions described here include antimony, arsenic, bromine, chlorinc, germanium, iodinc, mercury, tellurium and tin. Loddicotte17 obtained excellent decontamination from antimony, arsenic, sodium, tclIurium and tin, by using a similar but slightly simpler separation. The nuclidrs of arsenic, germanium, mercury and tin havc vcry differmt half-lives from that of selenium-81 , but interference could be serious from 21-minute antimony-124, 25-minute tellurium-131, 37-minute chlorine- 138 or 18-minutc bromine-80. In biological material we regardSeptember, 19631 I N BIOLOGICAL MATERIAL BY RA4DIOACTIVATION 725 bromine-80 as the most serious impurity since it has the same half-life as selenium-81, and bromine is much more abundant than either antimony or tellurium.Fortunately, decon- tamination from the halogens is reasonably good, as shown in Table 111. The only long- lived impurity we have been able to find in our samples was manganese-56. Because of the high cross-section of manganese, this nuclide constitutes a major source of radioactivity in activated vegetable material, and even after a distillation step traces may contaminate the final selenium precipitate.If hydrogen peroxide is added when the selenium precipitate is dissolved in nitric acid, we find that decontamination from manganese is markedly improved; acidified hydrogen peroxide is a well-known solvent for manganese dioxide. TABLE III PERCENTAGES OF SIX ELEMENTS FOUND IN RADIOCHEMICAL FRACTIONS Arsenic Bromine Manganese Sodium Zinc Selenium Fraction found, found, found, found, found, found, % Yo /Q 94 % % O / Valatilised during ashing . . 0 95-42 0 0 0 0-25 DistiIlation residue . . 99.30 0-04 99-88 99.74 99.78 0-95 First selenium supernate . . 0.69 4-54 0.12 0.26 0.14 0.55 Second selenium supernate 0.019 0.02 3 0.00005 0.0015 < 0.05 0-55 Final selenium precipitate 0.00037 0.0027 0.00019 0*0000081 < 0-03 97-70 TABLE I V SELENIUM CONTENT OF VARIOUS MATERIALS Sample Selenium content N.African phosphate rock. . . . .. 10.6 pg per g Superphosphate . . . . . . . . 3.77 pg per g Tomato leaf . . . . .. . . . . 0-088 p g per g Tomato seed . . . . . . .. 0.025 pg per g Tomato leaf* . . .. . . .. 1.25 pg per g Tomato fruit* . . .. , . .. 0.24 pg per g Human blood, mean .. . . I . 0.32 pg per ml Human blood, range of 8 samples * Grown in soil rich in selenium. . . 0-26 t o 0.37 pg per ml A small error is introduced in weighing the final precipitate, since this only weighs about 15 mg and ordinary balances are reliable to k0-2rng. When counting long-lived nuclides it is possible to obtain counting accuracy of within 1 per cent., by registering 10,000 counts.This may not be possible for rapidly decaying nuclides; a count rate of 10,000 counts per minute is given from samples containing about 1-5 pg of selenium, and many biological samples contain much less than this. Care must also be taken to record the exact time at which a count is begun, since an error of 1 minute corres- ponds to a 3.5 per cent. correction for radioactivity decay. RESULTS Some results obtained by this method of analysis are shown in Table IV. The results for plant tissues and blood are of the same order of magnitude as those determined by earlier worker^.^ The relatively high figures for the widely used fertilisers, phosphate rock and superphosphate, are a little disquieting in view of the toxicity of the element. Tag- werker3 states that any soil containing more than 0-5 pg per g of selenium is potentially dangerous and that chronic selenium toxicity may be caused by rations containing more than 5 pg per g of selenium.We thank Miss M. J. Dick for carrying out most of the counting involved in this work. REFERENCES 1. 2. 3. 4. Robinson, W. O., Dudley, H. C., Williams, K. T., and Byers, H. G., I n d . Eng. Chsm., Anal. Ed., Schwarz, K., and Foltz, C. M., J . Awzer. Chew. SOC., 1957, 79, 3292. Tagwerker, F., J . Agric. Vet. Clzem., 1960, 1, 23 and 78. Kelleher, W. J., and Johnson, M. J., Anal. Chem., 1961, 33, 1429. 1934, 6, 274.726 5. ti. 7. S. 9. 10. 11. 12. 13. 14. 3 5 . 16. 17. 18. 19. 30. 21. 22. 23. 24. 25. \Vatkinson, J . H., Ibid., 1960, 32, 981. Uonhorst, C. W., and Mattice, J.J., Ibid., 1959, 31, 2106. Ray, E. M., Goodyear Atomic Company Kcport GAT-372, 1961. HantIley, R., .4nal. Chem., 1060, 32, 1719. Okada, M., ,\7aluve, 1960, 187, 594. McConnell, 1.;. P., in “Modern Trends in Activation Analysis,” College Station, Tcxas, 1961, .\ndcrs, 0. U., :inul. Chew., 1961, 33, 1706. Gibbons, D., and Simpson, H., paper SM32/34 presented at the Conference on Short-Iivctl Karlio- Fineman, I., Ljunggren, K., Forsbcrg, H. G., and Erwali, L. G,, In!. -1. Appl. Rud. Isotopes. 1959, Erion, W. E., Mott, IV. I;., and Shedlovsky, J. P., Trails. L 4 m e ~ . Xucl. Soc., 1900, 3, S o . 1 , 253. Schintlewulf, U., Geochim. Cosmochitn. A cia, 19fi0, 19, 134. Gaittct, J . , Ann. Chim., Paris, 1960, 77, 1219. Iddicottc, G. \V., National Academy of Sciences Report NAS-NS 3030, 196 1 . Koch, R . C., and Koesmer, J . , J . Food Sci., 1962, 27, 309. Morris, D. F. C., and Killick, I<. :\., Talanta, 1963, 10, 279. Maddock, H. S., and Mcinkc, W. IY., I!nivcrsity of Michigan Progress Keport 1969, 8, 94. - - - , i b i d . , 1962, 11, 68. Yajima, S., Kamemoto, Y . , Shiba, K,, and Onocla, Y . , J . Chem. SOC. Japan, Pure Chem. Sect., Bowen, H. J. M., and Cawse, P. :I., U . K . Atomic Encrgy Rcscarch Establishment Rcport 212925, Gardner, D. G . , Nuclear Phys., 1962, 29, 373. Ro\v.cn, 1.1. J. M., U.K. Atomic Energy Rcscarch EsValdishment licport R4196, lIarwell, 1963. p. 137. isotopes, International Atomic Energy Agency, Vienna, 1962. 5 , 280. 1961, 82, 343. Harwell, 1959. Iieceivccl Abril 8th. 1963

ISSN:0003-2654    

DOI:10.1039/AN9638800721

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

September, 19631 RAHMAK 727 Determination of Thorium and Phosphorus Pentoxide In Solution and in Phosphate BY ABDUR RAHMAK* (Department o j Chemistry, Impevial Coltege of Science alzd Technology, Lotidoti) Methods are described for the gravimetric determination of thorium in the presence of substantial amounts of phosphate, and of phosphate in the presence of thorium. The application of these methods to the analysis of insoluble thorium phosphates is described. IN the course of studies on the preparation o€ thorium phosphates1 by solid-phase and precipitation reactions, a method for determining thorium and phosphate present together in solution or in an insoluble compound was needed. Accurate methods for determining thorium in its pure compounds or in the absence of interfering substances such as phosphate are readily found, but procedures for use in the presence of relatively high concentrations of phosphate have not been described.Some excellent review~~9~7~ on the determination of thorium are too voluminous to be summarised here. Although thorium can be precipitated as thorium iodate6 76 from nitric acid solution, and the separation can be achieved in the presence of phosphoric acid, the well-known method of determining thorium oxalate deserved attention in view of its sim- plicity, Accordingly, a procedure was established by which thorium could be determined by this method even in presence of a fairly large amount of phosphate. In preliminary work an attempt was made to separate thorium from phosphate by adsorbing the thorium ions on a cation-exchange resin (Zeo-Karb 225), initially in the hydrogen form. Thorium was strongly adsorbed by the resin, but all attempts to elute it afterwards, even by strong mineral acids, failed.This was not unexpected in view of the high charge on the thorium ion, Th4+, and hence the process was abandoned. The separation of thorium from phosphate by ion-exchange was apparently successful, but the quantitative recovery aimed at could not be achieved. Determination of phosphate was also attempted after separation of thorium on the same cation-exchange resin. The results were not reproducible, and about 3 per cent. of the total phosphate was lost on the resin column. Evidently some phosphate was retained in the form of a cation complex. Schoeller and Powell’s method7 for determining phosphoric acid in monazite sand was examined to see whether or not it could be used in presence of large amounts of thorium.Twenty-five millilitres of phosphoric acid solution (0-0908 g of phosphorus pentoxide) and 25 ml of thorium nitrate solution (0.1081 g of thorium dioxide) were placed in a 250-ml beaker. The thorium phosphate jelly was dissolved by adding 2 to 3 rnl of concentrated sulphuric acid. The solution was saturated with potassium sulphate and after thorough mixing was set aside overnight; the bulk of the thorium was precipitated as thorium sulphate.’ After a single precipitation of thorium by this method the amount of phosphate eventually recovered was 3-9 to 4.1 per cent. less than that originally taken; after double precipitation the amount of phosphate recovered was 1.2 to 1.7 per cent.less than that taken. Schoeller and Powell’s method, designed for use when only small amounts of thorium are involved, is evidently not successful when major amounts of thorium are present. METHOD REAGENTS- Standard thorium nitrate soldon-Prepared from AnalaR thorium nitrate. Thorium was determined as thorium oxide after precipitation by oxalate solution.6 The standard solution gave 0.1081, 0.1083 and 0.1079 g of thorium dioxide in 25-ml portions. Standard $hos@zoric acid solzdion-Prepared from AnalaR grade phosphoric acid. * Present address : Pakistan L4tomic Energy Commission, Karachi, Pakistan.728 R.4HMMAS DETEIIMIS.ITIOS OF THORIUM AXL) PHOSPIIOHCS [/lnd_l?$t, vol. ss Phosphorus pcntoside was determined 1)y doriblc precipitation with magnesia mixture and ignition of the magnesium ammonium phosphate to magnesium pyrophospliatc.Treatment of two 25-ml portions of standard solution give 0-1424 and 0-1424 g of magncsiiim pj-ro- phosphate (0.0908 and 0.0908 g of phosphorus pentoxide). Oxalic acid wash liquid--4 2 per cent. aqueous sollition of o d i c acid containing 1 rnl of concentrated nitric acid per 100 ml of solution. I)ETERXIIS.~TIOS OF THORILM IS THE PRESESC‘E OF I’EIOSI’IIATE IS A SOLI~TIOS - Yix 25 ml of thorium nitrate solution and 25 ml of phosplmric acid solution in a 4OC)-ml beaker. L)issoive the jelly formed bj- adding 10 ml of concentrated nitric acid and warming, Add about 25 ml uf water, and then add, dropwise, to the boiling solution 30 t o 40 rnl of a saturated solution of oxalic acid, with constant stirring.Finally, dilute the solution with water to ahout 300 ml to reduce the acid concentration to the admissiblc masimum of 3-5 ml of concentrated nitric acid per 100 ml of solution. Never dilute the solution before adding oxalic acid solution ; in the absence of oxalic acid the insoluble phosphate jt.115- reappears. Boil the diluted solution for about 5 minutes, and set aside overnight. Filter the solution through a Whatrnan KO. 40 filter-paper; allow most of thc precipitate to rcmain in the beaker. Wash the beakcr and thc precipitate on the filter-paper with about 150 ml of oxalic acid wash liquid. Transfer the precipitate, with the filter-paper, to the original beaker containing about 100 ml o f concentrated nitric acid, and xld about 5 ml o f conccntrated hydrochloric acid.Rrcak up the filter-paper, and gently boil t he solution in the covered beaker. After about 45 minutes the solution should be clear brown and much of the filter-paper shouId have been destroyed. Continue boiling the solution until its volume is rcduccd to about 25 ml. Transfer the beakcr to a steam-bath, cautiously cvaporate some of the remaining acid, and, when crystals begin to appear, cool the bcakcr. (It is important a t this stage not to evaporate to dryness, otherwise the products tend to become insoluble.) \Vasti the sidcs of the beaker and the clock glass with the minimum amount of water. Add about 20 ml of concentrated nitric acid, and c\.aporate the solution nrarly (but not com- pletely) to tlryncss on a steam-bath.Cool the solution, and then add about 2 ml of concen- trated nitric acid and 20 ml of a saturated solution of oxalic acid; dilute the solution as described above, and set aside overnight. Filter the solution through a V‘hatman No. 40 filter-paper, and thoroughly wash the precipitate of thorium oxalate with oxalic acid wash liquid. Transfer ttic precipitate and paper to a platinum crucible, dry, and then char the paper over a low flame. Finally, convert the precipitate into thorium dioxide by ignition over a full flame until it is white in colour and constant in weight. ‘ l h r i a was satisfactorily determined by this procedure; 26-ml portions of thorium nitrate solution (0.lOSl g of thoria) in the presence of 25-ml portions of phosplioric acid (0.0908 g of phosphorus pentoxide) gave the results tabulated below- Lh.iation, ol, .. I . . . -. 0.09 -0.18 - 0*09 Thoria found, g . , . . 0.1082 0*1070 0.1080 L)I<TEK?JIS:ITIOS OF fHOS1’EiATl: 1.V PRESESCE OF THORILM IS SOLUTIOX- I t was eventually found that thc most convenient method for determining plmsphorus pentoxide in the presence o f thorium was that reportcd by Hoffman and LundelP for deter- mining phosphorus pentoxide in phosphate rock. Satisfactory results wcrc obtained only when precautions similar to those suggested b y Hoffman and Lundell were taken- (a) :% large amount o f citric acid must be used in order to keep the thorium in soiution. ( b ) A concentrated solution of “magnesia mixture” is neccssary.(c) Ikfore precipitation of ammonium magnesium phosphate, it is necessary to chill the solution in ice for at least half an hour; after precipitation it must be vigorously shaken for 1 hour to break down super-saturation. I he rcbsults uf determinations of phosphorus pentoxide under different conditions of I t can seen that vigorous sliaking during precipitation is * - precipitation are shown in lable I. essential if reprodiicible results are t o be obtained.September, 19631 PENTOXIDE I N SOLUTION AND I N INSOLUBLE THORIUM PHOSPHATE 729 TABLE I THE EFFECT OF DIFFERENT CONDITIONS OF PRECIPITATION ON THE RECOVERY OF PHOSPHORUS PENTOXIDE Thoria Conditions of precipitation taken, g Solution chilled in ice and shaken vigorously 0.1104 Precipitation at room temperature Solution chilled in ice but not without shaking the solution 0.1104 shaken 0.1104 Phosphorus Phosphorus pentoxide taken, pentoxide found, Deviation, g g % 0.0933 0.0 0,0936 + 0-3 - 2-0 0.0933 - 1-9 0.0921 - 1.3 0.0933 [ 0,0931 - 0.2 0,0933 { 0.0916 - 1.8 DETERMINATION OF THORIUM AND PHOSPHORUS PENTOXIDE IN INSOLUBLE THORIUM PHOS- Grind the thorium phosphate to a powder in an agate mortar, and weigh into a platinum crucible.Add AnalaR sodium carbonate (5 to 6 times the weight of the sample), and mix the powders thoroughly. Fuse the mixture over a bunsen burner for about half an hour, cool, and transfer the crucible to a 150-mi beaker. Cover the crucible with water, and place the beaker on a steam-bath for about an hour, to disintegrate completely the fused mass.Filter the liquid through a 9-cm Whatman No. 40 filter-paper after adding macerated paper. Wash the crucible, the beaker and the residue on the filter-paper with about 250ml of a 1 per cent. solution of sodium carbonate, and collect the filtrate in a 500-ml conical flask. Acidify the filtrate with hydrochloric acid, and evaporate to about 75 ml. Transfer the residue, with the filter-paper, to a platinum crucible; burn o f f the paper, and then ignite the residue gently for 10 minutes. Mix with AnalaR sodium carbonate (about five times the weight of the original specimen), and fuse the mixture €or half an hour. Disintegrate the fused cake in water as described, and filter the solution through a 9-cm Whatman No. 40 filter-paper after adding macerated paper. Collect the fiItrate and washings (about 200 ml of 1 per cent.ammonium chloride solution) in the original conical flask. Add more hydrochloric acid t o keep the solution acidic, and reduce the volume to about 100 ml by boiling. Then determine phosphorus pentoxide by double precipitation with magnesia mixture by the conventional procedure. Transfer the filter-paper containing the residue to a weighed silica crucible, ignite, and heat until the weight of ignition residue is constant. Then add powdered potassium pyro- sulphate (ten times the weight of the residue), and fuse the mixture over a small bunsen flame for 15 to 20 minutes, or until the thoria has completely dissolved in the pyrosulphate melt. Transfer the crucible and fused cake to a 400-ml beaker, and cover the crucible with about 150ml of water and 5ml of concentrated nitric acid.Heat on the steam-bath for half an hour, until the fused cake has disintegrated and much of it has dissolved. Stir the solution at frequent intervals while heating for a further half hour to complete dissolution. Remove and wash the crucible, and adjust the volume of the solution to about 200rnl. (Whenever an attempt was made to dissolve the fused cake by boiling with water, an in- soluble crystalline precipitate resulted.) Boil the solution, and precipitate thorium oxalate by adding a saturated solution of oxalic acid, as described previously. From this stage the procedure corresponds exactly with that described for determining thorium in the presence of phosphate in solution. The results of several determinations on differeat synthetic specimens of thorium phosphate are presented in Table 11.It is noteworthy that the percentage of thorium oxide found after pyrosulphate fusion is always considerably less than the percentage of “ignited residue,” and that the thoria plus phosphorus pentoxide percentage is substantially less than 100 ; the thoria @AS ignited residue percentages, however, are closer to the expected values of very nearly 100. This suggested that the high concentrations of alkali-metal ions inevitably introduced by the fusion with pyrosulphate had interfered with the precipitation of thorium oxalate, and that the ignited residue values more riearly represent the thoria content of the samples. To elucidate this point and to exciude other sources of error (such as retention of thoria by slight PHATES-730 LAnalyst, vol.88 reaction with the silica crucible) , test determinations were carried out on pure thoria. Sodium pyrosulphate was used for some of the fusions, because of the possibility that sodium might caiise Less interference in the oxalate precipitation than potassium. RAHMAK : DETERMISATIOS OF THORIUM ASI) PHOSPHORUS \Vcight o f thorium phospliatt~ Ignited taken, residur, 0.53 2 8 5S.90 0.5IOO 74.2i 0.4 83 6 i3.20 0 ~5.7 ti2 71.54 ( 1 * 33 99 69-13 0.5464 tiG-8ti 0.' R TABLE I I RE s I- Lr s l'horia found 3ftUS pyrosulphat c fusion, :iCi*(itl 73.04 71.53 70.63 ti7.46 ti6.5i 0 I 'hosphorus pcntoxide fountl, 40.37 25.43 pfj.55 27.83 30.34 32.4 I 0 ' ,o '1hosi;i A phosphorus pcn toxide found, 97.05 !#*Pi 9 x - 08 98.36 98.80 97.88 "0 1 gnitctl residue L- phosphorus pcn tox i d e h u n d , :4J !)9*27 99-70 99-75 99-37 !W47 !I9 * 2 '7 Thc: entire fusion proccss, starting with the doul)le fpsion with sodium carbonate, was carried out on pure thorium oxide, prepared by igniting AnaiaR thorium nitrate at 800" C in a platinum dish for 5 to 6 hours.1)etcrminations were carried out in which both silica and platinum cruciblcs were employed for thc pyrosiilphate fusion. The results are presented in Tahlcs I11 and 11'. \Veight oi thoria taken, g 0.3050 I) - 2!) 2 4 0.3184 I? - "953 Fusion rcagcnt Potassium pyrosulphate Sodium pyrosulphate Thoria found 1l:cight of after potassium Loss in w i g h t of silica ignited residue, pyrosulphate fusion, crucible during fusion, g 0.3064 0.2925 0,3282 0.2 9 5 0 % !f7*34 97.50 97.75 9 7 *4!1 k-I'SIOX I N PLATINVM CRUCIBLES Weight of thoria \\'eight of taken, ignited rcsirlue, I3 g 0.3232 0*3P30 0.2960 0.2963 *.{ 0,3046 0-3044 0.3023 0.3025 * { 0.3139 0.3142 Thoria found aftrr pyrosu lphatc fusion, 97.58 97-08 97.3 1 97.38 97.74 O / /O These results show that the weight of thoria obtained after fusion with sodium or potas- sium pyrosulphate is always less by about 2-5 per cent.than the weight of thoria taken. Both sodium and potassium ions evidently haw an adverse effect on the precipitation of thorium oxalate. The ignited residue, however, corresponds accurately in weight with the thoria initially taken. Attempts to dissolve the ignited rcsidue by hating with concentrated sulpliuric acid, SO that the introduction of alkali mc>tals was avoided, werc unsuccessful.Thc insoluble thorium phosphates were thercfore fused twicc with sodium carbonate, as described, and phosphorus pentoxide was determined in the Lvater extract hy double precipitation with magnesia mixturt. The ignited residue was taken to represent the thoria prest.nt in the original sample. T thank the Ministry of Education, Government of Pakistan, for an Ovcweas Scholarship. I also thank Ilr. A. J. E. \Velch for supervising the work and Mr. L. S. 'I'heobald for advice.September, 19631 PENTOXIDE IN SOLUTION AND IN INSOLUBLE THORIUM PHOSPHATE 731 REFERENCES 1. 2. Moeller, T., Schweitzer, G. K., and Starr, D. D., Chem. Rev., 1948, 42, 63. 3. 4. 5. 6. 7. 5. Welch, A. J. E., and Rahman, A., in the press. Rodden, C. J . , Editov, “Analytical Chemistry of Manhattan Project,” McGraw-Hill Book Co. Inc., Fonseka, J. P. R., D.I.C. Thesis (Inorganic), 1955. Meyer, R. J., and Speter, M., Chem. Zeit., 1910, 34, 306. Vogel, A. I., “A Textbook of Quantitative Inorganic Analysis,” Second Edition, Longmans Green & Co. Ltd., London, 1953. Scboeller, W, R., and Powell, A. R., “The Analysis of the Minerals of the Rare Elements,” Third Edition, Charles Griffin, London, 1955, p. 111. Hoffman, J. I., and Lundefl, G. E. F., Bur. Stand. J . Res., 1937, 19, 59. New York and London, 1950. First received February 28th, 1962 Amended, December 191h, 1962

ISSN:0003-2654    

DOI:10.1039/AN9638800727

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

732 ‘ E E s SHORT P-II’ERS IJPP‘0X.I SHORT PAPERS A Field Method for determining 2,P.Tolylene Di-isocyanate Vapour in Air BY I>. A . HEiI,l,Y (Imperial Cliemiral Iiidaislri~s L l d , , 1)ye.duff.s ~)ivisz07z, Ilexagon House, Ulacitley, ;llo~ic/icstcr !I) TOLYLEXE UI-ISOCYAXATI: is wiclcly usctl in thc production of urethane products such as flexible foams, lacquers and synthetic rubbers. Lhring t h e manufacture of these materials the vapour of tolylene di-isocyanate may be c\wl\ml, and this can constitute a danger t o health if Proper care in storage, handIing and ventilation is not taken. The effcctiveness of precautionary measures can best be assessed by measuring rttinosplieric concentrations o f tolylcne di-isocyanate at appro- priatc times and places. Thc recent reduction in the maximum permissiblc concent ration o f tolylenc di-isocyana t e from 0.1 to 042 p.p.ni.v1v in air112 has meant that the widely used M.S..A. T . l ) , I , I)ctector is insufficiently sensitive for thc determination of concentrations in thc region of the new limit. I\:ith this detector, based on the w-ork of M a r ~ a l i , ~ a. 3-litrc sample of air is dratvn through 1.5 rnl of dilutc hydrochloric acid, when tolylcne di-isocyanate is hydrolysed to the corrcsponding diamine ; this is then diazotised with a sodium nitrite - sodium bromide solution, 1 he excess of nitrite removed by treatment with sulphaniic acid and the diazoniuin compound coupled with -Y-l -naphthylethyl- cnediamine to givc a pink coloured solution. ‘ihc colour of the solution is then nirttcIicd visuaIlv in turn against three strips of pink colonred plastic material rep- resenting concentrations of 0.06, I).10 and 0-20 p.p.ni. of tolylene di-isocyanate, respectively. This method is claimed t o dctermine primarily the 2,4-isorncr, as the 2,6-isomer is stated to react much more slowly. Concentrations in thc region of 0-02 p.p.111. can be detcr- mined with the existing method by incruasing the sarnple s i x from 3 to 15 litres, but this involves cithcr a n iriconvcniently long samp- ling time of 15 minutes or the use of a larger (but lcss portable) pump t o sarnpIe at a faster rate. This paper describes a modifica- tion of the earlier method in which a 3-litre sainple of air is drawn through a specially designed absorber containing 3 rnl of a mixtiire of dilute hydrochloric acid and i~,l\:di-n~ethlilfc,rrnamide.” The pink colour is developed as described above and matched visu;illy against inorganic colour standard solutions.If E TH o D -IPPARATVS- SampZing pump---Capable of drawing air a t a rate o f 1 litrc per minute through thc absorber. The M.S..\. 4-cylinder hand- cranked pump has been found con\.enient for this purpose. ,lltcrnativel>-, whcn coiiwnicnt, a n-atcr-filled aspirator may be used. l h e dimcm sions of the narrow lower part are important and should be adhered t o ; the upper part should be sufficirntly large t o prevent loss of the contents of the absorbcr by splashing during sampling. AZl-glass alsovbev-Of the type sho\vn in Fig. I . Tesl-tubes--lOO mm x 10 mm. Fig. I . :lll-glass absorber R 1: .A (; IISTS-- bromidc in about 80 mt of water, and dilute to 100 ml.S o d i z m nitrile - sodium byontide soZitlzon-lXssolvc 3 4 g of sodium nitrite and 5-0 g of sodium Szilfihamic acid solution, 10 per cent. W / T .September, 19631 SHORT PAPERS 733 Hydrochloric acid, diZute-Dilute 25 rnl of concentrated hydrochloric acid , sp.gr. 1-18, to 1 litre with water. N-l-Naphthylethylenedianzine dihydrochloride solution-Dissolve 50 mg of the dry salt in about 25 in1 of water, add 1 ml of concentrated hydrochloric acid, and dilute to 50 ml with water. Use within 48 hours. N,N-Dimethylfor~anzide-Distil through a water-cooled all-glass condenser, rejecting the first and last 10 per cent. of the distillate. Absovber solution-Add 25 ml of concentrated hydrochloric acid, sp.gr. 1.18, to 500 ml of water.To this solution add 250 rnl of N,N-dimethylformamide, and dilute to 1 litre with water. Cobaltous chloride calovinzetric solution-A solution in dilute hydrochloric acid containing 59.5rng of the analytical-reagent grade salt, CoCl,.GH,O (purity not less than 97.5 per cent.) per mL5 Cupric sulphate colovametric soZution-A solution in dilute hydrochloric acid containing 62.4 rng of the analytical-reagent grade salt CuSO,.SH,O (purity not less than 99-0 per cent.) per mL5 Colour standard solutions-Mix the volumes of the cobaltous chloride and cupric sulphate colorimetric solutions specified in Table I, and dilute t o 500 ml with dilute hydrochloric acid. These solutions have the same quality and depth of shade as the test solutions (containing the equivalent amounts of tolylene di-isocyanate) prepared as described below. Store in clean glass bottles having tightly fitting glass stoppers.COMPOSITION OF COLOUR STANDARD SOLUTIONS Cobaltous chloride Cupric sulphate solution, solution, Tolylene di-isocyanate, ml ml p*p.rn. 8.75 13.75 0-0 I 17.50 27-50 0.02 35.00 55-00 0.04 PROCEDURE- Draw 3 litres of the air sample at a rate of about 1 litre per minute through the absorber containing 3.0 ml of absorber solution. Disconnect and lift the inlet tube, allowing the liquid in it to drain into the body of the absorber. Expel the last drop by blowing gently on the inlet tube, and then withdraw this completely. ,4dd 8.1 rnl (3 drops) of sodium nitrite solution, close the absorber body with a glass stopper, and mix the solution by gentle shaking; set aside for 1 minutes.Add 0.1 ml (3 drops) of sulphamic acid solution, stopper the absorber, mix the solution by gentle shaking, and set aside for 14 minutes. Add 0.1 ml (3 drops) of N-l-naphthylethylene- diamine dihydrochloride solution, stopper the absorber, and mix the solution by gentle shaking. Fill 100-mm x 10-mm test-tubes with the three colour standard solutions to the same depth as that of liquid in the absorber. Between 1h and 2 minutes after adding the last reagent to the contents of the absorber, compare the colour of the test solution with each of the colour standard solutions in turn by looking downwards through the solutions towards a sheet of white paper held a few inches below the bottom of the tubes.Report the concentration of tolylene di-isocyanate as less than, equal to or more than that of the nearest colour standard. BASIS OF METHOD During the development of this method, a number of relevant factors were investigated. STRENGTH OF COLOUR STANDARD SOLUTIOXS- The composition of these solutions was arrived at by a process of trial and error. The quality of shade was first matched by mixing cupric sulphate and cobaltous chloride solutions in different proportions until the shade of the mixture was judged to be correct by two people. Solutions of different strengths, but with the two components in the same proportion, were next prepared and matched with test solutions prepared from 0.21, 0-42 and 0-84 pg of tolylene di-isocyanate (corre-734 SHORT PAPERS [Analyst, VOl.88 sponding to the weights present in 3 litres of tcst samples containing 0*01, 0.02 and 0-04 p.p.m. of tolylene di-isocyanate). IVhen the approximate strengths of the inorganic standards had been asccrtained, further s o h tions were prepared covering narrowcr concentration ranges, and the final concentrations were decided on. In preparing coloured solutions from tolylcnc di-isocyanate, a standard solution containing 7 4 pg per ml was prcparccl in aqueous acetic acid as described by MarcalP; 0-!5-, 1.0- and 2-O-ml portions of this solution wcre diluted to 50 ml with dilute hydrochloric acid, and 3-0-mI portions of thesc diluted solutions were then treated as described above. l<EACTIOS \VITH THE 2,6-ISOMER- Undcr the conditions described, the conclusion* that the reaction was spccific fur t h e 2,4-isomer was largely confirmed.There was a slow reaction with thc 2,6-isorncr, which produced a colour much bluer than that produced by the 2,4-isomer. The inorganic colour standards were matched against colours produced from a mixture containing 80 per cent. of the 2,4-isomer and 20 per cent. of the 2,6-isomer. The error when these standards were used for determining other commercial tolylene di-isocyanates, whose 2,4-isomer content might range from 60 to 100 per cent., would not be a scrious one. TIME REQUIRED 1’011 COLOUR DEVELOPMENT-- In the procedure it is recommended that the matching of the d o u r of the test solutions with that of the inorganic colour standards be done between 13 and 2 minutes after the addition of the -\r-l-naphthylethylencdiamine dihydrochloride.This time interval was adopted because, on further standing, the colour of the tcst solutions deepcncd slightly and became bluer in shade, presumably owing to slow reaction of the 2,6-isomer. r<ECOVEKY FROM A SAMPLE OF AIR- So attempts were made to produce accurately known atmospheric concentrations of tolylene di-isocyanate. It was found, however, that concentrations in the range 0.01 to 0.10 p.p.m. were produced when a stream of nitrogen was bubbIed through 20 ml of 2,4-tolylene di-isocyanate in a 15-cm x 2-5-cm test-tube a t 20” C, a t specds from 10 t o 100 ml per minute, and diluted with 10 litre5 of air per minute. Tests were then carricd out by sampling through two absortxrs in series and observing the ratio of isocyanate found in the first absorber to that found in the second, by measuring the optical densities at 544 m p with a Unicam SP600 spectrophotometer.Initial tests were carried out by using dilute hydrochloric acid without added it’,,~~-dimeth~lformamide as thc absorber solution; the results were- .\mount o f tolylenc di-isocyanate found in 2 absorbers, .\mount found in 1st absorber, as pcrccntage of total . . 66 66 ti7 69 69 63 pp.m. v/v .. * . . . .. .. . . 0.040 0.040 0.035 0,035 0-035 0.025 In later tests an absorber solution containing ~~,:,-~-diniethylformamide was used ; the results of thesc tests were- -4mount of tolylene di-isocyanate found in 2 absorbers, hmount found in 1st absorber, as percentage of total .. 93 SO 81 90 87 95 p.p.m. vjv . . .. . . . . . . . . 0.100 0.055 0.045 0.035 0.025 0.025 Cos CLU SION s The sensitivity of the test has been increased five-fold by reducing the volume of absorber solution used from 15 nil2 to 3 rnl. To ensure the absorption of a sufficiently high proportion of the total isocyanate in only 3 ml of liquid, an absorber giving a high ratio of depth to volume has been designcd. The retaining powcr of thc liquid was increased by incorporating 25 per cent. o f N,9-dimethylformamide in the absorber solution, resulting in the retention in the first absorber of 88 & 8 pcr ccnt. of the total amount of isocyanate trapped from a dry nitrogen stream. -4s this was intended as a rapid field method the accuracy and precision of these results was con- sidered to be adequate.September, 19631 SHORT PAPERS 735 REFERENCES 1. pheres, 1962 Amendment, H.M. Stationery Office, London. 2. 3. 4. Funke, W., and Hamann, F., Furbe u. Lack., 1958, 64, 120. 5 . Ministry of Labour, Safety, Health and Welfare Sevies No. 8, Toxic Substances in Factory Atmos- “Threshold Limits far 1961,” Arch. Enviran. Health, 1961, 3, 489. Marcali, K , , Anal. Chem., 1957, 29, 552. United States Pharmacopoeia, Fourteenth Revision, pp. 929 to 931. Received April Id, 1963

ISSN:0003-2654    

DOI:10.1039/AN9638800732

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

September, 19631 SHORT PAPERS 735 The Determination of Small Amounts of Sulphur in Toluene by Reduction with Raney Nickel BY R. H. REED (Imperial Chemical Industries Ltd,, DyestGffs Division, Hexagon House, Blackley, Munchester 9) ERANATELLI'S method, 1 in which sulphur compounds are reduced to sulphides by Raney nickel and the hydrogen sulphide produced on acidification is determined, is potentially attractive for the determination of total sulphur in certain types of material, particularly in highly refined hydrocarbons containing extremely small amounts of sulphur. However, when the method was tried in this laboratory on toluene solutions of known sulphur content the recoveries varied between 7 0 and 85 per cent. of theory. Modifications in the mode of preparation and use of the Raney nickel catalyst resulted in an increase in the percentage recovery from elementary sulphur, carbon disulphide and 2- and 3- methylthiophen to 95 f 5 per cent.at a level of 1 to 10 p.p.m. of sulphur. The modifications deal with the temperature at which the Raney nickel is activated, the detailed instructions for washing the activated catalyst and the immediate use of the catalyst after its preparation. METHOD REAGENT- catalytic nickel. NickeE - aluminium aEloy--50 per cent. of nickel, 50 per cent. of aluminium, for preparing Obtainable from British Drug Houses Ltd. ACTIVATION OF NICKEL CATALYST- Assemble an apparatus similar in dimensions to that described by Granatelli,l and grease the joints lightly with silicone grease. Remove the reduction flask, measure into it 10 ml of 2.5 M sodium hydroxide solution, and raise the temperature to 75" to 80' C.Add, in one portion, 0-5 0.05 g of nickel - aluminium alloy; a vigorous reaction ensues, and precautions should be taken against the mist of caustic liquor ejected from the flask. Set aside the Aask for 10 minutes. Decant the supernatant liquid as completely as possible from the flask, and wash down the neck of the flask with 10 to 15 ml of water from (for convenience) a polythene wash bottle. Swirl the water with moderate vigour to disturb the nickel residue slightly, but avoid entrainment of air. With the minimum delay required for the residue to settle, decant off the water as completely as possible. Repeat the washing of the neck of the flask and the nickel residue twice with water, and then with 10 ml of isopropanol, Decant off the isopropanol, and add a further 10 ml of isopropanol to the flask.Without delay transfer by pipette 10 or 25 ml of a sample expected to contain 1 to 10 p.p.m. of total sulphur, and replace the flask in the apparatus. PROCEDURE- Pass nitrogen through the sample solution at a slow rate (about 2 bubbles per second) for 10 minutes before heating is commenced, and then continue as described by Granatelli.1 The absorber may be connected to the apparatus at the start of the procedure; this de-aerates the con- tents of the absorber without the necessity of manipulating the gas flow and absorber later in the procedure.SHORT PAPEHS [Analyst, Vol. 88 L)ISCLTSSIOI\; OF *mT: METHOT) Treatment of the nickel - aluminium alloy with hot sodium hydroxide solution and the use of freshly prepared catalyst give improved efficiency of reduction of sulphur compounds in gencral. Certain compounds are affected by the alkalinity of the catalyst; if, for examplc, the nickel is washed j u s t once with water, only about 20 per ccnt. recovery of sulphur is obtained from 2- or 3- rncthylthiophcn.However, for the complete reduction of phenylvinylsulphone it is neccssary to Iimit the number of washings with water to one. Elcmcntary sulphur is completely reduced, irrespective of variations in the washing of the nickel. Limited cxperimcnts have shown that a short reduction with triplc washed Raney nickel, with subsequent addition of a small amount of sodium hydroxide solution and a further period of hcating will decomposc mixtures of compounds requiring different alkalinities for their reduc- tion; this may serve as a satisfactory arbitrary procedure that will eliminate the need for knowing what sulphur compounds are present in a new type o f sample. The modified method is eminently suitable for the routine determination of total sulphur in tolucne in the 1 to 10 p.p.m. range and should also be applicable to other aromatic and aliphatic hydrocarbons and to alcohols. The process is simple, quiet and rclativel y free from hazard, and is such that a routine operator can supervise two sets of apparatus a t the samc timc. KEFE RE Ic c: E 1. Granatelli, L., A n d . Chenz., 1959, 31, 434. Rcceiwd Afwil lsl, 1963

ISSN:0003-2654    

DOI:10.1039/AN9638800735

出版商:RSC    

年代:1963

数据来源: RSC