摘要:

MARCH, 1963 Vol. 88, No. 1044 THE ANALYST E,DITORIAL Accent on Youth IT has always been the policy of the Society to encourage the reading of papers and to promote discussion of subjects of analytical interest. Some come from industrial laboratories, some from academic institutions or municipal undertakings ; some are in private practice, some are in Government service. Many are acknowledged experts in their respective fields, others are at the beginnings of their careers-the experts of tomorrow. All have a common interest in some branch of Analytical Chemistry in its widest sense. Addressing a public scientific meeting for the first time is an occasion of some moment for the young researcher. How will his ideas be received by an audience that almost certainly will include “elder statesmen” of the profession-scientists of far wider experience than his own? Is he perhaps a little apprehensive, for we find that the post-graduate student is not always as ready as we would wish to come forward spontaneously and describe his work.Yet there is no doubt that it is from the work of these young men and women, in the Uni- versities and Technical Colleges throughout the country, that many of the most valuable advances will come and from whom many of the new techniques of the future will arise. The Society is very conscious of the importance of the work being done at the Univer- sities and Technical Colleges and the grade of “Junior Membership” was inaugurated some years ago so that the “younger generation” should be better represented in the Society’s ranks.But there is still a reluctance, or a reticence, on the part of many Junior Members and their contemporaries to present papers at our meetings, and the Programmes Committee has recently been making strenuous efforts to break down this resistance. We should like a t least one of our meetings each year to be set aside as a forum for discussion of work in progress at these centres of learning and thus to foster the active participation of the post-graduate students, who will, after all, be the practising analysts of the future. We have been in touch with several Universities and Colleges of Technology and, broadly speaking, found them in sympathy with our views. With their help we have been able to arrange a meeting, to be held on April 9th a t Chelsea College of Science and Technology, London, S.W.3, for which the programme will consist of six papers contributed entirely by these “up and coming” analysts. We look forward to hearing what they have to say and we are confident that an interesting and stimulating evening is in store. Speakers are drawn from far and wide. 153

ISSN:0003-2654    

DOI:10.1039/AN9638800153

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

154 PROCEEDINGS [Analyst, Vol. 88 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY XEW MEMBERS William Boardman, M.Sc. (Lond.), A.R.I.C., A.R.T.C.S. ; Kenneth Roy Capper, B.Pharm., Ph.D., D.I.C. ; Leslie Richard Chislett, A.R.I.C. ; Hugh Bernard Clarke, M.Sc.(Lond.), Din0 Coppini, L.D. ; Sven Lars Ake Danielsson, Fil.lic.(Stockholm) ; Kenneth Wallace De Witt, B.Sc.(Bris.) ; Christopher John Dodd; Peter James Duff, B.Sc.(Liv.) ; John Henry Dunn, B.Sc.(Lond.), A.R.I.C. ; Philip Sidney Hetherington, B.Sc.(Manc.), A.R.I.C. ; John Douglas Hobson, B.Sc., Ph.D. (Lond.), F.R.I.C., F.I.M., A.Met.(Sheff .) ; David Francis James; David Peter Lones, BSc., Ph.D.(Birm.); R. C. Madan, B.Sc.; Benjamin Max Milwidsky, B.Sc. (Rhodes) ; John Michael Orme, B.Sc.(Lond.) ; Philip Geoffrey Quartermain, R.Sc.(Lond.), A.R.I.C.; Dennis Noel Raine, B.Sc., Ph.D., M.B., B.S.(Dunelm.); Cyril Redshaw, B.Sc.(Lond.), D.L.C.(Loughborough) ; Trevor Salvage; Paschoal Senise, D.Sc.(Sao Paulo) ; Graham Robert Thomas, A.R.I.C. ; Eric George Towndrow, B.Sc.(Lond.), A.R.I.C. ; Terry Wallis; Joseph Westheimer, B.S. (New York). James Roger Cousin, BSc. (Hull) ; Michael Thompson, B.Sc. (Lond.), A.R.C.S., A.R.I.C. James Haythorn Baird, R.A.(Rutgers), M.Sc.(Brooklyn) ; John Alexander Baynes, BSc. (Southampton) ; Ronald Francis Bird, A.R.I.C. ; Lyn Davies, B.Sc. (Wales) ; James Sidney Double ; Andrew Gordon Gavin, A.H.-W.C., A.R.I.C., A.M.1nst.F. ; John Harry Greaves, BSc. (Lond.), F.R.I.C. ; Geoffrey James Holland, BSc. (Lond.), F.R.I.C. ; George Harold Jeffery, B.Sc., Ph.D.(Lond.), F.R.I.C.; William Arthur Lloyd; David Lyn Mack, L.R.I.C.; Gordon Nelson, A.R.I.C.; Ronald Platt ; Bryan Edward George Pledger, B.Sc.(Lond.) ; Richard John Marshall Ratcliffe, B.Sc. (Lond.) ; Gerard Aloysius Rimmer, B.Sc. (Liv.), A.R.I.C. ; Alan Sydney Waterhouse, B.Sc. (Lond.), A.R.I.C. ; Glynne Williams. NORTH OF ENGLAND SECTION THE thirty-eighth Annual General Meeting of the Section was held at 2.30 p.m, on Saturday, January 26th, 1963, at the Old Nag’s Head Hotel, Lloyd Street, Manchester, 3. The Chair was taken by the Chairman of the Section, Mr. J. Markland, B.Sc., F.R.I.C. The following appointments were made for the ensuing year: Chairman-Mr. C. J. House. Vice-Chairman -Mr. J. F. Clark. Hon. Secretary and Treasurer-Mr. G. F. Longman, Unilever Research Laboratory, Port Sunlight, Cheshire.Members of Committee-Mr. C. E. Davis, Dr. J. R. Edishury, Mr. B. Hulme, Mr. A. Hutchinson, Professor H. M. h’. H. Irving and Mr. A. N. Leather. Mr. A. A. D. Comrie and Mr. F. Dixon were re-appointed as Hon. Auditors. The Annual General Meeting was followed by an Ordinary Meeting of Section at which Mr. J. Markland, B.Sc., F.R.I.C., gave his Address as retiring Chairman. The Chair at this meeting was taken by the new Chairman of the Section, Mr. C. J. House, B.Sc., A.R.C.S., F.R.I.C. SCOTTISH SECTION THE twenty-eighth Annual General Meeting of the Section was held at 1.45 p.m. on Friday, January 25th, 1963, at More’s Hotel, India Street, Glasgow, C.2. The Chair was taken by the Vice-chairman of the Section, Dr. R. A.Chalmers, B.Sc. The following officer bearers were elected for the forthcoming year : Chairman-Dr. R. A. Chalmers. Vzce-Chairman- Mr. J. K. McLellan. Hon. Secretary and Treasurer-Mr. J. ITT. Murfin, Standards Department, Boots Pure Drug Co. Ltd., Motherwell Street, Airdrie, Lanarkshire. Members of Committee -Dr. D. M. W. Anderson, Mrs. D. A. Edrnond, Mr. J. C. Jack, Mr. W. J. Murray, Dr. J. Sandilands and Mr. A. F. Williams. Mr. J. S. Foster and Mr. R. A, Sutter were re-appointed as Hon. Auditors. ORDINARY MEMBERS ELECTED JANUARY 9TH, 1963 JUNIOR MEMBERS ELECTED JANUARY 9TH, 1963 ORDIKARY MEMBERS ELECTED FEBRUARY 6TH, 1963 WESTERN SECTION THE eighth Annual General Meeting of the Section was held at 5.30 p.m. on Thursday, January loth, 1963, in the University of Bristol. The Chair was taken by the Chairman of the Section, Dr. F. H. Pollard. The following appointments were made for the forthcorningMarch, 19631 PROCEEDINGS 155 year: Chairman-Dr. F. H. Pollard. Hosz. Secretary and Treasurer-Dr. T. G. Morris, Brockleigh, Clevedon Avenue, Sully, Glamorgan. ,Wembers of Committee-Dr. L. E. Coles, Dr. G. V. James, Mr. G. M. Telling, Mr. J. D. R. Thomas and Dr. W. J, Williams. Mr. S. Dixon and Mr. C. H. Manley were re-appointed as Hon. Auditors. The Annual General Meeting was followed by a Joint Meeting with the Bristol and District Section of the Royal Institute of Chemistry, at which a talk on “New Developments in Chelatometry-A Review of Methods and Reagents Related to EDTA” was given by T. S. West, B.Sc., Ph.D., A.R.I.C. Vice-Chairman-Mr. E. A. Hontoir. The Chair was once again taken by Dr. Pollard.

ISSN:0003-2654    

DOI:10.1039/AN9638800154

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

March, 19631 PROCEEDINGS 0 bituary MARY CORNER AFTER several months of suffering, Mary Corner died on Sunday, November 4th, 1962, a t St. Bartholomew’s Hospital. Born on March 25th, 1899, her schooling was delayed for some years by an unfortunate accident, the effect of which was to remain with her for the rest of her life. Her early education was received at Beulah House High School, Ralham, which she left in 1915. During the war years she worked in a pharmacy, and then, in 1922, entered Battersea Polytechnic, where she graduated. After spending a further year at London University, she took a post in 1928 at the British Cotton Industry Research Association, thus following Dr. (later Sir) Robert Pickard, whom she greatly admired and who, in 1927, had relinquished the Principalship of the Polytechnic to become Director of the Research Association. Miss Corner first served in the rayon department, and one of the papers pub- lished with her colleagues dealt with the microdetermination of metals in commercial rayon yarns (J.Text. Inst., 1933,24, 293). This work probably triggered off her interest in micro- analysis, and it was not long before she was made head of the microanalytical section. In 1933 she attended the first of two short courses on organic and inorganic micro methods held by Dr. Janet Matthews in the Plant Physiology Department of Imperial College. It was the members of these two courses who formed the Microchemical Club, which flourished until, in 1944, it became the nucleus of the Microchemistry Group of our Society.Early in 1945 Miss Corner left Manchester to take up a similar appointment with the British Leather Manufacturers Research Association. After a little more than two years, she was invited to become head of the newly formed Microanalytical Section of the Chemical Research Laboratory, later the National Chemical Laboratory. She was pleased to accept the post, which was conveniently sited, for she and her mother were now living in Teddington. An excellent microanalyst herself, she was not only good at instructing those who were assigned to assist her but also could instil into them her own high standard of integrity and her keen interest in microanalysis. More recently Miss Corner had devoted herself mainly to development and research problems. The oxygen-flask method had caught her interest and she quickly used it for the determination of organically-bound arsenic, phosphorus and boron.Always keenly interested in her work, she was ever ready to talk about it and, indeed, to discuss with anybody at any time any aspect of microanalysis. She took a prominent interest in the affairs of the Societies to which she belonged, being Vice-chairman of the Microchemistry Group at the time of her death, and having served on the Council of the Society for Analytical Chemistry during 1953-54. The work of the British Standards Institution also interested her, and she was Chairman of one of the Sub-committees dealing with microchemical apparatus. She was made a Fellow of the Royal Institute of Chemistry in 1946, having been elected to the Associateship in 1931. Miss Corner was an exceptionally fine character and it was a privilege to know her. Burdened with a severe disability, she had had, in addition, more than the usual share of suffering and trouble. Yet she never complained but, on the contrary, she was always cheerful and ever had a thought for the troubles of others. Even to the end she was intensely alive and, in spite of everything, she enjoyed life. In her early years she had even played games and was fond of horse-riding and gardening. A loyal and a generous friend, she has left all those who cherished her friendship much the poorer for her passing. Sub-micro methods of analysis were also being examined. G. R. DAVIES

ISSN:0003-2654    

DOI:10.1039/AN9638800155

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

156 ANALYTICAL METHODS COMMITTEE [Analyst, Vol. 88 Analytical Methods Committee PARTICLE SIZE ANALYSIS SUB-COMMITTEE Classification of Methods for Determining Particle Size A Review* FOREWORD THE information presented in this Review was collected by the Particle Size Analysis Sub- committee? of the Analytical Methods Committee of The Society for Analytical Chemistry during the period March 1961 to December 1962. It forms part of a comprehensive survey of this field of analysis that is being made by the Sub-committee. The initial task has been to list and classify existing methods; this classification in tabular form is presented here, together with short explanatory notes on the methods listed. Some form of appraisement or evaluation of methods or apparatus is envisaged for later publication.Meanwhile, it is hoped that this classification may be helpful as a guide to those whose work requires them to practise particle-size analysis, and in particular to those who are studying the subject for the first time. A. MEASUREMENT OF SIZE DISTRIBUTION A.I. Relative Motion between Particles and Fluid : ( Sedimentation [Gravitational 1 FLUID STATIC (sedimentation) pipettes . . . . . . . . - - Liquid . . . . . . . . Sedimentation JGas I Hindered settling columns Y Gravimetric . . Absorptiometric variation Radiometric . . . . . . . . . . , . I[ Density Sedimentation rate . . Fractional decantation . . L Centrifugal . I . . . . . . , . . . ..{?,id . . . . . . . . Gravitational . . . . . . .. . . . . . . Winnowing . . . . . . . . . . . . FLUID MOVING Method No.1, 2 3, 4, 5, 6, 7, 8 9 10 11, 12 13, 14, 15 16, 17, 18, 19, 20, 21, 22 23 24, 25, 26, 27 28, 29, 30 31, 32 33, 34, 35, 36, 37 38, 39 40, 41, 42, 43, 44 A.11. Image Formation : Individual sizing Automatic sizing and counting . . and counting . . 45, 46, 47 48, 49 . . . . . . . . 50, 51 .. . . . . . . 52 foptical microscope . . Flying-spot principle . . LIGHT BEAM ELECTRON BEAM-EleCtrOn microscope . . i [continued * Reprints of this paper will be available shortly. For details, please see p. 247. t The Sub-committee consisted of Messrs. E. Q. Laws (Chairman), R. de B. Ashworth, D. G. Beech, C. G. L. Furmidge, H. Heywood, H. W. Hibbott, J. F. Hinsley, R. Howes, R. Jackson and B. H. Kaye, with Miss C. H. Tinker as Secretary.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE A.MEASUREMENT OF SIZE DISTRIBUTION. continued Method No. A.111. Scattering of Radiation: LIGHT BEAM .. . . .. .. . . .. .. .. . . 53 AJV. A.V. A.VI. B.I. B.11. X-RAY BEAM . . .. . . . . . . .. .. LIGHT BEAM . . . . . . . . . . .. .. CONDUCTIVITY .. . . .. . . .. .. Diffraction of Radiation : Electrical Properties : Sieves : WOVEN WIRE (standardised) .. . . .. .. RECENT TECHNIQUES (not yet standardised) . . , . B. MEASUREMENT OF SURFACE Permeability to Fluid Flow: LIQUID FLOW . . . . . . . . . . . . . . GAS FLOW . . . . . . . . . . .. . . Adsorption Methods : GASES AND SOLUTES . . . . . . .. . . 54 .. . . 55 .. . . 56 * . .. 57 .. . . 58, 59 .. . . 60 .. . . 61, 62, 63, 64, 65, 66, 67 . . 68, 69, 70 . . , (g;;;mic.. . 71 Method No. A.I. 157 - - B.111. Absorption of Radiation: LIGHT BEAM .. . . . . . . . . .. . . .. . . 72 PENETRATIKG RADIATION . . . . . . . . . . .. . . 73 B.IV. Optical Measurement: . . . . . . .. . . .. . . 7 4 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. KEY TO CLASSIFIED TABLES Method Size Relative Motion between Particles and Fluid : range, FLUID STATIC (Sedimentation process)- P Page Andreasen pipette .. . . .. .. .. Fixed-depth pipette . . . . . . .. .. B.C.U.R.A. single tube . . . . .. .. . . Whitby column method . . . . .. .. .. Bostock sedimentation balance .. . . . . Sartorius sedimentation balance Shimadzu sedimentation balance Micromerograph . . . . . . Hindered settling apparatus .. .. Hydrometer method . . . . .. .. , . E.E,L. photosedimentometer . . .. .. .. “Bound Brook” photosedimentometer . . . . .. Incremental sedimentation (radiometric) . . . . Beaker-centrifuge (radiometric) . . . . . . Two-layer radiometric sedimentometer . . . . Beta back-scattering sedimentometer . . . . . . Gamma-ray absorption . . . . .. . . . . Beta-ray absorption . . . . .. . . . . Photography of particle track . . . . . . , . Liquid column method with sediment extraction . . . . .. .. .. .. .. .. } - - Diver method . . . . . . .. . . . . I.C.I. method . . . . .. . . .. .. Tracers in Andreasen and long-arm centrifugal pipettes Beaker method . . . . . . . . . . . . C.P.A.C. method . . . . * . .. .. * . W.H.O. method . . . . . . .. .. .. M.A.F.F. method .. . . . . . . . . . . Sharples centrifuge method . . . . . . .. Dietert centrifuge method . . .. .. .. Kaye disc centrifuge . . . . .. .. . . .. 2to60 162 . . 2 to 60 163 . . 3to76 163 .. 3to76 164 . . 0.05 to 100 164 . . 3 to 76 165 . . 1 to 60 165 . . 0*8to 250 166 . . 5 to 1000 166 . . 1 to25 166 . . 0.02 to 40 167 . . 1 to 50 167 . 1 to 50 168 . . 1 t o 76 168 .. 4to76 169 . . 0.05 to 76 169 . . 0.1 to76 169 . . 0.1 t o 5 170 .. 0.1 to76 170 . . 0.1 to 76 171 . . 0.1 to 5 171 .. 0-5to300 171 . . 10 upwards 171 .. 2 t o 7 6 172 .. 2to76 172 . , 2 t o 76 172 . . 0.05 to 50 173 . . 0.05 to 50 173 .. 1 t o 100 173158 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol. 88 Method No. Method Size range, A.11. A .III. A.IV. A.V. A.VI. B.I. P FLUID MOVING (Elutriation processes)- 3 1.Andrews elutriator .. . . . . .. .. . . 10to 76 32. Blythe elutriator . . . . . . . . .. . . . . 10to60 33. Gonell air elutriator . . . . . . . . .. . . 5to76 34. 5 to 76 Modified Gonell elutriator (Fuel Research Station apparatus) 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. Roller air elutriator . . . . Miniature elutriator . . . . Haultain elutriator (Infrasizer) . . Bahco centrifugal dust classifier B.C.U.R.A. centrifugal elutriator Walton horizontal elutriator . . Timbrell sampler . . . . . . Conifuge . . . . .. . . Conicycle . . . . .. . . Cascade impactor . . . . Image Formation : LIGHT BEAM- 45. Graticules . . . . . . . . 47. Zeiss-Endter particle size analyser 48. Casella instrument . . . . 49. Timbrell instrument . . . . 50. Rank Cintel instrument .. . . 51. Mullard instrument . . . . 52. Electron microscope . . . . 46. Konimeter (Conimeter) . . ELECTRON BEAM- Scattering of Radiation : 53. Goulden light scattering technique LIGHT BEAM- X-RAY BEAM- 54. Low-angle scattering . . . . Diffraction of Radiation : 55. Young’s diffraction rings . . Electrical Properties : 56. Conductivity (Coulter counter) . . Sieves : 57. WOVEN WIRE- British Standard specifications . . . . . . . . . . . . . . . . * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . RECENT TECHNIQUES (not yet standardised)- 58. Electroformed sieves . . . . . . . . 59. Air-jet sifting . . . . . . . . . . Permeability to Fluid Flow: 60.Carman method . . . . . . . . . . 61. Lea and Nurse method . . . . . . . . 62. Gooden and Smith method . . . . . . 63. Fisher subsieve sizer . . . . . . . . 64. Rigden method . . . . . . . . . . 65. Griffin surface area apparatus . . . . . . 66. Blaine fineness tester . . . . . . . . 67. Spillane method . . . . . . . . . . LIQUID FLOW- GAS FLOW- . . . . . . . . . . . . .. .. . . . * .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . 5to76 . . 10 to 76 . . 5 to 200 . . 5 to 100 . . 5to40 . . 2 t o 15 . . 5 to76 . . 0.5 to 30 . . 1 to 10 . . 0.7 to 20 . . 1 upwards . . 1 upwards . . 1 upwards . . 2 to 50 . . 1 to25 . . 1 to 5000 . . 1 to 5000 . . 0.001 to 20 . . 0.01 to 100 . . 0.002 to 0.1 5 to 50 . . 0.2 to 300 . . 53 to 3353 .. 16to 53 . . 53 to 3353 Page 173 174 174 174 174 175 175 175 175 176 176 176 177 177 178 179 179 179 180 180 180 180 181 181 181 182 182 183 183 183 184 184 . . . . . . 184 . . . . . , 184 . . . . . . 184 . . . . . . 185 . . . . . . 185March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 159 Method No. Method B.11. Adsorption Methods: STATIC- 68. Harkins and Jura absolute method (HJa) . . . . . . 69. Brunauer, Emmett and Teller method (BET) . . . . 70. Harkins and Jura relative method (HJr) . . .. . . 71. Dynamic capacitance method . . . . . . . . . . 72. Light beam technique . . .. .. . . . . . . 73. Penetrating radiation technique . . . . .. . . DYNAMIC- B.111. Absorption of Radiation: B .IV. Optical Microscope Measurement with Particles in Random Orientation: Random orientation of particles and projected area measure- ment .. .. .. .. . . . . . . . . 74. Page . . .. 186 . . .. 185 . . .. 186 . . . . 186 .. .. 187 . . . . 187 . . . . 187 METHODS FOR DETERMINING PARTICLE SIZE GENERAL INTRODUCTION POWDERS OF SOLID PL4RTICLES- The determination of the size distribution of particles in a powder is not as simple as it might at first sight appear. In addition to any experimental difficulties, there is the basic problem of defining both size and distribution. The variation in the answers obtained for the same powder when different definitions are used can be enormous, and therefore there should be a clear understanding of the definitions to be used before the size analysis is attempted. If powders contained only spherical particles there would be no difficulty in defining particle size, since the size of a sphere is uniquely determined by its diameter. But the particles of a powder are rarely spheres; usually they are irregular in shape and often far from spherical-sometimes being platelets or needle-shaped. Nevertheless, because of the work involved in sizing the large number of particles present in most powders, it is necessary for practical purposes to define the size of each particle by a single parameter.Since no definition can be fully satisfactory, it is usual to select the parameter most relevant to the problem under discussion. The difficulty in selecting a single parameter can be illustrated by a homely example. A standard brick measures 9 inches x 44 inches x 3 inches.In comparing the size of this with that of a non-standard brick, say 7 inches x 5 inches x 5 inches, the answer as to which is the bigger depends on the criterion laid down. For example, if the bricks are to be used as an edging to a path, the dimension that matters is the over-all length, then the standard brick is the bigger of the two. If the bricks are to be laid flat to form a pillar, then the larger brick as regards height of the pillar is the non-standard one. The volume or weight of the non-standard brick is also greater than that of the standard one, but the solid diagonal is less. Obviously, the parameter to be used depends on the purposes for which the bricks are intended. The same general arguments apply to all powder particles.If the particles are highly irregular, the problem is even more complex. Parameters often used are, for example, the maximum chord, the projected area, the distance between two parallel tangents and the maximum chord in a given direction. Many other parameters could easily be devised, each having its own value. For particles with the same shape and density, the free-falling velocity, an important measure in many industrial problems in which solids are mixed with liquids or gases, depends on the size of the particle. In such instances the diameter of the “equivalent sphere” is taken as a measure of size. The equivalent sphere is one with the same density as the particle, which also has the same free-falling velocity as the particle. If the diameter of160 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol.88 the equivalent sphere is calculated from the free-falling velocity by using Stokes’s law, this is usually referred to as the Stokes’s diameter. When the shape or density varies from one particle in a powder to another, the dimensional size is no longer the main controlling factor. The size distribution can be defined in terms either of the number or the weight of particles within a given size range. It is usually expressed either as the percentages in each of a series of size ranges or as the percentages of the powder under or over a series of specified sizes. The definitions of size and distribution adopted should therefore be selected first, accord- ing to the purpose for which the result is required, for example, whether the weight, covering power, free-falling velocity or some other parameter, is the most important in the problem and second, allied to this, on the particular form of instrument to be used to measure the size distribution.If the method is not well chosen the results may not be directly comparable with the particular property being investigated. In the foregoing tables-“Classification of Methods”-the size-determination technique is arranged first according to the parameter investigated and secondly according to the method. Finally, it should be appreciated that both the collection of a sample and also the act of measurement (for example, if this involves the use of ionising radiation such as X-rays) may alter the condition of the material. Fine particles tend to agglomerate into larger clusters that may not always be easy to distinguish as such or to re-disperse, before measurement, into the original individuals.Coarse particles are sometimes friable and break into smaller particles during collection or measurement. In all cases, and particularly when a powder is produced or used in the dispersed condition, considerable care should be given in inter- preting any of the results obtained by any measuring system. LIQUID DROPLETS- If the particles to be investigated are not solid but are in the form of liquid droplets, many methods do not apply and others may have to be modified. Sample collection is much more difficult and any method that allows the particles to touch and so coalesce is inadmissible. When collected on microscope slides or on an impervious surface, an extremely low density of droplets is necessary, since superposition must be avoided.The droplets will be distorted by contact with the surface, but correction for this may be made by determining the spread factor; since this will vary with droplet size it must be determined in the relevant size range. Collection on thin filter-paper may also be used and correction factors similarly determined. A more accurate method is to use relatively thick filter-paper or magnesium oxide or plaster of Paris blocks. In other techniques, droplets have been frozen in a stream of cold air before collection, and then treated as solid. Another ingenious way of solidifying drops is to allow them to fall into talc, wheat flour or other finely powdered material, and to determine the size of the lumps of agglomerated “dough” produced.Another method is to allow the drops to impinge on glass slides coated with magnesium oxide smoke from which the drops bounce, taking with them a piece of magnesium oxide and leaving a crater proportional to the size of the drop. This method is useful in the field, since the slides may be examined later at leisure, Some workers have caught water drops in a cell in which a density gradient has been produced by forming a carbon tetrachloride layer beneath light petroleum. Care must be taken to avoid too great a gradient so as to avoid distortion of the droplets; only a few droplets may be collected at a given time, since they may coalesce if they are too close.A combination of settling and collection on a moving belt of filter-paper has been used for accurate sizing, and the Conifuge (see (42)) has been used for similar work. SELECTION OF METHODS- Most of the methods given in this Classification cater for particles in the sub-sieve size range. Systems in which the individuals are resolvable by the naked eye are normally classified by sifting; the practical lower limits of woven-wire sieves is usually taken as 200 mesh -this corresponds to a nominal particle size of 76 p, but, in exceptional circumstances, this lower limit can be extended down to 300 mesh (53p). Often the naked eye or a low-power magnifying glass may be sufficient, but in general it is preferable to examine the material under a microscope. This examination should reveal the approxi- mate size range and distribution of the powder, its uniformity of composition, the shapes of the particles and the presence of aggregates.All these factors may influence the further Most methods supply only one parameter. Before beginning a size determination it is essential to look at the powder.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 161 steps to be taken. Of particular importance is the presence or otherwise of aggregates; it is necessary to know if they are present and in the same manner in the sample as in the original form. If information on this point is not available, this fact should be reported. The presence of aggregates may mean that these have to be broken down into their constituent parts or, on the other hand, that they must not be broken down because they are an essential feature of the material.Usually, the individual particles or aggregates have to be separated from their fellows in order to permit individual sizing. This is necessary when a microscope is being used, so as to avoid overlapping or obscuration of particles. In sedimentation, it is necessary to ensure that each particle behaves as if it were alone, since increasing concentration causes the mass to behave as a whole, not as individual particles. For each of the sizing techniques there is an appropriate method of preparation and a range of concentrations; these must be carefully observed. When deciding which method of sizing shall be used for a given problem there are both theoretical and practical considerations to be taken into account.From what has been said earlier, it is clear that the first and most important consideration must be the purpose for which the information is required. Knowledge of free-falling speeds, as in estimating the behaviour of powders in classifying or separating machines, can best be obtained from a sedimentation or elutriation technique ; covering power of a powder is related to the projected area size as seen, for example, under the microscope; other types of required behaviour are represented most nearly by the appropriate size, determined, possibly, by other techniques. When the most desirable type of technique has been selected, the choice can be narrowed by finding the instrument in this class that gives the necessary accuracy, covers the range of sizes of interest and is appropriate to the problem.Sedimentation, for example, requires a test fluid, and it may be difficult to find one that is both suitable for the apparatus and does not affect the powder. To take another example, light or X-rays can be used to size particles but are inappropriate for, say, silver bromide grains, since during the course of the test they would alter the nature of the material. Lastly, but by no means least, the availability of appropriate equipment is important. This may be the controlling factor in selection of a method, since only a single instrument may be available. It must then be fully appreciated that the “size” determined by this instrument is not necessarily the true size. In some instances the result may be correct; in others, a simple correction factor, say 4 2 , may produce a close approximation to the correct answer; in a number of instances, however, the determined “size” may bear only a vague relation to the one needed, and may be completely misleading.Such methods, if reproducible, may nevertheless be of value for process control purposes. In most methods of sizing, the powder has to be prepared in a suitable way. A.I. RELATIVE MOTION BETWEEN PARTICLES AND FLUID SEDIMENTATION PROCESSES Sedimentation analysis is a process in which the rate of settlement of particles in a static fluid is measured, the analysis being either with the particles uniformly dispersed throughout the whole of the fluid, or introduced at the top of the column, as in (5) and (10).The rate of sedimentation usually follows Stokes’s law, and settlement may be under conditions of gravitational acceleration or under centrifugal acceleration, imposed by rotation, in order to increase the rate of sedimentation. Stokes’s law is valid only in the regime of viscous flow, which sets an upper limit to the size of particle that can be tested by this means in a given fluid. The limit is determined by the magnitude of Reynolds number, which should not exceed 0.2 if the error when Stokes’s law is used is not to exceed 5 per cent. The con- centration of the suspension should always be as low as possible, except for method (lo), in order to avoid interference between particles, particularly since “clouds” of particles tend to move en masse and not individually.The method of measurement, e.g., by weighing, may dictate a lower limit of concentration beyond which inaccuracies become unacceptably high. Effective dispersion of the particles is an essential pre-requisite for all methods of sedi- mentation analysis. It is often necessary not only to stir the suspension vigorously, but to introduce the powder already mixed with a dispersing agent. A further agent to prevent162 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol. 88 flocculation may be needed, and it may be necessary to adjust the pH of the suspension to a suitable value determined by experiment. A suspension uniformly dispersed in an upright cylindrical vessel at time t = 0 begins to settle immediately. If two horizontal planes, @, and P,, are taken at depths h, and h, below the upper surface (h, > h2), then, at time t, the mass of suspension remaining between p , and p , will be determined by- (h, - h,) C plus particles that have fallen through @, minus particles that have fallen through P,, where C = mass per unit depth of original suspension, For incremental methods of analysis, the sample is theoretically taken from an extremely thin layer in which h, is nearly equal to h,; €or cumulative methods of analysis, the amount settling in a particular plane in a given time, normally at the bottom of the settling vessel, is measured.REFERENCES- Heywood, H., “Recent Developments in Mineral Dressing,” Institute of Mining and MetaZZurgy, 1952, Stairmand, C. J., “Symposium on Particle Size Analysis.” Supplement to Trans.I. Chem. E., 1947, - Donoghue, J. K., Brit. J. Appl. Phys., 1956, 7, 333. British Standard 3406: Part 2: in the press. p. 31. 25, 128. Pipette Methods In this technique a sample is extracted from the sedimenting suspension at appropriate intervals by means of a pipette. With the exception of the Whitby Column Method (5), these methods are all incremental. The sample is taken in one of two ways: (a) at a fixed position in the apparatus or (b) at a fixed depth below the surface of the suspension. It is assumed in both instances that no disturbance of the suspension takes place by eddies, etc., while the sample is being taken, that the sa.mple is representative of the suspension at the extraction point and that the sample taken is small.Method (a) must take into account any lowering of the level of the top surface of the suspension. (1) ANDREASEN PIPETTE- This is probably the most frequently used method of sedimentation analysis, the disadvantage being that samples withdrawn must be evaporated to dryness and weighed -a time-consuming procedure. The glass sedimentation vessel is about 5.5 cm in diameter and has a graduated scale, 0 to 20 cm, engraved on its side. The zero of the scale is positioned about 2.5 em from the base of the vessel, and the capacity when fiIled to the 20-cm mark is 550 to 600 ml. The stem of the pipette is fused to a bell-shaped dome having a ground-glass joint that fits the neck of the sedimentation vessel, the pipette being so positioned that its tip is fixed at the level of the zero mark of the scale; above the dome is a two-way tap and a side discharge tube.During an analysis, the sedimentation vessel is immersed in a constant-temperature bath up to the 20-cm mark. The vessel is filled with suspending liquid and powder to the 20-cm mark, shaken, and placed upright in the bath. At time intervals standing in a 2 to 1 progression, a 10-ml sample is withdrawn from the sedimentation vessel and is discharged into a series of tared dishes. The dishes are reweighed after the samples have been evaporated to dryness, and a deduction is made for the weight of dispersing agent added; thus, the weight of particles corresponding to each withdrawal time is determined. The size of particles is calculated from the height of fall and time elapsed, according to Stokes’s law, but allowance must be made for the decrease in the height of sedimentation column after each sample is withdrawn.The initial concentration in the suspension is calculated from the weight of powder and the volume of liquid in which it has been dispersed. This concentration may be up to 1 per cent. by volume for easily dispersed powders, but should be reduced if there is any tendency for flocculation to occur.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 163 SIZE RANGE- The apparatus is normally applicable to the particle size range 2 to 60p, and may be extended to larger sizes if viscous liquids are used, e.g., benzyl alcohol, or to smaller sizes if a constant temperature is maintained over the longer period required.One operator can perform four analyses simultaneously during a day. The weight of powder required is approximately 5 g. REFERENCES- Andreasen, A. H. M., Kolloid-Beih., 1928, 27, 349. -, IngenVidensk. S k y . , 1939, No. 3. Heywood, H., Proc. Soc. Mech. Eng., 1958, 140, 257. British Standard 3406 : Part 2 : in the press. Fixed position pipette method- (2) FIXED-DEPTH PIPETTE (B.S. 1377)- The apparatus consists of a 10-ml pipette fitted with a two-way tap and a washing bulb. The pipette is mounted on a sliding panel so that it can be lowered to any depth in the sedimentation vessel. The general method of procedure is similar to that of (1), except that the pipette is removed from the suspension after each sample has been withdrawn and is re-immersed to a fixed depth below the surface of the suspension liquid for the next sample.An advantage over the Andreasen pipette is the facility for washing the pipette with clear liquid to remove any deposited particles. SIZE RANGE- This is similar to (1)-2 to 60 p. of powder required is approximately 5 g. The operating time is also similar, and the weight REFERENCES- British Standard 1377 : 1948. British Standard 3406: Part 2: in the press. Fixed depth pipette method- Sedimentation Columns : (a) Liquid In these methods the powder may be initially dispersed in the sedimentation liquid or added, effectively instantaneously, at the upper free surface. The rate of fall of the powder is measured as the rate at which powder reaches the bottom of the column. I t may be physically removed from the column, if this has an openable base, or measured as the separating sediment in a column with a closed bottom.(3) LIQUID COLUMN METHODS WITH SEDIMENT EXTRACTION- The sedimentation vessel is of cylindrical form with a conical base leading to &-inch bore glass tube; just below the cone a branch tube, &-inch bore, joins the vertical tube at right angles and is connected to a liquid reservoir. The sedimentation vessel is filled with the liquid and dispersed powder, which is stirred until the beginning of the test by compressed air admitted through the outlet tube at the bottom. Immediately after the start of sedimentation, a centrifuge tube is connected to the outlet tube from the sedimentation vessel by means of a short length of rubber tubing fitted with a spring clip.Settled particles collect above this clip, and, at suitable time intervals, usually about 1, 2,4, etc., minutes from the beginning of the analysis, these particles are run off into the centrifuge tube, which is replaced by another tube in readiness for the next extraction. The liquid withdrawn is replaced from the clear-liquid reservoir, so that the height of the suspension column remains constant. The particles collected in the tubes are compacted in a centrifuge, washed if necessary, dried, and weighed to determine the weight of particles collected. The size of particles corresponding to each time interval is calculated by means of Stokes’s law, and the corresponding weights are determined by a tabular method of calculation, which allows for the fact that total sediment has been collected.It is essential for the validity of this calculation that the time interval be exactly in a geometrical progression of ratio 2 to 1.164 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol. 88 The apparatus is applicable to the size range 3 to 76 p, and the particle concentration The weight of powder required is 0.5 g or less. SIZE RANGE- may be of the order of 0.1 per cent. by volume. REFERENCES- Stairmand, C, J., “Symposium on Particle Size Analysis.” Supplement to Trans. I . Chew. E., 1947, 25. 128. British Standard 3406: Part 2: in the press. Liquid column method with sediment extraction. (4) B.C.U.R.A. SINGLE TUBE- The form of the tube is similar to that described under (3), with some detailed alterations. The conical lower portion contains no side tube, and the tip of the cone is ground away to leave a hole about 1 mm in diameter.When the tube is filled with liquid and the upper end closed by the tap, the liquid is held in the tube by the meniscus across the small hole at the bottom. When clear sedimenting liquid, contained in a small glass bucket, is brought into contact with the meniscus, this breaks down and sediment passes into the bucket. Removal of the bucket, after the appropriate time interval, takes away the sediment, but the sealing meniscus again forms. The process is repeated for the number of samples required. Initial dispersion of the powder by compressed air blown through the liquid, the treatment of the sediment samples and the final calculations are essentially similar to those for (3).SIZE RANGE- Any sediment collects at this hole. This is similar to (3)-3 to 76p. REFERENCE- Kobak, J., and Loveridge, D. J., J . Sci. In,;trum., 1960, 37, 266 (5) WHITBY COLUMN METHOD- The main features of this method are that it uses gravitational settlement followed by centrifugal sedimentation, and the sample suspension is superimposed on a column of clear liquid so that all the particles begin to settle from the same level, provided that the sample layer is thin. The suspending medium is chosen to have a density slightly less than that of the sedimentation liquid. The amount of solid sample required for the analysis is a few milligrams. The apparatus consists of a glass centrifuge tube, 14mm internal diameter, ending at the bottom in a graduated capillary tube, 1 mm internal diameter.The length of the upper portion is about 65mm, and the capillary, which is 30 mm long, is sealed to the upper i4-mm diameter tube by a smooth glass connection. The over-all length of the apparatus is approxi- mately 125 mm, including the junction. A feed-tube is provided for the introduction of the sample. This is a short metal tube that fits easily into the mouth of the glass centrifuge tube. This feed-tube is fitted with a collar that restricts its entry into the centrifuge tube to a depth of 10 mm. In operation, the centrifuge tube is filled with the sedimentation liquid to within 10 mm of the top. The feed-tube is then filled with the sample suspension and the upper end is closed by the operator’s finger. The feed-tube is then placed in the centrifuge tube with the suspension in contact with the column of clear sedimentation liquid; this starts the sedimen- tation.Gravitational settlement is allowed to proceed for 4 to 10 minutes, according to the particle-size range of the sample. The tube is then transferred to a suitable centrifuge, and a series of readings is taken at known intervals at several measured centrifuge speeds between 300 and 4000 r.p.m. The amount of material that has settled in the capillary tube is measured after each experiment by reading the volume of the sediment in the capillary tube. The relative volume is a direct measure of the amount of material having a particle size greater than that calculated from the time of settlement and the gravitational force applied.A feature of the method is that, since the suspension medium is less dense than the sedimentation liquid, an unstable density gradient is avoided, and the particles begin to settle according to their relative terminal velocities. A 40-mesh wire gauze is soldered across the bottom of the metal tube. The finger is then removed.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 165 SIZE RANGE- REFERENCE- 0.05 to 100 p. TZ’hitby, K. T., ‘‘A Rapid GeneraI-purpose Centrifuge Sedimentation Method for Measurement of Size Distribution of Small Particles,” A.S.H.A.E. J . Heating, Piping G. Air Conditzoning, 1955. Sedimentation Balances These instruments, sometimes referred to also as sediment accumulation devices, weigh the sediment as it accumulates on a weigh-pan at the base of the sedimentation column.The methods are cumulative ones. The powder may be dispersed initially in the bulk of the fluid or added instantaneously at the top. An advantage of this type of equipment is the absence of the conical base, needed in sediment extraction devices, upon the walls of which some sediment ma!‘ stick. The danger of particles sticking to the vertical walls is however still present. (6) ROSTOCK SEDIMEXTATIOS BALAXCE- The sedimentation vessel is cylindrical, water-jacketed and surmounted by a cylindro- conical pre-mixing and feed vessel. Below the open end of the sedimentation vessel is a balance pan supported by a torsion wire with an extended pointer, scale and optical magnifying system, such that 0.5 g of the material on the balance pan will cause full-scale deflection of the system. The liquid and dispersed powder are introduced into the sedimentation vessel via the feed vessel, and readings of the torsion balance (which represent the weights of particles settled on the pan) are taken at specified time intervals. The weights of particles corre- sponding to the various particle sizes, calculated from the time intervals, are determined by the tabular method as for (3) or by a graphical method given by references below.S I Z E RA4NGE- 3 to 76 p. The weight of powder required is 0.5 g. REFERENCES- Rostock, W., .J. Sci. Instrum., 1962, 29, 209. Ilonoghue, J . K., Brit. I . AppZ. Phys., 1956, 7, 333. British Standard 3406: Part 2: in the press.(7), (8) SARTORIL’S AXVD SHIMADZU SEDIMEXTATIOX BALAXCES- These instruments are automatic balances of the beam type. The balance pan is hung in a suspension of appropriate concentration, and the balances record the change in weight as powder settles on the pan. When the deflection of the beam reaches a certain point, corresponding to a given weight increment, the beam is brought back to equilibrium and the recording pen moves an appropriate distance on the weight axis of the chart. As counter- poise, the Sartorius balance has an auxiliary torsion beam and the Shimadzu balance has small steel balls. The instruments produce a step-wise record; a curve is then drawn through the tops of the steps and the cumulative size distribution is determined by the Oden tangent method.In order to apply this method, it is necessary to know the amount of material settling at infinite time (i.e., the total amount of material in suspension). With the Sartorius balance this may be determined directly, provided the material is not too fine, or may be calculated from the weight of the sample by applying an empirical correction for the material that escapes the balance pen. With the Shimadzu balance, it is recommended that the “un- sedimented powder” should be estimated from a sample of suspension siphoned off at the end of the experiment and added to the final weight of sediment to give the amount at infinite time. Provision is made in both instruments for varying the time scale. SIZE RANGE- REFERENCES- 1 to 60 p.Oden, S., Intern. Mitt. Bodenk., 1915, 5 , 257. Oden, S., in Jerome Alexander, Editor, “Colloid Chemistry,” Chemical Catalogue Co., New York. Rose, H. E., and Langmaid, R. N., Nature, 1957, 179, 774. 1926, p. 861.166 ANALYTICAL METHODS COMMITTEE : Sedimentation Columns : (b) Gas (9) MICROMEROGRAPH- [Analyst, Vol. 88 This commercial instrument is operated on the principle of sedimentation of particles in a static column of air. The powder is dispersed and injected at the top of the column by compressed nitrogen, so that all particles are assumed to begin settling from the same level. The settled particles are collected on a balance pan at the base of the air column. The weight of particles is recorded by an automatic servo-controlled, electrically-operated torsion balance, and a curve of weight of settled particles against time is obtained; this can easily be converted to relate weight and particle size.The sample weight required is 50 to 100 mg. SIZE RANGE- REFERENCE- 0.8 to 250 p. Eadie, F. S., and Payne, R. E. Bvit. Chem. Eng., October, 1956. Sedimentation Columns : (c) Hindered Settling (10) HINDERED SETTLING METHOD- In suspensions with a concentration of more than about 1 per cent. by volume, the particle fall becomes progressively hindered until, a t a concentration of about 18 per cent., the whole suspension settles en masse. A clear boundary forms between the settling suspension and the fluid draining out of the slurry. The rate of settlement of the boundary can be related to surface area of the powder by considering the situation to be the flow of fluid through a packed bed rather than the settling of particles.The mean specific surface diameter can be calculated from the formula- where d, = P = Po = g = E L . = uo = The method of the suspension € = mean specific surface diameter; porosity of suspension ; density of particles; density of fluid; acceleration due to gravity; viscosity of fluid; velocity of boundary. is easy to apply, since only two measurements are required-the porosity and the velocity of the boundary. It also has the advantage that a large sample of*powder is required so that sampling error is minimised. SIZE RANGE- REFERENCE- 5 to 1000 p. Orr, C., jun., and DallaValle, J. M., “Fine Particle Measurement,” The Macmillan Co., New York & London, 1959.Density Variation: (a) Gravimetric (11) HYDROMETER METHOD- The suspension of a powder is prepared as for other sedimentation analyses and poured into a 1-litre glass measuring cylinder. The variation with time of effective density,and hence of particle concentration, is determined from readings of a hydrometer immersed in the suspension. The hydrometer should be inserted in the suspension immediately before a reading is taken, and afterwards removed to avoid particle deposit on the bulb. The effective height of the suspension is from the surface level to the centre of volume of the bulb, which varies according to the density of the suspension. As the bulb length is an appreciable proportion of the sedimentation height, and not very small, as explained in the IntroductionMarch, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 167 to section A.1 (see p.l 6 l ) , the incremental method of calculating the size analysis is approxi- mate only. The method is frequently used for soils, and full descriptions can be found in any recognised text-book on this subject. SIZE RANGE- REFERENCE- Temperature and meniscus corrections should be applied. 1 to 25 p. British Standard 1377 : 1948. Methods of test for soil classification and compaction. (12) DIVER METHOD- This is a modification by Berg of the hydrometer method (11). Variation in effective density, and hence concentration, is measured by totally immersed divers. These are small glass vessels of approximately stream-line shape, ballasted to be in stable equilibrium, with the axis vertical, and to have a known density slightly greater than that of the sedimentation liquid.As the particles settle, the diver moves downwards in hydro-dynamic equilibrium at the appropriate density level. The diver indicates the position of a weight concentration equal to the density difference between the diver and the sedimentation liquid. Several divers of various densities are required, since each gives only one point on the size-distribution curve. The advantages claimed for this method over the hydrometer method are that, since the divers are relatively small and surface tension and deposition effects are avoided, the incremental theory is applicable. The method is simple to operate, and accurate when a set of divers has been prepared.Smaller divers can be used with centrifuge tubes, so that analysis can be extended down to particles of 0.02 p in size. A modification by Jarrett and Heywood permits the position of the diver in an opaque suspension to be determined electronically . Particles deposited on the upper surface of a spherical diver produce a state of unstable equilibrium such that a slight displacement causes the diver to rotate about a horizontal axis and shed some of the particles. This is an advantage in that the weight of the diver remains nearly constant. SIZE RANGE- REFERENCES- 0.02 to 40 p. Berg, S., IngenVidensk. Sky., 1940, No. 2 ; A.S.T.M. Special Technical Publication No. 234, 1958, Jarrett, B. A., and Heywood, H., Brit. J . Appl. Phys., Supplement No. 3, 1954, S21.p. 143. Density Variation : (b) Absorptiometric The presence of a suspension of particles in a cell through which a beam of light or other radiation is passed causes a reduction in intensity of the emergent radiation. This principle is used to follow the changes in particle concentration at various depths below the surface in a sedimentation cell. (13) “E.E.L.” PHOTOSEDIMENTOMETER- Light from a 36-watt lamp is projected through a system of lenses and stops, so that a beam of extremely small solid angle (0.00022 radian) is subtended in a position occupied by a sedimentation cell. A barrier-layer photo-cell receives the transmitted light; the output of this cell is passed to a galvanometer. The optical components are mounted on a movable carriage that can be traversed so that any one of six cells may be examined, the cells remaining stationary.One cell contains clear suspending medium, and the rest may contain five different samples, each dispersed in the medium. Readings of the galvanometer, in respect of each cell, are taken at fixed time intervals. SIZE RANGE- 1 to 50 p.168 ANALYTICAL METHODS COMMITTEE [Analyst, Vol. 88 (14) “BOUND BROOK” PHOTOSEDIMENTOMETER- This instrument employs a single light source sending a beam to two matched photo-cells. The sedimentation cell is placed in one of the optical paths and a reference cell of clear liquid in the other. The two photo-cell signals are passed to an automatic pen recorder consisting essentially of a slide wire and a resistor in series, the two signals being continually balanced and the ratio of the signals recorded, thereby preventing errors consequent on light source variation.SIZE RANGE- 1 to 60 p. Radiometric Techniques The specialised methods of measurement developed for use with radioactive tracers and other active materials are highly sensitive. They may be applied to particle-size analysis and are well adapted for use with sedimentation processes. (i) Penetrating radiation-Penetrating radiation has the merit of being effective when dispersing media opaque to visible light are being used. The types of radiation referred to below are of much shorter wavelength than visible light and have high energy and great penetrating power-extending to several inches of steel. (ii) Electromagnetic radiation-X-rays and gamma-rays form part of the spectrum of electromagnetic radiation and are propagated with the velocity of light.The usual source of X-rays is a tube in which a target is bombarded by high-velocity electrons. The type of radiation is continuous, with superimposed high-intensity lines characteristic of the target element. I t has a short wavelength limit related to the maximum electron velocity. Gamma- rays have a line-type spectrum ; they are conveniently available from radioactive isotopes and haire a constant quality, although the intensity diminishes with time. (iii) Atomic particles-Alpha-rays are high-speed helium particles travelling with up to one-sixteenth of the velocity of light ; beta-rays are electrons travelling with velocities approaching nine-tenths of that of light.Both these types of radiation are available from radioactive materials : alpha-particles carry a positive charge, electrons a negative charge. Neutrons, which are available from reactors, particle accelerators and special compact sources, carry no charge; “thermal” neutrons are most likely to be useful in particle-size determination, and have an equivalent wavelength of a few Angstroms. The material under examination may be initially radioactive or may have radioactivity induced in it by irradiation or by incorporation of tracers. Alternatively, radiations may be used as an external analytical probe. The techniques offer several advantages; some of these are- (1) Very high sensitivity permits the use of extremely low particle concentrations. (2) A settling system can be studied remotely without introducing disturbances asso- ciated with sampling devices.(3) Radiometric techniques are convenient for use in conjunction with a centrifuge for sizing particles down to 0-1 p-a range in which optical methods are not practical. (4) Sedimentation can be followed in opaque systems, such as liquid metal slurries. ( 5 ) The sedimentation rates of components of mixtures can, in favourable instances, be followed concurrently. (15) I.C.I. METHOD, INVOLVING USE OF X-RAYS- This method is suitable for material that is chemically homogeneous. An X-ray beam is split into two separate beams, each of which ultimately falls on to a device for converting X-rays into longer wavelength radiation, such as visible light. Fluorescent screens or crystal scintillators are suitable ; the longer wavelength radiation is conveniently monitored for inten- sity by a photo-electric device.The use of X-rays has the advantage over light methods in that absorption is directly related to the mass of a larger variety of absorbing material, with consequent increase in sensitivity.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 169 I t is thus possible, by the X-ray method, to complete a particle-size analysis of material such as fly-ash (size range 7 to 76 p) in about 30 minutes, whereas the normal sedimentation method takes 3 to 4 hours. The method also has the advantage of a direct continuous measure of the concentration of the settling powder with no disturbance of the suspension.Besides fly-ash, substances that have been examined include iron oxide and bismuth oxide, in concentrations between 0.2 and 0-3 per cent. by weight (or 0.15 and 0.04 per cent. by volume). SIZE RANGE- REFERENCE- 1 to 76 p. Brown, J. F., and Skrebowski, J. K., discussions associated with Jarrett and Heywood; “The Physics of Particle-size Analysis” (Institute of Physics, Nottingham, 1954, Supplement No. 3, S21). (16) INCREMENTAL SEDIMENTATION (RADIOMETRIC)- The technique consists in measuring the changing gamma-activity of a cross-sectional lamina of a sedimenting gamma-active suspension. The gamma-activity is directly propor- portional to the mass of active material in the lamina, which is defined by a collimating slit in a lead shield surrounding the sedimentation tube and a scintillation counter.A cumulative distribution curve of the mass-fraction of undersize particles is derived from the measurement of gamma-activity at appropriate time intervals. SIZE RANGE- REFERENCES- 4 to 76 p. Abraham, B. M., Flotow, H. E., and Carlson, R. D., Anal. Chew., 1957, 29, 1058. Connor, P., and Hardwick, W. H., “The Use of Radioactivity in Particle-size Determination,” Indust. Chemist, 1960, September. (17) BEAKER-CENTRIFUGE (RADIOMETRIC)- The activity of the tracer is induced by neutron irradiation. Portions of the irradiated suspension are placed in calibrated centrifuge tubes, and each sample is spun in a centrifuge for a pre- determined time. Portions of the supernatant liquid are removed and counted with a gamma- scintillation counter so that the percentage concentration of undersize particles in the sus- pension may be calculated. The method is stated to be sensitive to particles down to 0.05 p in size.The sedimentation rates of the constituents of a mixture of several particulate materials can be determined if neutron irradiation of the constituents produces a set of gamma-emitters that can be differentiated by gamma-spectroscopy, for example, by using a kick-sorter. A mixture of substances containing chromium, iron, zirconium and uranium has been size analysed by this method. SIZE RANGE- REFERENCES- A tracer technique is used in conjunction with Martin’s centrifugal method. 0.05 to 76 p. Martin, S. W., I n d . Eng. Chern., Anal. E d . , 1939, 11, 47. Bate, L. C., and Leddicote, G.W., Oak Ridge National Laboratory Microcard 57-1-116. (18) TRACERS IN ANDREASEN AND LONG-ARM CENTRIFUGAL PIPETTES- Low, and therefore free-settling, concentrations of thoria traced with protactinium-233 have been size analysed with an Andreasen pipette (1) and also Gupta’s long-arm centrifugal pipette. A fast and accurate procedure was evolved in which portions of the sample were transferred by pipette directly into flat-bottomed plastic centrifuge tubes, which were then spun in a centrifuge to ensure deposition of the suspended solid on the tube base. The activity of the sediment was determined by means of a gamma-scintillation counter. The measurements take the place of the sediment weights usually derived by evaporation to dryness of withdrawn samples.170 ANALYTICAL METHODS COMMITTEE : SIZE RANGE- 0.1 to 76 p.[Analyst, Vol. 88 REFERENCES- Andreasen, A. H. M., Kolloid-Beih., 1928, 27, 349. -, IngenVidensk. Skr., 1939, No. 3. Gupta, A. K., J . Appl. Chem., 1959, 9, 487. Connor, P., Hardwick, W. H., and Laundy, H. J., Indust. Chemist, 1960, September. (19) TWO-LAYER RADIOMETRIC SEDIMENTOMETER- It was developed by Connor, Hardwick and Laundy for determining the size distributions of thoria and other materials in the range 0.1 to 5 p. A few milligrams of thoria powder are dispersed in a dilute solution of trisodium polyphosphate in a small polythene bottle. The bottle, with contents, is irradiated in a flux of 1.47 x 10l2 neutrons per sq. cm per second for a few seconds, so that virtually the only effect is to produce beta-active thorium-233 with beta-energy of 1.23 MeV, After further agitation of the suspension, a portion containing about 2 mg of thoria is transferred, as in the Whitby Column method, by a special pipette on to the surface of a 9-cm column of concentrated sucrose solution contained in a tube machined from solid Perspex.The flat base of this tube is made thin (0.015 inch) to permit beta-radiation to pass through for recording by a beta-scintillation counter. The activity, and hence the relative mass of the sediment, is measured after various periods of centrifugation. A size distribution is easily determined from the results, because all the particles start from the same height. The method has been successfully applied to silica, lithopone and poly(viny1 chloride), the active isotopes being silicon-31, barium-139 and chlorine-38, respectively.SIZE RANGE- REFERENCES- This is an adaptation of the Whitby Column Method (5). 0.1 to 5 p. Marshall, C. E., Pvoc. Roy. Soc. A . , 1930, 126, 427. Whitby, K. T., A.S.H.A.E. J . Heating, Piping & A i r Conditioning, 1955, 27, 139. Connor, P., Hardwick, W. H., and Laundy, B. J., J . AppZ. Chem., 1958, 8, 716. (20) BETA BACK-SCATTERING SEDIMENTOMETER- In this sedimentometer developed by Connor, Hardwick and Laundy, an external source of beta-radiation is used. The mass of sediment on the bottom of a tube is determined by measuring the amount of radiation that it scatters back from a l-millicurie strontium-90 radioactive source. A special source holder and sedimentation tube are used in a fitting designed to maintain a consistent geometrical arrangement of these parts.For fine particles, less than 5 p in size, a centrifuge tube may be used to accelerate settling-a short settling height or a long-arm centrifuge being used. Since the penetrating power of beta-rays is relatively low, the shielding requirements are modest. A curve of cumulative weight deposited against the logarithm of the settling time is obtained. The size distribution is calculated by application of Gaudin's modification of an expression derived by Oden- d P w=p-- where P is the fraction deposited in time t and W is the fraction of material greater d ( W ) in size than the Stokes's diameter corresponding to t. The value of - dP is obtained by differentiating the curve.The method of drawing d ( W ) tangents (see (7) and (8)), the accuracy of which has been discussed by Donoghue, or numerical methods may be used. SIZE RANGE- 0.1 to 76 p.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE REFERENCES- Connor, P., Hardwick, W. H., and Laundy, B. J., J . Appl. Chew., 1959, 9, 525. Oden, S., Pvoc. Roy. SOC. Edin., 1915, 36, 219. Donoghue, J. K., Brit. J . Appl. Phys., 1956, 7, 333. Gaudin, A. M., Schuhmann, R., and Schlechter, A. W., J . Phys. Chem., 1942, 46, 903. (21) GAMMA-RAY ABSORPTION- 171 A sedimentometer making use of X-ray absorption has been described in (15). By using a stable gamma-ray source, an instrument independent of supply-mains fluctuations can be constructed. Ross has developed a method in which americium-241 is used as a source of gamma-rays.The strength of the source was 150 millicuries and the energy of the radiations corresponds to 60 keV, with a half-life of 470 years. SIZE RANGE- REFERENCE- 0-1 to 76 p. Ross, C. F., Anal. Chem., 1958, 31, 337. (22) BETA-RAY ABSORPTION- Connor and Hardwick have proposed a method of measuring the sedimentation rate of fine powders from a gaseous suspension. Their beta-sedimentometer incorporates a Perspex centrifuge tube in the extended base of which is placed a disc of organic phosphor. Above the disc is an aluminium shielding ring surmounted by a ring-shaped beta-ray source, both of these being let into the tube wall. A deposit on the base of the tube absorbs some of the beta-radiation that would otherwise reach the phosphor.By using a photomultiplier in conjunction with the phosphor, such decrease of activity can be recorded and related to the mass of the deposit. SIZE RANGE- REFERENCE- 0.1 to 5 p. Connor, P., and Hardwick, W. H., Indust. Chemist, 1960, September. Sedimentation Rate (23) PHOTOGRAPHY OF PARTICLE TRACK- The particles in extremely dilute concentration are allowed to settle in a glass cell and are powerfully illuminated by a light beam at right angles to the axis of an optical photographic system. The illumination is limited t o a definite plane perpendicular to the optical axis, and is restricted to a specific exposure time. The individual particles are photographed as tracks whose length is proportional to the sedimentation velocity. These tracks are measured on the photographic plate and the number - size relationship is determined for the particles.SIZE RANGE- REFERENCE- 0.5 to 300 p. Carey, W. F., and Stairmand, C. J., Trans. Inst. Chem. Eng., 1938, 16. Fractional Decantation Methods (24) BEAKER METHOD- The simple beaker method of separating a powdered material into fractions limited by selected particle sizes can be effective if skilfully performed. An appropriate amount of powder is dispersed in liquid in a tall beaker or large measuring cylinder and allowed to settle for a time corresponding to the smallest separating size selected. At the end of this time, the liquid is decanted without disturbing the collected sediment. Ideally, all particles removed in this liquid would be smaller than the selected size.Since many of these smaller particles remain in the sediment, an infinite number of decantations is required to effect complete separation. It has been shown by Heywood that a 90 per cent. separation of particles smaller than the selected size is given by 11 decantations.172 ANALYTICAL METHODS COMMITTEE [AnaZyst, Vol. 88 When the first stage of separation is regarded as sufficient, the process is repeated with a longer settling time according to the next selected size, and so on. The particles are collected and weighed. This method of separation cannot be regarded as an accurate sizing process, but it pro- duces fairly large amounts of approximately graded material. The procedure is improved if the liquid is decanted through a siphon tube, or as described for the apparatus in (25), (26) and (27).SIZE RANGE- REFERENCE- 10 p upwards. Heywood, H., “Recent Developments in Mineral Dressing,” Inst. Min. Met., 1952, 31. (25) C.P.A.C. METHOD- In the examination for suspensibility of water-dispersible powders, the method specifies that a suspension of known concentration is prepared under prescribed conditions with stan- dard hard water. If the manufacturer of the pesticide product recommends creaming when the material is prepared for use, then a sample in a beaker is mixed into a smooth paste with hard water before dilution to 250 ml to produce the suspension; when creaming is not men- tioned in the manufacturer’s instructions, a sample is added to hard water in a beaker and stirred to produce a similar suspension of 250m1.This suspension is placed in a glass-stoppered 250-ml measuring cylinder in which the distance between the 0- and 250-ml graduations is between 20 and 21.5 cm. The cylinder is kept at a thermostatically controlled temperature for the prescribed time; at the end of that time the top nine-tenths of the suspension is drawn off. The content of active agent in the bottom one-tenth is determined and hence that of the top nine-tenths can then be calculated, and the percentage suspensibility determined. SIZE RANGE- REFERENCE- 2 to 76 p. Collaborative Pesticides Analysis Committee of Europe (C.P.A.C.). Recommended Methods of Analysis for Pesticides, No. 1 l-“Determination of Suspensibility of Water-dispersible Powders.” F.A.O. Plant Protection Bulletin (in the press).(26) W.H.O. METHOD- This method is similar to the C.P.A.C. method (25), except that preparation of the sample by creaming is not included. SIZE RANGE- REFERENCE- (27) MINISTRY OF AGRICULTURE, FISHERIES AND FOOD METHOD- This method is used in the determination of the particle size distribution of free sulphur in agricultural chemicals. The general principles are as for the Beaker method (24), but the sedimentation vessel is a glass tube, 5.5 cm internal diameter and 25 cm long, fitted with a ground-glass stopper at the top and a tube outlet, with a tap, leading from the bottom. A side-tube with a tap is fitted near the bottom of the vessel; this has an upward-pointing nozzle, 1 mm in diameter, on the axis of the tube and 5 cm distant from the bottom.A weighed sample is dispersed in 1 litre of water and sufficient of the resulting suspension is poured into the sedimentation tube up to the graduation mark 20 cm above the top of the side-tube. The suspension is allowed to settle for a specified time at 22” to 25” C; the top 20 cm is run off through the side outlet-tube and its free sulphur content determined by the prescribed method. With a settling time of 166 minutes, the percentage distribution of sulphur particles of 6 p or less is determined; that of particles of 2 p or less is given with a settling time of 27 hours. SIZE RANGE- 2 to 76 p. “Specifications of Pesticides,” World Health Organisation, Geneva, 1961. 2 to 76 p.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE REFERENCE- Gray, J.R., J . Sci. Food Agric., 1956, 7, 3. M.A.F.F., Technical Bulletin No. 1, “Specifications for Pesticides,” Third Edition, 1958. 173 Centrifugal Methods (28), (29) SHARPLES AND DIETERT CENTRIFUGE METHODS- The Sharples instrument is a supercentrifuge that has been adapted for very small The Dietert instrument is a centrifugal classifier useful for production control of powdered particles, such as tobacco-mosaic virus (i.e., approaching molecular dimensions). materials when extremely small particles are involved. SIZE RANGE- REFERENCE- 0.05 to 50 p. Schachmann, H. K., J . Phys. Colloid Chem., 1948, 52, 1034. (30) KAYE DISC CENTRIFUGE- In this technique particles are spun in a centrifuge in a cylindrical transparent tank in which there is an entry port. The centrifuge is run up to speed partially filled with clear liquid. When steady-state conditions are attained the free surface of the fluid is in effect cylindrical, the axis of the surface being coincident with the axis of rotation.The powder to be examined is injected as a low-concentration suspension through the entry port, and a layer of suspension is formed on top of the free surface of the fluid in the tank. The particles travel outwards, and the concentration of each size is measured by means of a light beam passing through the tank at a given distance from the centre of rotation. The equipment is essentially a photosedimentometer employing centrifugal force to aid the sedimentation of fine particles; it is therefore subject to the limitations of the photo- sedimentomet er theory.SIZE RANGE- REFERENCE- 1 to 50 p. Kaye B. H., Ph.D. Thesis, London University; British Patent No. 895,222, 1962. ELUTRIATION PROCESSES Elutriation differs from sedimentation in that fluid moves vertically upwards and thereby carries with it all particles whose settling velocity by gravity is less than the fluid velocity. In practice, complications are introduced by such factors as the non-uniformity of the fluid velocity across a section of an elutriating tube, the influence of the walls of the tube and the effect of eddies in the flow. In consequence, any assumption that the separated particle size corresponds to the mean velocity of fluid flow is only approximately true; it would also require an infinite time to effect complete separation. Stokes’s law cannot always be used in the calculations, since laminar flow rates are often exceeded.Gravitational Met hods (31) ANDREWS ELUTRIATOR- This is a form of elutriator for use with liquid flow. It consists of a small-diameter The larger vessel contains a circulating device for break- Elutriation produces three grades of particles, one and a large-diameter vessel in series. ing down aggregations of powder. collected in the first vessel, one in the second and the other in the overflow. SIZE RANGE- REFERENCE- 10 to 76 p. Andrews, L., Pvoc. Inst. Eng. Inspection, 192711928, p. 25.174 ANALYTICAL METHODS COMMITTEE [Analyst, Vol. 88 The instrument consists of six beakers almost filled with water and connected in series by siphons; the diameter of the up-going tube of each siphon is 4 2 times that of the pre- ceding one.Each of the beakers is mechanically stirred, and water from a constant-head tank flows at a controlled rate through the system. A dispersed powder is washed into the first beaker, where the coarsest particles are retained. All particles below a specific size, determined by the water velocity up the first siphon, are carried over to the second beaker. The entraining velocity in the second siphon is only half that of the first so that, again, the coarser particles remain in the second beaker and the finer ones are carried along. It is thus possible to separate into seven grades. Though the process is lengthy, no attention is needed. SIZE RANGE- REFERENCE- (32) BLYTHE ELUTRIATOR- 10 to 60 p. Pryor, E.J., Blythe, H. N., and Eldridge, A., “Recent Developments in Mineral Dressing,” Inst. Min. Met., 1953, 11. (33) GONELL AIR ELUTRIATOR- This consists of a cylindrical brass tube (or a series of tubes) with a conical base. An air inlet is provided in this base on the axis of the tube. The sample of powder is placed in the inlet cone, and air is blown through the largest tube until separation is deemed complete, or for specified periods of time if working to the B.S. specification. The residue is removed, weighed, and transferred to a smaller-diameter tube, and the test is repeated. The tubes should have polished internal surfaces and should be periodically tapped or vibrated to disturb settled dust. Weight of powder required, about 1 g. Time for analysis at least 4 hours.SIZE RANGE- REFERENCE- 5 to 76 p. British Standard 893 : 1940. Method of testing dust extraction plant. (34) MODIFIED GONELL ELUTRIATOR (FUEL RESEARCH STATION AFP-4R.4TUS)- The shape of the dust reservoir was modified in the Gonell apparatus (33), and specified in B.S. 893; the changes are described by Hughes. It is suggested that elutriation should be started by an initial high-velocity blast sufficient to carry all the dust into the elutriating tube, but of such short duration that the particles do not have time to reach the top of the elutriating tube before the velocity falls to normal. SIZE RANGE- REFERENCES- 5 to 76 p. British Standard 893 : 1940. Hughes, T. H., Engineer (Lond.), 1957, 203, 860. Method of testing dust extraction plant. (35) ROLLER AIR ELUTRIATOR- This is a proprietary instrument of American origin.The special feature is a swan-necked tube through which air is admitted to the elutriator tube. This is vibrated and has a nozzle so designed that the powder contained in the entry tube is continuously circulated to break down agglomerations. SIZE RANGE- REFERENCES- The sample weight required is about 10 g. The general operating procedure follows that for the Gonell elutriator. 5 to 76 p. A.S.T.M. Specification No. B.293-54T; 1954. British Standard 3406: Part 3: 1963.2.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 175 This apparatus is not a general-purpose instrument, but was developed by Imperial Chemical Industries Ltd. for the rapid elutriation of powders that are difficult to disperse, and in the size range 10 to 76 p.Samples of 0.1 to 0.5 g are used. The special feature is the entrainment of the powder by a high-velocity air jet, which blows downwards into a thimble containing the powder at the base of the elutriating tube. The elutriating tube is smaller than that of other elutriators; it is 14 inches long and 1 inch in diameter. SIZE RANGE- REFERENCES- (36) MINIATURE ELUTRIATOR- The general operating procedure follows that for other elutriators. 10 to 76 p. Stairmand. C. 1 . . Enpzneerine, 1951, May 18th. British Standard 3406: Part 3: 1963. (37) HAULTAIN ELUTRIATOR, INFRASIZER- This elutriator was originally intended for mining and metallurgical laboratories in I t consists of six conical elutriating tubes in series, air flow entry being to the smal- Air enters through a conical seating supporting a golf ball, which, by rotation and The general operating procedure follows Canada.lest. impact, breaks down agglomerations of particles. that for other elutriators. SIZE RANGE- REFERENCES- 5 to 200 p. Haultain, H. E. T., Trans. Can. Min. Met., 1937, 40, 229. Price, E. \V., Ind. Eng. Chem. (Process Design & Development), 1962, 1, 79. Centrifugal Methods This is a proprietary instrument that is essentially a centrifugal air elutriator. Air and dispersed dust samples are drawn through the cavity of a rotating hollow disc in a radially inward direction against centrifugal forces. The dust particles are thus divided into under- and over-size fractions, collected, and weighed.Separation into different size-fractions is made by altering the air velocity. About 20 g of dust are required for the sample, and 8 size determinations can be made in 2 hours. SIZE RANGE- REFERENCE- (38) BAHCO CENTRIFUGAL DUST CLASSIFIER- 5 to l o o p Berm, E. G., Brit. J . Appl. Phys., Supplement No. 3, 1954, S208. (39) B.C.U.R.A. CENTRIFUGAL ELUTRIATOR- This consists of two parallel concentric discs of equal radius mounted on a motor-driven spindle within a short cylindrical co-axial chamber. The discs are spun at a controlled rate and air is drawn through the system, also at a controlled rate. The powder entrained in this air enters the chamber axially, flows outwards over the surface of the first disc, passes between the two from the periphery to the centre and leaves axially.The air and dust pick up tangential velocity and between the discs the dust is acted upon by two forces, centrifugal outwards and drag inwards. Coarse particles are collected from the circum- ference of the chamber, fine particles pass into the exit. There is appreciable turbulence between the periphery of the discs and the chamber wall, so that considerable scrubbing takes place, the coarse particles being cleaned of fines. The cut size can be calculated, but, because the flow is not laminar and uniform, the cut size differs from the theoretical by a constant that has to be determined empirically. SIZE RANGE- 5 to 40 p. REFERENCE- Godridge, A. M., Badzioch, S., and Hawksley, €’. G. W., J . Sci. Instrum., 1962, 13, 611.176 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol.88 Winnowing Methods The principle of these methods is to allow the powder particles to fall vertically (if gravity is used) in a stream of winnowing fluid moving horizontally. The particles therefore move under two velocities, a horizontal one equal to the fluid speed, and a vertical one depending on the particle size. The particles are thus sorted out along the floor of the winnowing chamber, the larger (or heavier) ‘being carried least and the smaller the greatest distance. The devices are often used, by suitable adjustment of the fluid viscosity and floor length, for carrying out a preliminary sorting of the coarse from finer particles. They may also be used for direct sizing of airborne particles without the need for intermediate collection with the associated dangers of aggregation or breakdown. (40) TIMBRELL SAMPLER- I n this apparatus, clean winnowing air is passed along a horizontal tube, cleaned, and recirculated.The dust particles, entrained in air, are drawn in through a narrow tube and liberated at one end of the horizontal tube; they are subsequently deposited along the floor of the tube according to their free-falling speeds. The apparatus is first calibrated by using glass spheres and examining the slide produced microscopically, thus achieving a curve for Stokes’s diameter zIersGs distance travelled. If all that is required is a curve of frequency versus Stokes’s diameter, this may now be determined on the sample dust, counting at various distances with a microscope, or by use of a densitometer.However, it is also possible to study the variation of projected diameter (by using a microscope with eyepiece graticule) with Stokes’s diameter in the sample, and the extent of true aggregation is obviously simple to determine. If the sampling is carried out for a longer period it is possible to obtain a weight-distribu- tion curve. The apparatus may be modified to give intermittent sampling of a continuous stream, such as flue gases. An excellent review, with references to original work, has been made by Timbrell. SIZE RANGE- REFERENCE- 2 to 15 p. Timbrell, V., Brit. J . AppZ. Phys., 1954, Supplement No. 3, S12 and S86. (41) WALTON HORIZONTAL ELUTRIATOR- This consists of a short horizontal duct containing horizontal shelves, thus producing a stack of low horizontal passages.The back of the duct is covered by a filter-paper. Air is drawn into the duct, and the larger particles of dust are deposited on the floors of each of the passages, the fine dust being caught by the filter-paper. The basic principles were described by Walton, and the commercial instrument, which differs slightly from the original design, is known as the Hexlet Collector. The design is used particularly for pneumoconiosis work, the front of the instrument simulating the nasal passages and the rear retaining particles similar to those found in the lungs. SIZE RANGE- REFERENCE- 5 to 76 p. Walton, W. H., “Horizontal Elutriators,” part; of a paper entitled “Theory of Size Classification of Airborne Dust Clouds by Elutriation,” Brit.J . AppZ. Phys., 1954, Supplement No. 3, S303. (42) CONIFUGE- This apparatus is essentially a conical self-pumping centrifuge, into which the air con- taining the sample of dust is drawn at the rate of 25 ml per minute and subjected to centrifugal force and the “winnowing” effect of the clean air recycled within the apparatus. Since theMarch, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 177 two forces acting on the particle are at an angle to one another, the resulting deposit takes the form of a “spectrum,” which can then be counted by one of the microscope techniques or determined photodensitometrically. The method is applicable to airborne particles of any type or shape, and is independent of whether the particle is solid or liquid.The apparatus is readily calibrated with solid spheres of known density, e.g., paraffin wax. Alternatively, Stokes’s law may be used to effect the calibration. SIZE RANGE- REFERENCE- 0.5 to 30 p for spherical particles of unit density. Sawyer, I<. F., and Walton, W. H., J . Sci. Instrum.., 1950, 27, 272. (43) CONICYCLE- This consists of a pair of concentric rotating cylinders. At each end are two discs, one closing the ends of the inner cylinder, the other being separated from the first by an annular space. A t the inlet end of the instrument, where the end disc has the same diameter as the outer cylinder, an extension of the outer cylinder forms a disc baffle almost to the axis, and divides the annulus so that air passes first towards the axis, then under the baffle and outwards to reach the space between the two cylinders.At the exit end, the end disc extends to a greater radius than the outer cylinder. When the instrument is rotating a t high speed ($000 r.p.m.) the difference in radii of the two end annuli causes the instrument to act as a centrifugal fan driving air out of the exit. Centrifugal forces on the particles entering the inlet oppose drag forces pulling them in, so that only particles less than a specific size (usually 10 p) enter the instrument. The particles that enter are deposited according to size along the inner wall of the outer cylinder, except for a fraction of fine particles (less than 1 p) that escape into the exit. The instrument was designed to simulate the action of the human body, the retained particles being similar to those captured by the lungs.SIZE RANGE- REFERENCE- 1 to 10 p. Carver, J., Nagelschmidt, G., Roach, S. A., Rossiter, C. E., and Wolff, H. S., Min. Eng., 1962, 121, 601. (44) CASCADE IMPACTOR- Particle-laden air is caused to pass through a series of graded jets or slits, one after another in order of size, the largest aperture being the first. Facing each jet is a glass slide, or similar collecting device. It is usual to employ four stages, but as many as seven have been used. The lower limit of particle size that can be captured depends on the velocity of the suspension through the orifice and an absolute limit is reached when this velocity becomes that of sound. The slides can be examined in the usual manner. SIZE RANGE- REFERENCES- 0.7 to 20 p.Gillespie, T., J . Colloid Sci., 1955, 266. Gillespie, T., and Rideal, E., Ibid., 1955, 281. Laskin, S., U.S. Atomic Energy Commission Report MDDC-1500, Oak Ridge ,Tenn., 1946. May, K. R., J . Sci. Instrum., 1945, 22, 187. Orr, C., jun., and DallaValle, J. M., “Fine Particle Measurement,” The Macmillan Co., New York, 1959. Pilcher, J . M., Proceedings of the 42nd Annual Meeting, Chemical Specialities Manufacturing Asso- ciation Inc., New York, December, 1955. Ranz, W. E., and Wong, J. B., A.M.A. Arch. Ind. Health, 1952, 5, 464. -- , I n d . Eng. Chem., 1952, 44, 1371. Wilcox, J. D., A.M.A. Arch. 2nd. Hyg., 1953, 7, 376. A.11. IMAGE FORMATION In methods involving image formation there is a common factor. Radiation normally propagated in rectilinear fashion, such as light or an electron beam, is interrupted by the particles under examination, and the pattern of the interruption can be observed in different ways, e.g., optically or photo-electrically.178 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol.88 Light Beam: (a) Optical Microscope The image produced may be viewed either directly by the eye or projected on to a screen. The latter method is usually the more convenient and involves less eyestrain; it is less satis- factory than direct observation for fine sizes near the limit of optical resolution. Sizing is commonly achieved by comparison of the particle images with a scale or graticule. Since the process is tedious, sizing is sometimes performed automatically. The smallest resolvable particle size is a function of the wavelength of the light used and may be about 0.5 to 1 p.Individual Sizing and Counting- (45) GRATICULES- The principle used is to observe the projected image of the particles formed in the eyepiece of a microscope and to compare their sizes with standard accurately drawn shapes held in the same plane. The earliest graticule for use in this type of work was designed by Patterson and Cawoodl in 1936. I t consists of a series of solid “globes” and lined “circles” of graded diameters aligned a t the top and bottom of a series of counting rectangles. The size of the particle can be determined by assessing whether it is larger, and protrudes behind a globe of given diameter, or is smaller than a circle of the same diameter. In this, the globes and circles were in a geometrical progression based on a 4 2 difference in diameter.This he considered to be the finest distinction that the eye can perceive, and it permits counts to be made at various magnifications to overlap in identical size ranges. Fairs also described two other graticules intended for use with larger particles. One of these contains three globes and circles in geometric progression and one larger circle in the centre. The other contains five circles arranged like a shooting target centred on a cross-wire. A further modification, usually referred to as the Porton Graticule, was described by May3 in 1945. This is essentially similar to Patterson and Cawood’s graticule, except that the diameters of the globes and circles increase in the 4 2 progression and this is continued in the right-hand portion of the counting rectangle by a series of ruled lines, thus extending the range by using only one magnification.It is most useful when sizing circular particles, such as droplets or stains left by droplets. Various other modifications have been described in the literature, notably, one described by Watson4 in 1951, which is a much simplified version of the Fairs graticule for use by inexperienced workers. All graticules are used in the same manner. The particles are deposited as evenly as possible on a glass microscope slide, either by use of a thermal precipitator or by making a smear, an inert liquid being used as carrier. The slide is then viewed at any suitable magnification and the image so formed compared with the image of the graticule, which is held in the eyepiece. The graticule may be previously calibrated by use of a stage micrometer.Several counts can be made at various magnifications and so a considerable size range may be covered. For simple circular particles an eyepiece micrometer may be used instead of a graticule. For irregularly shaped particles, a globe and circle graticle is almost essential. If the particles are long and rod-like, an eyepiece micrometer may be used to find their dimensions, but the projected area, as measured by the globe and circle graticule, is no longer meaningful. Since considerable experience is required in the use of these graticules, Watson5 has designedia test strip consisting of 100 numbered irregular particles embedded in a standard 3-inch by 1-inch microscope slide.It may be used to train workers, and should also be used to give frequent checks on workers already engaged on rou tinecounting work, as the judgment required in making an assessment will not only vary from worker to worker, but may vary with time for any given worker. Despite these difficulties, reasonable reproducibility may be achieved with practice. A modification of this basic graticule was described by Fairs2 in 1943. This is now commercially available. SIZE RANGE- 1 p upwards.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE REFERENCES- 1. 2. Fairs, G. I.., Chem. & Ind., 1943, 57, 174. 3. 4, 5. Patterson, H. S., and Cawood, W., Trans. Faraday SOC., 1936, 32, 1084. May, K.R., J. Sci. Instrum., 1945, 10, 187. Watson, H. H., Brit. J . Indust. Med., 1951, 9, 80. - , Brit. T. AppZ. Phys.. 1954, Supplement No. 3, S101. British Standard 3406: Part 4: in the press, p. 196. 179 (46) KONIMETER (CON1METER)- A screened sample is aspirated and allowed to impinge on to a surface. The sample In the Zeiss Konimeter, the operation of aspiration and examination are performed is then positioned under a microscope fitted with a special grid micrometer (graticule). in the same instrument by means of a rotatable stage. SIZE RANGE- REFERENCE- 1 p upwards. Kotze, K., Miners’ Phthisis Prevention Committee : Final Report. Union of South Africa, Johannes- burg, January loth, 1919. (47) ZEISS-ENDTER PARTICLE SIZE ANALYZER- This equipment is a semi-automatic device for microscope counting.The microscope slide or photomicrograph is examined with a spot of light adjustable in size by an iris dia- phragm. The spot of light is centred on a particle, the diaphragm adjusted to the size of the particle and a count of the size is then carried out automatically by depressing a foot- switch. This technique is quicker than manual counting and overcomes the difficulties of visually comparing particles with fixed circles some distance away. SIZE RANGE- REFERENCE- 1 p upwards. Endter, F., and Gebauer, H., Optik, 1956, 13, 97. Automatic Sizing and Counting- (48) CASELLA INSTRUMENT- The method assumes uniformly opaque material-although some variation can in practice be tolerated-and a fairly low (say, less than 4 per cent. by area) particle concen- tration on the slide is necessary.A determination takes 8 to 19 hours, depending on the material and on the number of size classes assessed. A magnified image of the particle plane is moved past a slit perpendicularly to its length. The light passing through the slit falls on a photomultiplier cell, which emits a signal pulse each time the slit is crossed by a particle, the pulse amplitude depending on the maximum light obscuration produced by the particle as it crosses. The pulses are discriminated at five different size levels simultaneously, and their numbers recorded by scaling units. From the counts at a number of different slit lengths the concentration and sizes of particles can be deduced. The method is referred to as “wide-track statistical scanning.” The metal shadowing technique offers a means of sizing material in which there is too wide an opacity variation for direct counting to be carried out.The instrument is less flexible than a visual observer in regard to the area of the slide sampled, but more flexible :*I regard to the spacing of the size classes. SIZE RANGE- REFERENCE- 2 to 50 p. Hawksley, P. G. W., Brit. J. AppZ. Phys., 1954, Supplement No. 3, S125. Hawksley, P. G. W., Blackett, J. H., and Fitzsimmons, A. E., Ibid., 1954, Supplement No. 3, S165. Morgan, B. B., and Meyer, E. W., J . Sci. Instrum., 1959, 36, 492.180 ANALYTICAL METHODS COMMITTEE [Analyst, Vol. 88 This instrument utilises the principle of double-image microscopy, and comprises a special microscope and an associated electrically operated particle-size analyser.The appara- tus can be connected to digital read-out and print-out on tape facilities for processing of the results by computer. SIZE RANGE- REFERENCE- (49) TIMBRELL INSTRUMENT- 1 to 25 p. Timbrell, V., Nature, 1952, 170, 318. Barnett, M. I., and Timbrell, V., Pharun. J., 1962, 379. Light Beam: (b) Flying-spot Principle The difference between this principle and that of the optical microscope methods is that the whole field is not illuminated simultaneously, but is scanned by a fine light spot. Inter- ruption of the illumination is measured electronically. (50) RANK CINTEL INSTRUMENT- A moving spot of light, produced on the face of a cathode-ray tube, is focussed on the sample by means of a suitable lens system.The amount of light passing through or reflected from the sample varies according to the optical density and configuration of the individual particles of the sample. These changes in light intensity are detected by a photomultiplier, the signal from which is fed into the counting and sizing circuits. A memory device prevents each particle of the sample being counted more than once, and sizing is accomplished by pulse subtraction. The method requires at least a 10 per cent. difference in contrast between particle and background. For maximum accuracy the particles should have well-defined edges, be regular in shape and their concentration should be low to avoid overlapping. The particles may be dark on a light background, or vice versa; they may be measured on a support (microscope slide, metal plate, glass cell, paper) or they may be measured from photographic negatives.The time taken to analyse a given sample depends on the particle density of the sample and its degree of heterogeneity; it is usually about 10 to 60 minutes. The total projected area of the particles in the sample and of an average particle may be measured in 2 to 20 minutes. SIZE RANGE- 1 p upwards by using transmitted light; 20 p upwards by using reflected light from opaque specimens. REFERENCES- Taylor, W. K., Brit. J . APpZ. Phys., 1954, Supplement No. 3, S173. Causley, D., and Young, J. Z., Research, 1955, 8, 430. Furmidge, C. G. L., Brit. J . Appl. Phys., 1961, 12, 268. (51) MULLARD INSTRUMENT- The basic principles of scanning, counting and sizing are similar to the Rank Cintel instrument (50).The main difference is that this instrument operates only for samples photographed on 35-mm film. A wide range of particle sizes may be measured by varying the magnification in the preliminary photographic process. SIZE RANGE- 1 p upwards by using transmitted light; 20 p upwards by using reflected light from opaque specimens. Electron Beam (52) ELECTRON MICROSCOPE- The electron microscope has much greater magnifying and resolving powers than has the optical microscope. It is used to produce a photographic image of a sample, the photo- graph being subsequently examined by normal optical methods. Special precautions are necessary for preparing the sample and for calibrating the magnification.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 181 SIZE RANGE- REFERENCE- 0.001 to 20 p.Cartwright, J., Brit. J . Appl. Phys., 1954, Supplement No. 3, S109. A.111. SCATTERING OF RADIATION Light Beam The intensity of light scattered from a parallel beam by an array of particles varies with direction and is dependent on the ratio h/D, where h is the wavelength of the incident radiation and D is the diameter of the particle. It also depends critically on the refractive indices of the materials considered. A comprehensive review on the relevant theory has been given by Van de Hu1st.l Several techniques have been developed for special systems that permit an estimate to be made of the particle size distribution and size from a study of the scattered radiation. (53) GOULDEN LIGHT SCATTERING TECHNIQUE- This method2 has been developed for measuring the size of fat globules in milk.SIZE RANGE- REFERENCES- 0.01 to l o o p . 1. 2. Van de Hulst, H. C., “Light Scattering by Small Particles,” John Wiley & Sons Ltd., London, 1957 Goulden, J., Brit. J . Appl. Phys., 1961, 12, 456. X-ray Beam (54) X-RAY METHODS INVOLVING LOW-ANGLE SCATTERING TECHNIQUES- In ordinary X-ray diffraction work, some information in respect of crystallite size or grain size can be obtained from line-broadening effects. Diffraction cameras, however, are mainly used for the study of crystalline substances. The type of light scattering referred to in A.IV (Young’s rings), however, also occurs when X-rays are scattered by a dispersion of small particles. Since X-rays are of extremely short wavelength, the scattering occurs with small particles in the size range of about 0.002 to 0.1 p (about 20 to lOOOA) and it reaches a maximum in directions that make angles of only a few degrees with the forward direction of the X-radiation.Special cameras have been designed to gain the maximum amount of information about this forward scatter. The methods may be equally well applied to both crystalline and amorphous substances. In contrast to ordinary diffraction cameras, long specimen to film distances are used, and special precautions are taken in the design of the collimeter and slit system to reduce instrumental scatter. Details of different designs and results of tests (including comparision with standard methods) are given in the references below.SIZE RANGE- REFERENCES- 0.002 to 0.1 p. Gerold, V., “Small-angle Scattering and Its Use in Particle-size Determination,” 2. angew. Phys., Kratley, P., Kolloid-Z., 1955, 144, 110. Guinier, A., Ann. Phys., 1939, 12, 161. Riley, D. P., “X-ray Diffraction by Polycrystalline Materials,” Inst. Phys., London, 1955. 1957, 9, 43. A.IV. DIFFRACTION OF RADIATION (55) YOUNG’S DIFFRACTION RINGS- If a point source of monochromatic light, wavelength A, is viewed through a screen of spherical particles with uniform diameter, d, the image is surrounded by a system of alternately light and dark diffraction rings. The dark rings are identified as first order, second order, . . . lzth order, by counting outwards from the source image.182 ANALYTICAL METHODS COMMITTEE : [Analyst, Vol.88 If 8 is the angle subtended at the point of observation by the mean semi-diameter of the ring of nth order, the following relation exists- ri + 0.22h sin 0 Particles with an acceptable degree of uniformity are vegetable spores, blood cells and The size of mist droplets is conveniently estimated by measuring the diameters For example, if the apparent diameter of the first order If the mean wavelength d = - mist droplets. of the minima between lunar haloes. dark ring is five times that of the lunar disc (31’ 7 ” ) , sin 8 = 0.02262. of visible light is taken as h = 0-57 p, d is 31 p. SIZE RANGE- REFERENCE- 5 to 50 p. Humphreys, H., “Physics of the L4ir,’’ McGraw-Hill Book Co., New York and London. A.V. ELECTRICAL PROPERTIES (56) CONDUCTIVITY (COULTER C0UNTER)- This commercially available equipment utilises the change in resistance of the system caused by the presence of a particle in an electrolyte.The suspension flows through a small aperture having an immersed electrode on either side, with particle concentration such that the particles traverse the aperture substantially one at a time. Each particle passage displaces electrolyte within the aperture, momentarily changing the resistance between the electrodes and producing a voltage pulse of magnitude proportional to the particle volume. The resulting series of pulses is electronically amplified, scaled and counted. SIZE RANGE- REFERENCE- 0-2 t o 300 p. Kubitschek, J., Research, 1960, 13, 129. A.VI. SIEVES Woven Wire Sieves Sizing by sifting, although simple in conception, has many practical and theoretical difficulties.The technique is covered by British and foreign standard specifications. (57) BRITISH STANDARD SPECIFICATIONS- British Standard 410 gives details of sieve construction. Part 1 specifies requirements for the construction of fine and medium test sieves; Part 2 refers to the wire cloth for the test sieves and includes tables of aperture sizes, wire diameters, screening areas and aperture tolerances; Part 3 specifies requirements for coarse test sieves; Part 4 deals with the con- struction of the perforated plates for coarse sieves. Appendices cover methods of examina- tion, measurement and calibration of sieves, and particulars of the American (A.S.T.M.), German (D.I.N.) and Institute of Mining and Metallurgy (I.M.M.) standard sieve sizes.The British specification covers two grades of wire sieves-the normal sieves with tolerances from 5 9 per cent. to &3 per cent., and the special sieves with tolerances from 5 7 per cent. to -&2 per cent. PARTICLE-SIZE RANGE- 76 to 3353 p (exceptionally, 53 to 3353 p). British Standard 1796 contains sections dealing with principles of sifting, equipment to be used, preliminary preparation of the sample, weight of samples to be used in tests, use of wet and dry methods for the preliminary removal of fine dust, hand sifting to the “end-point” by the rate test, machine sifting to the “end-point” by the rate test, loss of dust during sifting, sifting tests on materials with only a small amount of oversize on the sieve and routine sifting tests. Information is also given on the correct method of reporting results and limits of accuracy of the results obtained.Appendices give methods of subdividing bulk samples, and notes on sifting procedure.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 183 Corresponding American, Dutch, French and German standard specifications are avail- able: these are listed below, with the particle-size ranges catered for- U.S.A. A.S.T.M. E-11/39 (Fine series) 37 to 5660 p Netherlands N 480 50 to 850 p, France AFNOR N F Xll-501-1938 40 to 5000 p Germany DIN 1171 60 to 5000 p, Recent Techniques (Not yet standardised) (58) ELECTROFORMED SIEVES- Fine-mesh precision sieves formed of nickel by a photograving and electroplating tech- nique are available at a few aperture sizes down to 20 p.The regularity of aperture size is good (one measurement of a nominally 20 p aperture sieve gave 97.5 per cent. of the apertures in the range 20 & 1.5 p, none having an error greater than h2.5 p), but the sides of the aper- tures are less smooth than those of woven wire sieves. Nothing is known of the expected life of such sieves, SIZE R~INGE- REFERENCE- 16 to 53 p. Daeschner, H. W., A.S.T.M. Symposium, 1959, Special Technical Publicatioiz N o . 234, American Society for Testing Materials, Washington. (59) AIR-JET SIFTING- The Alpine “Air-Jet Sieving Machine” utilises the principle of a rotating jet blowing air up from beneath an enclosed sieve. This keeps the powder moving on the sieve while the undersize particles are drawn through the rest of the sieve by the air returning, via a suitable filter, to the suction side of a fan.SIZE RANGE- REFERENCE- 53 to 3353 p. Lauer, O., Stnub, 1960, 20, 69. B.I. PERMEABILITY TO FLUID FLOW The relevance of methods in this section to the measurement of specific surface depends on certain relationships that are assumed to hold between the rate of fluid flow, the pressure head, viscosity of the fluid, density and specific surface of the powder and porosity of the powder bed. Generally, the relationship assumed is the Kozeny equation or some modi- fication of it. I t must be stressed that permeability of the bed is the property that is actually measured; the validity of the estimates of specific surface derived therefrom depends on the validity of the relationships assumed.The methods give useful comparisions between different samples of materials of the same type, and hence are frequently used in control of raw materials. Anomalous results may be obtained if materials with widely divergent size or shape distri- butions are compared ; the results of permeability measurements should therefore be inter- preted with due caution. Some preparation of the sample, such as removal of coarse particles by sifting, may be necessary to achieve reproducible results. REFERENCES- Kozeny, J . , Siizber. Akad. Wiss. Wien. Math-naturw. KZ., 1927, IIa, 271. Rigden, I-’. J., J . SOC. Chem. Ind., 1947, 66, 130T. Liquid Flow (60) CA4RMAN METHOD- A liquid is allowed to flow, under constant pressure head, through a bed of the powder in a cylindrical tube. The rate of flow, pressure head, viscosity of the liquid, densityof the powder and porosity of the bed are measured; the specific surface of the powder may then1 84 ANALYTICAL METHODS COMMITTEE : [Analyst, VOl.88 be calculated. In its original form, the apparatus was applicable to corase-grained powders ; subsequent modifications, by increasing the pressure, are stated to have extended the range down to 2 to 3 p . This lower limit corresponds to a specific surface of about 10,000 sq. cm per g. REFERENCES- Carman, P. C., Trans. Inst. Chem. Eng., 1937, 15, 150; J..Soc. Chem. I n d . , 1938, 57, 225 and 19’39, 58, 1; Dzsc. Faraday SOC., 1948, 3, 72. Gas Flow (61) LEA AND NURSE METHOD- The apparatus consists of a cell, a manometer and a flowmeter.The powder is compacted into a bed in the cell and dry air is allowed to flow through it under a constant pressure differential, the rate of flow being measured. A series of measurements is made at several pressure differentials and the gradient of the pressure-flow graph is determined. Thence, from the density of the powder and the porosity of the bed, the specific surface may be calculated in accordance with the Kozeny - Carman equation. REFERENCES- Lea, F. M., and Nurse, R. W., J . SOC. Chem. Ind., 1939, 58, 278. British Standards 12 : 1958 and 915 : 1947. (62) GOODEN AND SMITH METHOD- The principle is the same as that of the Lea and Nurse method (61). It differs in having a fixed pressure on one side of the bed.Since the air also passes through a resistance flow- meter, the pressure on the downstream side of the bed varies with the rate of flow. When the manometer across the flowmeter becomes steady, the reading is taken and is used to obtain an “average particle size.” An automatic calculator is provided with the instrument. REFERENCE- Gooden, E. L., and Smith, C. M., Ind. Eng. Chem., Anal. Ed., 1940, 12, 479. (63) FISHER SUB-SIEVE SIZER- In this instrument, which is developed from the Gooden and Smith method (62), a standardised compaction procedure is used. The surface area is read directly from one of a family of curves on a chart incorporated in the instrument. REFERENCE- Fisher Scientific Company, Pittsburg, Pa., U.S.A. : Kek Ltd., Ancoats, Manchester 12.(64) RIGDEN METHOD- The powder is compacted into a bed in a B.S. 12 cell, and the two ends of the cell are connected to the two ends of a U-tube containing an oil of low vapour pressure. The oil is displaced initially by drawing up on one side; as it returns to its equilibrium level it forces air through the powder bed. The pressure differential thus diminishes as flow proceeds. The time for the oil to travel between two rnarks on the U-tube limb is measured and the specific surface then calculated. REFERENCE- Rigden, P. J., J . SOC. Chem. Ind., 1943, 62, 62 and 1947, 66, 130T. (65) GRIFFIN SURFACE AREA APPARATUS- This apparatus is based on the work of Rigden (64). The oil in a U-tube isdisplaced and, in returning to equilibrium, forces air through a powder bed.The manufacturers recommend that the bed be formed by compacting four or five successive scoop samples with a plunger. With the addition of a 0- to 760-mm mercury manometer, the rate of flow at reduced pressures can be used to give the mean pore radius. On a sample of Portland cement, a spread of &2 per cent. on the mean is claimed.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE 185 (66) BLAINE FINENESS TESTER- The powder is contained in a cylindrical cell, the lower end of which is connected to a U-tube manometer. Provision is made for the withdrawal of air from the space between cell and manometer, so that the level of oil in the manometer tube is raised. The time for the manometer liquid to travel between two lines on the tube is then measured.The method is therefore of the diminishing- pressure-differential type. REFERENCE- The principle is the same as that of the Rigden method (64). Blaine, R. L., A.S.T.M. Bulletin, 1943, No. 123, 51; A.S.T.M. Standards, 1958, Part 4, C204-55, p. 140. (67) SPILLANE METHOD- The principle is the same as that of the Rigden method (64). In effect, it gives a direct pointer reading of the time taken for mercury to travel between two points on a manometer tube. The dial may be calibrated, by means of samples already tested by the Rigdenor Lea and Nurse methods, to give direct readings of the specific surface of powders of uniform specific gravity. SPECIFIC su RFACE- REFERENCE- 1500 to 4500 sq. cm per g. Spillane, F. J., Analyst, 1957, 82, 712. B.11. ABSORPTION METHODS There are various methods for the determination of the surface area of solids based on the adsorption of a mono- or polymolecular layer on the surface of the solid.These methods do not measure the particle diameter or projected area as such, but measure the available surface per gram or millilitre of powder. With catalysts and adsorbent materials this is more relevant than particle size. Static : (a) Absolute Method (68) HARKINS AND JURA (ABSOLUTE METHOD)- In this methodl (designated HJa), the clean, de-gassed, finely-powdered solid is kept in the saturated vapour of a liquid that wets it. The solid adsorbs a polymolecular film whose outer surface, at equilibrium, has the same total surface energy, h f , as the liquid itself. The powder is then dropped into the same liquid in a calorimeter to obtain a measure of the heat developed by the disappearance of the adsorbed film.If water is chosen as the liquid, the energy change involved is fixed at 118.5 ergs per sq. cm at 95" C. The low energy available necessitates extremely accurate calorimetry, and measurement of temperature to -+0.00002" C is required. Extremely constant temperature control is also essential and, unless the specific surface is of the order of at least 3 to 9 sq. m per ml, high accuracy is not attainable. The method has been used to give an extremely accurate determination of surface area of reference samples; a sample of anatase with a surface area of 13.8 sq. m per g, determined by this method, is used in many laboratories for calibration purposes for other methods.Static: (b) Involving use of Adsorption Isotherms There are two main methods of determining surface area involving the use of absorption Both methods make use of a similar technique, but differ mainly in the treatment isotherms. of results. (69) BRUNAUER, EMMETT AND TELLER- In this method2s3 (designated BET), the surface area is not measured directly, but the number of molecules of the adsorbed substance required to give a monolayer (N) is deter- mined; if the mean area per molecule (a) of the adsorbed substance is known by other means, the area of the solid may be calculated as- E = No.186 ANALYTICAL METHODS COMMITTEE: [Analyst, Vol. 88 In this second method of Harkins and J ~ r a ~ ? ~ (designated HJr), the surface area is (70) HARKINS AND JURA- given directly after standardisation by the absolute method (68).Determination of Isotherm- Since both the “BET” (69) and “HJr” (70) methods use the adsorption isotherm, the experimental details of determining this are summarised as follows, and references are given for the treatment of results. The powder under investigation is thoroughly de-gassed in a vacuum and small amounts of gas or vapour are added at constant temperature; the resultant pressure is noted. With no solid present, the pressure would follow the gas laws; with powder present, the pressure is lower. A curve of volume of gas admitted v e m w pressure may be produced and, from this isotherm, the surface area may be determined by either method. Adsorihtion from Solution- A solid will adsorb a surface-active material or a dyestuff from solution and an isotherm may be constructed, use being made of concentration of adsorbate instead of pressure.Cali- bration with standards is essential. This method is limited in its application because solids soluble in the solvent cannot be used; further, sometimes (e.g., with electrolytes on charcoal or silica gel) the solvent may be adsorbed to a greater extent than the solute, producing apparent negative adsorption. The adsorption is generally believed to be a consequence of the solid presenting a solid - liquid interface that can accumulate a surface-active compound in the same way as can an air - liquid interface. The degree of adsorption increases with surface tension, so that water is usually used. As an example, a dyestuff, initially adsorbed from water, may often be removed from charcoal by alcohol or acetone. ModiJications for Control Purposes- The basic methods may be extremely accurate, but are tedious to operate. However, for control purposes, modifications may be made to simplify them and give relative values. Adsorption of both vapours and solutes may be modified to allow quick control methods to be carried out. If the properties of the powder are established, a size range is chosen so that a change in surface area may be monitored by a particular vapour or solute. The surface area may be determined by reference to results obtained by an absolute method. This method is most useful in the characterisation or control of finely divided surface catalysts. REFERENCES- 1 . 2. 3. 4. 5. 6. 7. Harkins, W. D., and Jura, G., J . Amer. Chem. SOC., 1944, 66, 1362. Rrunauer, C., Emmett, P. H., and Teller, IS., Ibid., 1938, 60, 309. Emmett, P. H., and Brunauer, S., Ibid., 1944, 66, 35. Harkins, W. D., and Jura, G., Ibid., 1944, 66, 1366. Jura, G., and Harkins, W. D., Ibid., 1944, 66, 1356. Wooten, L. A., and Brown, C., Ibid., 1943, 65, 113. Emmett, P. H., and Brunauer, S., Ibid., 1937, 59, 1553. Dynamic (71) DYNAMIC METHOD- This method is essentially a gas-chomatographic technique with the sample powder in place of the normal chromatographic column. A mixture of helium and nitrogen is passed through the sample and the concentration of nitrogen in the exit gas is measured by thermal- conductivity or gas-density methods, and plotted on a recording potentiometer. The change in nitrogen content of the exit gas when the sample is cooled in liquid nitrogen gives a measure of the quantity of nitrogen adsorbed on the sample surface. The adsorption measurement is repeated, with three or four different concentrations of nitrogen in the helium, and the surface area of the sample is calculated from the BET equation ( i e . , E = Nu). REFERENCE Nelsen, F. M., and Eggertsen, F. T., Anal. Chem., 1958, 30, 1387. Ellis, J. S., Forrest, C. W., and Howe, D. D., U.K. Atomic Energy Commission DEG Report 229 (CA), Daeschner, H. W., and Strass, F. H., Anal. Chem., 1962, 34, 1150. H.M. Stationery Office, 1960.March, 19631 CLASSIFICATION OF METHODS FOR DETERMINING PARTICLE SIZE B.111. ABSORPTION OF RADIATION (72) LIGHT BEAM TECHNIQUE- If a suspension of particles is irradiated by light, the particles of size 2 p and upwards will interrupt the beam in such a manner that the absorption is related to the projected area (see method (74)). For particle sizes less than 0.01 p, the absorption is related to the volume of the particle. For the intermediate range of sizes (between 0.01 and 2 p), an extremely complicated relationship is found. It is possible, therefore, to derive surface area of a uniform suspension by measurement of optical density, provided that particles in the size range 0-01 to 2 p are not present in large quantity. (73) PENETRATING RADIATION TECHNIQUE- If penetrating radiation (see “Radiometric Techniques,]’ p. 168) is used, the range of particle sizes in which absorption does not bear a simple relationship to surface area alters in accordance with the wavelength of the radiation used. Outside this range, the absorption depends on the mass of material (see method (15)). B.IV. OPTICAL MICROSCOPE MEASUREMENT WITH PARTICLES IN RANDOM ORIENTATION (74) area, in accordance with- According to the theorem of Cauchy, particle surface area is related to projected 4XA S =- n S approaches absolute surface as n increases, provided that there are no re-entrant C A is the sum of the areas of projected images of n convex particles in radom orientation I t follows that if a truly random dispersion on a slide can be achieved] the surface area A procedure for random dispersion of particles and for measuring the projected areas surfaces and there is negligible sub-microscopic roughness. (a condition not usually obtained on a microscope slide). of the particles is four times the mean projected area. is described by Pidgeon and Dodd. REFERENCES- Cauchy, A., Compt. Rend., 1841, 13, 1060. Pidgeon, F. D., and Dodd, C. G., Anal. Chew., 1954, 26, 1823.

ISSN:0003-2654    

DOI:10.1039/AN9638800156

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

188 FRYER, GALLIFORD AND YARDLEY: [Analyst, Vol. 88 The Furildioximes Part I. The Structure of the Isomeric Furildioximes BY F. A. FRYER, D. J. R. GALLIFORD AND J. T. YARDLEY (Hopkin & Williams Ltd., Freshwater Road, Chadwell Heath, Essex) The so-called a-furildioxime prepared by the usual procedures is shown to be an approximately 1 + 1 mixture of a and y isomers. When heated under reflux with n-pentanol, the y isomer is converted to a and p isomers, and the mixture is converted in part to the /3 isomer. Pure a-furildioxinie melts at 192" to 193"C, whereas the y and /3 compounds melt a t 182" to 183" C and about 150" to 152"C, respectively. All three isomers are shown to form nickel complexes. FL-RILDIOXIME has been extensively used a s a gravimetric and colorimetric reagent for n i ~ k e l , l + ~ ~ ~ ~ ~ p a l l a d i ~ m ~ , ~ and platinum6 for many years, and more recently it has found favour as a colorimetric reagent for r h e n i ~ m .~ The substance employed by all the workers cited has been called a-furildioxime and, in the absence of any information to the contrary, it seems likely that most samples were prepared by the oximation of furil with hydroxyammonium chloride and then aqueous recrystallisation by MacKair's method.8 This procedure is commonly quoted in the text- b 0 o l t s g ~ ~ 0 ~ ~ ~ and gives a product melting between 160" and 168" C, the higher figure being widely accepted as the melting-point of the pure a isomer. Subsequently, there has been confusion in the literature about the melting-points to be assigned to the several possible configurations.12 9 1 3 We found that repeated recrystallisation from water produced a substance that, after it had been dried, melted at 166" to 168" C; however, assay by treating with an excess of nickel solution and then weighing the precipitate gave an apparent purity of only about 55 per cent.(40 to 60 per cent. on material recrystallised only once). In contrast, by re- peatedly recrystallising the crude reaction product from ethanol, we were able to obtain a specimen melting sharply at 192" to 193" C that gave, when assayed by the excess of nickel method, a purity of approximately 100 per cent. These results led us to suppose that the initial reaction product was a mixture of isomeric furildioximes in which the a (anti-) compound was the minor constituent and that the melting- point most commonly attributed to this isomer was incorrect.Shinra and Tshikawa12 reached a somewhat similar general conclusion, but the configurations they assigned to the three isomers did not correspond with our own tentative findings. Further experiments were therefore conducted to test and amplify our own views, and the results are given below- 1. The "mixture" melting at 167" C: (ex-water) was separated by precipitating the red nickel complex with excess of slightly alkaline nickel solution and then evaporating the filtrate to small bulk, when crystals of a second ( y ) isomer separated. This compound was recrystallised from water, and the dried product was found to melt at 184" C and to contain 12-8 per cent.of nitrogen. (Cl()H8O,N2 requires 12.7 per cent. of nitrogen.) 3. A melting-point diagram for the system a - y isomers was obtained from synthetic mixtures derived from the two procedures outlined above. This gave a eutectic point a t about 164" C, corresponding to a mixture containing about 55 per cent. of the 01 isomer. 3. Ultraviolet and infrared spectroscopic examination showed the differences and similarities listed below- a Y Melting point of isomer, "C . . .. .. .. 193 183 Xmax., m p * * .. .. .. . . . . . . 270to 274 272 to 278 E . . . . .. . . . . . . .. .. 27,000 27,000 Absorption attributed to C : N stretching vibrations Absorption attributed to OH stretching vibrations . . 3210 cm-1 3275 cm-l NOTE-Infrared spectra were produced initially with h'uj 01 mulls in a Perkin Elmer Infracord spectrophotometer, but the absorptions quoted throughout this report are the similar but more precise values kindly determined for us by Dr.J. E. Page, . . 1630, 1580 cm-1 1630, 1560 cm-1March, 19831 THE FURILDIOXIMES. PART I 189 with a Perkin Elmer model 21 spectrophotometer (personal communication from Glaxo Research Ltd.). All ultraviolet spectra were obtained with alcoholic or aqueous solutions and a Beckman DU spectrophotometer. Attempts were then made to produce a sample of j3 (syn-) compound by heating the so-called a and y isomers under reflux in n-pentanol for 7 hours. Treatment of pure y isomer in this way gave a product that consisted largely of the sc compound. The mixed a - y product gave a treacly mass when the pentanol was removed, and this was dissolved in ether and chromatographed twice on silica gel (M.F.C. grade), with ether as eluent.Five fractions were collected, and, after the solvent had been evaporated, the residues were assayed with excess of nickel. The solid obtained from fractions 3 and 4 gave a brown alcohol-soluble nickel complex, which on weighing indicated an apparent purity of 90 per cent., calculated as a 1 + 1 compound of furildioxime and nickel. Other characteristics of this product were- Melting point, "C . . .. . . .. . . .. .. 150 t o 152 Xitrogen content, yo . . . . . . . . . . . . 11.4 (C,,H,O,N, requires 12.7) .Absorption attributed t o C : N stretching vibrations . . -\bsorption attributed t o OH stretching vibrations . . . .3200 t o 3260 cm-l Xrnax., mp - * . . . . .. . . .. . . .. 272 t o 276 ( E = 27,000) . . 1635, 1565 cm-l TABLE I DIFFERENCES IN THE XICKEL COMPLEXES OF THE THREE ISOMERS Nickel Conditions for Colour of precipitation of Amax., content, * Isomer complex complex Solubility mP E % a Red Nearly neutral Insoluble in ethanol 290 t o 292 52,000 11.8 P Brown Nearly neutral Soluble in ethanol 275 t o 287 36,000 24.8 Green Above pH 8 Slightly soluble in ethanol 274 t o 282 37,700 22.6 Y * A ratio of furildioxime t o nickel of 2 t o 1 corresponds t o a nickel content of 11.8 per cent. and a ratio of 1 t o 1 corresponds t o a nickel content of 21.3 per cent. The hydroxyl stretching frequencies shown by the parent dioximes suggest that a higher degree of hydrogen bonding exists in the Q isomer than in the y isomer, and this is consistent with their respective anti- and amphi- configurations.The intermediate position of the. p isomer suggests a moderate degree of intramolecular hydrogen bonding that might be expected with the syn-configuration. Such small differences as were observed with the C : N absorptions are also consistent with the previously reported results for a and /3 mon- oximes .I4 From the evidence offered there seems good reason to supp0s.e that the three compounds we have referred to as a, ,8 and y isomers (melting-points 193", 152" and 183" C) are, respec- tively, the anti-, syn- and aniphi-stereoisomers of furildioxime. All three form nickel com- plexes, that formed with the p isomer being least expected in view of the work of Meisen- heimer,15 but it is suggested that its structure may closely resemble that reported for the palladium - /3-benzildioxime complex.16 The ultraviolet spectra of the ,8 and y complexes (tabulated above) suggest closely similar structures.Other nickel - y dioxime complexes have been reported and shown to be 1 + 1 compoiind~.~~J7~~8 Nickel - a-furildioximate is stoicheiometric, and we found its ultraviolet absorption spectrum to be similar to those reported for nickel dimethylglyoximate and nickel - a-furildi- 0 ~ i r n a t e . l ~ However, unlike nickel dimethylglyoximate, it shows no infrared absorption around 1775 cm-l. Absorption in this region is considered20 to be due to hydroxyl stretching vibrations. We thank the Directors of Hopkin & Williams Ltd.for permission to publish this paper. 1 . 3. 3. 4. 6. 6. 7. 8. REFERENCES Harwood, H. F., and Theobald, L. S., Analyst, 1933, 58, 673. Taylor, C. G., Ibid., 1956, 81, 369. Soule, B. A,, J . Amer. Chem. SOC., 1925, 47, 981. Gahler, A. R., Mitchell, A. M., and Mellon, M. G., Anal. Chem., 1951, 23, 500. Ogburn, S. C., jun., J . Amer. Chem. SOC., 1926, 48, 2493 and 2507. Menis, O., and Rains, T. C., Anal. Chem., 1955, 27, 1932. Meloche, V. W., Martin, R. Id., and Webb, W. H., Ibid., 1957, 29, 527, MacNair, D. S., Annalen, 1890, 258, 226.190 9. 10. 1 1 . 12. 13. 14. 15. 16. 17. 18. 19. SO. FRYER, GALLIFORD AND YARDLEY [Analyst, Vol. 88 Rodd, E. H., Editor, “The Chemistry of Carbon Compounds, Volume IV A, Heterocyclic Com- Welcher, F. J., “Organic Analytical Reagents,” D. Van h‘ostrand Co. Inc., New York, 1947, Huntress, E. H., and Mulliken, S. P., “Identification of Pure Organic Compounds,” John Wiley Shinra, K., and Ishikawa, K., J . Chem. SOC. Japan, 1953, 74, 353. Yamasaki, K., Matsumoto, C. L., and Ito, R., Nippon Kagaku Zasshi, 1957,78, 126; Chem. Abstr. Palm, A., and Werbin, H., Canad. J . Chem., 1953, 31, 1004. Meisenheimer, J., and Theilacker, W., Annulen, 1929, 469, 128. Dwyer, F. D., and Mellor, D. P., J . PYOC. .Roy. SOC. N.S. Wales, 1934, 68, 107. Atack, F. W., J . Chem. SOC., 1913, 103, 1317. Hieber, W., and Leutert, F., Bey., 1929, 62, 1839. Sone, K., J . Amer. Chem. SOC., 1953, 75, 5207. Rundle, R. E., and Parasol, M. I., J . C h ~ m . Phys., 1952, 20, 1487. pounds,” Elsevier Publishing Co., Amsterdam, 1957. Volume 111. & Sons Inc., New York, 1941. 1958, 52, 7003. Received July 26th, 1962

ISSN:0003-2654    

DOI:10.1039/AN9638800188

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

March, 19631 FRYER, GALLIFORD AND YARDLEY 191 The Furildioximes Part 11. The Analytical Behaviour of Purified 01 -Furildioxime, with Particular Reference to the Determination of Rhenium in the Presence of Molybdenum BY F. A. FRYER, D. J. B. GALLIFORD AND J. T. YARDLEY (Hopkin & Williams Ltd., Freshwater Road, Chadwell Heath, Essex) Pure a-furildioxime is shown to behave satisfactorily in the colorimetric determination of rhenium, and a new method is developed for its deter- mination in the presence of 500 times as much molybdenum. AFTER the work described in Part I of this series had been carried out, it seemed desirable to ascertain whether or not pure a-furildioxime would exhibit any unexpected properties in typical analytical procedures that had previously involved use of a mixture consisting predominantly of the y isomer.Since the a isomer had been successfully assayed by precipitation of its nickel complex, this aspect was not pursued further, but it was decided to test the reagent in the procedure described by Meloche, Martin and Webb for the colorimetric determination of rhenium.l Accordingly, the procedure applicable in the absence of interfering ions was first carried out, and this yielded results identical with those previously reported. The molar extinction of the rhenium complex, calculated from our experiments, agreed with that stated by Meloche and co-workers to within 0.5 per cent. On the assumption that previous workers had used a reagent containing rather less than 50 per cent. of active isomer, it was thought worth while to repeat these experiments with half the previous amount of furildioxime, and, as expected, this was found to be adequate.This fact is particularly significant, because Meloche and co-workers had recognised that a large excess of reagent was essential for maximum colour development and had specified the least amount consistent with this obj ect ive. Although molybdenum does not form a coloured complex with furildioxime it interferes severely with the determination of rhenium, and Meloche, Martin and Webb utilised a pre- liminary extraction of the xanthate - molybdenum complex into chloroform. In this method the xanthate is first added while the test solution is at pH 9 to 11, and extraction is then carried out in the presence of a relatively high concentration of hydrochloric acid.The aqueous layer is washed twice with chloroform and then treated with a slight excess of 0-3 N potassium permanganate (about 10 drops, according to directions given by Meloche and co-workers) to oxidise any remaining xanthate or its decomposition products. This is necessary to prevent the subsequent colour development being hindered on the addition of stannous chloride and reagent solutions. We found that the aqueous layer was extremely turbid and required more than 50 ml of 0.1 N potassium permanganate to complete the oxidation. This made it impossible to restrict the final solution to the recommended volume, and, even when due allowance had been made for this, colour development was either completely inhibited or indefinitely delayed.Purification of the potassium ethyl xanthate did not improve colour development to any extent, and various modifications of the conditions of extraction (reduced acidity, for example) failed to prevent considerable decomposition of the xanthate ; however, it should be recorded that Kassner employed a similar separation as a preliminary to the determination of rhenium with 4-methylnio~ime.~ Various other separations were considered, and one or two were tried experimentally; the most promising was the extraction of the molybdenum into chloroform by means of 8-hydroxyquinoline (oxine) . Separation of molybdenum from rhenium as a gravimetric operation with oxine was first reported by Geilmann and Weibk,3 and then by Michailova, Posner and Ar~hipova.~ Later, Jasim, Magee and Wilson developed a micro-gravimetric separation based on the same prin~iple,~ and Gentry and Sherrington have reported that the molybdate - oxine complex can be extracted into chloroform in acidic solution.6 These observations provided the basis for the separation we ultimately developed.192 FRYER, GALLIFORD AND YARDLEY: [Analyst, Vol.88 Preliminary separation of the oxine - molybdate complex from a slightly acidic solution buffered with ammonium acetate, with subsequent washing of the aqueous phase with chloro- form, was found to yield a solution that failed to develop the full rhenium - furildioxime colour within a reasonable time, and this was attributed to the retention of traces of reduced molyb- denum by the aqueous layer. This conclusion was supported by the fact that further treat- ment with oxine in chloroform did not effect any improvement. Moreover, the presence of sexavalent molybdenum was known to result in the production of a blue colour when stan- nous chloride was added later, and this did not normally occur.The effect was presumably due to slight reduction of the molybdate by the oxine itself or by traces of impurities, It was therefore decided to oxidise any molybdenum compounds to molybdat e with permanganate after the first extraction with oxine, so that a second extraction with oxine in chloroform would be effective. Our expectation was realised in practice, and a satisfactory procedure was developed. METHOD REAGENTS- 8-Hydroxyquinoline (oxine) solution-Dissolve 5 g of analytical-reagent grade 8-hydroxy- quinoline in 100ml of chloroform.Stannous chloride solution-Dissolve 10 g of analytical-reagent grade stannous chloride in 10 ml of concentrated hydrochloric acid, and dilute to 100 ml with water. Ammonium acetate solution-Dissolve 15 g of analytical-reagent grade ammonium acetate in 100ml of water. or-Fuddioxima solution-Dissolve 0.35 g of pure or-furildioxime in 200 ml of acetone. (Meloche et al. used 0-7 g in 200 ml and reported that the solution turned yellow within 1 day. We observed no instability with a solution of the pure a isomer.) Stavldard rhenium solution-Dissolve 1-553 g of potassium perrhenate in 1 litre of water, and dilute 10-ml portions to 100 ml, as required. 1 ml = 100 pg of rhenium. Standard moly bdate solution-Dissolve 18 g of analytical-reagent grade ammonium molybdate in 1 litre of water.1 ml = 10 pg of molybdenum. (This solution was prepared for use during the development of the method. It is not required during routine determinations.) PROCEDURE- The sample solution should contain not more than 600pg of rhenium and not more than 50mg of molybdenum per 10m1, and both should be in the fully oxidised state, as perrhenate and molybdate, respectively. Transfer 10 ml of neutral sample solution to a 100-ml separating funnel, add 3 drops of 5 N sulphuric acid and 15 ml of ammonium acetate solution, and dilute to 45 ml with water. Mix, add 50 ml of oxine solution, and shake until the aqueous layer becomes clear. Reject the chloroform layer, and wash the aqueous phase with 15 ml of chloroform.Reject the washings, add 0-1 N potassium permanganate, dropwise, to the aqueous solution until a slight permanent pink colour is established, and then re-extract with a further 25 ml of oxine solution. Allow the two layers to separate, wash the aqueous layer with two successive 15-ml portions of chloroform, and filter it through a Whatman No. 41 filter-paper to remove any precipitated manganese dioxide. Collect the filtrate in a 100-ml calibrated flask, add 14 ml of 5 N hydrochloric acid, 26 ml of cc-furildioxime solution and 10 ml of stannous chloride solution, and dilute to exactly 100 ml with water. Shake, set the solution aside for 1 hour, and measure the optical density at 532 mp in a 1-cm cell. RESULTS In practice, we found that calibration curves constructed by treating aliquots of standard perrhenate solution alone (100 to 600 pg of rhenium), exactly as above, were almost identical with curves derived from similar amounts of perrhenate in the presence of 50 mg of molyb- denum, as molybdate.The two nearly congruent graphs were rectilinear and passed throughMarch, 19631 THE FURILDIOXIMES. PART II 193 the origin. Our results for the experiments, in which we used the most adverse ratios of molybdenum to rhenium (viz., rhenium was determined in the presence of 50 mg of molybdenum), were- Rhenium added, pg . . .. 100 200 300 400 500 600 Rhenium found, pg . . .. 104 904 291 393 491 605 Theoretically, there seems to be no upper limit to the ratio of molybdenum to rhenium that would prevent successful separation, but, because of the relatively large amounts of oxine solution and of chloroform required, the practical limit is set at about 500 to 1, by coincidence the same as that claimed by Meloche, Martin and Webb for their xanthate separation. Up to 100 mg of molybdenum were successfully separated from 100 pg of rhenium in isolated experiments (i.e., not part of a series), but the procedures were tedious. REFERENCES 1 . 2. 3. 4. 5. 6. Meloche, V. W., Martin, R. L., and Webb, W. H., Anal. Chem., 1957, 29, 527. Kassner, J . L., Ting, S. F., and Grove, E. L., Talanta, 1960, 7 , 269. Geilmann, W., and Weibk, F., 2. anorg. Chem., 1931, 199, 347. Michailova, O., Posner, S., and Archipova, N., 2. anal. Chem., 1932, 91, 25. Jasim, F., Magee, R. J . , and Wilson, C. L., Tulanta, 1960, 4, 17. Gentry, C. H. R., and Sherrington, L. G., Analyst, 1950, 75, 17. Received July 26tk, 1962

ISSN:0003-2654    

DOI:10.1039/AN9638800191

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

194 PRIESTLY AUTOMATIC DIGITAL THERMOMETRIC TITRATOR [Analyst, VOl. 88 An Automatic Digital Thermometric Titrator* BY P. T. PRIESTLEY (Research Laboratories, Kodak Limited, Wealdstone, Harrow, Middlesex) An apparatus is described for carrying out volumetric analysis wherein the cessation of heat change indicates the equivalence-point. Titrant is added at constant rate to the titrand so that the voltage output of a thermistor temperature transducer varies linearly with time up to the equivalence-point. The voltage change is electronically differentiated to give a square wave of time period equal to the titre. A timing counter is automatically started at the origin of this square wave and stopped when it begins to decline. The titre is thus given in a digital form. The titrator will operate with titrations giving rates of temperature change greater than 0.01" C per second, and results are given for several different titrations. THE fact that volumetric analysis can be performed by measuring the heat liberated or absorbed during a chemical reaction has been realised for some four decades.For example, when 1 ml of M sodium hydroxide is added to 30 ml of ~ / 3 0 hydrochloric acid the temperature rise is about 0.4" C. Similarly, if M tetrasodium ethylenediaminetetra-acetate and ~ / 3 0 magnesium chloride are substituted for alkali and acid under similar conditions the temperature falls by about 0.08" C. The energy changes occurring in such reactions are stoicheiometric, that is to say, they are proportional to the amount of substance formed, and so might be used as a measure of this amount.In practice it is unnecessary to measure the absolute heat of reaction, and in thermometric titration the cessation of heat liberation or absorption is taken as the indication of the equivalence-point. Dutoit and Grobet,l who used a Beckman thermometer and a Dewar flask, plotted temperature change against the volume of titrant added and thus observed a change in slope at the equivalence-point of the reaction. This method of recording tem- perature is tedious, and in the last decade research workers have used thermistors, resistors with high temperature coefficients of resistance, to detect the extremely small temperature changes encountered in these titrations. Several workers in America2 have suggested that this principle of measuring heat changes or thermometric titration could be of great value, since the equivalence-point of a reaction followed in this way would not be obscured by other effects such as colour of solution or poisoning of electrodes. If suitable precautions were taken, side effects owing to dilution or stirring could be avoided, since these would be included in the standardisation process.Jordan and Alleman3 used a syringe burette driven by a synchronous motor that de- livered titrant at a constant rate up to about 0-6ml per minute. The reaction vessel was thermally insulated by being contained in a Dewar flask, and stirring was by means of a 600 r.p.m. glass stirrer. The molar concentration of the titrant was usually about 100 times that of the titrand in order to minimise volume changes and the effect of slight temperature differences. Temperature changes were followed on a recording millivoltmeter, fed from a bridge circuit incorporating a thermistor, a t a chart speed of 1 inch per minute. The plot obtained showed a temperature ordinate and a titre (in millilitres or minutes) abscissa.The straight-line portions of the curve were extrapolated to give the initial and equivalence-points, and the distance between these points was measured along the time axis of the chart to give the titre. This method of evaluating the titre is not entirely satisfactory, especially for routine operation, since it is tedious and introduces error. If this equivalence-point could be detected automatically, then the burette could be stopped at this point and the volume of titrant delivered measured.Alternatively, the time between the initial and equivalence-points might be automatically measured and given in a digital form, since this time is proportional to the volume of titrant delivered. Under these conditions a difference of k0.3" C could be tolerated. * Communication No. 2281H from the Kodak Research Laboratories.Fig. 1. Digital thcrrnometric titratoriMarch, 19631 PRIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR 195 The apparatus described here does, in fact, perform these operations. The main require- ments of such a titrator are not only the ability to detect small temperature changes, but also the ability to discriminate between low orders in rate of temperature change (0.01" C per second) necessary for equivalence-point detection.As previously stated, thermistors have sufficient sensitivity, their characteristics satisfy the first requirement, and, consequently, they are utilised in the digital thermometric titrator. The other requirement, which is more difficult to attain, was satisfied by differentiation of the temperature - time curve. Zenchelsky and Segatto4 in their paper on derivative thermometric titrations pointed out that an increase in the rate of addition of titrant would be advantageous, since this would increase the rate of temperature change and thus permit the first or second differential of the temperature - time curve to be used to obtain sharper equivalence-point discrimination. They suggested, however, that a limit to such an increase in rate of addition of titrant would be set by problems of heat transfer, and gave 2 ml per minute as their estimate of the limit as applied to a volume of about 25 ml of titrand.During some earlier work in this laboratory, in which high-speed mixing was required, it had been found that a vibrator driving a Perspes paddle drilled obliquely with a number of holes was much more effective and convenient than other methods tried, and consequently was adapted for use in thermometric titrimetry. It has, in fact, been found that the efficiency of stirring allows use of a titrant addition rate of up to 10ml per minute, with a reduction in both titration time and the effect of small differences between temperatures of titrant and titrand.In the apparatus described here a delivery rate of 7 ml per minute has been utilised. GENERAL DESCRIPTION AND MODE OF OPERATION Fig. 1 shows the external appearance of the titrator, its general layout and the control positions. It consists of a constant-delivery-rate burette, the black component, and ,a white tower that acts as a stand for the titration vessel and also accommodates the transistor circuitry necessary for the operation of the titrator. Titrant, which is drawn from a reservoir into a glass syringe, is delivered into the titrand, held in the titration vessel, when the syringe- piston is driven into its barrel. A timing counter is automatically started when the tem- perature of the titrand begins to change and is stopped at the equivalence-point of the titration.The titre can then be read from the counter in a digital form. The mode of operation of the titrator may be more easily understood by reference to Fig. 2. When the titration is begun the temperature of the titrand begins to change at a constant rate with respect to time. The thermistor transducer converts this temperature change into an electrical signal, which takes the form of a linear decrease in voltage, for an exothermic titration, up to the equivalence-point. A filter circuit smooths out stirring noise, and a resistance - capacitance network differentiates the signal to give a square-wave voltage of amplitude proportional to the rate of temperature change and of time period equal to the duration ofithis change. This voltage is then amplified about 1000 times by a conventional lter and R.C.1 ~ , I I I I > L V I (d trl,,rPr I-{ differentiating circuits Burette and Equivalence-point Fig. 2. Basic schematic diagram196 PRIESTLEY AUTOMATIC DIGITAL THERMOMETRIC TITRATOR [Analyst, VOl. 88 transistor long-tailed pair amplifier to give an output of workable size. The equivalence- point detector consists of a transistor relay circuit that switches on at the change of voltage, i.e., as soon as the thermistor detects a change in temperature, and switches off when the temperature stops changing. Thus the detector relay automatically starts and stops the timing counter, in the burette, at the appropriate points. When the timing counter stops the burette ceases to deliver titrant and automatically refills.BURETTE- The constant-delivery-rate burette, which incorporates the timing counter and utilises a motor-driven interchangeable glass syringe, is housed in the black Perspex box (see Fig. 1) and can be easily removed from the white Perspex base, on which it is located by two spigots. The motive power is supplied by a synchronous motor having a constant speed of 3000 r.p.m. and the drive is transmitted to a micrometer screw having 40 turns per inch via a 30: 1 reduction worm gear, thus giving a screw speed of 100 r.p.m. The micrometer screw is supported a t its ends in bearings, and a nut is allowed to travel along the length of the screw when the latter rotates. A vertical pillar in contact with the end of the syringe piston is connected to this nut. Consequently, when the screw rotates the pillar moves in a longitudinal direction, pushing the piston into its barrel.Microswitches are suitably placed near the ends of the micrometer screw to limit the travel of the pillar to 1 inch and prevent damage. In order to allow withdrawal of the syringe piston, a stainless-steel clip connects the end of the piston to the pillar. A small gap (& inch) between the two allows the pillar to attain maximum velocity before it strikes the end of the syringe piston. The 5-ml interchangeable glass syringe is modified at its delivery end as described below (see Fig. 1). The Luer tip is removed, the hole is sealed, and two small holes are formed on the periphery of the barrel, close to the tip end, one above the other and on opposite sides.Two glass non-return valves, fitted with mercury-weighted ground-glass inserts, are then fused at right angles to the barrel, use being made of the holes previously blown, and are arranged in a vertical axis, one above the other, with an upward flow path through each when the svringe barrel is horizontal. The delivery arm, which directs titrant into the reaction vessel, has a short length (4 inch) of 0.2-mm capillary at its end to act as a jet to ensure a rapid delivery velocity of 370 cm per second, at a rate of 7 ml per minute, into the titrand. The jet is arranged to be in the “nine-o-clock” position when the titration vessel is viewed from the front. The lower valve is connected to a sintered-glass filter and thence to the titrant reservoir. CONSTRUCTIONAI, DETAILS PLI k*;% .5 Ski? Fig.3. Circuit diagram of burette (for values of components, see Appendix, p. 202)March, 19631 PIIIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR 197 When the piston is driven into the barrel, internal pressure causes the lower valve to close and the upper one to open, allowing titrant to enter the titration vessel. ,4s the piston is withdrawn from the barrel, reduced pressure in the latter causes the upper valve to close and the lower one to open, allowing fresh titrant to enter in preparation for the next delivery. EXPLANATIOX OF BUKETTE CIRCUIT (FIG. 3) All relay contacts and switches are shown in their normal (off) positions. I t is important that relay B should have a smaller time constant than that of relay A, so that at any instant relay B may be energised before the contacts of relay A start to change over. The motor is wired so that relay A when energised effects titrant delivery. When switch S, (situated at the rear of the burette) is in the position shown, automatic delivery is obtained in the manner described below.A momentary depression of the start button PB, energises relay A via contacts B,, the motor-start capacitor, C,, and the motor coil, and it is held on by contacts A, and S,. Contacts A, connect C, across one motor coil for forward motion (delivery), and A,, by-passing the stop microswitch, MS,, which is held open by the pillar when in its stop position, supplies mains current to the motor. The syringe piston will now be driven into the barrel, effecting delivery until the burette pillar hits the reversing microswitch, MS,.C, is then connected across the other motor coil for reversal and refill, and relay A is de- energised. Contacts A, open, but the motor does not stop since stop microswitch MS, is closed; A, open and A, revert to their normal positions preventing re-energisation of relay A if the start button is depressed again. The piston is now automatically withdrawn from the barrel effecting refill and, when the burette pillar strikes stop microswitch MS,, the latter opens and the motor stops. Since there is a time delay between depression of the start button and the instant at which the thermistor detects the heat change in the titration vessel, the timing counter is remotely started bj7 an impulse transmitted to the burette at points 4 and 5 of socket SKT,.Thus, when the burette is started and points 4 and 5 are automatically shorted in the tower by a temperature change, relay R is energised. Contacts B, change over, and the counter is connected to the neutral line causing it to register time at a rate of 50 counts per second. Contacts B, open and de-energise relay A, causing contacts A, and A, to open and A, to change over, thus ensuring that C, remains connected to the neutral line and that delivery continues uninterrupted. At the equivalence-point of the titration being performed, points 4 and 5 of SKT, are no longer shorted, and relay B is de-energised; this in turn stops the timing counter, reverses the motor and effects refill. W7hile the motor is reversing, the primary connections of transformer T, are both at the same potential, and consequently relay B cannot be re-energised if the points 4 and 5 of SKT, are again shorted.Transformer T, supplies 6.3 volts at 1-5 amps to the tower via points 1 and 2, and operates the vibrator unit used to effect stirring. The safety factors listed below have been taken into account- (1) The burette cannot be started if points 4 and 5 are shorted. (2) The counter will not operate until the burette is delivering titrant and points 4 and 5 are shorted. (3) The counter and delivery cannot be restarted, after points 4 and 5 have opened at the equivalence-point, until the burette has refilled and stopped. (4) If the start button is still depressed when the piston reaches the end of its forward travel, the pillar pushing the syringe piston will automatically reverse and effect refill.(5) If the equivalence point is not reached a t the end of the forward travel,, refill should occur since the thermistor will normally detect a break in addition of titrant and cause points 4 and 5 to open. In practice, a count greater than 22 seconds should be suspect, and the titration repeated with a smaller concentration of the titrand. I t should be noted that any proprietary constant-delivery-rate burette (such as the Metrohm or Meniscomatic) may be used in the apparatus described. Manual control of delivery is acceptable, since it is not essential for delivery to cease at the equivalence-point of the titration. Nevertheless, a timing counter, which can be remotely controlled, is required for determining the duration of the titration.Such a counter should be connected to pins 4 and 5 of PL, (see Fig. 4) via a suitable relay circuit, and 6 volts d.c. should be applied t o pins 5 and 3 of PL, for operation of the pilot lamp, ILP,.198 PRIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR [ A ?ZdySt, VOl. 88 EXPLANATION OF TOWER (FIG. 4) STIRKER- The paddle stirrer is driven by a vibration generator similar to that in a loudspeaker unit. The vibrator, which has a resistance of 3 ohms, is supplied with 50 c/s a.c. current from points 1 and 2 a t a voltage of 6-3 volts, and the amplitude of vibration is controlled by the 10-ohm variable resistor, RV,, situated low on the right-hand side panel of the tower. RV, is modified by taking off the last few turns of resistance wire so that, as this resistance is increased towards the maximum value, to reduce the stirring rate, the circuit is broken a t this value.The moving armature of the vibration generator is connected to a horizontal arm, 5 inches long and made of clear Perspex tube 8 inch outside diameter and inch inside diameter, by an aluminium clamp internally lined with sponge-rubber. One end of this arm is joined to a Tufnol mounting containing a ball-race that acts as a pivot. The other end of the arm is mitred to a vertical arm made of the same material and of length 14 inches. The horizontal arm passes through the slotted front panel of the tower. The small vertical arm acts as a holder for the thermistor, which is fitted with a short rubber sleeve so that the thermistor will remain attached to this arm.The thermistor leads are soldered (this is most important since two types of sockets have proved most troublesome in practice) to a length of coaxial cable that runs inside the horizontal arm and emerges at the pivot end. This arrangement ensures that there can be no corrosion of the thermistor leads, which could cause considerable electronic noise. PL3 TI Fig. 4. Circuit diagram of tower (for values of components, see ,4ppendix, p. 202) The circular paddle is of clear Perspex, &-inch x 14-inch diameter, obliquely drilled with holes, six of i-inch diameter being arranged in a hexagon around the centre of the wheel and others of &-inch diameter filling in the outer annulus.The paddle is secured to the end of the thermistor with a small rubber sleeve and positioned so that the temperature-sensitive tip of the thermistor protrudes through it. TITRATION VESSEL AND HOLDER- The titration vessels are cylindrical polyt hene beakers, without lips or protrusions, of 2 inches outside diameter x 2$ inches in length. Polythene has proved most suitable because of its good heat insulation and chemical resistance. These vessels are held in a clear Perspex tube, the inside diameter of which is slightly greater than the outside diameter of the vessels, that is mounted on the front of the tower at a height that allows the vessel to be insertedMarch, 19631 PRIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR 199 from below. titration vessel after a titration.the vessel securely in a vertical position during a titration. THERMISTOR TRANSDUCER- A thermistor, TH, (Standard Telephone and Cables Ltd. type F23), having a resistance of about 2000 ohms at 20" C and a temperature coefficient of resistance of 0.04 ohm per ohm per "C is used as the temperature-sensing element, and is incorporated in a simple temperature transducer together with resistance R,. The effective resistance of the com- bination varies nearly linearly (to within about 3 per cent.) with the temperature of the thermistor, and during an exothermic titration the voltage across R, drops at a rate of about 35 mV per "C. FILTER AND DIFFERENTIATING CIRCUITS- The signal from R, is smoothed by a conventional filter consisting of R, and C,, which also eliminates 50 c/s mains pick-up.The time constant of 100 milliseconds issufficient and causes only slight attenuation. The voltage variations occurring across C, are elec- tronically differentiated by the R.C. network consisting of C,, a 40-pF paper-insulated capacitor, and the input resistance of the amplifier, which is about. 1000 ohms. The time constant of this differentiating circuit is about 40 milliseconds, a compromise that has proved most suitable in practice. The sharpness of detection of voltage change could be increased by reducing the time constant of the differentiating circuit, but this gain would be associated with an attenuated and noisier differentiated output. Further stages of amplification would be required to raise the signal to the previous level and this would itself give rise to amplifier noise, and so the net effect of achieving sharper equivalence-point detection would be to give much more noise. This holder has two finger slots, one on each side, to facilitate removal of the A polythene clip, screwed to the top of the holder, holds TRANSISTOR AMPLIFIER- The magnitude of the differentiated signal developed across R, is increased by the 3-stage long-tailed pair amplifier comprising transistors VT, to 6.This amplifier has a gain of about 1200, and since direct coupling is used the base resistances of VT, and F7T2 must be adjusted to get the other transistors working on the correct parts of their characteristic curves. Re- sistors R, and R, facilitate this operation. The voltage between the collectors of VT, and VT6 can be brought to zero by the action of the 10-turn 15,000-ohm potentiometer, RV,, which is the amplifier balance control and is situated just below the balance meter at the top of the tower.The amplifier output can be attenuated by the sensitivity control, RV,, which is situated near the top of the right-hand side panel of the tower. This control, which covers a linear range of 4 : 1, limits the current into the meter and detector-relay circuit. Control RV, is fitted with a pointer that is surrounded by a plastic bezel linearly calibrated 0 to 10. When the pointer shows 0 the effective resistance of RV, is 1000 ohms; in the position 10 the effective resistance is zero and the sensitivity a maximum. The meter, which reads 50-0-50~4 and has a resistance of 850 ohms, always gives a deflection to the right during exothermic titrations and to the left for endothermic titrations.RANGE SWITCH- This switch, S,, has six poles and six positions and must be of the break-before-make variety. Poles A, B and C switch on the three batteries in positions 2 and 3, whereas in positions 4, 5 and 6 the battery voltages are individually shown on the meter. The meter needle should register between 38 and 50 pA on the scale, and the batteries must be renewed if necessary. Pole D switches the amplifier out of circuit to facilitate these battery checks. The remaining poles E and F transmit or reverse the output of the amplifier to the detector- relay circuit. RATTERIES- battery, BY,, are fitted with standard irreversible 2-pin plugs.compartment at the base of the tower. The two 4-5-volt batteries, BYza and BY,b, are in one container, and they and the 9-volt The batteries are kept in a200 PRIESTLEY AUTOMATIC DIGITAL THERMOMETRIC TITRATOR [Artalyst, Vol. 88 EQUIVALENCE-POINT DETECTOR- The detector circuit, which is the heart of the apparatus, consists basically of a sensitive differential relay circuit utilising a dual-coil Carpenter relay with bi-stable contacts. The relay contacts C, change state when the voltage across capacitor C, (mains hum suppressor) quickly changes by more than 30 mV. This voltage may rise to about 200 mV, and if there is no change in direction of this voltage the contacts C, remain in the same state, but they will revert to the other position if the voltage suddenly drops by more than 30 mV from a constant value.Consequently, before the start of a titration, contacts C, must be in the position where tags 9 and 10 are shorted and this is indicated by the illumination of signal lamp JLP,, which shines through the transparent casing of the meter. As a titration proceeds a voltage change suddenly occurs across C, that causes contacts C, to change over, and this in turn starts the timing counter and extinguishes the signal lamp. The diode, MR,, acts as a spark quenching element, since the switched load, relay B, of the detector Carpenter relay is inductive. The potentials of the sliding contacts of potentiometers RV, and RV, are set at -0.9 volt and -0-65 volt, respectively, and the collector potentials of VT, and VT, should be about -2.2 volts.The bias control, RV,, situated underneath the press button, PB,, on the side panel of the tower, allows the collector voltages to be made equal since the difference is shown on the meter when PR, is depressed. SETTING-UP AND OPERATION OF TITRATOR About 30 ml of titrand are placed in the titration vessel, which is then placed in the holder. The stirring rate is increased from zero until “spitting” of the titrand does not quite begin. The range switch is set to the required setting, and the sensitivity control is set to a suitable value. The amplifier balance control is rotated until the meter reads zero, making sure that the signal lamp that shines through the meter is alight. The bias-check press button is then depressed, and the bias control is rotated for zero meter reading. The burette counter is re-set to zero, and, when the burette start button is depressed, titrant is added to the titrand, and the meter shows a deflection as the counter automatically starts to register elapsed time.At the equivalence-point the meter deflection starts to return to zero and the counter stops. The stirrer is turned off and the titration vessel is removed; finally, the stirring assembly is washed by imrnersion in distilled water. The count is noted and compared with that for a standard strength titrand so that the titre may be evaluated. It should be noted that the detector circuit will respond to fluctuations in intensity of sunlight and, for this reason, the titrator should be shielded whenever necessary.RESULTS AND CONCLUSIONS The reproducibility of the results obtained has been evaluated by performing several titrations with M hydrochloric acid as titrant and 30 ml of ~ / 3 0 sodium hydroxide as titrand. The results of 15 replicates gave a mean count of 8-56 seconds and showed a standard deviation of 0.08 seconds. A second series of 15 replicate determinations containing 30ml of ~ / 1 5 sodium hydroxide gave a mean count of 1 7 . 1 6 seconds and the same standard deviation of 0.08 seconds. The stroke of the syringe piston limits the maximum volume of titrant to about 2.5 ml, which at the rate of 7 ml per minute corresponds to a count of 22 seconds. For good precision the minimum volume of titrant used was limited to 0.5 ml, and thus the titrator was operated with titrand concentrations giving counts (titres) in the range 4.4 to 22 seconds.In order to evaluate the relationship between the concentration of titrand and the count (titre) indicated by the titrator, series of similar titrations were performed. Each series included titrations of the same titrand at different concentrations with a standard titrant, e.g., 30 ml of ~ / 6 0 , M/40, ~ / 3 0 , ~ / 2 4 . . . ~ / 1 2 sodium hydroxide with M hydrochloric acid, giving counts between 4.4 and 22 seconds. Further series of titrations (see Table I) were performed with other titrands and M hydrochloric acid as titrant. Results for a range of titrants and titrands are shown in Tables I1 to V. The Ferranti Sirius Computer, suitably programmed, was utilised to evaluate, by the method of least squares, the best straight-line relationship between titrand concentration and count (titre) for each series of titrations. The zero intercept in units of time (seconds) is given in column 6 of the Tables, and its magnitude is influenced by the extent of delayedMarch, 19631 PRIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR 201 reaction between reactants or the solubility of the reaction product in the titrand.Even in titrations for which the intercept term is large, the method may still be utilised, since a calibration can be employed. The computer was also used to calculate the standard deviation of the line, i.e., the scatter of the results across the best straight line showing the relationship between concentration and count.The results in column 7 (headed “Sigma”) are in units of time (seconds) and give an indication of the thickness of the scatter of points around the TABLE I TITRATIONS WITH M HYDROCHLORIC ACID AS TITRANT Titrand NaOH . . . . &NO3 NH,OH Na2C03 . . NaHCO, . . Na2B407 . . Na,PO, . . Na2S203 . . KIO, . . . . . . KI + excess of Na,(EDTA) .. KIO, + excess of “It1 NaH,PO, NaHSO, . . . . . . Titrand Ca(N03)2 NiSO, . . CdSO, . . ZnSO, . . BaC1, . . c u s o , . . &NO, coc1, . . Cr(NO3) 3 Mg(NO,), Ce(S04)2 SnC1, . . AlC1, . . LiNO, . . NH,NO, Sr(NO3) 2 Fe(NO313 Hg 2 Fe(N03)2 MnC1, . . Titrand Pb AgPiO, . . HC1 . . cuso, . . Zn(N03)2 AlC1, . . .. .. .. .. . . . . .. .. . . .. . . . . . . . . . . . . .. . . . . . . .. . . . . . . . . . . . . Type + + + + + + + + Setting 0 0 0 8 10 10 5 10 No.of Intercept, titrations seconds 10 - 0.05 10 - 0.35 10 - 0.19 10 - 0.32 7 - 0.33 6 - 0.29 10 - 0.28 5 - 0.07 Sigma, seconds 0.09 0.08 0.17 0.07 0.05 0.12 0.09 0.13 ~ / 6 0 + 5 5 - 2.22 0.18 M/120 + 5 7 - 0.75 0.03 ~ / 3 6 0 + 5 6 -0.71 0.03 - (Insufficient heat t o give an end-point) - (Insufficient heat t o give an end-point) TABLE I1 TITRATIONS WITH M TETRASODIUM EDTA AS TITKANT Type + + -+ + + + + + + + - - - - + + + + + + - No. of Intercept, Sigma, Setting titrations seconds seconds 9 8 -0.15 0.16 8 11 -0.11 0.10 8 11 + 0.02 0.08 10 11 -0.16 0-07 10 5 - 0.01 0.06 10 10 - 0.01 0.14 5 4 - 1.15 0.32 10 4 - 0.93 0.05 0 7 - 0.27 0-14 8 10 -0.17 0.09 7 3 + 0.56 0.03 0 8 - 0.21 0.12 7 5 - 0.02 0.25 6 6 + 0.03 0.2 1 (Insufficient heat t o operate titrator) (Insufficient heat t o operate titrator) 5 7 TABLE 111 TITRATIONS WITH M SODIUM CARBONATE AS TITRANT No.of C m Type Setting titrations ~ / 6 0 + 10 4 ~ / 3 0 + 0 8 ~ / 6 0 - 6 7 ~ / 6 0 - 8 6 ~ / 3 0 + 2 6 M / 9 0 - 6 8 Intercept, seconds - 0.34 f0-15 0.00 - 1-48 - 3.20 -3.71 Sigma, seconds 0-2 1 0.17 0.08 0.17 0.13 0.44 Sigma, 0.6 0.5 1.1 0.5 0.3 0.8 0.6 0.8 1.2 0.2 0.2 % Sigma, 1.0 0.6 0.5 0.5 0.4 0.9 2.0 0.3 0-9 0-6 0-2 0-8 1.6 1-4 % Sigma, 1.4 1.1 0.5 1.1 0.8 3.0 70202 PRIESTLEY : AUTOMATIC DIGITAL THERMOMETRIC TITRATOR [ A ?ZdySt, VOl. 88 TABLE IV TITRATIONS WITH M / 4 CERIC SULPHATE IN SULPHURIC ACID AS TITRANT Titrand KI . . .. Na,SO,. . .. NaZS203 .. H ydroquinone Metol . . . . Phenidone . . Resorcinol . . Type + + + + + + + Setting 10 10 10 5 5 10 8 No.of titrations 8 13 8 8 7 11 8 Intercept, seconds - 3.08 - 0.54 - 1.33 - 0.80 - 0.77 SO-19 - 1.00 Sigma, seconds 0-24 0.1 1 0.06 0.06 0.06 0.2 7 0.12 TABLE V TITRATIONS WITH 2 M SODIUM THIOSULPHATE AS TITRANT No. of Intercept, Sigma, Titrand c m Type Setting titrations seconds seconds 1,in KI . . .. ~ / 6 0 + 10 8 - 0.05 0.24 Na,SO, in KI, (titration of ?xcess of Iz)* . . ~ / 1 0 0 0 + 10 6 + 0.02 0.06 NaHSO, in KI, (titration of excess of Iz)* . . ~ / 1 0 0 0 + 10 6 +0.01 0.08 Ce(SO,), in H,SO, ~ / 3 0 + 10 8 + 0.82 0.10 Hydroquinone in (titration of ex- KzCr29, + H,SO, cess of Cr,O,,-) t ~ / 4 0 0 0 + 0 4 -0.13 0.22 * Maximum concentration ~ / 1 2 . t Maximum concentration ~ / 5 0 . Sigma, 1.5 0.7 0.4 0-4 0-4 1-8 0-8 % Sigma, % 1.5 0-6 0.4 0-5 1.4 best straight line drawn through the points.Column 8 gives the coefficient of variation of the scatter for titrations giving a count of 15 seconds. Column 2 gives the minimum con- centration of C, of titrand that can be used with the standard titrant and the maximum concentration is in all cases 5 CnL, unless specified to the contrary. Column 3 (headed “Type”) indicates the type of reaction, + denoting exothermicity and - endothermicity. Column 4 (headed “Setting”) gives the values of the sensitivity setting utilised in the series of titrations ; column 5 shows the number of titrations in each series. Thus the first line in Table I indicates that sodium hydroxide was titrated against M hydrochloric acid in a concentration range of ~ / 6 0 to ~ / 1 2 with the titrator range switch at the position “EXO” and the sensitivity pointer at position 0.The intercept is -0.05 seconds, which means that if 0.05 is subtracted from the count for each of the ten titrations in this series then the straight-line relationship between concentration and count would pass through the origin. The coefficient of variation of the scatter for a count of 15 seconds is +O-6 per cent. The concentrations of titrants used were convenient values, but it should be noted that, although an increase would produce more heat due to reaction, this would not always give rise to a greater temperature change, since the heat of dilution might be in the opposite sense. Other titrants not included in the Tables have been utilised at concentrations ensuring a rate of temperature change of 0.01” C per second or more; for example, argentometric titrations with ~ / 5 silver nitrate as titrant and precipitation titrations with ~ / 5 sodium sulphide. Appendix LIST OF COMPONENTS USED IN THE CONSTRUCTION OF THE APPARATUS (Figs. 3 and 4) R,, R,, Rll = 2200-ohm resistors.Rz, R,, = 1000-ohm resistors. R3, R,, = 3300-ohm resistors. R,, R,, R,, R,,, R,, = 100,000-ohm resistors.March, 19631 PRIESTLEY AUTOMATIC DIGITAL THERMOMETRIC TITRATOR R5 RQ = Selected values (see text). RlO = 100-ohm resistor. R129 R,4 = 220-ohm resistors. R 1 3 2 R 1 5 p R17 = 150-ohm resistors. RIQ* R20 = 470-ohm resistors. R21 = 10,000-ohm resistor. R22 = 33-ohm resistor. R24 = l-megohm resistor. = 15,000-ohm 10-turn potentiometer (Colvern Ltd.) .= 1000-ohm l-turn potentiometers (Colvern Ltd.) . RV,, RV3 = 2000-ohm I-turn potentiometer (Colvern Ltd.) . = 5-ohm l-turn potentiometer (Colvern Ltd.). = l0-ohm l-turn potentiometer (Colvern Ltd.). = Stantel F23 thermistor, 2000-ohms a t 20" C. = 1.25-pF paper-insulated capacitor, 350-volt working. = 0- l-pF paper-insulated capacitors, 350-volt working. RVl RV4 RV5 THl Cl c2, c3 Rv6 c4 = 50-pF electrolytic capacitor, 12-volt working. c,, c; = 100-pF electrolytic capacitors, 6-volt working. = 40-pF paper-insulated capacitor, 150-volt working. = 2000-pF electrolytic capacitor, 6-volt working. VT,, VT,, VT,, VT, = Mullard OC202 transistors. VT,, ITT4, VT,, VT, = Mullard OCT75 transistors. RLA = 230-volt 50 c/s coil, 3-pole, C/O 5A (M.T.I. Ltd.). RLB = 6-volt d.c. coil, 2-pole, C/O 5A (Siemans Halske). RLC = Carpenter relay, l-pole, C/O 5C6 (Telephone Mg. Co. Ltd.). = 230-volt 50 c/s primary, 8-volt 0.5-amp secondary transformer. = 230-volt 50 c/s primary, 6-3-volt l.5-amp secondary transformer. = Westinghouse 15 RCI-1-16-1 rectifier. = Mullard OAlO diodes. MR,, MR, MS,, MS2 = Honeywell Brown 11 SM1-T microswitches. PB,, PI32 = Honeywell Brown 2 PB 12-T press buttons. ' 6 C, TI T2 MRl = Single-pole switch. = 6-pole 6-way rotary switch, break-before-make (Painton Ltd.) . = Vidor 6011 twin 4-5-volt battery. = Neon indicator lamp. Sl s2 BY1 = Vidor 6007 9-volt battery. ILP, = Pilot bulb, 6.3-volt 0.115 amp. 203 PL, ,A SKT, gk; :$} = $-way plug and socket (Plessey Ltd.). = 6-way plugs and sockets (Plessey Ltd.). MISCELLANEOUS COMPONENTS Counter F43/G3/ J46 (Lancashire Dynamo Co. Ltd.). Worm and worm wheel 30 : 1 (Muffets Ltd.). Micrometer screw (Moore and Wright Ltd.). 5-ml syringe (Chance Glass Ltd.). Synchronous motor FC84/0 (Evershed Vignoles Ltd.) . Vibration generator V47-3 (Goodman Ind. Ltd.). Meter, 50-0-50 pA, type 220 (Taylor Ltd.). Weight of burette = 7 lb (3.2 kg). Weight of tower and base = 14 Ib (6.4 kg). I thank Messrs. A. D. Johnson and D. M. Zeitlin of Kodak Ltd. who processed the results and programmed the computer. REFERENCES 1. 2. 3. 4. Dutoit, P., and Grobet, E., J . Chim. Phys., 1921, 19, 324. Zenchelsky, S. T., Anal. Chew., 1960, 32, 289 R . Jordan, J., and Alleman, T. G., Ibid., 1957, 29, 9. Zenchelsky, S. T., and Segatto, P. R., Ibid., 1957, 29, 1856. First received April 19th, 1962 Amended, October 19th, 1962

ISSN:0003-2654    

DOI:10.1039/AN9638800194

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

204 EVANS AND PAYNE : QUANTITATIVE DETERMINATION OF [AflalySt, VOl. 88 Quantitative Determination of Glucose and Maltose in Enzyme Reaction Mixtures BY W. A. L. EVANS AND D. W. PAYNE* (Zoology Department, University College, Cardiff) The application of a carbon-column selective-desorption technique is shown to be suitable for determining maltose and glucose in enzyme reaction mixtures. Precipitation of the enzyme protein and removal of the buffer ions is unnecessary for both the column separation and sugar determination by the copper colorimetric method adopted. In enzyme reaction mixtures containing initially 50 mg of maltose, the maltose and glucose concentrations can be determined, when the amount of hydrolysis is within the range 1 to 80 per cent., with a maximum error of about 2.5 per cent.MANY of the methods available for determining each component in glucose - maltose mixtures are time consuming and of limited accuracy. The accurate selective-desorption of sugars from columns of activated carbon by the method of Patterson and Savage1 and Patterson and Buchan2 9 3 appeared to be suitable, after modification, for the quantitative determination of glucose and maltose in enzyme reaction mixtures. However, the effect of enzyme prepara- tions and buffer ions on the determination was unknown, since Patterson and co-workers did not use enzyme reaction mixtures. Before determining reducing sugars in biological liquids it is customary to precipitate the proteins. Since the protein content in these enzyme reaction mixtures was small, it was decided to investigate the effect of not precipitating the protein, as such a procedure would considerably accelerate the determination.METHOD REAGENTS- Activated carbon-"Activated charcoal powder for decolorising purposes : washed with acid," obtained from British Drug Houses Ltd., Poole, Dorset, was used. Kieselguhr-The material purified with acid, obtained from British Drug Houses Ltd., was used. Somogyi* cojbper reagent-Dissolve 28 g of anhydrous disodium hydrogen orthophosphate and 40 g of sodium potassium tartrate in 700ml of hot water. Add 100ml of sodium hydroxide and 80 ml of a 10 per cent. solution of copper sulphate, CuS0,.5H20. Cool, make up to 1 litre, and then add 180 g of anhydrous sodium sulphate. 5" C, and filter, if necessary, after 48 hours.This reagent was preferred to a later Somogyi5 reagent chiefly because the latter formed persistent small bubbles after the addition of the Xelson reagent. These were difficult to remove and they seriously interfered with the colour measurement. Nelson6 chromogenic reagent-Dissolve 25 g of ammonium molybdate in 450 ml oi water, and add 21 ml of concentrated sulphuric acid. Ilissolve 3 g of sodium arsenate, Na2HAs0,.7H20, in water, and mix thoroughly with the molybdate - sulphuric acid solution. Make up to 500 ml, and incubate at 37" C for 48 hours. Filter, and store in a dark-brown bottle at room temperature. Universal b u e r mixture-Obtained from Hopkin & Williams Ltd., Chadwell Heath, Essex. Enzyme preparation-Two freeze-dried preparations were made from the gut of the desert locust, Schistocerca gregaria (Forsk), One consisted of foregut and midgut tissue and the other of foregut and midgut contents.The tissue was homogenised and spun in a centri- fuge (30009) for 10 minutes. The contents were shaken with a minimum of water, and then spun in a centrifuge in the same manner as for the tissue. The supernatant liquids were dialysed against distilled water for 48 hours at 2" C, and finally spun in a centrifuge Store at 25" * Present address : Biology Department, Chelsea College of Science and Technology, London, S.\T'.3.March, 19631 GLUCOSE AND MALTOSE IN ENZYME REACTION MIXTURES 205 for 45 minutes at high speed (22,000 9) before being shelled. Kjeldahl analysis of the enzyme preparations showed that the tissue preparation contained 11 per cent.of nitrogen and the contents preparation contained 7 per cent. of nitrogen. ENZYME REACTION MIXTURE AND PROCEDURE- Although the total volume of reaction mixture was varied, the proportion of the con- stituents were generally kept the same. A typical reaction mixture was 1 ml of a 5 per cent. solution of maltose, 2 ml of buffer solution at pH 5.7 and 1 ml of enzyme extract containing 1 mg of enzyme preparation. The carbon-column assembly was similar to that described by Patterson and Savage.l The temperature of the column was maintained at 38" 1" C, and the temperature of the eluting liquids in the reservoir at the top of the column was kept within this range by suitable lagging. The suction pressure on the column was arranged so that 100ml of eluate were collected in about 40 minutes.To avoid overloading the column with sugar, it was necessary to restrict the amount of reaction mixture placed on the column, so that the maximum amount of maltose was 12-5 mg. Equal volumes of eluate were collected for glucose and maltose determinations, the former being eluted first with water and maltose afterwards with 7 per cent. ethanol. The qualitative separation of sugar mixtures was examined by chromatographic analysis of the eluates after (a) a mixture containing 5 mg of glucose and 5 mg of maltose, (b) a mixture containing 5 mg of glucose, 5 mg of maltose and 5 mg of isomaltose and (c) a solution contain- ing 5 mg of glucose had been passed through the column. The eluates were concentrated by controlled evaporation to 0.2 ml. Appropriate sugar markers and approximately 2.5 pl of the concentrated eluates were placed on the chromatography paper (Whatman No.1 filter-paper) , which was then irrigated by an ascending-solvent technique with a mixture of n-butanol, pyridine and water in the ratio 6 to 4 to 3 parts by volume. The chromato- gram was then dried and developed to show the position of the sugars by Trevelyan, Proctor and Harrison's silver method.' Table I shows that qualitative separation of glucose and maltose was achieved, but isomaltose was eluted with maltose. TABLE I MOVEMENT OF SUGARS PRESENT IN COLUMN ELUATES COMPARED WITH STANDARD MARKERS RG value Sugar identification Markeys- Glucose . . .. . . .. . . . . 1.00 Glucose Maltose .. .. . . .. .. . . 0.64 Maltose Isomaltose . . . . .. . . . . 0.48 Isomaltose Eluate concentrates- 1. Glucose - maltose mixture- Water eluate . . * . .. . . 0.95 Glucose Ethanol eluate . . . . .. . . 0-65 Maltose Water eluate . . .. . . . . 0.97 Glucose Ethanol eluate . . .. .. . . 0.64, 0.47 Maltose, isomaltose Water eluate . . .. .. . . 0.94 Glucose Ethanol eluate . . .. . . .. 2. Glucose - maltose - isomaltose mixture- 3. Glucose solution- - - For the quantitative determination of glucose and maltose in their eluates, the colori- metric method of Somogyi4 and Nelson6 was modified. Four millilitres of eluate and 4 ml of Somogyi reagent were heated in a boiling-water bath for 15 minutes (glucose) or 20 minutes (maltose) in a 30-ml tube, the mouth of which was closed by a glass bulb.The tube was then immediately transferred to a cold-water bath for 20 minutes. Nelson's chromogenic reagent was added, the tube was well shaken, and the contents were made up with water to 10 or 25m1, according to the colour intensity. Maltose eluates were always made up to 25 ml. Eluates containing less than 0-25 mg of glucose were also made up to 10 ml, but eluates containing 0.25 to 0-65 mg of glucose were made up to 25 ml. The optical densities of the coloured solutions were measured after 25 5 minutes with a Unicam SP600 spectro- photometer at 500 mp. For the determination of sugars in eluates of non-enzymic solutions,206 EVANS AND PAYNE : QUANTITATIVE DETERMINATION OF [Analyst, vol. 88 which were used for calibration purposes, the spectrophotometer was adjusted so that zero optical-density readings were given by the reagent blank solutions containing 4 ml of water or 7 per cent.ethanol instead of 4 ml of water or ethanol eluate, respectively. The sugar content of the enzyme reaction mixtures was measured after samples had been immersed in a boiling-water bath for 1 minute to denature the enzyme. Reagent blank solutions for glucose determinations were made up from 4 ml of the aqueous eluate after denatured zero-incubation time samples of enzyme reaction mixture had been placed on the column. This procedure was not possible for the determinations of maltose, for which control reaction mixtures were made up with water instead of maltose solution, and samples were placed on the column. Four millilitres of the ethanol eluate were then used to make up the reagent blank solution.Eluates containing maltose and glucose below the range of the calibration curves were concentrated by controlled evaporation before the Somogyi reagent was added. Eluates containing glucose and maltose above the range of the calibration curves were suitably diluted with water. CALIBRATION- The calibration curve for maltose was h e a r , the maximum amount of maltose being 1.6 mg. Samples containing 1 mg of maltose gave an optical-density reading of 0-639. Two linear calibration curves were also obtained for glucose. When the final volume of coloured solution was made up to 10 ml (0-25 mg of glucose), 0.1 mg of glucose gave an optical-density reading of 0.296, and when the final volume of coloured solution was made up to 25 ml (0.25 to 0.65mg of glucose), 0.5mg of glucose gave an optical-density reading of 0.607.EFFECT OF ENZYME PROTEIN ON THE DETERMINATION OF GLUCOSE- Addition of enzyme to solutions of glucose increased their reducing power. Fig. 1 shows the effect of adding 1 mg of enzyme preparation (gut contents) to glucose solutions before treatment with Somogyi reagent, and then determining their reducing power. The change in reducing power is most marked in extremely dilute solutions of glucose when an increase of 7 per cent. was found compared with a 3 per cent. increase in solutions containing 0.2 mg of glucose. For solutions containing 0.2 to 0.6 mg of glucose, the increase in reducing power was less than 3 per cent. This enzyme protein effect, which was found to be approximately proportional to the amount of enzyme added, was about three times as great with the tissue preparations as with the contents preparation.In usual enzyme reaction mixtures, in which the relative amount of enzyme is a quarter of that used for obtain- ing the results in Fig. 1, the protein effect of the contents preparation was approximately 2 per cent. for extremely small amounts of glucose and less than 1 per cent. for amounts of glucose greater than 0-2 mg. The magnitude of the error is not unreasonable, particularly Glucose, mg Fig. 1. Calibration graphs for glucose: curve A, without enzyme preparation; curve B, with 1 mg of enzyme preparation (gut contents) added. Volume of solution used in determination 2.5 ml with final dilution to 10 mlMarch, 19631 GLUCOSE AND MALTOSE IN ENZYME REACTION MIXTURES 207 when the disadvantages of the alternative procedure of protein precipitation are considered, viz., time and dilution factors.The enzyme protein was therefore not precipitated from reaction mixtures before analysis. RECOVERY OF SUGARS FROM THE CARBON COLUMN- Table I1 shows a typical set of calibration results for the recovery of sugar from pure solutions and mixtures and from solutions containing buffer salts and denatured tissue enzyme preparation. When glucose and maltose solutions were used separately, the amount of sugar recovered was slightly less than that present in the original solution. The analysis of mixtures, however, indicated a further drop in maltose recovery, but an increase in glucose recovery.The results from many determinations indicate that there was at least 98 per cent. recovery of maltose and glucose when used separately, and that a recovery of 102 per cent. was obtained for glucose when used in a mixture with maltose, but the recovery of maltose (97.5 per cent.) was slightly less. These recoveries can be accounted for to a certain extent by the fact that, although the glucose preparation was chromatographically pure, the maltose preparation exhibited a trace of glucose on the chromatogram. The addition of buffer and enzyme preparation to glucose - maltose mixtures affected the recovery of sugar only to a small extent. When reaction mixtures containing ten times the usual amount of enzyme preparation were analysed, the average glucose gain was 2 per cent.and the average maltose loss 3 per cent. The presence of 0-25 mg per ml of denatured enzyme preparation and buffer ions in glucose - maltose mixtures therefore has a negligible effect on the recovery of the sugar. DETERMINATION OF SUGAR IN ENZYME REACTION MIXTURES- The hydrolysis of maltose can be followed quantitatively by measuring the decreasing maltose concentration and/or the increasing glucose concentration. Fig. 2 shows that the progress curves in relation to maltose disappearance and glucose appearance follow a corres- ponding pattern. However, the total sugar recovered decreases progressively with time. This was due to the formation of transfer sugars, the presence of which were established by paper-chromatographic analysis.8 Transfer disaccharides, trisaccharides and higher oligo- saccharides were formed (full details of which it is hoped will be published shortly), and these were presumably not removed from the column by elution with water and 7 per cent.ethanol. Time. hours Fig. 2. Progress curves for the hydrolysis of maltose at 37OC: curve A, maltose; curve B, glucose. Reaction mixture comprised 2 ml of a 5 per cent. solution of maltose, 4 ml of buffer solution at pH 5.8 and 2 ml of enzyme solution containing 2 mg of gut tissue enzyme preparation208 EVANS AND PAYNE [Analyst, Vol. 88 TABLE I1 DETERMINATION OF GLUCOSE AND MALTOSE IN SOLUTIONS AND MIXTURES The dilution factor for the controls, which were not run through the column, was the same as that for the eluates.The denatured reaction mixture contained buffer salts (pH 5.7) and 0.1 per cent. denatured enzyme preparation Amount of sugar in 4 ml of control, mg Solutions Sugar solutions- Glucose . . .. . . . . 0.430 Maltose . . . . . . . . 0.415 A (1 + 1) mixture- Glucose . . . . . . 0.430 Maltose . . .. .. 0.415 Denatured enzyme reaction mixtures- A (1 + 1) mixture- Glucose . . , . . . 0.448 Maltose . . .. .. 0.420 Amount of sugar in 4 ml of eluate, mg 0.421 0.408 0.437 0.405 0.456 0.408 For kinetic studies it is essential to remember that appreciable transfer action is possible as the concentration of glucose increases. Generally, it was found that in reaction mixtures containing 12.5m.g per ml of maltose, little transfer activity was measurable below 5 per cent. hydrolysis of the substrate. The change in concentrations of maltose and glucose in such reaction mixtures was readily measured by the technique given. One of us (D.W.P.) is indebted to the D.S.I.R. for a research scholarship, REFERENCES 1. 2. 4. 5. 6. Nelson, N., Ibid., 1944, 153, 375. 7. 8. Patterson, S. J., and Savage, R. I., Analyst, 1957, 82, 812. Patterson, S. J., and Buchan, J . I,., Ibid., 1960, 85, 75. Somogyi, M., J . Biol. Chem., 1945, 160, 61. ~ , Ibid., 1952, 195, 19. Trevelyan, W. E., Proctor, D. P., and Harrison, J. S., Nature, 1950, 166, 444. Evans, W. A. L., and Payne, D. W., Biochem. J . , 1960, 76, 5 0 ~ . 3. -, -, Ibid., 1961, 86, 160. Received September 12th, 1962

ISSN:0003-2654    

DOI:10.1039/AN9638800204

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

March, 19631 COOMES AND SANDERS 209 The Detection and Estimation of Aflatoxin in Groundnuts and Groundnut Materials Part I. Paper-chromatographic Procedure BY TREVOR J. COOMES AND J. C. SANDERS (Department of Scientific and Industrial Research, Tropical Products Institute, 66/62 G a y ’ s I n n Road, London, W.C. 1) A method for the detection and semi-quantitative determination of aflatoxin-a metaboIic product of the fungus Aspevgillus $av~.ts Link-is described. IT is now recognisedl that the toxicity associated with certain batches of groundnuts is due to the growth on them of a strain or strains of the fungus AspergiZZusJavzts Link ex Fries, which produce, in small amounts, a highly toxic factor to which the name aflatoxin is now being applied. A provisional chemical test, based on the isolation techniques described by Sargeant, Sheridan, O’Kelly and Carnaghan,2 has been in use for many month^.^ Subsequent work has, however, shown4j5 that the toxic factor is often composed of at least two components of different toxicities, and that the paper-chromatographic procedure originally reported3 does not distinguish between them.Nesbitt, O’Kelly, Sargeant and Sheridan6 have recently separated these two components by a counter-current distribution technique and for convenience refer to them as aflatoxin B and aflatoxin G, because of their fluorescent reactions in ultraviolet light when adsorbed on thin layers of aluminium oxide. The same workers6 report that both components are toxic to day-old ducklings, aflatoxin B exhibiting approximately three times the potency of aflatoxin G.This paper describes a method for detecting the commoner and more toxic component, aflatoxin B,* together with a procedure permitting the assessment of the likely biological effects of suspect materials. It must be emphasised that the results and correlations reported in this paper and in Part I1 of this series are based on the purest sample of aflatoxin B available a t the time. It is possible that further work on the isolation of this material may alter somewhat the scale of quantitative values given. It is hoped that these will prove to be at least approximately correct, but, for the present, categories rather than absolute values are preferred for reporting results. EXPERIMENTAL During the purification work on aflatoxin it was noticed repeatedly that biological activity of extracts was accompanied by a characteristic blue fluorescence, the presence of which afforded a fairly reliable indication of toxicity on a semi-quantitative basis.This led to the design of the provisional physico-chemical test referred to above.3 Chromatographic examinations of material eluted with methanol from the fluorescent spots obtained by this procedure with both alumina (Brockmann activity 111) and silica-gel - plaster of Paris (4 + 1) columns have shown4 that the material in fluorescent spots thus obtained is not homogeneous. Confirmation of this work has been obtained independently by thin-layer chromatography on a l ~ m i n a . ~ * Since the work described in this paper and in Part I1 was carried out, the work of Unilever in Holland (personal communication) has shown that the material designated aflatoxin B can be separated on Kieselgel plates, with a methanol - chloroform system, into two blue fluorescent spots having slightly different RF values.Of these, apparently, the material isolated from the faster running spot has been shown to be toxic and that from the slower spot to be non-toxic, but potentiating the biological activity of material comprising the faster spot. Hence the tests in these papers describe the detection of a potentiated toxin and i t is this material that is described as aflatoxin B. That this is a sufficiently reliable indication of over-all toxicity is, in the authors’ opinion, justified, pending further work, since it appears that the pro- portions of the two components of aflatoxin B in different toxic samples are approximately the same.This is supported by the correlations between total aflatoxin B content determined by these methods and the biological activity exhibited by total extracts of samples examined.210 [Analyst, Vol. 88 Descending-solvent paper chromatography with the benzene - cyclohexane - methanol - water (5 : 5 : 6 : 4 by volume) system proposed by Rhodes et a1.' on extracts of certain ground- nut materials demonstrated the existence of two blue-purple fluorescent components therein. Comparison chromatograms of the purest available material prepared from extracts of toxic groundnut meal and of the purest available material obtained by culture of a strain of A .flavus on a synthetic medium (both prepared in this Institute) confirm that the two fluorescent components of the groundnut extracts were aflatoxin B and aflatoxin G. Unfortunately, the development time for this solvent system is about 6 hours at 20" C, and a more rapid system seemed desirable to reduce the over-all time of analysis. The system benzene - toluene - cyclohexane - ethanol - water (3 : 3 : 5 : 8 : 5 by volume), to the upper layer of which is added acetic acid so that its concentration is 1 per cent. by volume, has been found efficacious. Descending-solvent development with this system yields a blue-purple fluorescent spot on paper at RF 0.57 to 0.61, corresponding to aflatoxin B, and from some samples a similar fluorescent spot at RF 0.28 to 0.31, corresponding to aflatoxin G, is obtained at the same time.The solvent front travels approximately 30 cm at 20" C in 3 hours during development. The method described below involves extraction of the comminuted groundnut material with methanol, de-fatting the methanol extract with light petroleum (boiling range 40" to 60" C), extraction of the toxin from the de-fatted extract into chloroform and then chromato- graphy of three portions of the purified chloroform extract on paper. Examination of the chromatogram in ultraviolet light (A 365 mp) permits the level of toxicity to be classified as very high (>2.0 p.p.m.), high (0.5 to 2.0 p.p.m.) medium (0.1 to 0.5 p.p.m.) or low or negative ((0.1 p.p.m.). These ranges have been esta.blished by reference to the chromatographic characteristics and fluorescence phenomena of the purest available sample of aflatoxin B.During the work described it has been observed repeatedly that the minimum detectable concentration of aflatoxin R on a payer chromatogram is 0-2pg. A more accurate deter- mination of the toxin content of a sample can be made by the technique of serial dilution until visual extinction of the fluorescent spot occurs. For practical purposes this extinction occurs at the 0.2-pg level, and, from a knowledge of the dilution required to achieve this extinction, the toxin content can be calculated. COOMES AND SANDERS: DETECTION AND ESTIMATION OF METHOD APPARATUS AND MATERIALS- and washed free from acid before use. joint glassware when appropriate. found suitable, and the manufacturer's reference numbers are quoted below in brackets.The glass apparatus detailed below must be thoroughly cleaned with dichromate mixture I t is convenient to use standard interchangeable- "Quickfit" (Quickfit & Quartz Ltd.) apparatus has been Soxhlet extractor, 200 ml (EX5/83/200). Coil condenser, effective length 200 mm (CX3/0S). Boiling $asks, short neck, 250 ml (FR250,/3S). Water bath, 6-hole to accept 250-ml flasks. L i p i d - liquid extractor, downward displacement type, 60 ml (EX10/23), Disc-bafle distributor, to fit extractor (EX10/20). Double-surface condenser, effective length 200 mm (CX5/35). Separating funnels, conical, 500 ml (D83/500). Chwomatographic column, with integral sinter, effective length 200 mm, bore 18 mm Calibrated $ask, 5 ml.Graduated micropipette, 100-pl capacity, in divisions of 5 p1. Graduated micropipette, 25-pl capacity, in divisions of 1.0 p1. Chromatography tank (22 inches x 12 inches x 7; inches) equipped with solvent trough Chromato4qraphy paper, Whatman No. 1, 46 cm x 57 cm. Extraction thimbles, 41 mm x 123 mm. Solvents-All solvents specified must be of analytical-reagent grade unless otherwise Chmmatograplzic alzmina-Woelm, neutral. (CR32/20). and supports (103 inches). stated.March, 19631 AFLATOXIN IN GROUNDNUTS AND GROUNDNUT MATERIALS. PART I 21 1 SAMPLING- Careful sampling is an essential preliminary to this method, and it is recommended that sampling procedures should follow closely the details laid down by I.U.P. A.C. for oilseeds.8 PROCEDURE- (a) Extraction-Extract 805' of finely ground (see Xote 1) material with methanol for 6 hours in a 200-ml Soxhlet extractor (syphon rate 4 to 6 cycles per hour).Cool, transfer the methanol extract quantitatively to a 500-ml separating funnel via a small funnel con- taining a plug of glass-wool, wash the extraction flask with two 10-ml portions of methanol, and add the washings to the contents of the separating funnel. Add 50 ml of light petroleum (boiling range 40" to 60" C) to the contents of the separating funnel, and shake vigorouslj7 for 2 minutes. Transfer the lower (methanol) layer to a second 500-ml separating funnel, and repeat the extraction first with a 25-ml portion and then with a 15-ml portion of light petroleum, finally running the methanol layer into a 250-ml boiling flask.Distil off the methanol at atmospheric pressure, and remo\'e the last traces in vaczco. Disperse the fat-free methanol-soluble residue in 60 to 80 ml of water, and extract with chloro- form in a downward-displacement-type continuous liquid - liquid extractor for 1 + hours (see Note 2). Reduce the volume of the chloroform extract to about 1 ml by distillation at atmospheric pressure, and place it on a chromatographic column containing neutral alumina (10 g ; Brockmann activity I). Elute with 100 ml of chloroform containing 5 per cent. v/v of methanol, concentrate the eluate to small volume, and make up to 5 ml in a calibrated flask with chloroform. Add approximately 0-25g of anhydrous sodium sulphate to the chloroform solution in the calibrated flask, shake, and allow to settle for 15 minutes.( b ) Chromafo~mphy--Shake the solvent system benzene - toluene - cyclohexane - ethanol - water (3 : 3 : 5 : 8 : 5 by volume) thoroughly in a separating funnel, and set aside overnight in a room at constant temperature (20" C). Separate, and add 1 volume of acetic acid to 100 volumes of upper layer. Use the lower layer in the bottom of the tank, and the acidified upper layer in the trough. Load 6.25-, 25- and 125-p1 portions of the chloroform solution of the toxin prepared and dried as described above on to a baseline 8 cm from the edge of a 23-cm x 50-cm sheet of W'hatman No. 1 chromatograph!. paper. Develop the chromatogram by descending-solvent flow for 3 hours at 20" C, spraying a jet of upper layer solvent just below the origin line on the paper immediately before develop- ment, as recommended bjr Rhodes et al.The broad band of saturation produced in the path of development in this manner improves the resolution of the components. At the end of the development period, remo't-e the paper from the tank, mark the position of the solvent front, and dry the paper in air. ( c ) '4 ssessment of toxicity levels-Examine the chromatogram obtained by the above procedure in ultraviolet light (A 365 mp), and observe the presence or absence of purple-blue fluorescent spots at K, 0-57 to 0.61 (aflatoxin B). The toxicitv level of the sample under examination ma!- be classified on the basis of the presence or absence of fluorescence at R, 0.57 to 0.61 after chromatography of the three aliquots referred to above.The classi- fication is shown in Table I. TABLE I CLASSIFICATION OF AFLATOXIS R LEVEL Fluorescence Fluorescence Aflatoxin B level Aliquot from absent, p.p.m. observed, p.p.m. if fluorescence 5 ml, pl of toxin of toxin is observed 6-25 < 2.0 > 2.0 Very high 25 < 0.5 0.5 to 2.0 High 125 < 0.1" 0.1 t o 0-5 Medium * Samples containing less than 0.1 p,p.m. of toxin are classified as of low or zero aflatoxin B level.212 COOMES AND SANDERS: DETECTION AND ESTIMATION OF [Analyst, Vol. 88 The chromatograms of certain samples may exhibit a second blue-purple fluorescent spot a t R, 0.28 to 0.31, corresponding to aflatoxin G. No sample so far examined has con- tained this second spot in the absence of aflatoxin B. As aflatoxin G is less toxic than afla- toxin B and is usually present in smaller proportion in naturally toxic groundnuts, the level of toxicity is assessed on the aflatoxin B content.NOTES- 1. It is recommended that representative samples of decorticated groundnuts be ground before examination in a vegetable slicing and shredding machine until the ground material passes a British Standard 10-mesh sieve. Representative portions of groundnut meals and cakes should be ground in a laboratory-type hammer mill so that the ground material passes a 10-mesh sieve. 2. The method of extracting the toxin into chloroform described below is suitable when only a few samples have to be analysed. Place the fat-free aqueous dispersion of the niethanol-soluble extract in a 500-ml separating funnel, and saturate with sodium chloride.Extract successively with 100, 50, 25 and 10ml of chloroform, and combine the chloroform extracts. RESULTS The results of the examination of ten typical samples of groundnuts or groundnut products for aflatoxin B by the proposed procedure are shown in Table 11, together with the biological test results for extracts of the same samples obtained by Mr. R. B. -4. Carnaghan of the Central Veterinary Laboratory, Weybridge. TABLE 11 CORRELATION BETWEEN AFLATOXIN B ASSESSMENT BY THE CHROMATOGRAPHIC PROCEDURE AND BIOLOGICAL ACTIVITY I N THE DUCKLING TEST Sample No. 411 920 57 1 607 498 260 553 268 493 552 Product Groundnut kernels Groundnut kernels Groundnut kernels Groundnut kernels Groundnut kernels Groundnut flour Groundnut flour Groundnut kernels Groundnut kernels Groundnut expeller cake Aflatoxin B level by chromatographic procedure Low or zero Low or zero Low or zero Low or zero Medium High High Very high Very high Very high Duckling test Duckling Hepatic mortality, lesions" 7 Nil Nil Nil Nil 0/3 0/3 3/3i 0/3 + t o + + 0,'3 +++ 0; 3 +++ 313 ++++ 3/3 ++++ 3/'3 ++++ Oj3 * Severity of hepatic lesions is assessed visually on an empirical + to + + + + basis.t The ducklings in this test were reported to have died as the result of an anaphylactic-type reaction. None of their livers showed lesions characteristic of aflatoxin. In order to test the efficiency of the method, recovery tests were carried out by adding suitable portions of a methanolic solution of the purest available specimen of aflatoxin B to 20-g portions of an Indian groundnut meal already shown to be toxin-free.Recovery of the toxin by the proposed method gave the results shown in Table 111. TABLE I11 RECOVERY OF AFLATOXIN B ADDED TO A NON-TOXIC MEAL Sample Aflatoxin B levd Aflatoxin B level No. based on toxin added (p.p.m.) based on toxin recovered Low or zero Low or zero Blank Zero (0.0)_ 1 Low (0.00) 2 Medium (0- 10) Medium 3 Medium (0.50) Medium 4 High (1.50) High 5 Very high (2.60) Very highMarch, 19631 AFLATOXIN IN GROUNDNUTS AND GROUNDNUT MATERIALS. PART I 213 CONCLUSIONS At the present time, in the absence of information about the constitution of metabolites produced in groundnuts and their products by the growth on them of strains of A . $anus, the method reported provides a semi-quantitative procedure suitable for the detection of toxicity at four levels. The correlation between the analytical results by this method and the biological response in ducklings based on the ten samples tested appears to be satisfactory. We thank Mr. R. B. A. Carnaghan of the Central Veterinary Laboratory, Weybridge, for the biological test results reported here and Mr. B. J. Francis of this Institute for preparing the extracts for biological testing. REFERENCE s 1. 2. 3. Tropical Products Institute Report No. 25, London, 1962. 4. 6. 6. 7. 8. Sargeant, I<., Sheridan, A , , O’Kelly, J., and Carnaghan, R. B. A , , Nature, 1961, 192, 1096. Sargeant, K., O’Kelly, J., Carnaghan, li. 13. A., and Allcroft, R., Vet. Rec., 1961, 73, 1219. Coomes, T. J., and Matthews, W. S . A , , unpublished work. Cornelius, J. A., and Shone, G., personal communication. Nesbitt, B. F., O’Kelly, J., Sargeant, K., and Sheridan, A., Nature, 1962, in the press. Rhodes, A., Boothroyd, B., McGonagle, M. P., and Somerfield, G. A., Biochem. J., 1961, 81, 28. “Standard Methods for the Analysis of Oils and Fats,” International Union of Pure and Applied Received September 13th, 1962 Chemistry Monograph, 4th Edition, 1954.

ISSN:0003-2654    

DOI:10.1039/AN9638800209

出版商:RSC    

年代:1963

数据来源: RSC

摘要:

214 BROADBENT, CORNELIUS AND SHOXE DETECTION AND ESTIMATION [AfldySt, 1701. 88 Glass plate I Spreading The Detection and Estimation of Aflatoxin in --Base plate --.- Glass or metal guide rod 1 Plasticine Groundnuts and Groundnut Materials Part 11. Thin-layer Chromatographic Method BY J. H. BROADBEKT, J. A. CORNELIUS AND G. SHONE (Department of Scienti,fic and Industrial Research, Tropical Products Institute, 56/62 Gray's Inn Road, London, W.C. 1) A semi-quantitative method capable of detecting 0.006 p g of aflatoxin €3 in groundnuts and groundnut products, especially groundnut meals, by thin- layer chromatography, is described. THIS paper describes a method more rapid than that proposed by Coomes and Sanders1 for detecting aflatoxin B in groundnut meals. The de-fatted meal is extracted with methano1,2 any toxin present in the methanol phase is extracted into chloroform, three aliquots of the chloroform extract are chromatographed on thin layers of aluminium oxide, and the chromato- gram is examined in ultraviolet light.The level of aflatoxin B in ground meals can be classified as very high (>.2 p.p.m.), high (0.5 to 2-0 p.p.m.), medium (0.1 to 0.5 p.p.m.) and low or zero (<0-1 p.p.m.) on the basis of this procedure. The sizes of the portions required for this classification have been worked out on samples of the purest toxin available. Aflatoxin G is also separated by the procedure described below and is characterised by a green fluorescence in ultraviolet light; it has a lower X, value than that of aflatoxin B, which exhibits a purple fluorescence.The predominant toxic metabolite in naturally infected groundnut materials is aflatoxin B and only in a few extremely highly toxic samples of groundnuts has aflatoxin G been found to be present to any significant extent; hence the method described here deals only with the estimation of aflatoxin R. I I Aluminium oxide slurry I I I G'ass'plate Base plate I Guide rod Spreading rod Fig. 1. Simple plate-coating apparatus METHOD PREPARATION OF THIN LAYERS OF ALUMINIUM OXIDE (CHROMATOPLATES)- slurry with distilled water, and apply to glass plates to give layers of thickness 740 Mix 90 g of aluminium oxide (M. Woelm; neutral) and 10 g of plaster of Paris into a 10 p.March, 19631 OF AFLATOXIN IN GROUNDNUTS AND GROUNDNUT MATERIALS. PART 11 215 Such layers are most convenientlv prepared with an apparatus such as the Camag thin-layer chromatography apparatus marketed by Carl Roth of Karlsruhe, Germany (obtainable from T.J. Sas & Son Ltd., London). Alternatively, these layers can be prepared by spreading the slurry with a glass rod on glass plates bounded by glass or metal rods of suitable diameter. The rods may be held on a baseplate with Plasticine (see Fig. 1). Heat the coated plates in an oven at 100" C for 2 hours, cool, and store in a vacuum desiccator. Reference compounds should be chromatographed periodically on the prepared chromato- plates to ensure that the activity of the adsorbent is reasonably constant (see Note 1). EXTKACTION AND CHROMATOGRAPHY OF THE TOXIN- Extract 20g (see Note 2) of de-fatted (see Note 3) groundnut meal with analytical- 1-PaPent grade methanol for 4 hours (see Note 4) in a 100-ml Soxhlet extractor (syphon rate K6 changes per hour). Concentrate the extract to 50 ml, and transfer to a separating funnel.Rinse the extraction flask with 25 ml of distilled water, and add the rinsings to the contents of the separating funnel. If an appreciable amount of solid material remains in the flask, add a further 5ml of water, and transfer to the separating funnel. Wash the flask with 25 ml of chloroform, transfer to the separating funnel, and shake gently; allow to separate, and run off the chloroform layer through a bed of anhydrous sodium sulphate. Repeat the extraction three more times with 25-ml portions of chloroform, and combine the extracts.Concentrate the combined extracts to 35 ml, spot 5- and 20-pl portions (see Kote 5) of this concentrate 1.5 cm from one edge of a chromatoplate, and develop with 1.5 per cent. of methanol in chloroform (see Note 6) until a solvent path-length of 10 cm from the base line has been obtained. Examine the resulting chromatogram (see Note 7) in ultraviolet light ( A 365 mp), and observe the presence or absence of a blue-purple fluorescent spot at R, 0.5. (The green fluorescence of aflatoxin G occiirs at R, 0.4.) If no fluorescence is visible in this region, concentrate the residual chloroform extract to 5 ml, and chromatograph 15 pl of this concentrate as before. Examine the chromatogram in ultraviolet light (see Note 8). LEVEL OF AFLATOXIN R- The aflatoxin B level (see Table I) of the meal being examined may be classified on the basis of the presence or absence of a blue-purple fluorescence at R, 0.5 after chromatography of the portions referred to above (see Note 9).It should be noted that a more absolute assessment of aflatoxin B content can be obtained by chromatographing various portions of the chloroform extract and noting the size of the smallest portion giving an observable fluorescence. This portion is then equivalent to about 0.006 pg of aflatoxin B. I t will be realisnd that aflatoxin R levels outside the range given in Table I (0.1 to 2-0 p.p.m.) can be assessed by this technique. XOTES- 1 . I t is recommended that the compounds listed below be run periodically with 1.6 per cent. of methanol in chloroform to ensure that the activity of the adsorbent is reasonably constant- Acridine orange-I?, 0.15.Phenanthraquinone-R*. 0.90. Alternatively, a groundnut extract containing aflatoxin B may be used if available (R, 0.5). 2. Groundnut meals have been found, on occasion, to be non-homogeneous, and care must therefore be taken to ensure representative sampling. 3. Meals that have been produced by expression of the oil from groundnuts will need to be de-fatted by extraction with aromatic-free light petroleum (boiling range 40" to 60" C) or diethyl ether for 2 hours in a 100-ml Soxhlet extractor (syphon rate 10 to 12 changes per hour). If groundnuts are to be tested, they will need to be crushed and de-fatted for a longer period (e.g., 4 hours). Some aflatoxin B (usually less than 1 per cent.of the total amount present) is extracted by light petroleum with the oil. This small amount of aflatoxin €3 can be recovered if the light petroleum extract is ad- justed to about 100 ml and shaken twice with 25 ml of aqueous methanol (25 per cent. v/v). The combined methanol extracts are then added to that obtained from the de-fatted meal.216 BROADBENT, CORNELIUS AND SHONE [Analyst, Vol. 88 4. If a rapid indication of whether or not a groundnut meal contains aflatoxin B is required, 20g of de-fatted meal can be extracted with chloroform for 2 hours," the extract concentrated to 1 ml, and 20 p1 of the concentrate run on a chromatoplate of aluminium oxide. The presence of a blue-purple fluorescent spot at the appropriate position would indicate that the meal contains aflatoxin B at a level greater than 0.1 p.p.m.5. When portions are being applied to the surface of the chromatoplate the solvent should only be allowed to spread to cover an area defined by 0.5 to 0.7 cm diameter. 6. A chromatographic tank of the smallest convenient size should be used. The sides of the tank should be lined with filter-paper, which should dip into the eluting solvent mixture covering the bottom of the tank (to a depth of 0.5 to 1 cm). Chi-omato- plates should be developed in the ascending-solvent manner by standing them on the bottom of the tank (resting against one side) with the lower end immersed in the solvent. 7. If more than one spot is to be examined at a time, it is recommended that each spot be viewed individually (beginning with that of least concentration) while the other spots are covered with black paper.8. Chromatograms should be examined immediately after development, as the fluorescence associated with aflatoxin B fades rapidly. 9. The faintest detectable spot is considered to be a positive in making this assessment. Different amounts of aflatoxin B were added to 20-g samples of a groundnut meal originally shown to be free from toxic material. These artificially fortified samples were de-fatted for 2 hours with aromatic-free light petroleum (boiling rage 40" to 60" C), and the aflatoxin B was recovered by the proposed procedure (see Table 11). RECOVERY TESTS TABLE I AFLATOXIN B LEVEL Aflatoxin B r 7 No fluorescence Fluorescence p.p.m. p.p.m. Size of portion observed, observed, 5 pl from 35 ml . . .. .. < 2.0 > 2.0 20 p1 from 35 ml .. .. < 0-5 0.5 t o 2.0 15 p1 from 5 ml . . .. .. < 0.1 0.1 t o 0.5 TABLE I1 RECOVERY OF AFLATOXIN B ADDED TO GROUNDNUT MEAL Sample Aflatoxin B level Aflatoxin B level Blank Zero (0.0) Low or zero 1 Low (0.05) Low or zero 2 Medium (0.20) Medium 3 Medium (0.40) Medium 4 High (1.25) High 5 Very high (2.50) Very high No. based on toxin added (p.p.m.) based on toxin recovered We thank our colleagues for their co-operation, and especially Miss Ann Player and Mr. T. W. Hammonds for their experimental assistance and Messrs. T. J. Coomes and W. S. A. Matthews for use of their aflatoxin B isolate. REFERENCES 1. 2. Coomes, T. J., and Sanders, J. C., Analyst, 1963, 88, 209. Sargeant, K., O'Kelly, J., Carnaghan, R. B. A., and Allcroft, R., Vet. Rec., 1961, 73, 1219. Received Sefitewaber 13th, 1962 * We are grateful t o Mr. W. V. Lee of Unilever Ltd. for this suggestion.

ISSN:0003-2654    

DOI:10.1039/AN9638800214

出版商:RSC    

年代:1963

数据来源: RSC