ISSN:0003-2654    

DOI:10.1039/AN95681FX013

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

No. 36 March, 1956 THE SOCIETY FOR ANALYTICAL CHEMISTRY BULLETIN ANNUAL GENERAL MEETING, FEBRUARY 29th, 1956 THE eighty-second Annual General Meeting of the Society was held at 4.30 p.m. on Wednesday, February 29th, 1956, in the Meeting Room of the Royal Society, Burlington House, London, W.l. The Chair was occupied by the President, Dr. K. A. Williams, F.R.I.C., A.Inst.P., M.Inst.Pet. The financial statement for 1955 was presented by the Honorary Treasurer and approved, and the Auditors for 1956 were appointed. The Report of the Council for the year ending February, -1956, was presented by the Honorary Secretary and adopted. The President announced that the following had hteu elected Officers for thc coming year- President-I<. A. Williams, BSc., Ph.l>., F.R.I.C., A.Inst.P., M.Inst.Pet.Past Presidents serving on the Council-Lewis Eynon, D. W. Kent-Jones, J . R. Nicholls Vice-Presidents-D. C. Garratt, J. Haslam and H. M. h:. H. Irving. Honorary Treasurer--J. H. Hamence. Honorary Secretary---N. L. Allport. Honorary Assistant Secretary--R. E. Stuckey. Other Members of Council (for the ensuing two years)--- S. G. Burgess, K. C. Chirnside, D. L). C. H. R. Gentry, W. C. Johnson, T. McT,achlan, R. F. Milton, Miss M. Olliver and S. A. Price, having been elected members of the Council in 1955, will, by the Society’s Articles of Association, remain Ordinary Members of the Council for 1956. J. R. Walmsley (Chairman of the North of England Section), F. .I. Elliott (Chairman of the Scottish Section), P. J. C. Haywood (Chairman of the Western Section), J.R. Leech (Chair- man of the Midlands Section), G. F. Hodsman (Chairman of the Microchemistry Group), J. E. Page (Chairman of the Physical Methods Group) and K. L. Smith (Chairman of the Biological Methods Group) will be ex-officio members of the Council for 1956. and George Taylor. Moir, F. J. C. Poulton, A. A. Smales and A. F. Williams. FORTHCOMING MEETINGS Ordinary Meeting of the Society, April 4th, 1956 AN Ordinary Meeting of the Society will be held at 7 p.m. on Wednesday, April 4th, 1!+56, in the Meeting Room of the Chemical Society, Burlington House, London, W.l. The following papers will be presented and discussed:- “The Determination of 4-CNoro-2-methylphenoxyacetic Acid in MCPA by a Differential “Paper Chromatography with Continuous C.hange in Solvent Composition.Part I : Part 11: Separation of Surface-active Agents,” by 1:. Refractometric Method,” by R. Hill, B.Sc., A.K.I.C. Separation of Fatty Acids. Franks, BSc., A.R.I.C.Joint Meeting of the Midlands Section and the Biological Methods Group, Aprii l l t h , 1956 A JOINT Meeting of the Midlands Section and the Biological Mcthods Group will be held at 7 p.m. on Wednesday, April l l t h , 1956, at the University, Edmurid Street, Birmingham, 3. The subject of the meeting will be “Microbiological Assays.” Ordinary Meeting of the Physical Methods Group, April loth, 1956 THE 53rd Ordinary Meeting of the Physical Methods Group will be held at 6.30 p.m. on Tuesday, *April loth, 1956, in the large chemistry lczture theatre of the Imperial College of Science and Technology, South Kensington, London, S.W.7.The subject of the meeting will be “Plant Instrumentation” and the following papers will be presented and discussed :- “Progress in Plant Analytical Control Methods,” by 13. W. Bradford, B.Sc., Ph.U., “The Sonic Gas Analyser,” by A. E. Martin, Ph.D., D.Sc. “Automation in the Laboratory,” by I). A. Patient, H.Sc., A.1nst.P. A. 1i.C. S., I). I. C., F. Inst . Pet. * Please note that the date and place of this meeting has been altered to the above from April 17th, 1956, at the Chemical Society, Burlington House, J,ondon, W.I. BRITISH STANDARDS INSTITUTION DRAFT SPECIFICATIONS A FEW copies of thv following draft specifications, issued for comment only, are available to members of the Society, and can be obtained from the Secretary, The Society for Analytical Chemistry, 7-8 Idol Lane, London, E.C.3. Draft Specification prepared by ‘Technical Committee ISE/18 -Sampling and Analysis CW(ISE)1354--Draft B.S. Methods for the Analysis of Iron, Steel and Ferro-alloys. Draft Specification prepared by Technical Committee FCC/4-Solvents and Allied Pro- of Iron and Steel. ducts. CW(FCC)117--Draft 1I.S. for ?z-l3ntanol (Kevisioii of B.S. 508). LIBRARY OF THE CHEMICAL SOCIETY Easter and Whitsun Closing, 1956 THE Librarian of The C.hemical Society has announced that the library will close for Easter at 1 p.m. on Thursday, March 29th and will re-open at 10 a.m. on Wednesday, April 4th. At Whitsun the library will be closed all day on Monday and Tuesday, May 21st and 22nd. PRlNTED B Y W. HEFFER SONS LTD. CAMBRIDGE. ENGLAND

ISSN:0003-2654    

DOI:10.1039/AN956810X015

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681BX017

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

No. 37 April, 1956 THE SOCIETY FOR ANALYTICAL CHEMISTRY BULLETIN FORTHCOMING MEETINGS Ordinary Meeting of the Society, May 2nd, 1956 AN Ordinary Meeting of the Society will be held a t 7 p.m. on Wednesday, May 2nd, 1958, in the Meeting Room of the Chemical Society, Burlington House, London, W. 1 . The following paper will be presented and discussed :-- “The Composition of Some Deposits and Muds in Estuaries, Rivers and Lakes,” by J. H. Hamence, MSc., Ph.D., F.R.I.C. Joint Meeting of the Society with the Food Group of the Society of Chemical Industry, May 231-4 1956 A JOINT Meeting of the Society with the Food Group of the Society of Chemical Industrv will be held a t 6.30 p.m. on Wednesday, May 23rd, 1956, in the Meeting Room of the Chemical Society, Burlington House, London, W.1 . Joint Meeting of the North of England Section, the Microchemistry Group and the A JOINT Meeting of the North of England Section, the Microchemistry Group and the Bradford Chemical Society will be held on Friday, May 25th, 1056. In the afternoon there will be a visit, by kind permission of the Coal Tar Research Association, to their laboratories at Gomersal, Leeds. Bradford Chemical Society, May 25th, 1956 The subject of the evening meeting in Bradford will be “~Iicro\.oliimetric -2n;dysis.” Ordinary Meeting of the Scottish Section, May llth, 1956 AN Ordinary Meeting of the Scottish Section will be held at 7 p.m. on Friday, May I lth, 19X, in the George Hotel, George Street, Edinburgh. ,4 lecture will be given on: “Cornplexones,” by I<. E. Stiickc!-, H.Sc., I ’ l ~ .l ) . , b.P.S., F.R.1.C. Joint Meeting of the Physical Methods Group with the Photo-electric Spectrometry Group, May 25th, 1956 A JOINT Meeting of the Physical Methods Group and the Photo-tktric Spectrcmwtry Group \\ill be held on Friday, May 25th, 1956, in Oxford. The subject of the Meeting will bc “Nuclear and Paramagnetic Reionance ”BRITISH STANDARDS INSTITUTION DRAFT SPECIFICATIONS A FEW copies of the following draft specifications, issued for comment only, are available to members of the Society, and can be obtained from the Secretary, The Society for Analytical Chemistry, 7-8 Idol Lane, London, E.C.3. Draft Specifications prepared by Technical Committee FCC/4--Solvents and Allied CW(FCC) 118-Draft B.S. for Dieethyl Ether (Technical) (Revision of B.S.579). CW(FCC)llg-Draft B.S. for isoPropyl Alcohol (Revision of B.S. 1595). Products . COMMUNICATIONS ACCEPTED FOR PUBLICATION IN THE ANALYST THE following communications have been accepted for publication in The Analyst, and are expected to appear in the near future. I t is not possible for the Editor to enter into corre- spondence about any of them. “A Simplified Form of Coulometric Titration,” by J. Hetman. “The Determination of Deuterium by the Mass-spectrometric Method, ” by A. MacDonald “The Microbiological Assay of Inositol with a Strain of Sckizosacckaromyces pombe,” by “A Method for the Detection and Determination of Iodide in the Presence of Chloride,” “A Micro-scale Spot-test for Nicotine,” by Policarpo Luis. “Chromatographic Separation, Detection and Determination of Selenium,” by E. G. “The Determination of Non-fat Dry Milk Solids in Bread by Paper Chromatography,” and R. I. Reed. F. W. Norris and A. Darbre. by M. 2. Barakat, M. F. Abd El-Wahab and M. M. El-Sadr. W eat herle y . by H. Zentner. “Paper Chromatography with Continuous Change in Solvent Composition. Part I : Separation of Fatty Acids; Part 11: Separation of Surface-active Agents,” by F. Franks. “The Determination of Uranium by High-precision Spectrophometry,” by A. Bacon and G. W. C. Milner. “A Simple Colorimetric Method for Determining Blood Glucose,” by N. Wahba, S. Hanna “Factors Influencing the Polarimetric Determination of Lactose in Concentrated Whey “The Volumetric Deternination of Plutonium with Ethylenediaminetetra-acetic Acid,” ‘Ton-ferrous Metallurgical Analysis,” by G. W. C. Milner. and M. M. El-Sadr. Products,” by M. A. House. by G. W. C. Milner and J. L. Woodhead. (Review.) PRINTED BY W. HEPFER 13 SONS LTD.. CAMBRIDGE

ISSN:0003-2654    

DOI:10.1039/AN956810X021

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681FP023

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681BP031

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

MARCH, 1956 Vol. 81, No. 960 THE ANALYST PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY NEW MEMBERS ORDINARY MEMBERS Marcus Falk Alexander, BSc. (Bristol) ; James Blenkin, B.Sc. (Wales) ; George Brewer; Malcolm Terence Brooke, A.R.I.C., A.C.T. (Birm.) ; Sidney Landsman, A.R.I.C. ; Mary Fauriel Lockett, M.B., B.S., M.D. (Lond.), M.R.C.P., Ph.D. (Cantab.); Lance Patrinos, Dip.Ind.Chem. (Brisbane) ; Ian Robert Reid, B.Sc. (Glasgow), A.R.I.C. ; Dennis Oswald Singleton, B.Sc. (Lond.) ; Joy Stephens, B.Sc. (Bristol) ; George Francis Sutcliffe, B.Sc. (T.C.D.), M.1 .C.I. ; Leonard George Townsend, MSc. (Lond.) , F.R. I.C. ; John Williams, B.Sc., Ph.D. (Liv.), F.R.I.C. JUNIOR MEMBERS Barbara Gartside, B.Pharm. (Lond.) ; Alan Lever; David Robb Milne; Leslie Herbert Ruddle, B.Sc.(Lond.) , A. R. I .C. We record with regret the deaths of DEATHS Horatio Ballantyne Vincent Binns. SUMMARIES OF PAPERS PRESENTED AT MEETINGS OF THE SOCIETY THE following are summaries of the papers presented at the Ordinary Meeting of the Society organised by the Microchemistry Group on Friday, January 27th, 1956, in London. A first report appeared in The Analyst, 1956, 81, 74. The papers were: “Microchemical Methods in the Art Gallery and Museum,” by A. E. A. Werner, M.A., M.Sc., D.Phil., A.R.I.C. ; “The Ring-oven Technique and its Application in Archaeology,” by H. Weisz, Dr. techn. Dip1.-Ing. MICROCHEMICAL TECHNIQUES IN THE ART GALLERY AND MUSEUM DR. WERNER said that in the examination of paintings, chemical analysis was only qualitative in scope, being concerned with the problem of obtaining information about the pigments present and the nature of the binding medium.In the case of mineral pigments, the number of metallic ions to be identified was not unduly large. The use of organic reagents for metals in conjunction with Feigl’s spot-test technique had proved that these reagents were simpler and more rapid than the methods previously used, which depended on the formation of precipitates whose crystalline appearance under the microscope was characteristic for a given metallic ion. Further, one could test for a number of metallic ions successively; and capillary analysis could be used to separate interfering ions. A technique for the preparation of cross-sections has been evolved by J. Plesters (Mzts. J . , Lond., 1954,54, 97), who used Marco Resin S.B.26 C as the embedding material.This permitted the study of the individual paint layers, and it was possible to carry out microchemical spot-tests on the cross-section so as to detect pigments present in each layer. This method was used in the analysis of the dark brown coating on the 129130 PROCEEDINGS [Vol. 81 Piltdown canine (A. E. A. Werner and R. J. Plesters, BUZZ. Brit. Mus., Nut. Hid., 1955, 2, [6], 271). In the analysis of organic pigments, spectrophotometry has proved useful. It was used by M. Farnsworth ( J . Chem. Educ., 1951, 28, 72) to confirm rose madder among pigments found at Corinth dating from the second century B.C. A comparison of the absorption curves of the Greek madder with a modern sample showed that the two samples were practically identical.The same method was also used by K. A. Freeman and C. Graichen (Far East. Ceram. BUZZ., 1954,6, [ 2 ] , 5) to detect the presence of three synthetic coal-tar dyes in samples of red paint from two Chinese ceramic objects, so proving their modern origin. The microchemical tests for the analysis of paint media had not been so refined as those available for pigments. However, the technique of partition chromatography promised to bring a greater degree of analytical precision into this field. For nitrogenous media, the methods of partition chromatography that had been used for protein analysis were directly applicable. These had been used recently to prove that a painting attributed to Franz Hals was a fake, since it was painted in glue-a medium never used by that artist.In the National Gallery laboratory, a system of reverse-phase partition chromato- graphy for the identification of natural resins had been devised (J. S. Mills and A. E. A. Werner, Nature, 1952, 169,1054; J . Oil CoZ. Chem. Ass., 1954,37, 131). Each natural resin gave a distinctive chromatogram characterised by number, developed colour and RF values of the spots produced by the separated components. The nature of the analytical problems that arose in a museum laboratory was best illustrated by reference to a few typical examples. In the quantitative analysis of ancient bronzes and coins, colorimetric methods were used where possible, but copper was determined gravimetrically by micro electro-deposition. In a study of the phenomenon of “weeping” glass, i.e., the loss of transparency and deposition of tears of moisture on glass objects exposed to damp conditions, the ratio of sodium to potassium in the tears was determined with a flame photometer.The value of a spot-test for phosphate to classify English bone porcelain was described (H. J. Plenderleith, Burlington Magazine, 1927, 51, 142), and also the use of the azide-iodine reagent for the identi- fication of the black material called “niello” used to inlay ornamental designs on ancient metal objects. THE RING-OVEN TECHNIQUE AND ITS APPLICATION I N ARCHAEOLOGY DR. WEISZ said that an investigation had been carried out whose objective was to determine the nature of the materials used to prepare early Egyptian objects d’art.It involved the qualitative analysis of a number of metal statues. All these statuettes, which belong to the Kunsthistorisches Museum in Vienna, were votive offerings dating from the sixth century B.C.; only one of them-a fragment-was younger, and dated from the Greco - Roman period. The archaeological value of the objects permitted the removal of only very small amounts of sample for analysis. A microscope slide with a roughened section in its centre was rubbed across the specimen several times so that a streak containing several micrograms of the metal was produced on the rough glass surface (R. Strebinger and H. Holzer, Mikrochemie, 1930, 8, 264). The metal was dissolved on the slide by means of a drop of aqua regia. The drop was evaporated to dryness and the residue was dissolved in 2N hydrochloric acid.Obviously, only microchemical methods could be used for the analysis of this one drop of solution containing only a few micrograms of solid test substance. The following ions alone were of interest : lead, bismuth, copper, cadmium, tin, iron, zinc, nickel and cobalt. The procedure used was the ring-oven method (H. Weisz, Mikrochim. Acta, 1954, 140 and 376; Chem. Age, 1954, 71, 1039; and C. A. Bank and W. van der Eijk, Chem. WeekbZ., 1955,51, 351), a simple technique for separating ions or groups of ions in a single drop. An apparatus called the ring oven has been designed to elute soluble materials from a spot on filter-paper and to concentrate them in a sharply bounded circular ring zone, where they can be detected.A 1.5-pl drop of solution derived from the sample was placed by means of a self- filling capillary pipette in the centre of a 5-5-cm quantitative filter-paper. All theMarch, 19561 PItOCEEDI NGS 131 lead, bismuth, copper, cadmium and tin was fixed in position by treatment with hydrogen sulphide in a suitable glass apparatus. The filter-paper was next transferred to the ring oven; there all soluble elements, the iron, zinc, nickel and cobalt were washed with 0-1N hydrochloric acid into the ring zone and concentrated there. After drying, the inner spot, which contained all the lead, bismuth, copper, cadmium and tin as sulphides, was punched out. The remaining piece of filter-paper contained in a sharply outlined ring zone all the iron, zinc, nickel and cobalt; this zone was called Ring I.The small punched-out disc of filter-paper containing the sulphides was then held in bromine vapour to oxidise all the sulphides to sulphates, made ammoniacal by being held over ammonia solution, dried and placed in the centre of a fresh filter-paper. This, together with the disc, was placed on the ring-oven and eluted with ammonia solution as if the filter-paper disc were nothing but a normal spot. All the copper and cadmium was dissolved, absorbed into the underlying filter-paper and once more concentrated there in the outer ring zone, which was called Ring 11. The little filter-paper disc then contained only the lead, as sulphate, and the tin and bismuth as hydroxides or basic salts; it was again placed on a new filter-paper and, by means of the ring oven, eluted with yellow ammonium sulphide. This treatment transferred all the tin into the outer ring zone of the supporting paper.After drying, this filter-paper contained (in Ring 111) all the tin, again as sulphide, while the little disc contained all the lead and bismuth, also as sulphides. In this way all the ions had been separated into four groups (three rings and one disc), so that they could be detected in the presence of one another. The entire procedure required only one drop of test solution and was complete in about 15 minutes. Each of the three rings was then cut into several sectors. When developed with suitable reagents, the individual ions, when present, yielded circular arcs of the appropriate colours. All tests were conducted on paper by means of spot-reactions, which were often modified for this particular purpose.In this way it was found that all the specimens without exception were bronzes. The lead content of the specimen of later date (the fragment) was distinctly higher than that of any of the others. However, no essential qualitative difference could be estab- lished in the composition of the pieces originating from the various places (H. Weisz, J . Chem. Educ., 1955, 30, 70). This one example should show the use of the ring-oven technique, which was not confined to qualitative analysis only. Semi-quantitative procedures had been developed for several of the commoner metal ions (H. Weisz, Mikrochim. Acta, 1954, 460 and 785), which permitted their determination with an accuracy of 5 per cent., only minute volumes of solution (a few microlitres) being used. Combinations of the ring-oven and other established analytical methods have been developed, but details of these have not yet been published. All these methods, derived from the ring-oven technique, should be useful when only limited and generally extremely small quantities of sample are available, as in investigations in art and, especially, archaeology.

ISSN:0003-2654    

DOI:10.1039/AN9568100129

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

132 THE ESTIMATION OF VITAMIN B,, [Vol. 82 Analytical Methods committee REPORT PREPARED BY THE VITAMIN-B,, PANEL * The Estimation of Vitamin BI2 THE Analytical Methods Committee has received the following report from its Vitamin-B,, Panel. The Report has been approved by the Analytical Methods Committee and its publication has been authorised by the Council. REPORT In March, 1953, the Sub-committee on Vitamin Estimations of the Analytical Methods Committee appointed a Vitamin-B,, Advisory Panel “to survey the methods already proposed for the estimation of vitamin B,, and to report to the Sub-committee on the work required to establish a standard method or methods.” The Panel made its report in September, 1953, and in it stated that an organism and a technique should be selected that would overcome the non-specificity attaching to the techniques generally employed, e.g., those involving Bacterium coli mutant and Lactobacilli%s leichmannii, and recommended for this purpose the use of Ochyomonas malhamensis, as described by Ford.1 The Analytical Methods Committee implemented the recommendations of this report and in November, 1953, appointed a Vitamin-B,, Working Panel under the Chairmanship of Dr.A. J. Amos “to establish, if possible, an agreed method for the estimation of vitamin B,, in foods and feeding stuffs as suggested in the report of the Vitamin-B,, Advisory Panel.” The members of the Panel were: Dr. W. F. J. Cuthbertson, Dr. J. E. Ford, Dr. F. W. Norris, Mr. S. A. Price, Mr. G. E. Shaw, Dr. R. E. Stuckey, Mr. G. Sykes and Dr. F.Wokes, who acted as Honorary Secretary. INTRODUCTION- The ultimate goal of a working panel appointed by the Analytical Methods Committee is the establish~ent of a reliable analytical method. When the test substance has biological activity, it is not the function of the Panel to supplement its analytical recommendations by interpretations of that activity or to institute relevant researches in clinical medicine and animal nutrition. The Panel responsible for the present report has accordingly restricted its deliberations and experiments to establishing for the determination of vitamin B,, an analytical method that, in our present state of knowledge, is least affected by any known vitamin-B,,-like substances co-existing with vitamin B,, ; the correlation between concen- trations of vitamin B,, so determined and haematopoietic activity in man on the one hand or growth-promoting activity in animals on the other is not the concern of this Panel.Most of the methods available for the measurement of vitamin B,, are open t o criticism on the grounds of insensitivity and non-specificity, and hence have a limited application. Several of the established microbiological assay techniques involving bacteria as test organisms are highly susceptible to interference by vitamin-B,,-like compounds and, in some instances, by other non-specific interfering substances. Ochromoutas malhamensis has more animal-like specificity in its response to vitamin B,, than the other accepted test organisms and may be used to measure the vit\amin even when related compounds are present.EXPERIMENTAL WORK- Under its terms of reference the Panel was required to investigate the microbiological method of Ford,, in which 0. malhamensis is used as the test organism. In order to check the accuracy of this method the assay procedure was applied t o a solution of vitamin-B,, the concentration of which was unknown to members of the Panel. The results showed * Although the results of the assays in this report are expressed in terms of cyanocobalamin, the Panel considers that in the light of present knowledge the term “vitamin E,2” should be retained.March, lc356j THE ESTIMATION OF VITAMIX BIZ 133 excellent agreement between all the laboratories and also were in satisfactory agreement with the figure furnished by a spectrophotometric determination of the potency.The method was then applied to a sample of dried-liver powder, but the figures returned for the vitamin B,, content showed too wide a spread. This suggested that the method of extraction was not sufficiently closely defined and a modified extraction procedure was therefore devised. In the second and third series of collaborative tests the analytical procedure as published and the modified method of extraction were applied to two samples of liver powder and a sample of fish solubles. The vitamin-B,, contents reported for each of these test substances showed acceptable agreement , particularly as most members of the Panel were unfamiliar with this assay, and this satisfied the Panel that the modifications introduced into the extraction procedure had been successful.As a final test of the method, collaborative assays were performed upon two substances that presented particular difficulties ; one was a sample of chick mash having a very low vitamin-B,, content and the other was a sample of desiccated cow manure, a substance very rich in vitamin-B,,+ke compounds. All the standard vitamin-B,, solutions used throughout this investigation were checked spectrophotometrically before and after use in the microbiological assay and no significant changes in concentration were detected. Full details of the modified extraction procedure and of Ford's analytical technique, as recommended by the Panel, are given in the A4ppendix. RESULTS OF COLLABORATIVE TESTS- :3f the extraction procedure.Table I shows the individual results of the collaboi-ative assays that followed modification TABLE 1 RESULTS FOR VITAMIN B,, I N COLLABORATIVE ASSAT SAMPLES Laboratory r------------------A 7 Sample Liver powder A* . .{ Mean . . . . Liver powder 13 1 Mean . . . . Fish solubles 1 Mean . . . . f I i Chick mash Mean . . . . r Dried cow manure Mean . . . . A, Pg/g 0.91 0-87 0-95 0.9 1 0.98 1-18 1.06 1.05 2-07 0.20 0.22 0.19 0.20 0.20 0.01 0.01 0.01 0.01 0.01 0.20 0.1 1 0.19 0.21 0.18 .- - C, Pg/g 0.82 0.97 0.94 0.91 1.39 1.09 1.27 1.25 0.29 0.29 0.26 0.28 0.010 0.010 0.017 0.009 0.012 0.13 0.12 0.10 0.13 0.12 - - - - D, Pglg - - - - 1.08 0-89 0.95 0.97 0.18 0- 16 0.13 0.14 0.13 0.020 0.027 0.020 0.018 0.021 0.1 1 0.12 0.13 0.11 0.13 - - - E, Pg/g 0.94 0.84 0.75 0.84 1.43 1.20 1.20 1-16 1.25 0.28 0.29 0.27 0.27 0.28 0.013 0.013 0.012 0.012 0.012 0.23 0.23 0.18 0.20 0.2 1 -- - 0.96 1.36 1.12 1.17 1.05 0.99 1.04 1-17 __ - __ 0.27 0.24 - 0.25 - 0.24 0.25 - - - 0.022 - 0.015 - 0.012 __ 0.014 - 0.013 - 0.015 - 0.2 3 - 0.17 - 0.12 - 0.11 - 0.11 0.20 0.15 * In this first collaborative trial, four of the laboratories encountered practical difficulties in establishing the assay and for this reason no results are reported.134 THE ESTIMATION OF VITAMIN B12 [Vol.81 SUMMARY- The Panel accepted published evidence of the non-specificity of microbiological assays of vitamin B,, with B. coli, L. Zeichmamii or EugEevta as test organisms, and a more specific technique with the protozoan 0. malhamevtsis was selected for study. In view of the inherent difficulties encountered in microbiological assays and the low level of vitamin B,, in some of the samples examined, the results provided by these collaborative trials were considered to be satisfactory.APPENDIX RECOMMENDEL) PKOCEDURE FOR THIS MICROBIOLOGICAL IIETEKMINATION OF VITAMIN B,, PREPARATION OF TEST EXTRACTS- Vitamin B,, may occur naturally in complex-bound forms, which may escape detection because of non-quantitative extraction or because of their unavailability to the test organism- -\ further problem is that the vitamin may be present as hydroxocobalamin, which is relatively unstable to heat and may thus be lost in the processes of measurement. Both problems can be largely resolved by the use of cyanide in the preparation of the test extract.The effect of cyanide is twofold. Firstly, it appears to facilitate extraction by displacing protein or polypeptide groups originally linked to the vitamin. Secondly, it converts the cobalamin to the stable cyano form. In general it is sufficient to add from 0.5 to 5.0 mg of sodium cyanide (as a 1 per cent- aqueous solution) to each gram of test sample, the amount depending upon the nature and potency of the test substance. It is most important, however, that the amount of sodium cyanide used should not be sufficient, even if no cyanide is lost during the extraction and sterilisation processes, to give a concentration of sodium cyanide in the final growth medium greater than 10 pg per ml. Hence, the lower the potency of the material the smaller should be the amount of cyanide used.To 1 g of the test material in a 125-ml conical flask add 30 ml of water and from 9-05 to 0.5 ml of freshly prepared 1 per cent. aqueous sodium cyanide, mix well and adjust the pH to between 4-6 and 5.0 with N hydrochloric acid. Allow the mixture to stand for 30 minutes at room temperature with occasional shaking and re-adjust the pH if necessary. Place the flask in a boiling-water bath and leave it for 30 minutes after the extract has reached 90" C. Cool, transfer the liquid to a 100-ml calibrated flask and dilute to the mark with distilled water. Clarify by centrifuging; then take a 50-ml aliquot of the cleared liquid and dilute it until the concentration of vitamin B,, is of the order of 0-0002pg per ml. Starchy samples may yield turbid extracts, but they can be cleared by treatment with t akadiast ase. Some materials, notably preparations of gut mucosa, have the property of "binding" vitamin BI2 and rendering it unavailable to the test micro-organism.The foregoing simple extraction method may not liberate the vitamin from such materials; for them a more elaborate procedure involving digestion with cyanide-activated papain should be used, as described by Gregory.2 Very few other natural materials call for this enzymic extraction procedure, however, and in the main the simple method recommended will prove adequate. The extraction procedure recommended is- ~IAIKTENANCE OF STOCK CULTURES- To 400 ml of an aqueous solution containing 0.00025 pg of vitamin B,, per ml add 100 ml of basal medium. Dispense 10-ml portions of this diluted and enriched basal medium into 50-ml conical flasks, plug the flasks and sterilise by autoclaving for 15 minutes at 10 lb pressure, covering the plugs with greaseproof paper during the autoclaving. Inoculate one of the flasks with a culture of 0.malhamevtsis (Pringsheim strain)" and incubate at 27" to 30" C, one foot below a 60-watt tubular tungsten-ffilament lamp. Having established the culture, prepare sub-cultures at intervals, usually every 4 to 5 days, by transferring 0.5 ml of the current growth to a flask of sterile medium. When the organism is sub-cultured, the parent culture should be dense and yellow-brown in colour. * Obtainable from the Curator, Culture Collection of Algae and Protozoa, Downing Street, Cambridge.March, 19561 THE ESTIMATION OF VITAMIN BIZ 135 PREPARATION OF BASAL MEDIU&I- To 150 ml of distilled water add the following ingredients- “Vitamin-free’’ casein hydrolysate .. . . 5.0 g Glucose . . . . .. . . . . . . 10.0 g Diammonium hydrogen citrate .. . . 0.8 g Magnesium sulphate . . . . . . . . 0.2 g Calcium chloride (anhydrous) . . .. . . 0.15 g DL-Tryptophan . . .. . . . . . . 0.1 g DL-Methionine . . . . . . . . . . 0.2 g Potassium dihydrogen phosphate . . . . 0.3 g Sodium molybdate (Na,Mo04.2H,0) . . . . 0.05 g Tween 80, T.B. culture grade7 . . . . . . 1.0 ml “Metals” solution . . . . . . . . 10.0 ml “ Vitamins ” solution . . . . . . . . 2.0 ml Sodium cyanide solution . . .. . . 0-2 ml Obtainable from Messrs. Honeywill and Stein Ltd ., Devonshire House, Rlayfair l’lace, Piccadilly, Stir 0.1 g of L-cystine into 15 ml of hot water, and add 6 N hydrochloric acid dropwise Add the liquid to the foregoing solution, adjust the pH This basal medium, which is five times single strength, should be stored frozen in screw- London, W.l.until the amino acid is dissolved. to 5.5 with 4 N aqueous potassium hydroxide, and dilute to 200ml. capped polythene containers. The solutions listed as ingredients of the basal medium are prepared as follows- hot water. “Metals” solution-Dissolve 5 g of ethylenediaminetetra-acetic acid in 900 ml of Then in the solution dissolve the following- Manganese sulphate, MnS04.4H,0 . . . . 6.0 g Zinc sulphate, ZnS04.7H,0 . . . . . . ll*Og Ferrous sulphate, FeS0,.7H20 . . . . . . 1.0 g Cobalt sulphate, CoS04.7H20 .. . . . . 0.3 g Copper sulphate, CuS0,.5H20 . . . . . . 0.04 g Boric acid, H3BO3 .. . . . . . . 0.06 g Potassium iodide, KI . . . . .. . . 0.001 g Make up the solution to 1 litre with water. “Vitamins” solution-Weigh out 1 g of inositol, 200 mg of choline chloride, 100 mg of 9-aminobenzoic acid, 200 mg of thiamine hydrochloride and 1 mg of biotin, and transfer to a 200-ml calibrated flask. Dilute to the mark with 20 per cent. v/v aqueous ethanol. Store in a refrigerator. Sodium cyanide solzttion-Dissolve 1 g of sodium cyanide in water and dilute to 100 ml. The solution should be renewed at fortnightly intervals. SETTING UP THE ASSAY- Dispense each of the quantities 0-5, 1, 2 and 4 ml of a standard vitamin-B,, solution (0-0002 pg of cyanocobalamin per ml*) into each of four bacteriological test tubes (19 mm x 150 mm), and similarly put up in quadruplicate the same volumes of the test extract.To each of these 32 tubes and to each of 5 empty tubes add sufficient distilled water to give a total volume of 4 ml in each tube. To each of the 37 tubes now add I ml of the basal medium. Cap or plug the tubes, place them in suitable racks or baskets and autoclave for 10 minutes at 10 lb pressure. Cool the tubes to about 30” C by standing them in cold water and then inoculate all the tubes except one of the five containing only water with one drop of a five-day culture of 0. malhamensis. This step should be performed as speedily and as uniformly as possible. Place the racks or baskets of tubes in a shaking machine installed in a dark incubator maintained at 29” to 30” C and keep the shaker in motion for 72 hours.MEASUREMENT OF RESPONSE- At the end of the incubation period remove the tubes and steam them for 5 to 10 minutes. Cool, mix the contents thoroughly and measure instrumentally with a suitable filter the * Prepared by diluting a 5 pg per ml stock solution of cyanocobalamin, which must have been The stock solution will The dilutions should be prepared from 1 ml of standardised spectrophotometrically, with 20 per cent. v/v aqueous ethanol. keep indefinitely if stored in the dark in a refrigerator. the stock solution to which 1 drop of sodium cyanide solution has been added.136 ELWELL AND WILSON : THE COLORIMETRIC DETERMINATION [Vol. 81 turbidity of the contents of each tube using, where necessary, the uninoculated tube as a “blank.” A suitable procedure is to make the measurements in 4-cm cells in a Spekker absorptiometer, with Hilger H508 filters. CALCULATION OF RESULTS- A “standard” curve can be plotted from the dose - response relationship provided by the tubes containing 0, 0.5, 1, 2 and 4 ml of standard vitamin-B,, solution and from this curve can be read the quantities of vitamin B,, corresponding to the turbidities found in the tubes containing test extracts. REFERENCES 1. 2. Ford, J. E., Brit. J. Nutr., 1953, 7, 299. Gregory, M. E., Brit. J . Nutr., 1955, 8, 340.

ISSN:0003-2654    

DOI:10.1039/AN9568100132

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

136 ELWELL AND WILSON : THE COLORIMETRIC DETERMINATION [Vol. 81 The Colorimetric Determination of Phosphorus in Steel and Copper-base Alloys BY W. T. ELWELL* AND H. N. WILSON (Presented at the meeting of the Society on Wednesday, October 5fh, 1955) A colorimetric procedure is recommended for the determination of phosphorus in all classes of steel; the procedure has also been successfully applied to two standard bronze samples. The new procedure permits a considerable saving in time, particularly in the examination of highly alloyed steels, when the determination can be completed in under 14 hours, compared with about 8 hours required by the existing gravimetric method. The determination is based on the formation of phosphovanadomolybdic acid, which is soluble in isoamyl alcohol.The yellow colour, which is proportional to the amount of phosphorus present, does not fade and can be measured in any convenient way. THE phosphorus content of steel is usually determined gravimetrically after solution of the sample in acid, removal of arsenic by volatilisation and precipitation of ammonium phospho- molybdate. The precipitate is filtered off and the weight of phosphorus that it contains can be determined in several ways, e.g., by dissolving it in aqueous ammonia, precipitating the molybdenum as lead molybdate and calculating the phosphorus from the molybdenum content .I This procedure is reproducible, but very special attention to detail is necessary ; particularly in the examination of high chromium-bearing steels containing the metals tungsten, niobium, titanium, vanadium or tin.In the examination of such steels additional steps must be introduced in order to achieve complete recovery of phosphorus.2 This gravimetric method is a tedious determination, which may take up to 8 hours to complete and may spread over 2 days. These facts led us to investigate and develop a colorimetric procedure, for application to alloy steels, that would demand less personal skill and produce accurate results in a much shorter time. Various methods based on the reduction of phosphomolybdic acid to molybdenum blue have been put forward for the colorimetric determination of phosphorus in iron and steel,3 but the reaction is too much influenced by factors such as acidity to make it suitable for routine work; moreover, it is too sensitive.In 1908 Misson4 put forward a colorimetric method for the determination of phosphorus based on the reaction of molybdate with a solution containing vanadate and phosphate, but for many years this reaction was neglected. Recently, it has been put forward for the determination of phosphate in fertilisers5; several papers have also been written on its applications in the analysis of iron ore and plain carbon ~ t e e l s . ~ , ~ , ~ * Present address : Research Dept., I.C.I. Ltd., Metals Division, Kynoch Works, Witton, Birmingham.March, 19561 OF PHOSPHORUS I N STEEL AND COPPER-BASE ALLOYS 137 It is formed very readily and has an intense orange-yellow colour, which obeys Beer’s law, as it exists in true solution (in contrast to “molybdenum blue,” generally considered to be a colloid and of uncertain composition9).Because of its definite composition, one can confidently expect it to be useful in accurate colorimetric analysis. In the analysis of steels, ores and such like, there is a complicating factor that is not present in fertiliser analysis-the solutions are often coloured by other ions. Kitson and Mellon,8 for example, emphasise that the method is applicable to simple steels if special attention is given to “background” colour. When, however, other elements, such as chromium, nickel and cobalt, are present, the “background” colour is much greater and becomes too variable for the method to be directly applied to an aqueous solution of the samvle. The composition of the compound concerned is H,PO,.VO,.llMoO,.0.15- 1. 0.10 of E - w curve of 0.10 mg of phosphorus (as phosphovanadomolybdic acid-the equivalent of 0.02 per cent. of phosphorus) in acid of the same concentration. Hence, it is clear that the phosphorus colour cannot be accurately measured in the presence of iron, chromium and nickel. There is, however, a very simple and rapid means of separating the phosphorus complex, since it is readily soluble in isoamyl alcohol, leaving the other constituents of the steel in the aqueous solution. A small amount of, presumably, vanadic acid is also extracted, but under suitable conditions interference from this cause can be made constant, and hence allowed for in the blank. By use of this method of extracting the coloured complex from all interfering compounds, we have been able to devise a method suitable for a wide variety of phosphorus contents and applicable to a very wide range of metallurgical materials.EXPERIMENTAL INFLUENCE OF IRON- After a few preliminary experiments of a semi-quantitative nature to decide the best conditions of acidity for use of the reagent described by Hanson for fertiliser analy~is,~ it was shown that complete extraction of the coloured complex could be achieved by a single extraction of the acid aqueous solution with isoamyl alcohol. The optical density of the isoamyl alcohol extract was proportional to the amount of phosphorus present (at 4260 A, Spekker photo-electric absorptiometer and Ilford No. 601 filters). But when spectro- graphically pure iron was added to the solution, the optical density of the alcohol layer became less as the concentration of the iron increased (Fig.2). I t was realised that this was probably due to the formation of ferrimolybdate complexes in solution, which resulted in insufficient molybdate being present to complete the reaction. Fig. 3 illustrates the effect of increasing the amount of molybdate, the amounts of iron and phosphorus and the acidity being constant. It is thus necessary to have at least 650 mg of[Vol. 81 138 ELU'ELL AND WILSON : THE COLORIMETRIC DETERMINATION ammonium molybdate present per 500 mg of iron if the effect of iron is to be minimised and a sensibly constant optical density achieved per unit quantity of phosphorus in samples of various iron contents. In the recommended procedure we have specified 1 g of ammonium molybdate with a total sample weight of 250 mg, an ample excess even if a 500-mg sample is used.INFLUENCE OF CHROMIUM- When the reaction was applied to steels of the 18/8 class, dissolved in perchloric acid and evaporated to fumes, results were low and variable. The low results were shown to be caused by sexavalent chromium; when the chromium was reduced to the tervalent state 0.30 r / 0.24 0.22 0.26 I I 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Amount of ammonium molybdate, g Fig. 3. Influence of molybdenum on solution containing 0.5 g of iron, 0.025 g of ammonium vanadate and 0.05 mg of phos- phorus No additions 250 mg of iron and 25 mg of nickel 0.6 o 500mg of copper 0cr'0.02 0.b4 0.b6 008 0:IO O;I2 0:14 Amount of phosphorus, mg Fig. 4.Calibration graph for the Spekker absorptiometer with Calorex H503 and Ilford No. 601 filters by sulphur dioxide (excess of which must be removed by boiling or oxidation before adding the ammonium vanadate - molybdate reagent), there was no longer any interference. Table I shows the effect of adding various amounts of chromium (as potassium di- chromate) to a solution containing 500 mg of pure iron and 0.2 mg of phosphorus (added as potassium dihydrogen phosphate)-equivalent to 0.040 per cent. of phosphorus calculated on the weight of iron taken. TABLE I EFFECT OF CHROMIUM Chromium added, percentage on weight of iron Phosphorus found, Remarks % Nil 0.042 - 1 0.037 o.028 }Chromium present as CrVI 5 1 o.041 0-04 1 }Chromium reduced to CrIII 5 Nil 0.04 1 - In the procedure, the sulphurous acid is mostly removed by boiling, but to make sure that no ferrous iron is present, a little more nitric acid is added and the boiling is continued for a few minutes.Besides re-oxidising any ferrous iron, this step adjusts the acidity of the solution t o the optimum for the rapid formation of the coloured complex. INFLUENCE OF ARSENIC- It may be expected that arsenate will react similarly to phosphate and it is not certain that arsenic will be completely reduced by sulphurous acid in a fairly acid solution. It was therefore essential to show to what extent arsenic did interfere. Measured volumes of a standard arsenate solution were examined in the absence of iron and phosphate, both by the recommended procedure and by a similar procedure, the reduction stage with sulphurous acid being omitted, Le., all the arsenic was present as arsenate.March, 1956 OF PHOSPHORUS I N STEEL AND COPPER-BASE ALLOYS 139 TABLE I1 ARSENIC SOLUTIONS ALONE Arsenic Phosphorus equivalent added, Arsenic added, calculated from mg yo (for 0.5-g sample) ElC* Elcm, yo 0.25 0.5 1.0 2.5 0.012 0.0 12 0.022 0.037 0.045 0.052 0-078 0.092 0-05 0.10 0-20 0.50 0.0005 0.0005* 0*0009 0.0015* 0.0018 0.002 1 * 0.003 1 0*0037* .* In the absence of sulphurous acid. The results obtained are summarised in Table 11, which shows that in unreduced solutions arsenic has an effect equivalent to about one-hundredth of its weight of phosphorus; in reduced solutions the effect is even less. As the percentage of arsenic in steel is rarely as great as that of phosphorus, no significant error will arise.Table I11 demonstrates the insignificance of this effect, by the addition of arsenate to plain carbon (standard) steels, before solution of the sample in acid. TABLE I11 !IRSENIC ADDED TO STANDARD STEELS Arsenic (As) added, percentage Sample Type of steel equivalent B.C.S. No. 154 Carbon Nil B.C.S. No. 154 Carbon 0.05 B.C.S. No. 154 Carbon 0.10 B.C.S. No. 235 18/8 stainless Nil B.C.S. No. 236 I8/8 stainless 0.05 B.C.S. No. 235 18/8 stainless 0.10 Certified value for phosphorus, % 0.036 0.036 0.036 0.020 0.020 0.020 Phosphorus found, 0.036 0.036 0.036 0.022 0.021 0.021 Y O INFLUENCE OF TUNGSTEN- In both samples the tungstic acid precipitated was filtered off and the solution was examined for phosphorus. The insoluble matter was dissolved in ammonia; the solution was acidified with nitric acid and examined for phosphorus.The results of these experiments are given in Table IV. It is concluded that loss of phosphorus in the tungstic acid precipitate is insignificant. Two steels with high tungsten content were examined. TABLE IV TUNGSTEN-BEARING STEELS Phosphorus found -.- Certified value for in in Sample composition phosphorus, solution, precipitate, 0’ % % /O B.C.S. No. 167 W, 16%; Cr, 3%; V, 0.8y0; Co, 4%; Mo, 0.026 0,025 0.0009 B.C.S. No. 220 Cr, 5%; V, 1%; Co, 0.7%; W, 7%; Mo, 4% 0.024 0.026 0*0009 INFLUENCE OF COPPER- There seemed no reason to expect that bivalent metals would have any influence on this method of analysis (unless they have a fairly high reducing power and reduce the reagent), but it was thought desirable to check this point (a procedure has already been published for the direct colorimetric determination of phosphorus in copper-base alloys1*).The fact that nickel is without effect is shown by the good results obtained with 18/8 stainless steels (see Tables VI and VII). By dissolving two standard bronze samples (0.5 g) in a suitable mixture of nitric and hydrochloric acids, such that all the tin was retained in solution and the correct amount of140 ELWELL AND WILSON : THE COLORIMETRIC DETERMINATION [Vol. 81 acid was present, results were obtained that are in good agreement with certified values (see Table VII). The absence of chromate made the sulphurous acid addition unnecessary. These experiments confirm that copper (and tin) do not interfere.STABILITY OF THE COLOURED COMPLEX- Gericke and Kurmiesll have stated that the colour of the phosphovanadomolybdic acid obtained from solution of fertilisers is stable for prolonged periods, but it was desirable to check this observation by using metallurgical samples. It was also desirable to know the stability of the coloured complex after extraction with the organic reagent. Stability of the complex irt aqueous solution-Solutions of phosphate, in the presence and absence of iron, chromium and nickel, were set aside for various periods, after addition of the reagent and before extraction of the complex. The optical densities of the isoamvl alcohol solutions are given in Table 1’. TABLE V STABILITY OF COMPLEX IN AQUEOUS SOLUTION El cm, with standing period before extraction Metals present, mg A r ~ r-------A--.--- - Iron Chromium Nickel 10 minutes 30 minutes 60 minutes None None None 0.596 0.600 0.594 500 - - 0-586 0-586 0.572 350 100 60 0-552 0.558 0.552 As the complex is usually extracted within 15 minutes, no serious error will be experienced on this account.Stability of the complex in isoamyl alcohol-Experiments were made on solutions obtained from an 18/8 steel (as in the recommended procedure). The two layers were allowed to separate and, after 5 minutes, the aqueous layer was removed, i.e., the normal procedure. The optical density of the isoamyl alcohol layer was measured at intervals, the results being as follows- Standing period after extraction, minutes . .10 25 40 55 240 E l c m * - . . .. .. .. . . 0.360 0.358 0-352 0.354 0.348 These results show a very slight diminution in optical density, but if the measurement is made within 20 minutes of the extraction, no serious error will be introduced. CHOICE OF WAVELENGTH- Reference to Fig. 1 shows that by choosing a suitable wavelength almost any desired degree of sensitivity can be obtained, but, unless a spectrophotometer giving monochromatic light is available, the choice of wavelength is limited by the availability of filters. An Ilford No. 601 filter is suitable for use with a Spekker photo-electric absorptiometer \mercury-vapour lamp). An additional practical consideration is the small amount of vanadic acid that is extracted by the isoamyl alcohol. Since the amount extracted is reason- ably constant, the optical density of this blank becomes more significant if smaller samples are used, even if the measurement is made at a lower wavelength.EXTINCTION COEFFICIENT- Fig. 4 shows the results obtained when a Spekker absorptiometer was calibrated by using various solutions, typical of those that would be obtained in the course of examining metallurgical samples. The graph clearly indicates that, by using the recommended procedures, the extinction coefficient is uninfluenced by iron, chrorniumI1I, nickel or copper. APPLICATION OF THE METHOD TO METALLURGICAL SAMPLES- At the time of this investigation we had occasion to examine a number of 18/8 steels for phosphorus. These samples were examined by a gravimetric procedure2 and also by the proposed new method.Results obtained are shown in Table VI. Table VII shows the results obtained when the recommended procedure was applied t o a number of miscellaneous metallurgical British Chemical Standards (issued by the Bureau of Analysed Samples Ltd.).March, 19561 Sample marked OB oc OD OE OF OG PL ST sx SY Sample OF PHOSPHORUS I N STEEL AND COPPER-BASE ALLOYS MISCELLANEOUS METALLURGICAL SAMPLES Pure iron granules . . .. Plain carbon steel . . .. Plain carbon steel . . . . Plain carbon steel . . .. Chrome - vanadium steel . . Alloy steel (W, 16% ; Cr, 3%) . . Iron ore “,4” . . .. .. Bronze “A” . . .. .. Ferromolybdenum (72 yo of Mo) Bronze “C” . . .. . . Chrome steel (13% of Cr) . . Plain carbon steel . . . . Plain carbon steel . . .. Stainless steel 18/8/Ti .. .. TABLE \’I COMPARISON OF RESULTS Phosphorus by gravimetric method, % 0.030 { 0.028 0.025 0.024 0.027 0.0 16 0.021 0.024 0.027 0.019 { 0.022 I 1 I { 0.021 Phosphorus by colorimetric method, 0.030 0.031 0.025 0.024 0.027 0.029 0.019 0-018 0.027 0.028 0.021 0.020 % 0.022 ’ 0.023 0.021 0.0 17 { 0.017 0.020 ; 0.021 { 0.019 TABLE VII 0.021 0.820 0.021 0.021 0.020 0.018 0.020 0.019 .. .. .. .. .. .. .. .. .. . . .. . . .. .. High-speed tool steel (W, 6.7% ; Mo, 4.2% ; Cr, 4.6% ; Co, 0.7% ; V, 1.4%) Chrome - vanadium steel . . . . .Alloy steel (Ni, 1 . 7 0 / , ; Cr, 0.3%; Mo, 0.3%) . . . . . . . . . . Cast steel. . . . . . . . .. Stainless steel 18/8/Ti/W . . . . Plain carbon steel . . . . . . White cast iron . . . . . . . . High-duty cast iron .. .. .. Stainless steel 18/12/Nb . , . . lS/Cr/ll/Ni niobium-bearing steel . . 141 B.C.S. NO. 149 154 161 163 165 167 175 183 200 207 209 211 213 218 220 223 224 225 230 235 239 847 261 123At Phosphorus Certified by new value for procedure, phosphorus, Remarks o / O/ /O 0.009 0.035, 0.036 0.053 0.047, 0.047 0.024, 0.026 0.026, 0.025 0-057 0.24 0.028 0.053 0.017, 0.017 0.018 0.029 0.045, 0-046 0.026, 0.024 0.45 0.013, 0.012 0.021, 0.020 0.035 0~021,0*022 0.02 1 1.16 0.022 0.030, 0.029 /O 0.010 0.036 0*058* 0.049 0.024 0.026 0.056 0.25 0.03* 0.055 0*018 0.017 0.029 0.045 0.024 0.45 0.012 0.02 1 0-04* 0.020 0.023 1-19” < 0*02* 0.0351. 0.5-g sample 0-5-g sample in 100 ml; 10-ml aliquot taken By procedure for copper- base alloys 0.25-g sample in 100 ml; 10-ml aliquot taken 0.25-g sample in 2.70 inI; 10-ml aliquot taken * These samples have not been standardised for phosphorus.t N.B.S. number. In some cases only a single tentative figure is reported. ,411 other values for phosphorus are approved certified values. Single phosphorus figure reported-no details.142 ELWELL AND WILSON THE COLORIhlETRIC DETERMINATION [lTol. 81 METHOD Recommended filters for use with the Spekker photo-electric absorptiometer (mercury-vapour lamp) are Calorex H503 and Ilford No. 601. REAGENTS- implied throughout. Optical densities are measured at 20°C at a wavelength of 4 2 6 0 ~ . ,411 reagents must be the highest purity obtainable. Redistilled isoamyl alcohol-The reagent supplied for milk testing is recommended. SuZphurous acid-Prepared by saturating water with sulphur dioxide.Ammonium vanadate - molybdate reagent-Dissolve 1 g of ammonium vanadate in about 300 ml of water and then add slowly 140 ml of concentrated nitric acid, sp.gr. 1-42. To this solution add a solution of 40 g of ammonium molybdate, (NH,),Mo,O,~~H,O, dissolved in about 400ml of water. Dilute to 1 litre with water and mix well. This reagent must be freshly prepared. Standard phosphate solzttion-Dry sufficient potassium dihydrogen phosphate at 105" C until the weight is constant (cool in a desiccator). Dissolve 4-393 g of the dry solid in water and dilute with water to 1 litre and mix well. Dilute 10.0 ml of this solution to 1 litre, then- Both solutions must be freshly prepared. PROCEDURE FOR STEELS- Transfer 0.25 g of the millings to a small (150-ml) beaker and dissolve in a mixture of 5 ml of concentrated hydrochloric acid, sp.gr.1.18, and 5 ml of concentrated nitric acid, sp.gr. 1.42 (Note 1). Add 5ml of perchloric acid, sp.gr. 1.54, and evaporate to fumes of this acid. Cover with a watch-glass and continue to heat to fumes for 10 to 15 minutes, at such a temperature as to maintain a steady reflux of acid on the side of the beaker. Cool, add 25ml of water (Note 2), 10ml of sulphurous acid and a few small glass or fish-spine beads and boil for 10 minutes (Note 3). To the boiling solution add 5 ml of 20 per cent. v/v nitric acid and continue to boil for 2 to 3 minutes. Remove the beaker from the source of heat, and add to the hot solution 25 ml of the ammonium vanadate - molydate reagent and then cool to 20" C.Transfer to a 100-ml graduated separating funnel, dilute with water to 80 ml and mix well. Add 10.0 ml of isoamyl alcohol, shake vigorously for 2 to 3 minutes, and then allow- the two layers to separate for 5 minutes. Run off the lower aqueous layer and then dry the inside of the stem with rolled filter-paper. Transfer the coloured organic layer, via a dry filter-paper, to a 1-cm cell, and measure the optical density. Use the organic extract from the blank in the compensating cell. The use of distilled water is 1 ml = 0.01 mg of phosphorus. Perform a blank on the reagents; proceed exactly as described in the procedure. E,, x 0.20 PROCEDURE FOR COPPER-BASE ALLOYS- Transfer 0.25 g of the millings to a small (150-ml) beaker and dissolve in a mixture of 2.5 ml of concentrated hydrochloric acid, sp.gr.1.18, 2-5 ml of concentrated nitric acid, sp.gr. 1.42, and 5 ml of water. Dilute with 25 ml of water, warm to about 80" C, add 25 ml of the ammonium vanadate - molybdate reagent, cool to 20" C and proceed as outlined for steels. NOTES- 1. This weight is suitable for samples containing normal amounts of phosphorus (0.025 to 0.048 per cent.). Samples containing phosphorus outside this range can usually be examined by simple adjustment of the sample weight or size of absorption cell. The method of solution can also be modified to meet individual requirements, but the presence of nitric acid during dissolution of the steel must be ensured. 2. In the examination of tungsten-bearing steels, add only 10 ml of warm water, mix well and filter.IVash about five times with a total volume of about 30 ml of warm water. Add 10 ml of sulphurous acid to the filtrate and proceed as outlined. Omit the addition of sulphurous acid in the examination of plain carbon steels, cast iron and so on, and samples nominally free from chromium, Le., add 5 ml of 20 per cent. v/v nitric acid and continue to boil for a further 2 to 3 minutes. = mg of phosphorus (Note 4), i.e., Elcm x 0.080 = percentage of phosphorus (for a 0-25-g sample). Discard the residue. 3.March, 19561 OF PHOSPHORUS IN STEEL AND COPPER-BASE ALLOYS 143 This factor was obtained on solutions of high-purity iron (as in the procedure) to which had been added various amounts of the standard phosphate solution. It must be determined for each individual instrument and filter.,4 suitable range of standards is obtained by using accurately measured volumes of the standard phosphate solution (1 ml = 0.01 mg of phosphorus) as follows: 0, 2.5, 5.0, 7-5, 10.0, 12-5 and 15.0 ml. When samples of the same class are under examination, it is an advantage to examine a reference sample, of similar composition, a t the same time as the instrument is being calibrated. 4. (See also Fig. 4.) The major part of the experimental work recorded in this paper was carried out by Mr. S. R. Hill. 1. 3. 4. 5. 6. 7. 8. 9. 10. 11. > I . REFERENCES British Standard 1121 : Part 1 : 1943. British Standard 1121 : Part 9 : 1948. Zinzade, C., I n d . Eng. Chern., Anal. Ed., 1935, 7, 227; 7, 320. Misson, G., Chem.Ztg., 1908, 32, 633. Hanson, W. C., J . Sci. Food Agric., 1950, 1, 172. Murray, TI’. M., and Ashley, S. E. Q., I n d . Eng. Chem., Anal. Ed., 1938, 10, 1. Bogatzki, G., Arch. Eisenhiittenw., 1938-39, 12, 195. Kitson, R. F., and Mellon, M. G., Ind. Eng. Chem., Anal. Ed., 1944, 16, 379. Sidgwick, N. V., “The Chemical Elements and their Compounds,” Oxford University Press, ,1950, p. 1046. A.S.T.M. “Methods for Chemical Analysis of Metals,” 1950, p. 294. Gericke, S., and Kurmies, B., 2. am,?. Chem., 1952, 15, 137. RESEARCH DEPARTMENT IMPERIAL CHEMICAL INDUSTRIES LIMITED BILLINGHAM DIVISION BILLINGHAM, Co. DURHAM August 5th, 1965 DISCUSSION DR. W, STROSS said he appreciated that with highly alloyed steels it was necessary to extract the yellow complex in order to avoid interference by certain alloying elements, but he wished to know for which type of copper-base alloys the extraction was necessary for the same reason.With copper-base alloys that were not highly alloyed, he had obtained satisfactory results by the photometric molybdate - vanadate method without extraction, the technique published by the A.S.T.M. (“Methods for Chemical Analysis of Metals,’’ 1950, pp. 293-296) being essentially followed. He also asked whether there was any interference by silicon. MR. ELWELL, in reply, stated that the new procedure was primarily intended for the examination of steel samples. In the examination of copper-base alloys by the A.S.T.M. procedure, background inter- ference was not serious, but the proposed extraction procedure would permit a larger sample weight to be taken with obvious advantage.Referring to Dr. Stross’s second question, Mr. Elwell said that it seemed reasonable to conclude from the variety of samples examined that silicon was without effect. MR. C. H. R. GENTRY asked if the possible interferences with this method had been studied with the aid of solutions of known compositions, and also if there were any interfering elements or whether the method was specific for phosphorus. MR. ELWELL replied that solutions of known compositions had not been examined, but he referred Mr. Gentry to Table VII, and added that in view of the wide variety of metallurgical samples examined, there was no evidence of interference by any of the common elements. MR. K. A. WICKHAM enquired if the low results for phosphorus in the presence of chromate could be explained, as the normal extraction of a yellow solution would lead one to expect high results for phosphorus.MR. ELWELL stated that low results were presumably caused by an oxidation reaction involving reduction of chromate, but it was not a simple reaction, otherwise interference by chromate would be apparent before the 5 per cent. chromium level. DR. J. H. HAMENCE congratulated the authors on having made a valuable contribution to the methods for the determination of phosphorus. He was not so much interested in the determination of phosphorus in steel as in the determination of phosphorus in fertilisers. He had found that the colorimetric vanado- molybdate method as described by Hanson gave a useful approximation of the amount of phosphorus present in the fertilisers, but the method was insufficiently accurate for many purposes. He had been fortunate in having had prior sight of the method from the authors and so he had had a chance of applying the method to fertilisers and the results so far obtained were a considerable improvement over those given by the Hanson technique. The three main advances made by the authors were: (1) the use of the larger144 COOK AND SPEIGHT: THE DETERMINATION OF SMALL AMOUNTS [Vol. 81 quantity of reagent, (2) the extraction of the coloured complex by isoamyl alcohol, which eliminated inter- ference from natural colouring matter present in solution, such as often occurred with basic slag, and (3) the measurement of the colour a t a specific wavelength instead of an over-wide waveband, as had been used by earlier workers. In dealing with fertiliser he preferred to use three extractions with isoamyl alcohol and dilute the extracts to the standard volume instead of using the one extraction as recommended by the present authors for steel examination. Dr. Hamence thought it was quite probable that in the near future this method might become a routine one for the determination of phosphorus in fertilisers.

ISSN:0003-2654    

DOI:10.1039/AN9568100136

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

144 COOK AND SPEIGHT: THE DETERMINATION OF SMALL AMOUNTS [Vol. 81 The Determination of Small Amounts of Carbon in Steel by Low-pressure Analysis BY R. M. COOK AND G. E. SPEIGHT (Presented at the meeting of the Society on Wednesday, October 5th, 1955) A simplified low-pressure method is described for the determination of carbon in steel. The use of a normal combustion furnace in conjunction with the low-pressure analytical apparatus permits samples to be analysed with a precision of -J= 0.0001 per cent. of carbon up to 0.036 per cent. of carbon, the deviation being slightly greater with higher carbon contents. The method lends itself readily to the analysis of carbon contents on a semi-micro basis when only small weights of sample are availablk. With medium and high-carbon steels, analyses may be carried out on samples weighing as little as 0-05 g with a degree of accuracy a t least equal to that given by the normal gravimetric combustion method operated under the best conditions, 2.729-g samples being used.THE development of silicon - iron alloys and low-carbon stainless steels in recent years has necessitated more precise methods for carbon determination. Although the normal combus- tion method, in which a weighed absorption tube is used, is generally satisfactory at the higher carbon levels, it suffers serious limitations when applied to low-carbon contents. Some improvement in accuracy is possible by the adoption of a semi-micro technique, but for precise analysis the methvd is considered to be unsatisfactory. Review of published work on the determination of carbon in recent years has led to the conclusion that, of the various methods available, the low-pressure method is potentially the one of highest precision.This method, which was first developed by Yensen,l consists briefly in burning the sample in a purified stream of oxygen, combustion being effected in a specially designed wire-wound furnace. The gases after combustion are passed over heated copper oxide to ensure complete oxidation of carbon to carbon dioxide, and the latter is removed from the oxygen stream by means of a freezing trap immersed in liquid air. After removal of oxygen by means of vacuum-pumps, the carbon dioxide is evaporated into a calibrated volume and measured in a McLeod gauge. The carbon content is then calculated from the volume of carbon dioxide collected.Subsequent worker^^^^^^^^ have used certain modifications of this method, the major alteration being the introduction of a high-frequency induction furnace for combustion of the sample. Murray and Ashley6 describe an apparatus comprising four analytical units in conjunction with a common high-frequency generator. The foregoing methods are undoubtedly capable of high precision, but they employ complicated apparatus requiring specially skilled operators, and are therefore not suitable for normal routine laboratory analysis. An important advance in this direction was made by Wells,7 whose excellent paper on the low-pressure determination of carbon shows that results of high accuracy are readily obtained by a much simplified procedure. The encourage- ment offered by the foregoing paper has stimulated further work on the determination ofMarch, 19561 OF CARBON I N STEEL BY LOW-PRESSURE ANALYSIS 145146 COOK AND SPEIGHT: THE DETERMINATION OF SMALL AMOUNTS [Vol.81 carbon by the low-pressure technique, and an apparatus has been evolved that is similar in basic principle to that described by Wells. Certain modifications have been made in order to simplify still further the design and operation of the apparatus, whilst maintaining the high precision afforded by the basic principle. METHOD APPARATUS- The complete apparatus consisting of combustion furnace, freezing trap and measuring gauge is illustrated in Figs. 1 and 2. Although the collecting and measuring apparatus is one self-contained unit, the com- ponent parts will be described separately for convenience.Oxygen sufpZy-The oxygen supply to the combustion furnace is passed through a silica tube containing platinised asbestos maintained at 600" C, which serves to oxidise carbonaceous impurities to carbon dioxide. The latter, together with water vapour, is removed by subsequent absorption in soda asbestos and magnesium perchlor ate. Blank values obtained from the oxygen supply purified in this manner are extremely low, being of the order of 0-00005 per cent. of carbon on the basis of a 2-729-g sample. Combustion furnace-The furnace consists of a Mullite combustion tube 26 inches x Q inch diameter heated over the central 9-inch length in a wire-wound furnace capable of maintaining a temperature of 1200" to 1250" C.Combustion of the sample is carried out in the con- ventional type of refractory boat, with the addition of specially prepared red-lead flux. Early tests indicated the need for some form of atmosphere trap to prevent the entry of carbon dioxide into the furnace during the introduction of samples. This consists of a short length of wide-bore glass tubing sealed axially to a short length of narrow-bore tubing, closed by means of a rubber stopper. A second tube sealed through a t right angles connects with the oxygen purification system. The boat containing the sample is placed in the mouth of the combustion tube, which is closed by connecting the atmosphere trap. Oxygen is passed to sweep out entrained air and, by temporarily removing the rubber stopper, the boat may be transferred into the hot zone by means of a steel rod inserted through the narrow glass tube.Owing to the positive pressure of oxygen within the combustion tube, further ingress of air is entirely prevented. The gases after combustion pass through an asbestos-wool filter to remove solid particles, and are partly dehydrated by means of magnesium perchlorate. Sulphur gases are removed by passage through precipitated manganese dioxide, and final removal of water vapour is ensured by means of magnesium perchlorate and phosphorus pentoxide traps. The three-way tap TI, shown in the diagram, serves to by-pass the oxygen supply a t the end of the com- bustion cycle. The gases, consisting of carbon dioxide together with excess of oxygen, pass into the freezing trap through tap T,, which serves to isolate the vacuum system.Freezing trap-Carbon dioxide is completely removed from the oxygen stream by means of a multiple-coil freezing trap immersed in liquid oxygen contained in a small vacuum-flask. Vacuum measuring system-The conventional diffusion pump and combined Toepler and McLeod gauge system has been modified in order to reduce the time required for collection and measurement of the carbon dioxide. Four calibrated volumes are provided by means of three bulbs connected through the three-way tap T, with the U-tube, which in turn is isolated from the freezing trap by means of tap T,. The four volumes measured from T, to the calibration marks are, respectively, 2-99, 5-24, 10.75 and 23-20m1, covering a range of carbon contents up to approximately 0.2 per cent.of carbon on the basis of a 2.729-g sample. A pressure difference of 1 mm is equivalent to 0.00007, 0.00012, 0.00025 and 0.00055 per cent. of carbon, respectively, on the basis of the above sample weight. Evacuation of the system is effected through the measuring limb by means of a two-stage mercury-diffusion pump backed by a rotary oil pump, which also serves to control the mercury in the reservoir through tap T,. During the combustion of the sample, the oxygen passing through the freezing trap is conducted via T, to the flowmeter, which is used to regulate the oxygen flow. A second phosphorous pentoxide trap prevents back-diffusion of water vapour into the freezing trap. PROCEDURE- The sample boats, which normally contain appreciable amounts of absorbed carbon dioxide, are stored in a muffle furnace a t 1000" C and removed immediately before use.Fig.2. General view of apparatusMarch, 19561 OF CARBON IN STEEL BY LOW-PRESSURE ANALYSIS 147 Similarly, the red-lead flux is stored in a muffle furnace at 450" C and removed in 2-g portions when required. These special precautions are necessary to ensure consistent low blank values. Samples in the form of millings or drillings must be washed in ether, otherwise completely erroneous results may be obtained. For the determination of carbon contents up to 0.2 per cent., the procedure is as follows. Transfer an empty boat into the hot zone of the combustion furnace and pass oxygen for 2 to 3 minutes to ensure complete oxidation of carbonaceous matter. Remove and allow to cool for 1 minute.Weigh out 2.729 g of the sample, mix with 2 g of red lead and transfer to the combustion boat. Insert into the mouth of the tube, replace the atmosphere trap, and pass oxygen for 2 minutes to sweep out entrained air, by-passing the stream through T,. Return T, to the normal position and, with T, and T o en, and T, connected to the flowmeter, adjust the oxygen flow to 750 ml per minute. (3ItPs assumed that the diffusion and rotary pumps are in operation in order to maintain the measuring system under vacuum from T4.) Raise the liquid-oxygen flask to immerse the freezing trap and transfer the boat into the hot zone of the furnace in the manner previously described, regulating the oxygen supply during combustion to maintain a slight positive pressure.At the end of the combustion period, usually Q to 1 minute, re-adjust the flow to 750 ml per minute. After a timed interval of 4 minutes from the commencement of combustion, by-pass the oxygen stream through T,, close T, and pump out the remaining oxygen by slowly turning T, to connect with the calibrated bulbs. At this stage remove the boat from the furnace and insert a new boat in readiness for the next sample, the oxygen flow being maintained. After evacuating the oxygen from the freezing trap and connecting tubes, which requires approximately 1 minute, isolate the pump by rotating T, through go", and transfer the liquid-oxygen flask to the U-tube. By warming the freezing trap, the collected carbon dioxide is rapidly evaporated and recondensed in the U-tube.At this stage, remove the fresh combustion boat from the furnace, allow to cool for 1 minute and replace in the mouth of the combustion tube after the addition of the next sample. For the measurement of the collected carbon dioxide, close tap T3 to isolate the calibrated volumes from the freezing trap, and open T, to connect with the bulbs. By opening T, slowly to air, allow the mercury to fill the bulbs to the top calibration mark. This operation isolates the pumps from the calibrated volumes, as the measuring limb is now filled with mercury to the same level. Remove the liquid oxygen from the U-tube and warm to room temperature. The timed application of hot air, followed by cold air from a hair drier, enables this to be effected in approximately 1 minute. Raise the mercury to the highest calibration point that can be reached, and note the corresponding height of mercury in the measuring limb by means of the fixed scale behind the latter.From this reading, subtract the base reading obtained for this particular calibration point when the system is under vacuum, to obtain the true pressure of the carbon dioxide. From the known values of pressure, volume and temperature, the volume of carbon dioxide at S.T.P. is readily calculated. To remove the carbon dioxide in readiness for the next test, slowly open T, to vacuum, so that when the mercury falls below the junction with the measuring limb, the carbon dioxide is removed by means of the pumps.Care is required in this operation, particularly when the amount of carbon dioxide to be removed is relatively high. In such cases it is preferable to compress most of the carbon dioxide into the volume between T, and T,, and after closing T, to lower the mercury to give free passage between the calibrated bulbs and the measuring limb. When exhaustion of the gas is complete, isolate the diffusion pump at T, and re-admit the oxygen supply via TI, T,, T, and T, in readiness for the next test. The proper timing of the various operations permits one test to be carried out every 12 minutes, the approximate timing for the complete cycle being as follows- The remaining carbon dioxide is then removed on re-opening T,. 0 minutes. 4 minutes, 5 minutes. 7 minutes.Insert sample foF combustion. Isolate freezing trap and pump out oxygen. Remove boat containing sample. Insert fresh boat. Transfer carbon dioxide from freezing trap to U-tube. Weigh next sample and flux. Remove fresh boat from furnace. Isolate U-tube and evaporate carbon dioxide into calibrated volume.148 COOK AND SPEIGHT: THE DETERMINATION OF 8 minutes. 10 minutes. 11 minutes. 12 minutes. Place next sample in mouth of furnace. Adjust mercury level and read pressure. Pump out carbon dioxide. Open taps, adjust oxygen flow and replace Insert next sample for combustion. Blank determinations are carried out in SMALL AMOUNTS [Vol. 81 liquid oxygen on freezing trap, the manner described with a combustion boat containing 2 g of red lead only. CALCULATION OF RESULTS- Assuming that a pressure reading of P,mm is obtained at a volume of V,ml and temperature T'K, with a zero vacuum reading of P2 mm at the particular calibration mark, the true volume of carbon dioxide at S.T.P.is given by- Volume of carbon dioxide = - *1--2x273xJ7,ml, T 760 PI - P2 273 44.01 X - X & X - T 22,430 " Hence, weight of carbon dioxide = ___ 760 273 '1 - '2 x - x V , x 44.01 x 10 per cent. and carbon in a 2-729-g sample = - T 22,430 760 EXPERIMENTAL During the development of the method, several factors were examined in considerable It is proposed to discuss these more detail in order to ensure the highest degree of accuracy. fully under separate headings. FURNACE TEMPERATURE- between 1050" and 1300" C ; 1100" C is a convenient working temperature for these materials.temperature, about 1200" to 1250" C. For plain carbon and low-alloy steels, consistent results were obtained at all temperatures Stainless and highly alloyed materials, however, were found to require a higher EFFICIENCY OF FREEZING TRAP- Various types of freezing traps were investigated in order to obtain maximum efficiency at a moderately fast rate of flow. Early forms consisting of straight U-tubes and helices of varying dimensions were found to be efficient only at a very low rate of flow, less than 100 ml per minute. The present form of trap, which consists of six vertical coils of narrow- bore tubing, is found to be completely efficient even with a rate of flow in excess of 1 litre per minute. Above a certain limiting value, the depth of immersion in the liquid oxygen is not critical.Tests carried out in the manner described with an additional flask of liquid oxygen placed on the U-tube show no loss of carbon dioxide either during the passage of gas or during evacuation. Similarly, no loss of carbon dioxide can be detected on prolonged evacuation, provided the trap remains immersed in liquid oxygen. BLANK VALUES- The major source of blank in the combustion procedure is the flux employed, the boats being virtually blank-free after the pre-ignition treatment described. Various fluxes, includ- ing lead foil, lead powder, tin foil and tin powder have been investigated, but high and somewhat erratic blank values have precluded their use. Red lead treated in the manner described has been found to give the lowest and most consistent blank values. Numerous determinations carried out over a period of several months have given total blank values between 0.0003 and 0.0005 per cent.of carbon, on the basis of a 2-729-g sample. This blank is considerably lower than that claimed by Wells,' and is of a similar order to the reported values obtained by workers in U.S.A., employing highly complicated apparatus.March, 19561 OF CARBON I N STEEL BY LOW-PRESSURE ANALYSIS 149 EFFECT OF WATER VAPOUR- The method described by Wells7 differs from the present method in that the carbon dioxide is measured when saturated with water vapour, for which a correction is made. It was considered highly desirable to collect and measure the carbon dioxide in the dry state, in view of the ready absorption of water vapour by glass surfaces. This has been achieved successfully by the use of a final phosphorus pentoxide trap immediately preceding the freezing trap.The efficiency of this desiccant has been clearly demonstrated by replacing the liquid-oxygen flask by one containing freezing acetone (- 95" C), at which temperature water is completely retained. After the carbon dioxide had been pumped off, no trace of water vapour was ever detected. TABLE I CARBON DETERMINATIONS WITH LOW-PRESSURE APPARATUS ON 2.729-g SAMPLES Carbon determined with Sample furnace, furnace, Carbon determined with normal tube high-frequency induction Yo % S.P.L. iron 0.0005 Mark 2 0*0005 (high-purity iron) 0*0005 0.0004 0.0005 0*0005 -4verage 0-0005 SG 6129 (plain carbon steel) 1-3.I. S . 13. A. MGS 92 (plain carbon steel) H . I. S . R. A. MGS 93 (18/8 stainless steel) B.S.I.R.A. MGS 94 (1 8/8 stainless steel) So. 9281 (18/13 stainless steel containing Mo and Nb) 0.0081 0.0081 0.0081 0.0081 0.0080 0.008 Z Average 0.0081 0.03 18 0.03 19 0-03 19 0.03 18 0.0319 0.0320 hverage 0.03 19 0.0231 0-0232 0.0230 0.0232 0-023 1 0.0232 Average 0.0231 0.0361 0.0360 0.0361 0.0360 0.036 1 0.036 1 0-0361 Average 0.0361 0.0606 0.0604 0.0603 0.0602 0.0606 0.0605 lZvera ge 0.0604 0.024 0-023 0.024 0.024 0.024 0.023 Average 0-023(7) 0.037 0.036 0.037 0-036 0-036 0-036 0-036 Average 0.036(3)150 vacuum conditions was checked for prolonged periods. dioxide could be detected. COOK AND SPEIGHT: THE DETERMINATION OF SMALL AMOUNTS [c’ol. 81 Although the phosphorus pentoxide trap is not evacuated in practice, its stability under No evolution of water or carbon EFFECT OF SULPHUR GASES- The efficiency of manganese dioxide for the removal of sulphur dioxide and trioxide from the gases after combustion was verified by the introduction of a second trap.Pressure - time curves determined for the evaporation of the carbon dioxide collected also showed no trace of either sulphur dioxide or trioxide. FORMATION OF CARBON NONOXIDE- tures, was considered to be a distinct possibility. The formation of some carbon monoxide, especially at the higher combustion tempera- In order to investigate this effect, a platinum TABLE I1 c:ARBON DETERMINATIONS WITH LOW-PRESSURE APPARATUS : XORMAL TUBE FURNACE Accepted carbon Sample Sample content, weight, % g D 6158 1*03(5) (plain carbon steel) 0.05 0.5 SG 5231 (plain carbon steel) 0*39(5) 0.05 i Carbon determined, % 1-04 1.03 1.03 1.04 1.04 1.04 Average 1.037 1.033 1.038 1.036 Average 1.036 0.396 0.398 0.392 0.395 0.394 0-396 Average 0.395 0.396 0.395 0.396 Average 0.396 catalyst was inserted between the manganese dioxide and magnesium perchlorate traps to ensure complete oxidation of carbon monoxide to carbon dioxide.Numerous check deter- minations carried out showed the complete absence of carbon monoxide. RESULTS OF TESTS- Table I shows the results of several determinations on six samples of plain carbon and stainless steel types. Check determinations on two stainless steel samples, with a Lindberg radio-frequency induction furnace in place of the normal tube furnace, are also included.Table I1 shows the results of tests on two plain carbon steels with various sample weights. Results obtained on a variety of miscellaneous samples analysed over a period of several months have also shown excellent reproducibility. COMMENTS ON RESULTS- The results of six determinations on each of the low-carbon steels given in Table I show that a reproducibility of -t. 0.0001 per cent. of carbon is readily obtained with plain carbonMarch, 19561 OF CARBON I N STEEL BY LOW-PRESSURE ANALYSIS 151 and stainless steel samples containing up to 0.036 per cent. of carbon. Sample KO. 9281 shows a slightly wider spread (2 0.0002 per cent. of carbon), which would be expected, as the sensitivity of the method is slightly lower at the larger calibrated volumes.The possibility of incomplete combustion of highly alloyed steels, at the maximum temperatures possible with the normal tube furnace, can be discounted in the case of samples MGS 93 and 94, since check determinations with the Lindberg radio-frequency induction furnace have not given significantly higher results. It should be pointed out that owing to the design of the latter instrument sample weights were limited to 1 g, and the sensitivity of the method was thereby decreased. In order to establish the absolute accuracy of the method, it would appear invidious to make a comparison with results obtained by other methods, as the low-pressure method is potentially far more accurate. Results given in Table I1 confirm the accuracy of the method, as the determinations on small and large weight samples are in very close agreement. TABLE I11 MISCELLANEOUS DETERNINATIONS WITH LOW-PRESSURE APPARATUS: FORMAL TIJBE FURNACE Material White cast iron .... High-carbon steel . . .. Plain carbon steel . . . . Plain carbon steel . . .. Plain carbon steel . . .. 80 Ni - 20 Cr alloy . . . . Plain carbon high-sulphur steel Plain carbon high-sulphur steel Sample weight, g .. 0.1 0.1 .. 0.1 0.5 .. 1 0.5 .. 3.799 1 0.5 0.2 . i 2.729 2.729 .. 2.729 2.729 2-729 2-729 .. 2.729 2.729 .. 2-729 2.739 Carbon determined, % 3-28 3-28 0.999 0.996 0.388 0.384 0.186 0.186 0.185 0.185 0-0920 0.0920 0.0240 0.0242 ' 0.0236 0.0240 0.107 0-107 0.120 0.120 APPLICATION OF THE METHOD The method described permits carbon determinations to be carried out on low-carbon steels with a high degree of accuracy.Owing to the simplified design of the apparatus com- pared with other forms of low-pressure apparatus described in the literature,l 9 2 9 3 9 4 9 5 s6 special operating skill is not required and an output of five tests per hour is readily obtained. This time cycle, which compares favourably with that of the normal gravimetric combustion method in which a single-tube furnace is used, enables the method to be used on a normal routine laboratory basis. The high sensitivity of the method is of particular advantage for the analysis of medium and high-carbon steels when only a small weight of sample is available. In this respect, a slight variation in technique is adopted in order to obtain the highest degree of accuracy. The red-lead flux is pre-ignited in the combustion boat in order to form a glaze before the addition of the sample.This procedure, which however is only applicable to low-alloy steels, results in a very consistent blank value of 0.0001 per cent. of carbon, on the basis of a 2.729-g sample. The authors express their thanks to Mr. F. H. Saniter and the Directors of The United Steel Companies for permission to publish this paper, and to Mr. TV. E. Bardgett, Research Manager, for his interest in the work.152 COOK AND SPEIGHT REFERENCES 1. Yensen, T. D., Trans. Electvochenz. SOG., 1920, 37, 327. 2. 3. 4. 5. 6. 7. Ziegler, N. A., Ibid., 1929, 56, 231. Wooten, L. A., and Guldner, W. G., I n d . Eng. Chem., Anal. Ed., 1942, 14, 835.Stanley, J. K., and Yensen, T. D., Ibid., 1945, 17, 699. Naughton, J. J., and Uhlig, H. H., Anal. C h e w , 1948, 20, 477. Murray, W. M., and Ashley, S. E. Q., I n d . Enp.. Chem., Anal. Ed., 1944, 16, 242. Wells, J. E., J . Iyon & Steel Inst., 1950, 166, 113. [Vol. 81 THE BRITISH IRON & STEEL RESEARCH ASSOCIATION HOYLE STREET SHEFFIELD July 7th, 1955 DISCUSSION MR. D. MAPPER said that he had been greatly interested in this paper, as a t Harwell they also had worked on this problem and had published their methods and results in an Atomic Encrgy Research Establishment report, “-4 Low-pressure Method for the Micro-determination of Carbon in Metals,” by G. A. Rarnett, J. N. Gregory and D. Mapper (A.E.R.E. Report C/R1315, January, 1954). The author had emphasised the simplicity of design and ease of working of his apparatus, and indeed, for routine work on the scale described in this paper, one readily admitted the advantages.On the other hand, it seemed that he had dealt rather harshly with those workers who had favoured the use of high- frequency heating and the “normal” vacuum technique (see report cited above). This type of apparatus might be slightly more involved to construct, but was hardly less easy to operate. When a large number of small samples had to be dealt with, very low blanks (on their apparatus, equivalent to 1 mg of carbon) were essential, and on 20-mg samples or less they had obtained standard deviations equivalent to 50.012 per cent. on a 3 per cent. nickel steel (0.035 per cent. of carbon).The author had given 12 minutes as the time for one analysis; he asked if this involved the complete operation from weighing of sample to calculation of result. MR. COOK acknowledged the excellent paper by Mr. Mapper and his co-workers, which they had studied with great interest during the early stages of their work. He fully agreed that high-frequency heating provided a rapid and convenient method, and he could appreciate the reasons for their preference of this form of heating for the particular type of analysis on which they were engaged. However, he stressed that the apparatus described here was developed primarily with the object of providing a method suitable for use in a works laboratory, without the need for specially skilled operators. Under these conditions he considered that the expense of a high-frequency generator would scarcely be justified.With the apparatus described, the time for one complete determination from weighing the sample to calculating the result was of the order of 12 minutes. In practice it had been found that an operator could readily achieve an output of 5 tests per hour. MR. W. T. ELWELL asked the author if he would indicate the smallest weight of sample he would require for the examination of a nominal 0.03 per cent. carbon highlyalloyed steel in order to give an accuracy of f0-002 per cent. MR. COOK said that a l-g sample would be adequate to give an accuracy of f0.002 per cent. with a highly alloyed steel containing nominally 0.03 per cent. of carbon, but he would prefer to mix the sample with 1 g of a low-carbon steel to ensure complete combustion. MR. J. E. STILL asked if the author had detected the diffusion of carbon dioxide through aMullite furnace tube a t 1200” to 1300°C. He and his colleagues had repeatedly measured this diffusion by a conductimetric method of detection. MR. COOK replied that he had not detected any appreciable diffusion of carbon dioxide through a Rlullite furnace tube, although occasionally he had experienced failure of these tubes owing to pin-holing. Comparable tests had been carried out on the samples reported, a high-frequency furnace with a cold tube being used, and with every sample there was good agreement between results with each type of heating. MR. R. H. A. CRAWLEY drew attention to the controversy in the U.S.,4., reported by Naughton and Uhlig (Anal. Chem., 1948, 20, 477). In the 0.001 per cent. (and lower) carbon range, low-pressure methods, the Yensen-type apparatus being used, gave lower results than methods in which an ordinary combustion tube was used. It was possible that the increase in temperature that occurred when a steel was burnt would temporarily increase the porosity of the combustion tube (in such a system as Mr. Cook’s). This phenomenon would not occur during the blank determination, so that a slightly high value would be expected by this method. MR. COOK said that possible errors caused by the increase in temperature of the tube during combustion must be extremely small, since tests carried out on progressively smaller samples had not shown any tendency towards higher or lower results. He also pointed out that the effect of diffusion of carbon dioxide through the hot furnace tube would result in lower values by this method, since the combustion tube was under positive pressure throughout the test.

ISSN:0003-2654    

DOI:10.1039/AN9568100144

出版商:RSC    

年代:1956

数据来源: RSC