ISSN:0003-2654    

DOI:10.1039/AN96691FX033

出版商:RSC    

年代:1966

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN96691BX035

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

iv SUMMARIES OF PAPERS I N THIS ISSUE [September, 1966Summaries of Papers in this IssueComparison of Particle-size Analysis Results Obtained by Usinga Centrifugal Photosedimentometer with those Obtained withCentrifugal Pipette EquipmentSize analysis results obtained with a centrifugal disc photosedimentometerare compared with those obtained with the Slater - Cohen disc centrifuge.I t is shown that for three of the powders tested, agreement between thepowders was good. The discrepancies noted with the fourth powder areprobably due to dispersion difficulties.M. W. G. BURTAtomic Weapons Research Establishment, Aldermaston, England.and B. H. KAYEI.I.T. Research Institute, 10 W. 35th Street, Chicago, Illinois, U.S.A.Analyst, 1966, 91, 547-552.A Colorimetric Method for the Determination of Oxides of NitrogenA method is described for determining oxides of nitrogen, and it has beenapplied to the gaseous products derived from initiating compositions.Theoxides of nitrogen are absorbed from the sample into sulphuric acid, iron(I1)sulphate is added, and the pink colour is measured. The interference effectsof a number of gases such as hydrogen sulphide and sulphur dioxide have beeninvestigated. The range of the method is 0.006 to 5 per cent. of oxides ofnitrogen (calculated as nitrogen dioxide).GEORGE NORWITZPitman-Dunn Laboratories, Frankford Arsenal, Philadelphia, Pennsylvania, U.S.A.Analyst, 1966, 91 , 553-558.Flame-spectrophotometric Determination of Calciumin Human SalivaA method is described for the determination of calcium in saliva by usingthe Unicam SP900 flame spectrophotometer with an acetylene - air flame.Interference by added orthophosphate is negligible.This, together withthe good recoveries of calcium added to saliva, indicates that the methodis applicable to the direct determination of calcium in saliva and that nospecial precautions need be taken to insure against interferences.J. G. JONES and J. D. R. THOMASDepartment of Chemistry, Welsh College of Advanced Technology, Cardiff, Wales.Analyst, 1966, 91, 559-562.Colorimetric Determination of Sodium Isethionate by Means ofAmmonium Ceric NitrateA direct colorimetric method is described for determining sodiumisethionate (2-hydroxyethane sulphonate) in aqueous solution by meansof its red complex with ammonium ceric nitrate.Ethylene glycol, if present,is removed from the sample by extraction with ethyl acetate before colori-metric determination.D. W. G. DICKER and T. H. NEWLOVEUnilever Research Laboratory, Port Sunlight, Cheshire.The over-all precision of the method is within f l per cent.Analyst, 1966, 91, 563-666SUMMARIES OF PAPERS IN THIS ISSUE [September, 1966The Spectrophotometric Determination of Vitamin D inFresh-water Fish Liver OilsThere are many difficulties associated with the determination of vita-min D, especially in natural products such as fish liver oils. Vitamin A isthe chief interfering material; it masks the absorption of vitamin D both inthe ultraviolet region and in the antimony trichloride colour test, makingthe determination of vitamin D almost impossible. In addition to vitaminA,, fresh-water fish liver oils contain vitamin A,, which also interferes in directspectrophotometry.A method for determining vitamin D is described, inwhich vitamins A, and A, are eliminated by converting them to anhydro-vitamins A, and A, and separating them from vitamin L> by chromatographyon an alumina column.R. K. BARUA and M. V. K. RAODepartment of Chemistry, Gauhati University, Jalnlrbari, Assam, India.Analyst, 1966, 91, 567-570.The Gas-chromatographic Analysis of Gases extracted fromMetals by Vacuum FusionThe equipment necessary to use a micro-ionisation detector of theLovelock type with helium carrier gas is described.Although not used a t itshighest sensitivity, the apparatus can measure quantitatively 1 per cent. v/vof components in a mixture of carbon monoxide, nitrogen, hydrogen andmethane of total volume 5 x cm3 a t S.T.P.M. T. LILBURNEMetallnrgy Division, National Physical Laboratory, Tcddington, Middlesex.Analyst, 1966, 91, 571-575.The Determination of Boron in Mild SteelA method of controlling the reaction between acetic anhydride andwater to permit the development of the boron - curcumin complex withoutthe previous separation of iron is described. The effect of other elementsin amounts normally found in mild steel has been examined. The method hasbeen applied to a series of British Chemical Standards and also to coinmcrcialsamples.T.S. HARRISON and W. D. COBBApplcby-Frodingham Steel Company, Central Laboratory, Scunthorpc, Lincoln-shire.Analyst, 1966, 91, 576-581.Distillation Method for Determining Total Carbon in SodiumA procedure is described for measuring total carbon in sodium by thcremoval of the alkali metal by distillation, combustion of the residue inoxygen and manometric determination of the resultant carbon dioxide.It has been shown by radio-tracer techniques that there is no loss of carbonfrom the sample during the distillation stage, and that recovery of variousforms of carbon is essentially complete. The coefficient of variation of asingle determination a t the 20 p.p.m. level is about 10 per cent. The biasis believed to be less than 10 per cent.; the average blank value is about1.5 p.p.m. of carbon.V.M. SINCLAIR, J. L. DRUMMOND and A. W. SMITHDounreay Experimental Reactor Establishment, United Kingdom Atomic EnergyAuthority, Thurso, Caithness, Scotland.Analyst, 1966, 91, 582-586... Vlll SUMMARIES OF PA4PERS IN THIS ISSUEThe Use of Diphenylcarbazone for the Determination ofMicrogram Amounts of LeadThe conditions necessary for extracting the crimson diphenylcarbazone -lead complex into xylene from an aqueous cyanide solution have beenexamined. A procedure is described for the spectrophotometric determinationof lead isolated by means of diphenylthiocarbazone.N. TRINDERMinistry of Agriculture, Fisheries and Food, National Agricultural Advisory Service,Kenton Bar, Newcastle upon Tyne 3.Analyst, 1966, 91, 587-590.[September, 1966Polarographic Determination of 0.01 to 0.10 per cent.ofBismuth in LeadShort PaperJ. BASSETT and J. C. H. JONESChemistry Department, Woolwich Polytechnic, London, S.E. 18.Analyst, 1966, 91, 591-502.The Photometric Determination of Excess of Cadmiumin Cadmium OxideShcvt PaperV. J. NORMANAustralian Defence Scientific Service, Defence Standards I.aboratories, Departmentof Supply, Adelaidc.Analyst, 1966, 91, 593.The Refractive Index of Aqueous Perchloric AcidShort PaperJ. R. McLEAN and G. S. PEARSONMinistry of Aviation, Rocket Propulsion Establishment, Westcott, Aylesbury, Bucks.Analyst, 196G, 91, 594-595.Semi-quantitative Determination of Organophosphorus Insecticidesby the Ring- oven Technique with Preliminary Thin- layerChromatographyShort PaPevI.PEJKOVIC-TADI~, M. B. CELAP, T. J. J A N J I ~Institute of Chemistry, Faculty of Science, University of Belgrade, 16 Studentskitrg, Belgrade.and S. LJ. VITOROVICInstitute for Application of Nuclear Energy in Agriculture, Veterinary and Forestry,Zemun, 15 Baranjska, Yugoslavia.Analyst, 1966, 91 , 595-597.Comments on “The Effect of Nitrilotriacetic Acid Impurity on theStandardisation of Solutions of Ethylenediaminetetra-acetic Acid”Short PaperR. G. MONKU.K.A.E.A., Atomic Weapons Research Establishment, Aldermaston, Berkshire.Analyst, 19GG, 91, 507-508x SUMXIARIES OF PhPERS I N THIS ISSIjE [September, 1966A Simple Multi-purpose TitrimeterShort PaperW.JENNISON and M. L. CLARKTJ.K.A.E.A., Analytical Group, 1)ounrcay Expcrimcntnl licactor l.:xtablishmcnt,Thurso, Caithnc4s, Scotlantl.Amdyst, lt)Mi, 91, 598-600.Some Anomalous Results given by Phase-solubility AnalysisShort PaperJ. V. WILKINSON and J. S. WRAGGAnalytical Development Group, Standards Dcpartment, Hoots Pure Drug Co. Ltd.,Station Street, Nottingham.A~?nZyst, 19G6, 91, GOO-603.A Simple Chromatograph for the Analysis of Air,Chlorine and Hydrogen ChlorideShort PaperD. M. RUTHVEN and C. N. RENNEYDepartment of Chemical Engineering, University of Cambridge, l’embrolie Street,Cambridge.Analyst, 1966, 91, 603-606.Official, Standardised and RecommendedMethods of AnalysisCompiled and Edited forTHE ANALYTICAL METHODS COMMITTEEOF THE SOCIETY FOR ANALYTICAL CHEMISTRYby S. C. JOLLY, B.Pharm., BSc., A.R.I.C., M.P.S.PP. xx+577Price 326 6s. netMembers of the Society for Analytical Chemistry are entitled t obuy copies at the special Members’ price of L4 4s. provided theyorder direct from:The Editor, The Analyst, 14 Belgrave Square, London, S.W.I.Remittances made out t o “Society for Analytical Chemistry”must accompany Members’ orders.Published for the Society for Analytical ChemistryW. HEFFER & SONS LTD., PETTY CURY, CAMBRIDGEb

ISSN:0003-2654    

DOI:10.1039/AN96691FP179

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

September, 3.9661 THE ANALYST ... XlllCLASSIFIED ADVERTISEMENTSThe rate for classic advertisements is 7s. a line (or spaceequivalent of a line), with an extra charge of 2s. for theuse of a Box Number. Semi-displayed classifiedadcertisernents are 80s. for single-column inch.Copy required not later than the 20th of the month pre-ceding date of publication which is on the 16th of eachmonth. Advertisements should be addressed to l h eAnalyst, 47 Gresham Sfreet, London, E.C.2. Tel.:MONnvch 7644.ANALYTICAL CHEMISTSPreferably with 2-5years expericncc in inorganic analysis. Knowledge ofX-ray Fluorescmce and Atomic Absorption Spectrometrydesirable.Graduate chemists or equivalent.A full benefits programme is available. Duties and startingApplications containing full details should be addressedsalary commensurate with qualifications and experience.to :Mr.J. Gillies,Gen. Supervisor of Laboratorirs,Noranda Mines Limited,‘NORANDA’ Quebec, Canada.NEW ZEALANDDEPARTMENT OF SCIENTIFIC ANDINDUSTRIAL RESEARCHApplications are invited for the undermentioned vacancy:lracancy B 13/18/50/3234: Scientist. Applications arcinvited for the position of chemist in the Division’s AucklandLaboratory. The appointee will be required to work with asmall group of specialists in the field of industrial chemicalresearch and service, including problems in plastics, watersand corrosion. A wide range of chemical and physicaltechniques is available while independent research isencouraged.Salary: Up to k2115 pa.is offered according to qualifica-tions and experience with opportunity for further advancc-rnent on scientific merit up to J;2575 p a .Qualifications desired: B.Sc. (Hons.).Passages: Fares for appointee and his wife and family, ifmarried, will be paid.Incidental expenses: Up to J;35 for a single man and El00for a married man can be claimed to cover the cost of takingpersonal effects to New Zealand..4pplication forms and general information are availablefrom the High Commissioner for New Zraland, New ZealandHouse, Haymarket, London, S.W.1, with whom applicationswill close on September Xkh, 1966.Please quote reference B 13/18/50/3234 when enquiring.IRELANDLOCAL APPOINTMENTS COMMISSIONVacancy forANALYTICAL CHEMIST, DUBLIN HEALTH AUTHORITY.SALARY: 41,640-L2,000 on certain conditions.First orSecond Class Honours University Degree in Chemistry orequivalent essential. Further vacancies, if they arise, maybe filled from this competition. Application forms, etc.from Secretary, 45, Upper O’Connell Street, Dublin, I.Latest time for receiving completed Application Forms: 5.30p.m. on SEPTEMBER 29th, 1966.Proposed UNIVERSITY OF BRADFORD(Bradford Institute of Technology)POSTGRADUATE SCHOOL OFPOWDER TECHNOLOGY-4 short course on“Particle Size Determination”will be hcld from Monday, October 10th to Friday, October28th, 1966.The three weeks’ course on the measurement of the sizeand surface area of solid particles includes lectures on thetheory of size measurement and experimental work in a wellequipped laboratory.Further details and application forms may be obtained fromthe Registrar, Bradford Institute of Technology, Bradford, 7.OPPORTUNITIES INCANADANoranda Research Centre, which is situatedin the Montreal area a t Pointe Claire, Quebec,is the Research Division of the Noranda MinesGroup of Companies.The Research Centreoffers excellent career opportunities in a num-ber of fields including metallurgy, chemistry,chemical engineering, solid state physics andtechnical economics. Candidates are presentlybeing sought for the following positions:SENIOR ANALYTICAL CHEMIST, withtraining and experience in modern physico-chemical methods as applied to metallurgicalanalysis together with capability and interestin the investigation of difficult and challenginganalytical problems, with a minimum of super-vision.Experience in atomic absorptionspectroscopy an asset. Requirements: l-’h.D. orequivalent plus several years relative cxperi-ence.ANALYTICAL CHEMIST to superintendroutine and development work of a groupengaged in thc analysis of a variety of metal-lurgical products by chemical and instrumentaltechniques. Rcquirernents: R.Sc. or equivalentwith five years relevant experience.Send complete resum6 in confidcncc to:Head, Department of Administration,Noranda Research Centre,240 Hymus Boulevard,Pointe Claire, Quebec,Canada.AccuratepH controlby unskilled staffThe Johnson range of COMPARATOR &UNIVERSAL test papers permits immediateaccurate readings within the range pH 1.0 topH 11.0.Inexpensive and simple to use, even byunskilled operatives, they are eminentlysuitablefor many aspects of works control in a varietyof industries.Comparator Books. Each of 20 leaves withsix colout matches inside cover.No.1035forpHl.Oto 3.5instepsof0.5pHNo. 3651 ,, pH 3.6 ,, 5.1 ,, ,, 0.3 pHNo. 5267 ,, pH 5.2 ,, 6.7 ,, ,, 0.3 pHNo. 6883 I , pH 6.8 ,, 8.3 ,, I , 0.3 pHNo. 8410 ,, pH8.4 ,, 10.0 ,, ,, 0.3pHUniversal Books. Each of 20 leaves with11 colour matches inside covers.Range:- pH 1.0 to pH 11.0 in steps of 1.0 pHHENDON - LONDON MWxxii THE ANALYST [September, 1966REPORTS OF THE ANALYTICAL METHODS COMMITTEEOBTAINABLE FROM THE SECRETARYThe Reports of the Analytical Methods Committee listed below may be obtained direct from theSecretary, The Society for Analytical Chemistry, 14 Belgrave Square, London, S.W.1 (not through TradeAgents), a t the price of 1s. 6d. to members of the Society and 2s. 6d. to non-members. Remittances mustaccompany orders and be made payable to “Society for Analytical Chemistry.”Additives in Animal Feeding Stuffs Sub-committee :Certain Rcports published before 1946 have been omitted from this list, but are still available.Report of the Antibiotics Panel : The Determination of Penicillin, Chlortetracycline and OxytetracyclineReport of the Hormones Pancl: The Determination of Stilboestrol and Hexoestrol in CompoundReport of the Prophylactics I’anel : The Determination of Nitrofurazone in Compound Feeding Stuffs.Report of the Vitamins (Water-soluble) Panel : The Determination of Water-soluble Vitamins inin Diet Supplements and Compound Feeding Stuffs.Feeding Stuffs.Compound Feeding Stuffs.Report of the Vitamins (Fat-soluble) Panel: The Determination of Fat-soluble Vitamins in DietSupplements and Compound Feeding-Stuffs.Animal Feeding Stuffs.Analytical Standards Sub-committee :Chlorine in Organic Compounds Sub-committee :Essential Oils Sub-committee :Report of the Prophylactics in Animal Feeds Sub-Committcc : The Determination of Amprolium inReport of the Prophylactics in Animal Feeds Sub-committee : The Determination of Sulphaquinoxaline.Sodium Carbonate as a Primary Standard in Acid - Base Titrimetry.The Semi-micro-determination of Chlorine in Agricultural Technical Organic Chemicals and theirFormulations.Report So.14. Solubility Test for Ceylon Citronella Oil. (Gratis.)Report No. 16.Fiore Method for Determining Linalol: Amendment.Application of Gas - Liquid Chromatography to Essential-oil Analysis: Interim Report on the Deter-Spectral Characteristics of Engenol.Analysis of Meat Extract.Determination of Gelatin in Mcat Extract and Meat Stocks: Interim Report.Nitrogen Factors for Pork and Nitrogcn Content of Rusk Filler (as onc reprint).Nitrogen Factors for Beef.Nitrogen Factors for Chicken.Nitrogen Factors for Liver.Nitrogen Factors for Veal.Nitrogen Factors for Turkey.Nitrogen Content of Rusk Filler.Report No.4.Determination of Lead in Foodstuffs : Tentative Method.Methods for the Destruction of Organic Matter.Sotes on Perchloric Acid and its Handling in Analytical Work.The Determination of Lead.The Determination of Small Amounts of Arsenic in Organic Matter.The Determination of Small ilmounts of Copper in Organic Matter.The Determination of Small iZmounts of Mercury in Organic Matter.Determination of Linalol in Essential Oils.(Gratis.)mination of Citronellol in Admixture with Geraniol.Meat Products Sub-committee (formerly Meat Extract Sub-committee) :Metallic Impurities in Foodstuffs Sub-committee :Determination of Zinc.Metallic Impurities in Organic Matter Sub-committee :Sub-committee on the Determination of Unsaponifiable Matter in Oils and Fats and ofUnsaponified Fat in Soaps:Report No. 6.Report on the Microbiological Assay of Riboflavine and Nicotinic Acid.The Determination of Carotene in Green-Leaf Material.The Determination of Carotene in Green-Leaf Material.The Chemical Assay of Aneurine (Thiamine) in Foodstuffs.The Microbiological Determination of Thiamine.The Estimation of Vitamin F3,2.Determination of Phenols in Soaps.Sub-committee on Vitamin Estimations :Part 1. Fresh Grass.Part 2. Green-Leaf Materials other thanGrass. (Gratis.)I~ Vitamin-E Panel:The Determination of Tocopherols in Oils, Foods and Feeding Stuffs.Tragacanth Sub-committee:Report No. 1.Report No. 2.Examination of Detergent Preparations.Determination of Small Amounts of Total Organic Chlorine in Solvent Extracts of Vegetable Material.Evaluation of Powdered Tragacanth.Evaluation of Flake Tragacanth.Soapless Detergents Sub-Committee:Pesticides Residues in Foodstuffs Sub-committee

ISSN:0003-2654    

DOI:10.1039/AN96691BP189

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

SEPTEMBER, 1966 THE ANALYST Vol. 91. No. 1086 Comparison of Particle-size Analysis Results Obtained by Using a Centrifugal Photosedimentometer with those Obtained with Centrifugal Pipette Equipment BY M. W. G. BURT (Atomic Weapoizs Research Establishment, A lderwzaston, England) AND B. H. KAYE (I.I.T. Research Institute, 10 W . 35th Street, Chicago, Illinois, l7.S.A .) Size analysis results obtained with a centrifugal disc photosedimentometer are compared with those obtained with the Slater - Cohen disc centrifuge. it is shown that for three of the powders tested, agreement between the powders was good. The discrepancies noted with the fourth powder are probably due to dispersion difficulties. SEVERAL new techniques for measuring the size distribution of powders by centrifugal sedimentation have been reported recent1y.l v 2 v 3 9 * From theoretical reasoning alone it is difficult to predict the relationship between results obtained with the various instruments because of the different physical phenomena used to measure concentration changes within the settling suspension.The experiments reported in this paper form the first part of an investigation of the performance of centrifugal particle-size analysers. The instruments compared are the Slater - Cohen disc centrifuge1 and the centrifugal disc photosedimento- meter developed by Kayeq2 Fig. 1. Slater - Cohen disc centrifuge The basic system of %he Slater - Cohen centrifuge is shown in Fig. 1. It is essentially a centrifugal analogue of the Andreasen pipette method. The centrifuging chamber is a hollow disc with a central pillar.Several holes are drilled through the base of the disc from the central pillar and terminate in a series of apertures equally spaced around the perimeter of a circle concentric with the axis of rotation. The disc is partly filled with suspension and rapidly accelerated to the speed at which the analysis is to be carried out. Under centrifugal forces the free surface of the suspension is a cylinder concentric with the axis of rotation. When the suspension has been spun in the centrifuge for a given time, a sample of suspension is withdrawn through the base by applying suction to the drainage pillar. The solids con- centration of the sample is then determined directly. The process is repeated at a series 547548 of times. A discussion of the method and the treatment of the results has been presented fully.l In the centrifugal disc centrifuge developed by Kaye the disc has transparent portions through which a beam of light is passed to measure the concentration of the particles as they are centrifuged outwards.On the top plate of the disc there is a cylindrical entry port at the axis of rotation through which the suspension is injected. In the experiments reported in this communication the two-layer analytical procedure was When this technique is used the centrifuge is run up to speed partially filled with clear liquid. When steady-state conditions are attained, the free surface of the fluid approximates to a cylinder, its axis being coincident with the axis of rotation.The powder to be examined is injected as a low concen- tration suspension through the entry port. Initially the incoming suspension does not have the energy to penetrate the free surface of the clear fluid, which has high angular momentum. During this initial period, which appears to last for about 2 or 3 seconds, the suspension forms a thin uniform layer over the free surface of the clear fluid. Therefore, all the particles start off at essentially the same distance from the centre of rotation, and it follows that each size of particle reaches the light beam at the measuring zone at a definite time. By measuring the changes in light intensity as the particles cross the beam, the concentration of each size group is measured continuously. Claims of absolute accuracy have not previously been made for the centrifugal disc photosedimentometer because of the difficulty of interpreting light scattering measurements of concentration changes. The light obscured by particles suspended in the path of a parallel beam of light is related to the concentration by the equation- BURT AND KAYE: COMPARISON OF PARTICLE-SIZE ANALYSIS RESULTS [Analyst, Vol.. . .. * * (1) log&= RE^^^ nidi2Ki . . I t where I, is the intensity of light beam entering the suspension; I t is the intensity of light beam leaving the suspension; R is a constant, dependent on the dimensions of the beam of light and on the shape ni is the number of particles of diameter d i ; Ki is the extinction coefficient of a particle of diameter di; s is the smallest particle present; and nz is the largest particle present.of the particles ; By definition, Ki is the ratio of light obscured by a particle to the light which it would have obscured if the laws of geometric optics were valid for the system under consideration. If, for a given system, the ratio of the particle diameter to the incident wavelength (A) is greater than 100, the value of Ki is constant, and can be taken as unity. For smaller ratios, Ki becomes a complex function of dl/A, the relative refractive index of the particle with respect to that of the supporting medium and the shape of the particle. The quantity, log,l,/It, is usually termed the optical density of the suspension, and changes in the optical density with time can be related to the concentration of particles in specific size groups.The need to know B is usually eliminated by carrying out the analysis until all the particles are removed from the suspension, and then calculating the percentage concentrations. The difficulty in absolute interpretation of the results arises from the need to know Ki. In theory its value can be either calculated or measured experimentally; however, insufficient results are available for the analyst to do this often. The calculations are so complex that in the past the writers6T7’8 have recommended that a white-light source be used in conjunction with a relatively wide-angle acceptance receiver for the transmitted light to minimise fluctuations in Ki, and then assume that Ki = 1 . By making this assumption, the equation can be manipulated readily, although the results can only be regarded as having a high precision, but an unknown accuracy.Recent workg ,lo has demonstrated the high precision that can be obtained with the centri- fugal photosedimentometer, however, as it is well known that particles of the same order of magnitude as the wavelength of light scatter light more effectively per unit-weight than larger particles in comparable systems. There is a possibility that the percentage weight of fines within a powder could be over-estimated when using the centrifugal disc photo- sedimentometer.6 I t was to explore the influence of the light-scattering properties of the small particles that the performance of the centrifugal disc photosedimentometer was com- pared with the Slater - Cohen centrifuge, as in this centrifugal equipment the concentration changes in the suspension are measured gravimetrically.September, 19661 OBTAINED BY USING A CENTRIFUGAL PHOTOSEDIMENTOMETER 549 EXPERIMENTAL Four powders that had been analysed on the Slater - Cohen disc centrifuge and the These four powders were appropriate analytical results were made available by Dr.Cohen. analysed on the centrifugal disc photosedimentometer. -der Fig. 2. Optical circuit of the Atomic Weapons Research Establishment centrifugal pho tosedimentometer The equipment used in this investigation was based upon the original specification by Kaye,2 and was designed and built at the Atomic Weapons Research Establishment, Alder- maston, by the Electronics Engineering Department. I t consists of a shallow cylindrical stainless-steel tank with a system of slotted windows set at 3 fixed radii (5, 7 and 9 cm) such that light can pass through the tank and its contents at these radii.The tank can rotate synchronously at either 750 or 1500 r.p.m. An incandescent lamp and lens unit directs a parallel beam of white light into four separate 45" mirrors. Three of these mirrors are able, one at a time (as selected by the move- ment of a shutter), to pass this measuring beam through the tank and its contents at the required radius on to a similar mirror beneath the tank, and thence on to a photoelectric ~ cell. The fourth mirror reflects a reference beam, through a slot in the rim of the tank, avoiding the contents of the tank, on to another mirror below the tank that directs the beam on to the same photoelectric cell. The system of slots is arranged to allow the cell to pick up the reference beam for one half cycle while the measuring beam is blocked, and the measuring beam on the next half cycle while the reference beam is blocked.The cell is thus continmusly comparing the two beams. In the path of the reference beam is a manually controlled optical wedge (the reference wedge). A similar wedge is placed in the measuring path, and is calibrated in optical density over the range 0 to 0.4, and automatically controlled by a continuously self-balancing servo- system. The movement of the latter optical wedge is recorded on a 10-inch chart recorder. With the rotating tank partly filled with clean liquid, the position of the pen recorder is adjusted to zero by operating the reference wedge manual control.The sample of powder suspension is then injected into the tank. Provided the solids concentration is low (see later discussion), the suspension forms a thin layer on the free surface of the fluid and the particles move outwards under centrifugal force. When the particles are in the path of the measuring beam, previously focused at the selected sampling radius, the intensity of the beam decreases. The strength of the beam passing through the disc is continuously compared, by means of the550 BURT AND KAYE: COMPARISON OF PARTICLE-SIZE ANALYSIS RESULTS [A?ZaiySt, VOl. 91 photocell, with that of the constant reference beam. Any difference between the two beams constitutes an error signal that is amplified and fed to the servo-system.This servo-system causes the measuring wedge to take up a new position to restore the balance between the two beams. The movement of the wedge is recorded on the pen recorder, thus providing a record of optical density at the sampling zone with time. A diagram of the optical circuit is shown in Fig. 2; further technical data concerning the instrument have been published.1° PROCEDURE- All four powders were analysed in the Atomic Weapons Research Establishment disc centrifuge by using water containing 0.01 per cent. of Cetavlon as the dispersed phase and spinning in a centrifuge at 750 r.p.m. It has been found experimentally that for the injected layer to be stable, very low solids concentrations have to be used in the injected s u s p e n s i ~ n .~ ~ ~ The value of the solids concentration used was determined by carrying out successive analyses at increasing solids dilution until the measured distribution was independent of the concen- tration used. The concentrations used in these experiments were- Percentage by volume Barium sulphate . . . . . . 0.002 Silica . . . . . . . . . . 0.008 Titanium dioxide . . . . . . 0.0015 Zinc oxide . . . . . . . . 0.003 Suspensions were made up as follows: The dry powder was coned and quartered, and the amount needed to make approximately 100ml of suspension was taken and weighed. The sample was gently mixed with water into a paste by using a rubber-tipped rod; it was then washed into a beaker. The suspension was made up to the required volume and viewed under a microscope to determine whether adequate dispersion had occurred.With moderate stirring, 20-ml samples of suspension were withdrawn from the bulk by using a syringe. (The syringe was moved about in the suspension during this process.) All the materials, with the exception of the zinc oxide, were readily dispersed in 0.01 per cent. Cetavlon solution, although some gentle working in a paste was required to breakup agglomerates each time. It was necessary, however, first to disperse the zinc oxide powder in a few drops of an 0.01 per cent. solution of sodium hexametaphosphate, and rather vigorous paste-mixing was required to break up some agglomerates that were quite large. Only the zinc oxide had both compounds added to it, the other three materials were dispersed with the Cetavlon alone.CALCULATION OF SIZE DISTRIBUTIONS Treasurell has shown that the range of sizes present in the sampling zone is proportional to dt and that, therefore, the optical density of the zone at time t is directly proportional to d!. TABLE I CALCULATION OF THE PROPORTION UNDERSIZE, PER CENT. w/w Area under optical Stokes’ diameter dcnsity z w s u s dt curve, Percentage by weight Proportion undersize, limits, p cm2 or inch2 within d , limits per cent. w/w d , t o d, a1 Wl =3 x 100 100 d, to d , a, w, =& x 100 100 - w1 A A d3 to d4 a3 w3 =a. x 100 A 100 - w, - w, etc. etc. etc. etc. 100 - 2el - 7 v 2 . . - La,,& a72 A w, = - x 100 >d, a , where A = total area under curve dl>d, . . . .September, 1966j OBTAINED BY USING A CENTRIFUGAL PHOTOSEDIMENTOMETER 55 1 By using the optical equation (1) to determine d t was constructed.This density zlersm time results given by the recorder, and applying the values of dt at various times, a graph of optical density against is the weight - frequency distribution curve for the powder. This curve is integrated by areas to give the percentage by weight within chosen particle diameter limits, and hence, cumulative proportions undersize, per cent. w/w, can is illustrated in Table I. The results of the two separate analyses on photosedimentometer are given in Table 11. given in Table IT. The data from the Slater TABLE TI DATA FOR THE SLATER - COHEN DISC CENTRIFUGE Zinc oxide Barium sulphate Titanium dioxide (rutile) - Cumulative Cumulative Cumulative proportion proportion proportion undersize.undersize, undersize, r------7 be calculated. This the centrifugal disc - Cohen analysis are Silica -7 Cumulative proportion undersize, Stokes’ per cent. w/w Stokes’ t 1 i ame t cr , f-A-, diameter , 4.0 90 94 95 2.0 3.0 87 91 92 1.5 2.0 80 85 86 1.0 1.5 74 74 75 0.9 1.0 61 48 46 0.8 0-9 56 40 37 0.7 0.8 5.0 32 28 0.6 0.7 43 25 21 0.5 0.6 34 18 15 0.4 0.5 25 13 10 0.3 0.4 16 8 6 0.2 0.3 7 5 3 - 0.2 2 2 1 - P A B B’ P per cent. w/w r-+ ( A B B’ 96 98 98 92 95 95 82 80 79 77 74 73 72 67 66 64 57 56 53 46 44 41 34 32 27 22 20 13 11 12 3 5 4 - - Stokes’ per cent. w/w cliameter, f---h---, ( FL A B B’ 2.0 95 98 98 1.6 92 96 95 1.0 81 86 86 0.9 77 79 ’78 0.8 70 71 60 0.7 61 62 60 0.6 48 48 47 0.5 37 30 31 0.4 17 18 19 0.3 5 1 0 10 0.2 4 5 - - - - Stokes’ per cent.w/w liameter, f-A-, P A B B’ 8.0 92 96 95 5.6 80 86 85 4.0 68 71 70 2.8 53 53 52 2.0 36 33 33 1.4 23 16 16 1.0 13 9 10 0- 7 6 4 5 0.5 3 2 2 - - - - - - - - *A, = Analysis as determined on the Slater - Cohen disc centrifuge. The data for the Slater - Cohen disc centrifuge are published by permission of the management B and 13’ = Analysis on the centrifugal disc photosedimentometer. of Simon Carves Ltd. DISCUSSION The agreement between the two separate analyses on the centrifugal disc photosedimento- meter further confirms the precision attainable with this instrument. The agreement between the results from the two instruments for the barium sulphate, titanium dioxide and silica is of the order to be expected from differences in the dispersion techniques and sampling variations.An important factor governing the scattering power of a small particle is the difference in refractive index between the particle and the surrounding medium.12 Taking the appropriate values of the refractive indices to be 1.64 for barium sulphate, 1-55 for silica, 2-76 for titanium dioxide (rutile) and 1.33 for water, the valuesll for the difference in refractive indices are 0-31, 0.28 and 1.43. Over the wide range of an important variable, and although the barium sulphate and titanium dioxide both contained a high percentage of particles, below 1-p Stokes’ diameter there is no indication that analysis by centrifugal disc photo sedimentometer over-estimates the fines content as compared to techniques in which the particle concentrations are measured gravimetrically.The analysis of the zinc oxide by the centrifugal disc photosedimentometer gives a coarser estimate of the particle-size distribution than by the Slater - Cohen disc centrifuge. As reported earlier we had difficulties in dispersing this powder. It has been brought to our attention since we completed this work that the mixture of sodium hexametaphosphate with Cetavlon solution eventually used to disperse the zinc oxide was an unfortunate choice because Cetavlon contains quaternary ammonium compounds that are incompatible with soluble phosphates. The resultant precipitations of the insoluble quaternary phosphate probably interfered with the efficient dispersion of the zinc oxide. In the absence of informa- tion on the dispersion technique used by Dr. Cohen, no further comment on this discrepancy is possible.552 BURT AND KAY In conclusion, it can be said that the results reported here indicate that the accuracy of the centrifugal disc photosedimentometer is probably comparable with that of the Slater - Cohen disc centrifuge, and the the fines content is not over-estimated for the powders considered. 1 . 8. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. REFERENCES Slater, C., and Cohen, L., J . Scient. Instrum., 1962, 39, 614. Iiaye, €3. H., “improvements in or Relating to Centrifuges,” British Patent 7173 (May 2nd, 1962). Donoghue, J . I<., and Bostock, W., Trans. Inst. Chem. Engrs (London), 1955, 33, 72. Lambert, G. &I., Res. Dev., 1963, No. 23, 42. Kaye, B. H., “Physical Problems of Particle Size Analysis,” Ph.D. Thesis, TTniversity of London, €<aye, B. H., and Allen, T., Analyst, 1965, 90, 14i. Kaye, B. H., PaiM Oil and Colour J . , 1965, 148, 1211. Groves, hl. J . , Kaye, B. H., and Scarlett, B., Brit. Chem. Engng, 1964, 9, 742. Burt, M. W. G., U.K. Atomic Energy Authovity Report, A.W.R.E., N o . O-i6/64, 1964. Treasure, C. R. G., Technical Paper No. 50, Whiting and Industrial Powders Research Council, 1962. -, Ibid., 1965, 148, 1156. 1964. van de Hulst, H. C., “Light Scattering by Small Particles,” John Wilcy & Sons Inc., N e w Yorl;, Received May loth, 1965 1957.

ISSN:0003-2654    

DOI:10.1039/AN9669100547

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

.dnaZyst, September, 1966, Vol. 91, $9. 553458 553 A Colorimetric Method for the Determination of Oxides of Nitrogen BY GEORGE NORWITZ (Pitman-Duniz Labovatories, Fvankford Arsenal, Philadelphia, Pennsylvania, U.S. A .) A method is described for determining oxides of nitrogen, and it has been applied to the gaseous products derived from initiating compositions. The oxides of nitrogen are absorbed from the sample into sulphuric acid, iron(1r) sulphate is added, and the pink colour is measured. The interference effects of a number of gases such as hydrogen sulphide and sulphur dioxide have been investigated. The range of the method is 0*005 to 5 per cent. of oxides of nitrogen (calculated as nitrogen dioxide). A METHOD was required for determining oxides of nitrogen in gases from initiating com- positions.The oxides of nitrogen include nitric oxide, nitrogen dioxide, nitrogen trioxide and nitrogen tetroxide (but not nitrous oxide). The determination of oxides of nitrogen is usually made as the nitrate by the phenoldisulphonic acid method,l v2 which is time-consuming. Another method for determining oxides of nitrogen involves the formation of nitrite by their absorption in alkali followed by diazotisation in which acid and a-naphthylamine or similar compounds are ~sed.1,~9~-~,6 This method has the disadvantage that the reaction is not stoicheiometric. Methods have also been proposed for determining the individual oxides of nitrogen in the presence of each other by titration79* and by physical means2~9~10,11~12 but, in general, these methods lack sensitivity.It seemed that a relatively simple colorimetric method could be developed for deter- mining oxides of nitrogen based upon absorption of the gases into sulphuric acid to form nitrate and nitrite, followed by the addition of iron(r1) sulphate to form a pink colour. The formation of a pink colour with iron(r1) sulphate has previously been used as the basis of a method for determining nitrate in many materials.13 to METHOD APPARATUS- The apparatus (Fig. 1) consists of a gas bulb (A), an adaptor (B) and a solid cap (C). The cap with the opening (D) is needed only if standard samples are to be prepared to check the method. The gas bulb has a capacity of about 800 ml (the volume to the top stopcock should be determined by filling with water).The ground-glass joints of the gas bulb, adaptor and caps have hooks for the purpose of attaching springs. The stopcocks are greased with Chorofluorolube grease (obtainable from Hooker Chemical Corporation, Niagara Falls, New York) as other greases are attacked by oxides of nitrogen. REAGENTS- SuZPhuric acid (10 + 3, v/v)-Place 600 ml of water in a Pyrex bottle, add 2 litres of concentrated sulphuric acid (spgr. 1-84) and cool the resulting mixture to room temperature. Iron(11) sulphate solution-Add 4.0 g of iron(I1) sulphate (FeS04.7H,0) to a mixture of 55 ml of water and 5 ml of concentrated sulphuric acid, and stir to dissolve. Add 200 ml of concentrated sulphuric acid and cool to room temperature. Prepare daily. Standard potassium nitrate solution-Dry analytical-reagent grade potassium nitrate a t 100" C for 2 hours.After cooling it, dissolve 2.1973 g of it in water and dilute the solution to 1 litre in a calibrated flask. 1 ml E 1.0 mg of nitrogen dioxide. Sodizm hydroxide solution (0.4 per cent. w/v).554 NORWITZ: A COLORIMETRIC METHOD FOR THE [Auzazyst, VOl: 91 \ Inner 14/35 Vacuum stopcocks u 3 inches A = Gas bulb, 800-mi capacity (approx.) B = Adaptor C = Solid cap D = Cap with opening opening 14/35 Fig. 1. Diagram of the apparatus used for the determination of oxides of nitrogen and preparation of standards PREPARATION OF CALIBRATION GRAPH- Transfer 2.0, 3.0, 4.0, 5.0, 6-0 and 7.0-ml portions of standard potassium nitrate solution, measured with a semi-microburette, to 250-ml beakers and add 5 drops of sodium hydroxide solution (0.4 per cent.).Evaporate thc solutions to dryness at low heat on the hot-plate. Cool the beakers to room temperature and add to each 25.0 ml of sulphuric acid (10 + 3, v/v) measured with a 25-ml graduated cylinder with 0-2-ml divisions. Allow the solutions to stand for several minutes to ensure dissolution of the salts. Add 25.0 ml of iron(1r) sulphate solution to each beaker, swirl the solutions,-and allow them to stand for 10 minutes. Transfer a portion of each solution in turn to a 1-cm cell and read the transmissions within 2 hours at 520 mp with a spectrophotometer that has been set to 100 per cent. transmission with the reagent blank. Plot milligrams of nitrogen dioxide against percentage transmission. Carry out a determination on a reagent blank.PROCEDURE- Collect the gas to be examined in the recommended gas bulb (A) (see Fig. 1) so that the pressure is less than 500mm and the content of nitrogen oxides (as nitrogen dioxide) is less than 0.7 per cent., calculated at atmospheric pressure. For higher percentages of oxides of nitrogen use a correspondingly lower pressure. If desired, the gas can be collected in any suitable device and transferred to the gas bulb. Measure the pressure in the gas bulb by connecting it to an open-end manometer, observing the difference in the mercury levels and subtracting this difference from the atmospheric pressure. Attach the adapter (B) and add 25ml of sulphuric acid, while manipulating the stopcock so that a little sulphuric acid remains in the adaptor.After all of the sulphuric acid has been added, turn the stopcock to admit air to bring to atmospheric pressure. Detach the adaptor and place it in a clean beaker. Attach the solid cap (C) to the stopcock and shake the bulb vigorously for 10September, 19661 DETERMINATION OF OXIDES I N NITROGEN 555 minutes. Re-attach the adaptor, open the second stopcock and add 25-0 ml of iron(I1) sulphate solution. Shake the bulb gently once and allow it to stand for 5 minutes. Transfer a portion of the iron(I1) sulphate to a 1-cm cell and read the transmission within 2 hours at 520mp with a spectrophotometer that has been set to 100 per cent. transmission with the reagent blank (equal volumes of sulphuric acid and iron(I1) sulphate solution). Convert the reading obtained to milligrams of nitrogen dioxide by reference to the calibration graph.Calculate the corrected volume of the gas as follows- where V c is the corrected volume of gas, ml; Vb is the volume of gas bulb, ml; P is the pressure of gas in bulb, mm; and 7' is the room temperature, O absolute. Calculate the percentage of nitrogen dioxide as follows- W x 24.1 x 100 46 Vc! Percentage of nitrogen dioxide = where IV is the nitrogen dioxide found, mg; and V c is the corrected volume of gas, ml. (The formula weight of nitrogen dioxide is 46 and the gram molecular volume at 20" C and 760 mm pressure is 24.1.) Combining the above equations- 136 x W x T v, x P Percentage of nitrogen dioxide = To prepare for the next run, wash out the bulb with water and heat it over a Meker burner to drive off most of the moisture; while still hot connect the bulb to a vacuum pump.RESULTS APPLICATION OF THE METHOD TO THE DETERMINATION OF NITRATE AND NITRITE IONS- For the direct application of the iron(I1) sulphate method to the sulphuric acid solution of the oxides of nitrogen, the complexes of nitrate and nitrite with iron(I1) sulphate must be the same. The relationship between the two complexes was therefore investigated by using potassium nitrate solution (2.1973 g per litre) and sodium nitrite solution (1.500 g per litre), both solutions containing the equivalent of 1.0 mg of nitrogen dioxide per ml. After treating 1 to 5ml of the solutions with 5 drops of sodium hydroxide, the solutions were evaporated to dryness in beakers on the hot-plate.The beakers were cooled to room temper- ature, 254ml of sulphuric acid were added to each beaker and the solutions were then allowed to stand for a few minutes to dissolve the salts. Iron(1r) sulphate solution was then added to each beaker in 25-ml portions and after a few minutes the transmissions were measured at 520 mp. Nitrite solutions are not oxidised during the evaporation on the hot-plate.20 The potassium nitrate dissolved in the sulphuric acid without loss as was expected, considering that up to 1 g of potassium nitrate may be dissolved in sulphuric acid (94.5 per cent.) in the nitrometer method for determining nitrate.21 The sodium nitrite dissolved in the sulphuric acid in the range indicated (up to 5 mg of nitrogen dioxide) without loss, but with amounts larger than 5 mg of nitrogen dioxide, some loss seemed to occur. During the course of the work an improvement was made to the method for preparing the iron(I1) sulphate solution.It is recommended that it should be prepared by dissolving the iron(r1) sulphate in dilute sulphuric acid (5 parts of sulphuric acid + 55 parts of water, v/v) and then adding concentrated sulphuric acid. The iron(I1) sulphate dissolved im- mediately without oxidation and clear water-white solutions were always obtained. The sulphuric acid and iron(I1) sulphate solution were added by means of a 25-ml graduated cvlinder with 0.2-ml divisions. By using this technique the necessity of transferring to, and diluting in, calibrated flasks was avoided; this was convenient when working with gas lml bs.556 NORWITZ: A COLORIMETRIC METHOD FOR THE [Analyst, Vol.91 The results for the potassium nitrate and sodium nitrite solutions showed that the transmissions obtained with the two solutions were the same, indicating that the complexes with the nitrate and nitrite were similar. Further confirmation that the complexes were identical was obtained by making a spectrophotometric curve for the nitrite complex; this curve was the same as that obtained for the nitrate complex.16 The apparent explanation for the colour complexes being the same is presumably that the iron(I1) ion reduces the nitrogen to the tervalent state (nitrous acid) in the concentrated sulphuric acid, and that the tervalent nitrogen then reacts with the iron(I1) ion to form the pink complex. If the nitrogen is already in the tervalent state no reduction is involved. These findings concerning the similarity of the two complexes are at variance with the conclusions of English1* who stated that the complexes were not the same. English, however, used conditions that were different from those used by the present investigator.APPLICATION OF THE COLORIMETRIC PROCEDURE TO OXIDES OF NITROGEN- In order to check the application of the method to the analysis of oxides of nitrogen, it was necessary to prepare standard gas mixtures. In this investigation standard samples were prepared by using a Hamilton gas syringe (KO. 1001), small cylinders of nitric oxide and nitrogen dioxide, and a special gas bulb (Fig. 1). The cylinders of nitric oxide and nitrogen dioxide were of the standard type, 2 inches in diameter and 15 inches in height, (Matheson size KO.9). They were equipped with a Matheson KO. 59 valve, which is a combination of a stainless-steel valve with a vertical stainless-steel tube (12 inches in height and & inch in diameter).22 The procedure in preparing the gas mixture was as follows : the gas bulb (A) was evacuated and the adaptor (B) and the cap with the 1-5-mm opening (D) were attached. The end of the stainless-steel tube from the valve of the cylinder of nitric oxide or nitrogen dioxide was connected by a short piece of Tygon tubing (obtainable from U.S. Stoneware Co., Tall- madge, Ohio) to a piece of glass tubing that was several inches in length, and dipped below the surface of water contained in a small Erlenmeyer flask.The stainless-steel tubing was flushed at a moderate rate (under a hood) with nitric oxide or nitrogen dioxide, the gas-flow reduced to a very slow rate, and the Tygon tubing finally disconnected. The Hamilton syringe was inserted into the stainless-steel tube, filled and emptied, and then filled and emptied again. The syringe was filled with gas to the appropriate mark, inserted through the 1.5-mm opening of the cap of the adaptor, and depressed. The syringe was withdrawn and a finger immediately placed over the 1.5-mm opening. The top stopcock was opened a little, the finger withdrawn, the nitrogen oxide flushed into the system by the air, and the stopcock closed. Up to five 1-ml portions of the gas could be added by using this technique.After adding the nitric oxide or nitrogen dioxide to the gas bulb, the cap was detached, and 25.0 ml of sulphuric acid was added while manipulating the stopper so that a little sulphuric acid remained in the adaptor. After all of the sulphuric acid had been added, air was admitted to bring the bulb to atmospheric pressure and the analysis was carried out as described in the section headed Method. Experiments showed that vigorous shaking of the bulb for 10 minutes was sufficient. It is necessary that at least half of the nitric oxide should be oxidised (an equimolar mixture of nitric oxide and nitrogen dioxide behaves like nitrogen trioxide and dissolves in sulphuric acid to form nitrous acid, while nitrogen dioxide alone dissolves in sulphuric acid to form nitrous acid and nitric acid). It would be expected that much more than enough oxygen to oxidise half of the nitric oxide to nitrogen dioxide would be present if the recom- mended method is followed, which requires that the initial pressure of 500 mm or less in the 800-ml gas bulb should be raised to atmospheric pressure by the admission of air.This conclusion was checked by filling two evacuated gas bulbs to 500-mm pressure with nitrogen, carefully admitting 0.1 and 1.0 ml of nitric oxide and proceeding as in the described method. Transmissions of 95 and 66 per cent. were obtained for the 0.1 and 1.0 ml of nitric oxide, respectively. When the same experiment was repeated without the initial nitrogen being present, transmissions of 96 and 66 per cent. were obtained.Experimental calibration graphs were prepared with 0-1 to 4 ml of nitric oxide and 0.1 to 2.5 ml of nitrogen dioxide, The upper limits cited gave a transmission of approximately 20 per cent. Some vacuum should remain in the gas bulb after the flushing. These experimental calibration graphs followed Beer’s law.September, 19661 DETERMINATION OF OXIDES I N NITROGEN 567 The volumes used for the preparation of the experimental calibration graph for nitric oxide were corrected to S.T.P., and the theoretical amount of nitrogen dioxide that would be produced was calculated by using the gram molecular volume. Actual recovery averaged 97.4 per cent. of theoretical (Table I). This is an excellent result considering the uncertain TABLE I RECOVERY OF NITRIC OXIDE Nitric oxide Nitrogen dioxide 7 r------h---- ~ ----- h added, corrected to S.T.P., found, theoretical, Percentage ml ml mg mg recovery 0.10 0.20 0.50 1.00 1.00 2-00 3.00 4.00 0.09 0.18 0.46 0.9 1 0.9 1 1-82 2-73 3.64 0.18 0-36 0.90 1.85 1-85 3.68 5.50 7.30 0.19 0.37 0.94 1.87 1.87 3.74 5.61 7-45 Average 94.7 97.3 95-7 98.9 913.9 98.4 98.0 97.6 97.4 state of purity of nitric oxide.gv10 The reproducibility and the ratios of the nitrogen dioxide recovered in the experiments with nitrogen dioxide were good (Table 11).Application of the gram molecular volume calculations to nitrogen dioxide is uncertain because of the equilibrium between nitrogen dioxide and nitrogen tetroxide, the impurities that are present and the fact that nitrogen dioxide does not behave as a perfect gas.Accepting that nitrogen tetroxide is 20 per cent. dissociated into nitrogen dioxide at 27" C at atmospheric pressure,23 the recoveries in Table I1 are approximately 90 per cent. of theoretical values. TABLE I1 RECOVERY OF NITROGEN DIOXIDE Nitrogen dioxide added, ml 0.10 0.20 0.50 0.50 1.00 1.00 2.00 2.00 2.50 corrected to S.T.P., ml 0.09 0.18 0.46 0.46 0.92 0.92 1.84 1.84 2-30 ------l found, mg 0-30 0-60 1-50 1-58 3.00 2.96 5.91 5.9 1 7.48 The calibration curve obtained with nitric oxide could be readily used for calculating the percentage of nitric oxide in a gas sample, especially if all volumes were referred to fixed conditions, e.g., 760 mm pressure and 20" C. The graph obtained with nitrogen dioxide cannot be used advantageously for calculating the percentage of nitrogen dioxide, as a mixture of nitrogen tetroxide and nitrogen dioxide rather than nitrogen dioxide is used for plotting the graph.The graphs for nitric oxide and nitrogen dioxide prepared as above were found to be useful for testing interferences. The method in which the 800-ml bulb and 500-mm pressure are used will give good results for up to 0.7 per cent. of oxides of nitrogen, calculated as nitrogen dioxide For the analysis of larger amounts of nitrogen dioxide it is only necessary to decrease the pressure. The miiiimum amount of oxides of nitrogen that can be detected (by using as a criterion the amount that will give 98 per cent, transmission) is approximately 0.005 per cent. of nitrogen dioxide. If it is known that nitric oxide is the only oxide present, the calculation can, of course, be made as nitric oxide.Low results were obtained each time on shaking with the iron(I1) sulphate solution without first dissolving the nitrogen oxides in sulphuric acid.558 NORWITZ INVESTIGATION OF THE EFFECT OF OTHER GASES- A study of possible interferences (Table 111) showed that carbon monoxide, carbon dioxide, hydrogen, oxygen, nitrogen, methane and nitrous oxide did not interfere. The presence of as much as 0.6 per cent. of sulphur dioxide (calculated at atmospheric pressure) did not cause interference, but the presence of more than 0.04 per cent. of hydrogen sulphide (calculated at atmospheric pressure) gave low results. TABLE TI1 TESTS FOR INTERFERENCES Present (in 805 ml volume) Nitrous oxide (500 mm pressure) .. . . . . . . . . Nitrous oxide (100 mm pressure) 1- 1.00 ml of nitric oxide Gas mixture* (300 mm pressure) + 1.00 ml of nitric oxide Sulphur dioxide (1 ml) + 0.50 ml of nitrogen dioxide Hydrogen sulphide (0.1 ml) + 1.00 ml of nitrogen dioxide Hydrogen sulphide (0.3 ml) + 1.00 ml of nitrogen dioxide Hydrogen sulphidc (0.5 ml) + 1.00 ml of nitrogen dioxide Hydrogen sulphide (1 ml) 1- 1.00 ml of nitrogen dioxide . . . . Nitrous oxide (100 mm pressure) + 1.00 ml of nitrogen dioxide . . Gas mixture” (300 mm pressure) + 1.00 ml of nitrogen dioxidc . . Sulphur dioxide (200 mm pressure) t- 0.50 ml of nitrogen dioxide . . . . . . . . . . . . . . . . . . . . . . . . . . Volume found, ml N & & z - G z c dioxide oxide . . 0.00 - . . - 1-02 . . 1.00 - . .- 0.99 . . 0-98 - . . 0-52 - . . 0.41 - . . 1.00 - . . 0.98 - . . 0.68 - . . 0.36 - * Contains : carbon dioxide, 40.0 pcr cent. ; nitrogen, 44.0 per cent. ; oxygen, 2.9 per cent. ; hydrogen, 4.0 per cent. ; carbon monoxide, 8.7 per cent. : methane, 0.6 per cent. GASES DERIVED FROM INITIATING COMPOSITIONS- obtained were 0.56, 0.021, 1-21, 1.07 and 0-53 per cent. of nitrogen dioxide. The gases derived from 5 different initiating compositions were analysed. The results The author thanks Mr. Samuel Sitelman of this laboratory for his suggestions. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. REFERENCES American Society for Testing Materials, A.S.T.M. Stand., 1961, Philadelphia, Pa., 1961, Part 10, pp. 1600 and 1611. Clear, A. J., and Roth, M., in Kolthoff, I .M., Elving, P. J., and Sandell, E. B., Editors, “Treatise on Analytical Chemistry,” Part 11, 1701. 5, lnterscience Publishers Inc., New York and London, 1961, p. 217. Internal Division of Pure and Applied Chemistry, “Methods for the Determination of Toxic Substances in Air,” Butterworth and Co. Ltd., London, 1961, Method 16.1. Strafford, N., Strouts, C. R. N., and Stubbings, W. V., Editors, “The Detcrrnination of Toxic Substances in Air: A Manual of I.C.I. Practice,” W. Heffer & Sons Ltd., Cambridge, 1956, p. 195. Jacobs, XI. B., and Hochheiser, S., Agzalyt. Chein., 1958, 30, 426. Saltzman, I3. E., Ibid., 1954, 26, 1949. Johnson, C. L., Ibid., 1952, 24, 1572. Whitnack, G. C., Holford, C. J., Gantz, E. St. Clair, and Smith, C. R. J,., Ibid., 1951, 23, 464. Fricdel, R. A , , Sharkey, A. G., Shultz, J. L., and Humbert, C. R., Ibid., 1953, 25, 1314. Saier, E. L., and Pozefsky, A., Ibid., 1954, 26, 1079. Nicksic, S. W., and Harliins, J., Ibid., 1962, 34, 985. Comer, S. IV., and Jensen, A. V., Ibid., 1964, 36, 799. Bowman, F. C., and Scott, W. W., J . I?ad. Engng Chem., 1915, 7, 766. English, F. L., I ~ z d . Engng Chewt. Autalyt. Edpz, 1947, 19, 850. Laccetti, M. A., Seniel, S., and Roth, M., AFzaljJt. Chem., 1959, 31, 1049. Norwitz, G., Ibid., 1962, 34, 227. ___ , Analyst, 1962, 87, 839. Semel, S., Laccetti, M. A., and Roth, M., Analyt. Claern., 1959, 31, 1050. Swann, M. H., and Adams, M. L., Ibid., 1956, 28, 1630. Sisler, H. H., irz Sneed, M. C., and Brasted, K. C., Editors, “Comprehensive Inorganic Chemistry,” Military Standard, “Propellants, Solid : Sampling, Examination, and Testing,” MTL-STD-286A, Matheson Co. Inc., “Compressed Gases and Fluid Controls,” East Rutherford, New Jersey, 1963. Ephraim, F., “Inorganic Chemistry,” Sixth Edition, revised by Thorne, P. C. L., and Roberts, E. I?., Oliver and Boyd Ltd., Edinburgh, and Tnterscience Publishers Inc., New York, 1954, p. TOO. Received January 5th, 1966 D. Van Nostrand Co. Inc., New York, Toronto and London, 1956, Volume 5, pp. 72 and 7 7 . Washington, 1963, Method 209.3.2.

ISSN:0003-2654    

DOI:10.1039/AN9669100553

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

Analyst, September, 1966, Vol. 91, p$. 559-562 550 Flame-spectrophotometric Determination of Calcium in Human Saliva BY J. G. JONES AND J. D. R. THOMAS (Depavfmeizt of Chemistry, Welsh College of Advanced Techpiology, Cavdi’; U’ales) A method is described for the determination of calcium in saliva by using the IJnicam SP900 flame spectrophotometer with an acetylene - air flame. Interference by added orthophosphate is negligible. This, togethcr with the good recoveries of calcium added to saliva, indicates that the method is applicable to the direct determination of calcium in saliva and that no special precautions need be taken to insure against interferences. THE determination of calcium in biological fluids continues to be a subject of intense interest. The methods that have been developed range from the classical oxalate - permanganate titration and oxalate - carbonate gravimetric method, through the more recent complexometric and colorimetric techniques, to the physico-chemical methods of emission flame photometry and atomic-absorption spectroscopy.Saliva has been examined for calcium by several of the above methods, but, unfortunately, most are subject to certain disadvantages. For example, the oxalate methods require large amounts of saliva, while the flame-spectroscopic methods are frequently subject to inter- ferences, especially by refractory-forming materials. In this latter connection, the presence of phosphorus is a major source of interference in both the flame-photometric1 and atomic- absorption spect roscopic2 met hods. Acetylene - air mixtures give hotter flames than either coal gas - air or propane - air mixtures, a feature which facilitates the dissociation of the refractory compounds that tend to arise in the presence of phosphates and other materials present in biological fluids.Acetylene - air mixtures are, therefore, used on a wide scale in flame spectroscopic methods. However, there does not appear to be any reference in the literature to the determination of calcium in saliva by using these mixtures in conjunction with the flame-spectrophotometric method. In view of the rather complex measures that frequently have to be taken to overcome interferences in the determination of calcium in saliva, particularly with regard to phosphorus, the present investigation was undertaken to explore the possible application of the use of the acetylene - air flame in conjunction with the flame-spectrophotometric method for the direct determination of calcium in whole saliva.EXPERIMENTAL APPARATUS- is used. warm-up period of several hours in accordance with the manufacturer’s instructions. A Unicam SHOO flame spectrophotometer, fitted with an acetylene - air burner unit To ensure sound reproducibility of results, the instrument should be allowed a PRE PAKATI ON o F SOLUTION S- Standard calcium solution-Dissolve 0.2518 g of calcium carbonate in 60 ml of 0.1 M hydrochloric acid and dilute to 1 litre with de-ionised water. This stock solution contains 100 pg of calcium per ml and is diluted to the required concentration with de-ionised water, or as otherwise described for the phosphorus interference studies.Standard ovthophosphate soEution-Dissolve 1-8700 g of disodium hydrogen orthophos- phate, Na,HP04.12H,0, in 1 litre of de-ionised water. This stock solution contains 500 pg of orthophosphate per ml (Po43-), and for the phosphorus interference studies it is added in appropriate aliquots when preparing the standard calcium solutions. However, for the study of the effect of orthophosphate concentration at 1500pg per ml (as PO:-), prepare the solution directly by dissolving 1.4020 g of the disodium hydrogen orthophosphate in de-ionised water. Add 5 ml of the stock calcium-containing solution, and dilute to 250 ml with de-ionised water.560 JONES AND THOMAS FLAME-SPECTROPHOTOMETRIC [AnaLJ'St, VOl.!)I SETTING-UP AND CALIBRATIOX PROCEDUKE- Set up the flame spectrophotometer. Calibrate the instrument at 4227 A& and a slit width of 0.065 mm, with the acetylene - air flame controlled at 28-5 p.s.i. for the air and at 13.2 inches pressure for the acetylene. With the voltage selector at 1, the filter selector at 3 and the electrical band-width a t 1, adjust the galvanometer controls to give a zero galvanometer deflection while de-ionised water is sprayed into the flame, With a calcium solution containing 10 pg per ml being sprayed into the flame, adjust the instrument gain selector to give a galvanometer deflec- tion of 80. Continue adjusting the galvanometer controls and the instrument gain selector, whilst water and the calcium solution are alternately sprayed into the flame, to a galvanometer deflection of zero and 80, respectively, until no further adjustment is necessary.Switch the electrical band-width control to 4, and by using standards containing 0 to 10 pg of calcium per ml, check the linearity of the calibration. DETERMINATION O F CALCIUM IN WHOLE SALIVA- After following the setting-up and calibration procedure described above, dilute 1 ml of saliva to 25 ml (or alternatively, to 10ml) with de-ionised water and spray into the acetylene - air flame. When the dilution factor has been allowed for, the calcium content can then be determined directly from the galvanometer deflection. INTERFERENCE CHECKS- Phosphorus as orthophosphate-Check the calibration for absence of interference by phos- phorus, in the form of orthophosphate, on solutions containing 2 pg of calcium per ml together with varying amounts of orthophosphate ranging from 0 to 100 pg of orthophosphate per ml.The solution containing 1500 pg of orthophosphate per ml is prepared by the method described above. As a further check on the absence of interference by added orthophosphate, take readings on various calcium-containing solutions, each containing 100 pg per ml of added ortho- phosphate. De-proteinisation-Check whether de-proteinisation has any effect on the apparent calcium content of the saliva by adding 5 ml of an aqueous 4 per cent. solution of trichloro- acetic acid to 1 ml of saliva and spinning the solution in a centrifuge. Dilute the supernatant liquid to 25ml with de-ionised water and determine the calcium content in the manner described above.COMPARISON OF PROPOSED FLAME-SPECTKOPHOTOMETRIC METHOD WITH THE CLASSICAL De-proteinise3 a 10-ml aliquot of saliva with 1.0 ml of 30 per cent. trichloroacetic acid. Spin the solution in a centrifuge. Use the supernatant liquid to determine the calcium content by the oxalate - permanganate titration m e t h ~ d . ~ Compare the result with that obtained for a 1-ml aliquot of the same sample of saliva by the proposed flame-spectrophotometric method. OXALATE - PERMANGANATE TITRATION METHOD- RESULTS CALIBRATION- Setting the galvanometer deflection to zero with de-ionised water required the galvano- meter controls to be set a t 8.6, while the instrument gain selector was set at 3-0 at the deflection of 80 for the solution containing 10 pg of calcium per ml.The check on the setting-up calibration for calcium gave the following linearly related figures- Calcium concentration, pg per ml . . 0 2 4 6 8 1 0 Galvanomcter deflection . . . . . . 0 16 32 48 64 80 Calibration checks on a solution containing 2 pg of calcium per ml and varying amounts of orthophosphate gave the following galvanometer deflections, which indicate negligible interference by phosphorus under these conditions- Phosphorus concentration, pg of per ml . . 0 10 30 50 7 0 100 1500 Galvanometer deflection . . . . . . . . 16 17 15 17 17 17 I 7September, 19661 DETERMINATION OF CALCIUM I N HUMAN SALIVA 561 The negligible interference by phosphorus in the form of orthophosphate was confirmed by the following galvanometer deflections obtained on solutions, each containing 100 pg of orthophosphate per ml in addition to calcium- Calcium concentration, pg per ml .. 0 2 4 6 8 10 Galvanometer defection . . . . . . 0 17 30 45 63 78 CALCIUM IN WHOLE SALIVA- The procedure for the determination of calcium in whole saliva was tested on spat saliva collected from a number of students. This was carried out 2 hours after the previous meal, no brushing of teeth, rinsing of mouth, smoking or any other treatment having occurred in the intervening period. Results were obtained on saliva collected on each of 4 different days, and the calibration curve obtained as above for pure calcium-containing solutions was used. The results quoted in Table I were obtained on 1 in 25 dilutions of saliva in de-ionised water.It was ascertained several times that 1 in 10 dilutions gave identical figures for the calcium content of the whole saliva. On the other hand, direct determinations on undiluted saliva are not feasible, as the concentrations of calcium are then beyond the range of the flame spectrophotometer when used with the acetylene - air flame. TABLE I CALCIUM CONTENT OF WHOLE SALIVA BY THE METHOD DESCRIBED Calcium content, pg per ml of whole saliva s tutlcnt 1 3 4 5 6 7 8 9 1 0 > I Day 1 50 50 54 50 49 50 57 56 50 47 Day 2 50 50 53 50 49 50 57 56 49 48 Day 3 49.5 50 54 50 50 51 58 55 49 46 Day 4 50 50 54 50 49 50 57 57 51 48 The highei- dilution for the saliva (1 in 25) was selected for the investigation as it gave saliva solutions with calcium content nearer the recommended working range of the SP900 flame spec t rophot omet er .Readings taken on each side of the region of calcium emission, i.e., at 4180 and 4280 A, gave zero galvanometer deflections for both 1 in 25 and 1 in 10 diluted samples. This indicates the absence of direct sodium interference. Apart from the determination of calcium on ordinary diluted samples, the saliva taken on Day 1 was also de-proteinised and treated in the manner described above. The apparent calcium content obtained in this way corresponded exactly with the values quoted in Table I for this day. A blank determination on the trichloroacetic acid treated in an analogous manner gave a zero content for calcium. However, a solution containing 50 pg of calcium per ml prepared from the stock solution and treated in the same way as the saliva gave a reading for the calcium content corresponding to 50 pg per ml.In a selected number of instances, 1 in 25 diluted solutions of saliva were prepared to contain 150 pg of added orthophosphate per ml. On no occasion was there any significant difference in the apparent calcium content over that observed for the normal uncontaminated diluted solutions. Determination of the calcium content of a series of samples of saliva by the proposed method and by the classical oxalate - permanganate titration method indicated a favourable measure o f agreement between the two methods- Proposed method (calcium in pg per ml) . . . . . . 47 52-5 55 58 42 42 Oxalatv - pcrmanganate method (calcium in pg per ml) .. 45 53 53.5 57.5 42 40.5562 JONES AND THOMAS RECOVERY OF CALCIUM ADDED TO WHOLE SALIVA- Statistical experiments were designed to determine the standard deviation of percentage recoveries of added calcium to whole saliva. For a series of 20 samples, the standard deviation was found to be +1.62 per cent., and for a further series of 100 samples, +la50 per cent., the added calcium ranging from 25 to 100 p.p.m. for each series. DISCUSSION Flame spectrophotometry used in conjunction with the acetylene - air flame has revealed a quick method for the determination of calcium in saliva. Reproducibility is excellent, provided an adequate warm-up period has been allowed for the instrument. The negligible effect of added orthophosphate, even in considerable excess, on the flame- spectrophotometric response for standard calcium solutions suggests that the acetylene - air flame is effective in dissociating refractory compounds of calcium, and it indicates that no special precautions need to be taken to overcome possible interference by phosphorus.This is confirmed by the reproducibility of the calcium readings on saliva, even in the presence of orthophosphate. The consistency of the results of untreated whole saliva and de-proteinised whole saliva, the good recoveries obtained for calcium added to whole saliva, and the zero galvanometer deflections each side of the calcium emission, all provide further confirmation that the method is apparently free from interference effects. A propane - air flame can cope with larger concentrations of calcium, but, as indicated in the opening remarks, interference by phosphorus creates a problem so that the advantages of being able to use undiluted saliva are lost. The disadvantages of the time consumed in preparing diluted samples is greatly outweighed by the advantage of overcoming interference effects. A final point calling for comment is that dilution of saliva at the different levels of 1 in 10 and 1 in 25 produces no difference in the apparent calcium content. This is contrary to what might have been expected in the light of Newburn’s remarks2 to the effect that dilution with water reduces the apparent calcium content in the determination by atomic-absorption spectroscopy. However, there is no indication of Newburn having used an acetylene - air flame. We thank the authorities of the Unilever Research Laboratory, Isleworth, for suggesting the project. REFERENCES 1. hlalinowski, J., Cornvvtissaviat 2. Newburn, E., Nature, 1961, 192, 1182. 3. Schmidt-Nielson, B., Acta Physiol. Scand., 1946, 9, 162. 4. L’Energie Atomique (France), Iiappovt IL’o. 1699, Centre cl’Etuclcs Nucleaires de Saclay, Gif-sur-Yvette, 1960. Varley, H., “Practical Clinical Biochemistry,” Heinemann, London, Third Edition (Reprinted), Iieceived February 1 lth, 1966 1964, p. 360.

ISSN:0003-2654    

DOI:10.1039/AN9669100559

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

Analyst, September, 1966, Vol. 91, j5p. 563-566 563 Colorimetric Determination of Sodium Isethionate Means of Ammonium Ceric Nitrate BY D. W. G. DICKER* AND T. H. NEWLOVE ( Unilever Research Laboratory, Port Sunlight, Cheshire) A direct colorimetric method is described for determining sodium isethionate (2-hydroxyethane sulphonate) in aqueous solution by means of its red complex with ammonium ceric nitrate. Ethylene glycol, if present, is removed from the sample by extraction with ethyl acetate before colori- metric determination. The over-all precision of the method is within kl per cent. SODIUM 2-hydroxyethane sulphonate (isethionate) is manufactured commercially by the addition of sodium bisulphite to ethylene oxidel and is used for the synthesis of surface- active agents by esterification with fatty acids.No method of analysis appears to have been published for this compound, and it is our object to describe a suitable direct colorimetric method. This method is based on the use of ammonium ceric nitrate (ammonium hexa- nitratocerate (NH,),Ce(NO,),) for the colorimetric determination of alcohols. The formation of coloured complexes of alcohols and ammonium ceric nitrate was first reported in 1940 by Duke and Smith2 as a qualitative test for the alcoholic hydroxyl group. Later, Duke3 described how the reaction could be used for determining relatively small amounts of alcohol in mixtures with approximately 5 per cent. accuracy. In 1952, Reid and Truelove4 reported the use of ammonium ceric nitrate for the precise determination of alcohols in dilute aqueous solution.A further paper by Reid and Salmon5 described a modified procedure for trace amounts (up to 0.1 per cent. w/w) of alcohols, and this is the basis of the method described in the present report. It may be noted that the ammonium ceric nitrate reagent has also been shown to give colour reactions with certain thiophen derivatives and other heterocyclic compounds6 and has been used for the determination of 1-chloropropan-2-01.~ MECHANISM OF REACTION- A study of the mechanism of the reaction was made by Duke and Smith.2 These workers carried out potentiometric titrations of aqueous potassium ceric nitrate and of ammonium ceric nitrate in absolute ethanol with aqueous and absolute ethanolic standard solutions of sodium hydroxide, respectively. They deduced that the basis of the test was the formation of the red complex anion according to the equation- [Ce(NO,),]- + ROH = [Ce(OR)(NO,),]- + H+ + NO3- They also found that alcohols with up to 10 carbon atoms were responsive to the test.METHOD REAGENTS- of standardised 4 N nitric acid. sintered-glass funnel. solution, and adjust the strength to 0.36 N, if necessary. A cetoTze-Re-distil from anhydrous calcium chloride. Ethyl acetate. All the reagents should be of the highest purity obtainable. Spectrophotometer-A suitable instrument is the Unicam SP600. Ammo~zium ceric nitrate reagent4-Dissolve 20 g of pure ammonium ceric nitrate in 100 ml Allow to stand overnight and then filter through a No. 4 Standardise by titration with 0.1 N ammonium ferrous sulphate APPARATUS- * Present address : Unilever House, Blackfriars, London, E.C.4.564 DICKER AND NEWLOVE COLORIMETRIC DETERMINATION OF SODIUM [Af~aZyst, VOl.91 PROCEDURE- Accurately weigh about 4 g of the solution containing approximately 50 per cent. w/w of sodium isethionate into an evaporating dish and add 5 ml of dry acetone. Place the dish on the steam-bath and gently evaporate the acetone. Add a further 5ml of acetone and evaporate again while stirring with a glass rod, taking care to avoid loss of any solid material by spitting of the solution. Remove the dish from the steam-bath and allow the solid residue to cool. Extract the residue by stirring with 10 ml of ethyl acetate and filter the suspension through a sintered-glass funnel (No. 4 porosity) with suction.Transfer quantitatively any residue from the dish to the filter with three 5-ml portions of ethyl acetate and then wash the sinter with two 5-ml portions of this solvent. Continue air suction for about 5 minutes to ensure that the solvent has been removed. Wash the glass dish with warm distilled water (at a temperature of about 40" C) and pass the solution through the funnel into a clean Buchner flask. Use about 200 ml of warm water in this way and finally transfer the solution to a 500-ml calibrated flask. Wash the Buchner flask four times with 25-ml portions of distilled water, and transfer these to the calibrated flask. Allow to cool, then make up to volume with distilled water. Transfer by pipette 5 ml of this solution (containing about 0.02 g of sodium isethionate) into a small flask and add by pipette (with a safety bulb) 2 ml of the ammonium ceric nitrate reagent.Mix the solutions well and, after exactly 5 minutes, read the optical density of the resultant solution at 486 mp in a 1-cm cell, with 5 ml of distilled water and 2 ml of the ammonium ceric nitrate reagent in the reference cell. Read off the amount of sodium isethionate in the sample from a previously prepared calibration graph (which should be linear and should pass through the origin) relating optical density to concentration. Alternatively, calculate the amount of isethionate by multiplying the optical density by a factor given by the slope of the line. Discard the filtrate. DISCUSSION OF RESULTS SPECTROPHOTOMETRIC EXAMINATION- Spectrophotometric studies of the red anionic complex formed by sodium isethionate and the reagent were made with a pure sample of sodium isethionate, that had been re- crystallised 4 times from acetic acid to a constant melting-point of 196" to 198" C, and a Cary model 14 (ultraviolet and visible) recording spectrophotometer.The absorption spectrum of the red complex was measured over the range 400 to 600 mp, during the period of 4.5 to 6 minutes after mixing the isethionate solution and the reagent, the reagent mixed with water (in place of the sodium isethionate) being used in the reference cell. The spectrum showed a fairly sharp peak at 448 mp. Reid and S a l m ~ n , ~ however, specified 486 mp as the wavelength for measurement of simple aliphatic alcohol complexes with the reagent.Further tests with the reagent alone against water in the reference cell showed rapidly increasing optical absorption with decreasing wavelength in the region of 486 to 448 mp. By using the SP600 spectrophotometer recommended for the procedure, results could not be obtained at wavelengths below 460mp because of the high absorption of the blank. The wavelength 486 mp was therefore chosen for our measurements ; subsequent absorption measurements were made with a Unicam SP600 spectrophotometer. THE EFFECT OF TIME ON THE OPTICAL DENSITY OF THE COMPLEX- The colour of the complex formed by sodium isethionate solution (0.25 per cent. w/v) and the reagent fades linearly with time at a rate corresponding to a reduction of 0.002 in optical density per minute.Provided that the readings are taken within t-1 minute of a given time interval after mixing the solution of the sample and the reagent (5 minutes is the recommended time), the variation in optical density may be neglected. THE EFFECT OF TEMPERATURE ON THE OPTICAL DENSITY OF THE COMPLEX- A rise in temperature from 10" to 30" C had the effect of raising the optical density of the complex (with 0.30 per cent. w/v sodium isethionate solution) from 0.423 to 0.440, but with a further rise from 30" to 40" C, the optical density fell from 0.440 to 0.407, showing instability of the complex.September, 19661 ISETHIONXTE BY MEANS OF AMMONIUM CERIC NITKATE 565 REAGENT STABILITY- Tests were made with 5-ml portions of 0.25 per cent. w/v solution of sodium isethionate and 2 ml of ammonium ceric nitrate reagent after the reagent had been allowed to age.For up to 67 days there was no significant change in the optical density observed. TNTERFEKING SVBSTAXCES- Tests were made to determine whether certain substances that are liable to be present in commercial sodium isethionate would interfere with the determination. The comparisons which follow refer to the standard 0.4 per cent. w/v strength recommended for sodium isethionate solution, although isethionate was absent from the test solutions. Sodium szdPlznte-The effect was negligible even at the same weight concentration as the sodium isethionate. At higher concentrations there is an effect, but this is of no practical significance as, in an isethionate sample of the type envisaged, sulphate would approximate to 1 per cent.of the isethionate content. Sodium chloride-This substance interfered, but the effect was approximately 1/20th of that of sodium isethionate at the same weight concentration. As the level of chloride present in commercial sodium isethionate would be about 0.05 per cent., the effect is negligible. Sodiztm sztlphite-This substance caused a reduction in optical density by bleaching. I t was found that 0.0002 g of sodium sulphite in 5 ml of solution reduced the optical density by 0.012, which is negligible. This amount of sodium sulphite would correspond to 0.5 per cent, in the sodium isethionate sample when using the specified amount of the latter for test. Hence sodium sulphite can be tolerated up to at least 0.5 per cent.Ethylem gZycoZ-This substance produced optical densities approximately twice those given by sodium isethionate solutions at the same level of concentration. As approximately 1 per cent. of glycol is generally found in commercial sodium isethiona te, a procedure for its removal before determining isethionate was sought. It was found that extraction of the total solid material in the isethionate sample with ethyl acetate was effective in removing the glycol, and was accordingly adopted in the recommended procedure. CALIBRATION GRAPH- A calibration graph was prepared by treating 5-ml samples of standard solutions of pure re-crystallised sodium isethionate with 2-ml portions of the reagent, according to the standard test procedure. The resulting optical densities are given in Table I.TABLE I oPTICAI, DENSITIES OF STANDARD SOLUTIONS OF SODIUM ISETHIONATE \freight of sodium iscthionate in 5-ml sample, mg 5 10 15 20 25 30 Optical density ( I a n cell), 486 mp* 0.158 0.296 0.438 0.580 0.721 0.858 * Measured n-ith the SP600 spectrophotometer. The above results fall on a straight lineand, within thelimitsof accuracy of the instrument, this line passes through the origin. ACCUKACY AND PRECISION- The accuracy and precision of the method were measured by subjecting to the recom- mended procedure mixtures (all of which were made up to 4-0 g with water) containing a known amount of pure sodium isethionate, together with various amounts of sodium sulphate, sodium chloride, sodium sulphite and ethylene glycol. The results are given below. No.of 7-- c-----7 {-----+--.JL-_-_- -7 determinations taken, g found, g absolute, g relative, per cent Sodium isethionate Standard deviation 11 2.00 1.99 (mean) 0.015 0.75566 DICKER AND NEWLOVE The relative precision, as defined by standard deviation, is within '.1 per cent. With amounts normally of about 50 per cent. w/w of sodium isethionate on 4 g of test solution, the standard deviation would be 0-38 per cent. and the 95 per cent. confidence limits +O-84 per cent. It may be noted that in the solutions tested, ethylene glycol was present up to 3.75 per cent. and the recoveries of sodium isethionate were still satisfactory. COMPARATIVE TESTS- The ammonium ceric nitrate colorimetric method described in this paper was applied to three manufactured samples of sodium isethionate (of unknown composition) and the values thereby obtained for sodium isethionate content were compared with the respective values calculated by subtraction of the separately determined sodium sulphate, sodium chloride, sodium sulphite and ethylene glycol from the total solid content in each sample.The results are given in Table 11 and show satisfactory agreement. TABLE I1 COMPARISON OF RESULTS BY THE COLORIMETRIC AND WEIGHT DIFFERENCE METHODS Sodium isethionate, per cent., found by- Sample Colorimetric mcthod Weight difference method 1 51.7 2 57.6 3 59.5 52- 1 58.0 59.2 CONCLUSIONS A rapid colorimetric method for the determination of sodium isethionate in commercial samples of this substance has been described, and has been shown to give results that are both accurate and of good precision. We thank the Directors of Unilever Ltd. for permission to publish this work, and acknowledge the assistance of Mr. C. J. Goodwin in performing the analyses. REFERENCES 1. 2 . 3. Duke, F. R., Ibid., 1946, 17, 572. 4. 5. Reid, V. W., and Salmon, D. G., Ibid., 1955, 80, 704. 6. Hartough, H. D., Analyt. Chem., 1951, 23, 1128. 7. Hi-ivnBC, M., Z. analyt. Chem., 1962, 189 (2), 219. Lauer, IT:. M., and Hill, A., J . Amer. Chem. SOC., 1936, 58, 1873. Duke, F. R., and Smith, G. F., Ind. Engng Chem., Analyt. Edn, 1940, 12, 201. Reid, V. W., and Truelove, R. K., Analyst, 1952, 77, 325. Received January 4th, 1966

ISSN:0003-2654    

DOI:10.1039/AN9669100563

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

,-lnalyst, September, 1966, Vol. 91, p p . 567-570 567 The Spectrophotometric Determination of Vitamin D in Fresh-water Fish Liver Oils BY R. K. BARUA AND M. V. K. RAO (Department of Chemistry, Gauhati Univevsity, Jalukbari, A ssnnz, India) There are many difficulties associated with the determination of vita- min D, especially in natural products such as fish liver oils. Vitamin A is the chief interfering material; it masks the absorption of vitamin D both in the ultraviolet region and in the antimony trichloride colour test, making the determination of vitamin D almost impossible. In addition to vitamin A,, fresh-water fish liver oils contain vitamin A,, which also interferes in direct spectrophotometry. A method for determining vitamin D is described, in which vitamins A, and A, are eliminated by converting them to anhydro- vitamins A, and A, and separating them from vitamin 11 by chromatography on an alumina column.A METHOD for determining vitamin D in fresh-water fish liver oils is described, in which both vitamins A, and A, are eliminated by converting them to the anhydrovitamins A, and A,, by treatment of the unsaponifiable material from the liver oils with 0-033 N ethanolic hydrogen chloride. The light petroleum extract of the resulting product is chromatographed by placing it on a column of weakened alumina containing 8 per cent. w/v of water. On developing the chromatogram with light petroleum, both anhydrovitamins A, and A, flow through the column. Three distinct yellow bands develop on the column; the colourless portion between the second and third yellow zones contains all the vitamin D.The colourless portion is extruded and eluted with diethyl ether, the eluate is evaporated to dryness under reduced pressure, and the colour developed with antimony trichloride - acetyl chloride reagent is measured within 30 seconds of adding the reagent. Over 90 per cent. of the vitamin D is recovered. The determination of vitamin I> in natural products like fish liver oils is beset with many difficulties. The presence of various interfering materials such as vitamins A,, A, and their congeners, sterols and provitamins D, free fatty acids, phospholipids and wax alcohols make the determination of vitamin D by the spectroscopic method almost impossible. Of these, vitamin A seems to be the chief interfering material as it masks vitamin D in both the ultraviolet absorption and colour tests.Vitamin A,, in the liver oils of fresh-water fish predominates over vitamin A,, from both European, and Indian, waters. Vitamin A, has an absorption maximum at 351 mp and a subsidiary maximum at 286 mp with an inflection at 276 mp. In fresh-water fish liver oils the presence of vitamin A, will therefore cause an additional interference in the spectrophoto- metric determination of vitamin D, which absorbs at 265 mp. Although fresh-water fish liver oils have a preponderance of vitamin A, over vitamin A,, it has been found that vitamin A, is invariably present in most of these oils in varying proportions3 and, as such, these oils present a problem for the determination of vitamin D after elimination of both vitamin A, and A,.Vitamin A, is a highly labile compound. Gillam et aL4 reported that vitamin A, decomposed when absorbed on alumina giving rise to a considerable amount of material that exhibited a strong 650-mp band in the antimony trichloride test. On treatment with ethanolic hydrogen chloride, vitamin A, yields ethoxy anhydrovitamin A2,536 which is more strongly adsorbed than anhydrovitamin A, on columns of alumina. In an earlier paper,’ we described a method for determining vitamin D in the presence of vitamin A, by converting vitamin A, to anhydrovitamin A, and separating it from vita- min D by chromatography on a column of alumina (weakened by addition of 8 per cent. w/v568 BARUA AND RAO : SPECTROPHOTOMETKIC DETERMINATION jL4naZyst, Vol.9 1 of water). In the present paper, a method is described for determining vitamin 1) in the presence of both vitamins A, and A,, by converting these two vitamins into their anhydro products, and separating them from vitamin D by chromatography. The anhydrovitamins A, and A, will flow down the column, and all of the vitamin D will be located during develop- ment of the chromatograms in a colourless zone between two yellow zones at the top of the column. The colourless zone is extruded and vitamin r> is eluted and determined by the Zimmerli - Nield - Russel antimony trichloride reagent. METHOD APPAKATU S- photometer. The spectrophotometric determinations were performed on a Beckman DK-2 spectro- REAGENTS- Calciferol-B.P. grade.Other reagents (alcohol, light petroleum, diethyl ether, anhydrous sodium sulphate, antimony trichloride reagent, alcoholic potassium hydroxide, weakened alumina and ethanolic hydrogen chloride) should be purified or prepared by the procedures described previously. EXTRACTION OF LIVER OIL- The liver oils were obtained from the livers of four fresh-water fishes, viz., WaZZagzt attu, Hagarius bagarim, Mystas seenghala and Silonia siiondia. Mix the livers with acid-washed silver sand and anhydrous sodium sulphate in a glass mortar. Extract the thoroughly ground material repeatedly with light petroleum until the extract gives no blue or green colour with antimony trichloride reagent. Then combine the extracts and dry over anhydrous sodium sulphate. Remove the solvent under reduced pressure and store the oils obtained in amber coloured bottles in a refrigerator.Determine the vitamin A, and A, by the method described by Cama and Morton.* The results for these oils are given in Table I. TABLE I VITAMIN A, AND A, CONTENT OF FKESH-WATER FISH LIVER OILS Vitamin A,, Vitamin A,, i.u. per g i.u. per g Bagarius bagarizts . . . . 4000 143,200 Wallagu attu . . . . 19,550 113,000 Mystus seenghala . . . . 11,420 41,730 Silonia silondia . . .. 2436 2 13,200 DETERMINATION OF VITAMIN D- SaponiJcatiozz-Take a convenient weight of the oil which contains not less than 300 i.u. of vitamin D and saponify with freshly prepared potassium hydroxide solution under a slow stream of nitrogen for 10 to 15 minutes. Maintain a ratio of 2 6 g of potassium hydroxide to 1 g of oil.Dilute the ethanolic soap solution with an equal volume of water and extract 4 times with suitable volumes of light petroleum. Combine the extracts and wash them with water until the washings are neutral to phenolphthalein. Dry the solution over anhydrous sodium sulphate. Evaporate the solution (a known volume) to dryness and treat the residue with a convenient volume, 10 to 40 ml, of ethanolic hydrogen chloride for 40 minutes under an inert atmosphere. Neutralise the excess of acid with the least amount of solid sodium hydrogen carbonate. Extract the solution with petroleum as described above. W7ash the extract free from alkali and dry it over anhydrous sodium sulphate. ChromatogYaj5hy-Evaporate the light petroleum solution to a small volume (1 to 2 ml) and transfer to a chromatographic column (1 x 9cm) packed with alumina weakened by the addition of 8 per cent.of water. Develop the chromatogram with light petroleum. Maintain the rate of flow at approximately 2 ml per minute. Anhydrovitamin A, will flow out in the first 25 ml of the eluate. Anhydrovitamin A, will flow out in the next 40 to 50 mlSeptember, 19661 OF VITAMIN D IN FRESH-WATER FISH LIVER OILS 569 of the eluate. A t this stage several yellow bands can be observed on the column. Further development of the column with about 100ml of the solvent elutes all the yellow bands except 3, and a colourless portion between the second and third yellow bands. Described from the top of the column they will be- (;) A thin yellow band. This substance exhibits a broad absorption band at 318 to 330 mp and a subsidiary band at 271 mp.The antimony trichloride reaction product exhibits absorption maximum at 650 mp and at 580 mp. (ii) A relatively wide yellow band about 3 to 4 mm below the first. This material has values for A,,,, of 335, 349, 368 and 390 mp. The antimony trichloride reaction product has an absorption band at 650mp. (iii) A colourless portion, 4 to 5 cm long, containing vitamin I). (iv) A wide yellow band about 4 to 5 cm below the second yellow band. This material shows a broad absorption band at 325 mp, with inflections at 300, 310, 347 and 367 mp. The antimony trichloride reaction product of this material has an absorption band at 650mp. From a set of preliminary experiments it has been found that all the vitamin D was contained in the colourless portion (iii) of the column.EXTRACTION AND DETERMINATION OF VITAMIN D- Extrude the column, remove the upper half of the colourless portion between the second and third yellow bands and elute it with two 15-ml portions of diethyl ether. Evaporate the eluate to dryness under nitrogen, and dissolve the residue in a convenient volume of light petroleum (10 ml). Evaporate a 5-ml portion of this to dryness under nitrogen in a small flask. Add 4 ml of antimony trichloride reagent to the residue and gently swirl the flask for 30 seconds and transfer the solution quickly to an optical cell. Measure the extinction at 500 mp. Calculate the amount of vitamin D present, assuming Ei:i, value at 500 mp for pure vitamin D to be 1800.RESULTS Table 11 shows the results of the determination of vitamin I> in fresh-water fish liver oils by the above method. TABLE 11 VITAMIN I) IN FRESH-WATER FISH LIVER OILS Name of fish Bagarius bagarius W a l l a g ~ attu . . Mystus seenghala Siloiiia silondia Weight of oil, f3 . . 2.0670 2.1210 2.0240 . . 0.3421 0.3562 0.36 1 I .. 0-5426 0-5212 0.5365 6500rn1.~ 0.34 0.34 0.32 0.65 0.67 0.69 0.58 0.55 0.57 El% ~ c m 9 at 500m~ 0-01 3 17 0.01 31 3 0*01265 0.1521 0.1505 0.1529 0-08551 0.08443 0.08504 . . 0.3741 0.68 0.1455 0.3561 0.64 0.1438 0.3420 0.6 1 0.1427 Vitamin D present, i.u. per g 292.7 291.8 281.1 3379 3345 3398 1901 1876 1900 3232 3197 3171 RECOVERY EXPERIMENTS- To assess the accuracy of the method described for determining vitamin I) in the fish liver oils, recovery experiments were carried out after adding vitamin D to these oils, as recovery experiments are considered the most critical proof that vitamin I) is, in fact, being measured.Measured volumes of an ethanolic solution of calciferol of known strength were added to different weights of these oils and the method outlined above was followed to determine vitamin D. The recoveries of total vitamin D are shown in Table 111.570 BARUA AND RAO TABLE I11 RECOVERIES WITH ADDED CALCIFEROL TO THE LIVER OILS Name of fish Bagarius bagariws Wallagti attu . . Mystus seenghala Silonia silondia Weight of oil, g . . 0.5112 0.5235 0.6702 0.7104 0-1477 0-1548 . . 0-4121 0.3802 0- 1244 0.1652 . . 0,3132 0.3201 0-4649 0.4456 . . 0.2124 0.2090 0.3462 0.3451 Vitamin D, i.u.per g r present 288 288 288 288 288 288 3374 3374 3374 3374 1892 1892 1892 1892 3200 3200 3200 3200 added 982 959 700 985 2993 2842 1635 1660 13,460 5501 2618 2624 1161 1212 4341 4412 1863 1869 total 1270 1247 988 1273 3281 3130 4909 5034 16,834 8875 4510 4516 3053 3104 7541 7612 5063 5069 1 found 1201 1207 995 1221 2943 2872 4680 4325 15,940 8878 4302 4228 2885 2981 7001 7516 4790 4837 Recovery, per cent. 94.5 96.8 100.7 95.9 89.7 91.7 95.3 86.9 94.7 100.0 95.4 96.6 94.5 96.0 92.8 96-5 94.6 95.4 The recoveries of total vitamin D were usually over 90 per cent. and show the reliability of this method for the determination of vitamin D in fish liver oils. DISCUSSION In all of these experiments in the determination of vitamin D in different fish liver oils, the chromatograms were found to be similar, with the three yellow zones and a wide colourless portion between the second and the third yellow cones.In all instances, only the eluates from the upper half of the colourless portion showed an absorption band at 500 mp when treated with antimony trichloride reagent. The intensity of this band remained constant even after 5 minutes, but a weak second absorption band developed at 420 mp (406 mp with bagarius bagarius liver oil) as shown by the spectra recorded 5 minutes after adding the reagent. With antimony trichloride reagent, a weak band at 650 m p also appeared, the intensity of which increased with time. Both of these bands made little contribution at 500 mp at which wavelength the extinctions for vitamin I> were measured. The 420 mp band was of very low intensity and initially absent 30 seconds after adding the antimony trichloride reagent.The 420 mp band may be due to sterols and, as these have little effe~t,~,lO the extinction at 500mp gives a measure of the vitamin L) present if all extinctions are measured within 30 seconds after adding the antimony trichloride reagent. In these investigations no attempt has been made to determine the bio-potency of the oils, but from the results of the recovery experiments, and also from the analysis of the fresh-water fish liver oils that gave reproducible results, it may be suggested that this method can be used for the determination of vitamin D in fish liver oils. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. REFERENCES Lederer, E., and Rosanova, V. A., Biokhirniya, 1937, 2, 293. Balasundaram, S., Cama, H. R., Sundaresan, P. R., and Varma, T. N. R., Biochem. J., 1956, 64, 150. Morton, R. A., “The Application of Absorption Spectra to the Study of Vitamins, Hormones and Coenzymes,” Second Edition, Adam Hilger Ltd., London, 1942, p. 78. Gillam, A. E., and Heilbron, I. M., Jones, W. E. and Lederer, E., Biochew. J., 1938, 32, 405. Shantz, E. M., J . Biol. Chew., 1950, 182, 515. Cama, H. R., Dalvi, P. D., Morton, R. A., and Salah, M., Biochem. J., 1952, 52, 540. Barua, R. K., and Rao, M. V. K., Analyst, 1964, 89, 534. Cama, H. R., and Morton, R. A., Ibid., 1953, 78, 74. Zimmerli, A., Nield, C. H., and Russel, W. C., J . Biol. Chew., 1943, 148, 245. Mueller, A., J . Anzer. Chew. Soc., 1949, 71, 924. Received April 25fh, 1966

ISSN:0003-2654    

DOI:10.1039/AN9669100567

出版商:RSC    

年代:1966

数据来源: RSC

摘要:

,4nalvst, September, 1966, Vol. 91, p p . 571-575 57 I The Gas -chromatographic Analysis of Gases Extracted from Metals by Vacuum Fusion BY M. T. LILBURNE (12IetnlZurgy Divisiom, National Physical Laboratory, Teddington, Middlesex) The equipment necessary to use a micro-ionisation detector of the Lovelock type with helium carrier gas is described. Although not used at its highest sensitivity, the apparatus can measure quantitatively 1 per cent. v/v of components in a mixture of carbon monoxide, nitrogen, hydrogen and methane of total volume 5 Y cm3 at S.T.P. THE purpose of this paper is to describe the application of gas chromatography and Lovelock ionisation detectors1 to the analysis of gas mixtures obtained from vacuum fusion experiments on metals. A typical gas mixture extracted by the vacuum fusion procedure contains carbon monoxide (from oxygen in the metal), nitrogen, hydrogen and, occasionally, methane.With present-day trends towards high purity metals and smaller samples being made available to the analyst, existing methods of gas analysis have been found to be inadequate. These methods remove the reactive gases from the mixture in succession, either chemically, (e.g., by absorption of carbon monoxide in ammoniacal cuprous chloride solution) or physically, (e.g., by diffusion of hydrogen out of the system through a palladium thimble), and the gas remaining is considered to be nitrogen. As a result, the determination of nitrogen is not as precise as the determination of oxygen and hydrogen. Moreover, in order to determine oxygen, a commonly used physical method involves oxidation of the carbon monoxide in the mixture to carbon dioxide by exposure to hopcalite, the resulting carbon dioxide then being frozen out in a cold trap.The hopcalite, however, absorbs small amounts of hydrogen, and, as a result, the oxygen determination can be in error. This error is not readily apparent until the total gas content in the sample falls below 10 p.p.m. w/w. An obvious way of solving this problem is to remove the hydrogen first, but it is not always convenient to do so. The sensitivity of the chemical methods of analysis is such that 3 x 10-1 cm3 a t S.T.P. of the mixture is required in order that a 1 per cent. v/v component may be accurately deter- mined; for physical methods 3 x 10-2cm3 at S.T.P.is required. Several workers2.3y4 have used gas chromatography to separate the gases in the mixture and calibrated thermal-conductivity detectors for their determination. As the gases have been separated before reaching the detector, the precision of determination depends only on the response of the detector for the particular gas and the subsequent recording and integrating systems. Because it is no longer obtained by difference the nitrogen figure is more reliable. However, when using a thermal-conductivity detector, a choice has to be made between high sensitivity for carbon monoxide and nitrogen, and low sensitivity for hydrogen, or vice zwsa. The former is obtained when helium is used as carrier gas, and the latter when argon is used. The thermal-conductivity detector acts as a balanced device comparing heat transfer in the pure carrier gas with that in the carrier gas plus sample, and hence for high sensitivity, a requirement is for the thermal conductivities of the carrier and sample gases to be quite different.As can be seen from Table I, if the gas mixture contains carbon monoxide, nitrogen and hydrogen, it is impossible to obtain simultaneously a high sensitivity for these three gases. Of the two choices of carrier gas available, helium TABLE I THERMAL CONDUCTIVITIES OF SOME GASES Gas Helium . . . . . . . . Carbon monoxide . . . . Methane . . .. . . Hydrogen . . I . . . Argon . . . . . . . . Nitrogen . . . . . . Thermal conductivity x cal. per second per cm per "C 41.3 34.3 3-89 5.58 6-47 5.81572 LILBURNE : GAS-CHROMATOGRAPHIC ANALYSIS OF [Analyst, Vol.91 is the better because approximately 10 times the volume of hydrogen per p.p.m. by weight in the sample is released compared with the other gases, thus diminishing the effect of low sensitivity of the detector for this gas. The minimum amount of gas required is the same as that for the physical methods, vix., 1 to 3 x 10-2cm3 at S.T.P. of gas mixture. In principle, the Lovelock type of micro-ionisation detector, which has an approximately equal sensitivity to all permanent gases, can be used with helium as carrier gas. The sensitivity of this detector can be as much as 1000 times that of the thermal-conductivity type. The linear range of the ionisation detector is three orders of magnitude, and the maximum concen- tration of permanent gases in the helium carrier gas that can be handled is 1 in lo?.Thus in order to use this detector over its full range, the carrier gas must have a total impurity content of no more than 1 p.p.m. by volume. Water vapour, probably the most difficult impurity to eliminate from a gas system, alters the mode of operation of these detectors thereby producing a disastrous effect on sensitivity and reproducibility. The purification of the carrier gas and the design of the train to introduce the sample mixture are the critical points in the application of ionisation detectors to permanent gases. A description is now given of the apparatus developed at the National Physical Laboratory. EXPERIMENTAL HELIUM PURIFICATION- Berry5 was the first to apply ionisation detectors successfully to the detection of the permanent gases in helium.His success was due to the use of a complex chemical purification train with chilled molecular-sieve traps (- 196" C) and heated titanium and manganese dioxide traps (900" and 350" C, respectively). Initial experiments at the Xational Physical Laboratory were conducted with a system following that of Berry. Although excellent results at high sensitivity were obtained for limited periods, the performance of the system was erratic, and it had a lifetime of only 2 months, presumably because the traps were not large enough. A helium diffusion cell to the design of McAfee (from a private communication) consisting of a large bundle of capillary silica tubes, housed in a steel tube that was capable of with- standing high pressures and temperatures, was then obtained from Electron Technology Inc., Kearny, U.S.A.When the cell operates at a pressure of 700 p.s.i.g. and a temperature of 350" C, a flow-rate of 100 cm3 per minute at 10 p.s.i.g. is obtained on the low pressure side. The only helium flow involved is that which diffuses through the silica barrier. This type of cell allows diffusion through the silica of small amounts of hydrogen and neon in addition to helium, but with the helium supplies commonly available, the concentration of these impurities will never reach the level of 1 p.p.m. by volume. This device has been operating successfully for more than 9 months. A helium flow-rate of 100 cm3 per minute was used. SAMPLING VALVE AND APPARATUS TUBING- Stainless-steel connecting tubing was used throughout, and joints were made with neoprene O-rings that were later replaced by silicone O-rings.The carrier gas exhibited evidence of being impure when both of these types of O-ring were used, probably owing to gas diffusion through the elastomer itself. Finally, all joints were made with compression fittings of the "Swagelok" type with indium ferrules. The ferrules were made from a strip of indium that was wrapped around the tube and placed into the fitting immediately before screwing down the backing nut. Joints made in this manner were found to be more reliable than those made with the stainless-steel ferrule provided. I t was found necessary to outgas the tubing progressively (from the diffusion cell to the detector) after the apparatus had been assembled and with the full helium flow passing through, taking care not to overheat any fitting containing indium.The sampling valve was designed so that the primary gas-tight seal was between stainless-steel and PTFE plates. This valve consisted of two circular plates, one of stainless steel and one of PTFE, whose mating surfaces were polished flat and clamped together via a thrust bearing so that the PTFE plate was free to rotate. The polished surfaces provided an excellent gas-tight joint after a very small application of vacuum grease. The stainless-steel plate contained 4 ports: two, close together, to allow the entry and exit of carrier gas; the third, to allow Any O-rings present were of secondary importance.September, 19661 GASES EXTRACTED FROM METALS BY VACUUM FUSION 573 the sample gas to enter; and the fourth, to enable the valve to be evacuated.The vacuum port is drawn in Fig. 1 (a) and it shows the internal pumping lines necessary to evacuate the annular rings a t the centre and edge of the flat plates. Thus when a collar containing suitably placed O-rings is fixed round the plates, the mating surfaces of the plates can be isolated from the atmosphere. The PTFE plate has three equally-spaced slots each of 2 x cm3 volume. The slots are long enough to connect the gas inlet and exit ports, Fig. 1 ( b ) . In operation, slot 1 is connected to the sample gas contained in a small reservoir at a pressure of about 20 torr; slot 2, to the carrier-gas ports; and slot 3, to a vacuum line.On being rotated in the correct direction through 120°, a volume of 5 x 10-4cm3 a t S.T.P. of sample gas is injected into the carrier stream; an evacuated slot is carried to the reservoir; and the third slot is evacuated. ; be S t a maring .inless-steel plate PTFE plate (4 Collar (b) Fig. 1. The sampling valve: (a), the vacuum port; (b), the PTFE plate with the positions of the ports in the stainless-steel plate shown In a vacuum-fusion experiment the extracted gases are trapped a t low pressure in the Eore-line of a high-speed mercury diffusion pump, and then removed to calibrated volumes by a Toepler pump. After measuring the volume of the gas, it is collected over mercury at atmospheric pressure in a small transfer vessel fitted with a greased vacuum tap.This cnables separate analyses by chemical and gas-chromatographic methods to be undertaken. 5 x cm3 at S.T.P. of this gas mixture is transferred to the small sample reservoir of 1.5 cm3 volume attached to the chromatograph through a modified three-way glass stopcock. The pressure in the sample reservoir is then 20 to 25 torr. The usual “T” bore of the three-way stopcock was replaced by a straight-through bore of 2 mm diameter and 5 x PO-2cm3 volume so that the gas could be trapped across the tap between mercury columns. Thus, when the tap is rotated through 90°, the trapped gas is allowed to enter the reservoir. CHROMATOGRAPH- 120” c. Analytical C0~2L?n?2-L4 column 120 cm long and 0.5 cm in diameter, coiled and used a t574 LILHURNE : GAS-CHROMATOGRAPHIC ANALYSIS OF [ A naLvst, Vol.91 Stationary phase-Molecular-sieve (Linde 5A) powder of 36 to 60 mesh. After crushing, the “fines” were floated off by stirring the powder supported in a tall beaker vigorously with a water jet. The sieve was activated by heating at 400” C overnight in a stream of carrier gas. Detector and power suPpZy-Type I.E. 103B, available from Gas Chromatography Ltd., Maidenhead, England. Integrator-Type 92314 available from Electro Methods Ltd., Stevenage, England. This was driven by a voltage delivered from a transmitting slide-wire on the recorder. I t was found necessary to integrate the area under the peaks because the carbon monoxide peak showed some “tailing.” RE su LTS All other peaks were gaussian in shape.The apparatus was calibrated by injecting, in turn, known amounts of the pure gases required, and noting the response. The sensitivity was found to be highly dependent on the voltage applied to the detector and was more than 60 coulombs per mole at 500 volts. In the present application it was only necessary to operate the detector at 220 volts when the sensitivitv was about 1 coulomb per mole (Fig. 2 ) . So long as the peak height was greater Volume (cm3 x at S.T.P. Fig. 2 . Calibration curves (1 count = 3.6 x coulombs); A, carbon monoxide; R, methane; C, nitrogen; D, hydrogen than 50 per cent. of the scale reading, the integrated area of the peaks obtained was repro- ducible to within 4 per cent. Frequent calibration checks over a period of several months established that there was no variation in sensitivity.Table I1 gives the results of several analyses, chosen at random from a large number, compared with the analyses obtained by chemical means. The time taken for analysis by gas chromatography was 2 minutes and compared favourably with the 20 minutes taken by the chemical method. TABLE I1 COMPARISON OF GAS-CHROMATOGRAPHIC AKD CHEMICAL METHODS Gas chromatography, p.p.m. by weight in sample (--L_- 7 volume used h 5 x lo-* cms a t S.T.P., x Sample Oxygen Hydrogen Nitrogen Nimonic . . . . 194 3.8 17.7 Chromium . . . . 18.0 1.0 6.2 Pure iron . . . . 16.1 0.2 7.6 Chromium flake . . 56-0 1.3 14.7 Chromium flake . . 16.2 0.8 18.4 Iron-manganese . . 7 . 2 0.4 3.9 Chemical method, p.p.m. by weight in sample r-.p--L - Oxygen Hydrogen Nitrogen 185 4.9 18.6 18.6 0.8 7.3 15-9 0.2 7.6 57.3 1-3 13-3 15-2 0.8 19.5 6.6 0.3 4.0 yolumc used 2 2 x 10-1 cm3 at S.T.P., 7September, 19661 GASES EXTKACTEI) FKOM METALS BY VACUUM FUSION 575 I)ISCUSSION -4ND CONCLUSIONS The speed and accuracy of gas analysis by gas chromatography have been demonstrated particularly for samples having low gas contents.The result obtained by this method of analysis is superior to those obtained by chemical and other physical methods, in that a direct determination of nitrogen is made, and it is therefore of higher accuracy. The sensitivity of the method is not as great as that obtainable by using mass-spectrometric methods but can, if necessary, be increased by a factor of more than 10. On the other hand, these latter methods have to use an indirect approach as both carbon monoxide and nitrogen have an atomic mass of 28. I thank Dr. K. B. McAfee, jun., Bell Telephone Laboratories, Sew Jersey, for much helpful correspondence and information freely given. Acknowledgment is also given to Mr. H. G. Short for helpful discussion. The work described in this paper has been conducted as part of the research programme of the National Physical Laboratory and is published by permission o f the Director. REFERENCES 1. 2. 3. 1. 5 . Lovelock, J . E., =inulyt. Clieun., 1961, 33, 162. Feightinger, H., Raechtold, H., and Brauner, K., Schweizer Awlr. nizgcw. Wiss. Tech., 1963, 28, 125. Suzuki, S., Bunseki Kugaku, [Japan A~zaZvst], 1962, 11, 618. Tyou, P., and Hans, A., Revue Me‘tall., Paris, 1961, 58, 187. Berry, R., in van Swaay, M., Editor, “Gas Chromatography 1962,’’ Ihitterworth and Co. (Pub- Receivcd Jmzuary loth, 1966 lishers) Ltd., Laondon, 1962, p. 321.

ISSN:0003-2654    

DOI:10.1039/AN9669100571

出版商:RSC    

年代:1966

数据来源: RSC