ANALYTICAL CHEMISTRY1. INTRODUCTIONAs in our previous Report, attention has been concentrated on originalpapers rather than abstracts. The division into sections is on the samegeneral lines as last year except that we have decided to combine the Generaland Miscellaneous sections under the one heading of General Analysis. Thesections are therefore as follows : General. Conventional Qualitative andQuantitative Inorganic and Organic Analysis. Physical Methods of Analysisunder the headings : (a) electrical methods; (b) polarography ; (c) chromato-graphy; (a) absorption spectroscopy; (e) emission spectroscopy. Micro-chemical Methods. Radio-chemical Methods. Apparatus.Under each heading we have tended to mention inorganic applicationsfirst, following these by organic applications.In the course of preparation of the Report, we have noted certain matterswhich seem to merit special attention.There is no doubt that there is agreat tendency to place many conventional methods of analysis on anautomatic basis, in particular in order that plant processes may be moreefficiently controlled. The interest in gas-liquid and gas-solid chromato-graphy is being maintained at a high level, the greatest emphasis beingprobably on more sensitive means of detection of separated fractions. Inpaper chromatography we have noted a tendency to change the method ofapplication of the test, whilst in polarography a great deal of work is beingcarried out using alternative electrodes to the conventional droppingmercury electrode. Non-aqueous titrimetry and flame photometry arebecoming increasingly important, whilst there is a great tendency todetermine innumerable elements by final absorptiometric procedures at aparticular wavelength in the visible region. The possibilities of toluene-3,4-dithiol as an analytical reagent are beginning to be realised, whilstanalysts in many fields are finding the flask combustion method of pre-liminary combustion in oxygen to be an excellent and simple method ofopening up organic samples.2.GENERALAs mentioned above, the automatic instrumental analysis of processstreams is becoming more and more common. In this respect the contribu-tion of Bertram and his co-workers1 is significant. They have devised anautomatic instrument composed essentially of a derivative polarographand a rotary-valve proportioning system, which will analyse process streamscontaining high concentrations (100-200 g./l.) of uranium.The uraniumconcentration of the process stream is recorded every 15 minutes with a highorder of accuracy and reproducibility, attributed mainly to the resistance-capacitance method of derivative polarography using a reversible scanningAnalyt. Chem., 1958, 308 354.1 H. W. Bertram, M. W. Lerner, G. J. Petretic, E. S . Roszkowski, and C. J. Rodden390 ANALYTICAL CHEMISTRY.procedure. The principle of an automatic titration unit designed by Brownand Weir,2 which enables them to take a definite volume of ammonia solutionfrom a plant process, to dilute this solution with a selected volume of water,and then to titrate the diluted solution with standard acid to a definite pH,is likely to be of value in other plant control tests.The method adopted bythese authors enables them to record the result and to repeat the sequenceof operations at regular time intervals.Yet another example of automatic analysis is described by Crumley,3who has designed an instrument for the automatic determination of sulphurtrioxide and sulphuric acid vapour in flue gases. The SO, is absorbed inpropan-2-01, and the sulphate precipitated as barium salt which is measuredphotoelectrically. In the range 10-100 p.p.m. the instrument givesresults which are more consistent than those normally obtained by manualdeterminations made by established methods.Also on the subject ofatmosphere analysis, Helwig and Gordon have described an apparatusfor continuous sampling, analysis, and recording of the sulphur dioxidecontent of the air. The method is based on the production of the red-violetcolour caused by reaction of the gas with pararosaniline hydrochloride inacid media.A useful contribution to the several improvements made recently on theWinkler method for the determination of dissolved oxygen in water has beenpresented by Griffiths and Ja~kman.~ In their procedure the iodinetitration is carried out potentiometrically, a glass and platinum electrodepair being used, and the glass electrode is used as reference in a highly acidmedia. The end-point potential difference is reproducible and thus thetitration can easily be put on an automatic and recording basis.Briggset aZ.6 have also developed a novel method for the continuous recordingof the dissolved oxygen content of natural waters and sewage effluents inthe range 0-15.0 p.p.m. by weight of oxygen; the method can be used withwater flowing past the electrodes. The principle of polarographic reductiona t an applied voltage of -0.5 v with respect to a zinc anode is used in thetest. The cathode is quite unconventional, being a capillary of wide internaldiameter, i.e., 0.8 mm., sloping upwards in the solution under test.Chemists concerned with the regular analysis of amino-acids will find themethod of Spackman et a1.' of interest. They have designed an automaticinstrument which records the ninhydrin colour value of the effluent fromion-exchange columns.The eluted material, mixed with a measured flowof ninhydrin reagent, is heated, and the absorbance measured continuouslyat 570 and 440 mp. The instrument will deal with protein and peptidehydrolyzates in 24 hr. and the analysis of blood plamsa, urine, etc., can becompleted in 48 hr. The precision of the method is &3% for the range0.1-3.0 pmoles of each amino-acid separated by the ion-exchange columns.Also constructed for the greater part with well-known commercial instru-2 J. F. Brown and R. J. Weir, Analyst, 1958, 88, 491.4 H. L. Helwig and C. L. Gordon, Analyt. Chem., 1958, 30, 1810.5 V. S. Griffiths and M. I. Jackman, Analyt. Chim. Acta, 1957, 17, 603.6 R.Briggs, G. V. Dyke, and G. Knowles, Analyst, 1968, 83, 304.7 D. H. Spackman, W. H. Stein, and S. Moore, Analyt. Chem., 1958, 30, 1190.P. H. Crumley, J . Inst. Fuel, 1958, 212, 378HASLAM AND SQUIRREL : GENERAL. 391ments, a simple apparatus has been designed by Barrollier et aL8 for thesemi-automatic tracing of the extinction curves of chromatograms andpherograms. A galvanometer light spot reflected from a curved steelmirror is registered as an extinction curve by means of a recorder. The useof this apparatus for quantitative determination of amino-acids by directphotometry of the chromatograms is described.Included in the many papers of general interest is that which describesa new principle put forward by Specker et aL9 for the determination oflithium in alkali chlorides which may have been recovered as such as aresult of various analytical operations.The dried alkali chlorides areextracted with cyclohexanone or acetone, and the lithium in the extracts nowdetermined either photometrically (at 366 mp) or by conductometrictitration with standard copper perchlorate solution. The basis of the testis that, under the experimental conditions, an intense reddish-orangecomplex of the formula Li(CuCl,),nCOR, is produced. Small amounts ofoxide in such substances as lithium nitride and calcium carbide have beendetermined by Juza, Pruff, and Witt.lo The preparation is dissolved inwater-free methanol plus water-free acetic acid, and the water resultingfrom the oxide titrated at 0" with Karl Fischer reagent.Interestingobservations are made on the preparation of water-free methanol, acetic acid,and pyridine.Geilmann l1 has developed methods of determination of certain readilyvolatile elements such as thallium, cadmium, indium, and lead. He hasused the principle to determine " nanograms," i.e., thousandths of amicrogram, of, e.g., lead in copper and filter-paper and thallium in tobacco.The sample is heated in a stream of gas such as hydrogen or nitrogen, andthe volatilised element, e.g., lead or thallium, collected on a cold finger.The cold finger may take the form of an electrode if spectrographic deter-minations are to be carried out subsequently. Alternatively, a quartz coldfinger may be employed if the sublimate is to be dissolved and determinedby chemical procedures.Ion-exchange membranes based on artificialresins l2 are likely to prove of value in analytical separations, e.g., of mag-nesium from aluminium and in preparative chemistry involving thepreparation of complex compounds such as the cobaltammines.The measurement of chloride-ion concentration in dilute solutions( 1 . 0 ~ to ~O+M) has been carried out by Stern et aZ.,I3 and the results expressedas pC1. The measurements are made on a standard pH meter, with asilver chloride indicator electrode and a saturated calomel referenceelectrode. The instrument is standardised by using known chloride solutions(O-~N-KC~ = pC1 1-11; O-O~N-KC~ = pC1 2-04) and measurements can thenbe made on as little as 1 ml. of sample solution.The pC1 can be convertedinto chloride-ion concentration by reference to the appropriate relationship8 J. Barrollier, J. Heilmann, and E. Watzke, J . Chromatog., 1958, 1, 434.9 H. Specker, H. Hartkamp, and E. Jackwerth, 2. analyt. Chem., 1958, 163, 111.10 R. Juza, H. Pruff, and H. Witt, ibid., 159, 277.11 W. Geilmann, ibid., 160, 410.12 E. Blasius and G. Lange, ibid., p . 169.13 M. Stem, H. Shwachman, T. S. Light, and A. J. de Bethune, Analyt. Chem., 1958,30, 1506392 ANALYTICAL CHEMISTRY.curve. The method has been applied to the determination of chloride insweat and no pre-treatment other than dilution of the sample is required.For the turbidimetric determination of small amounts of sulphate,Vosloo and Sampson14 prefer to use a nephelometer rather than anabsorptiometer.In an investigation of the method they conclude that,although the turbidity produced by precipitation as barium sulphate isreproducible, yet it is not strictly proportional to the amount of sulphatepresent, and it is dependent on the volume at which precipitation is carriedout and on the crystal size of the barium chloride used to effect precipitation.The turbidity is, however, independent of temperature and is stable for8 hour. The method, described in detail, appears superior to other formsof the test. The work of Tufts and Lodge l5 on the identification of halideand sulphate in submicron particles using the electron microscope will nodoubt be of use to analysts faced with problems connected with air pollutionand the examination of fine particles.The particles are identified bycharacteristic spot tests after collection on an appropriate substrate, thereagent for chloride being mercurous fluorosilicate and for sulphate a mixtureof barium and lead nitrates. The appearance of the spots resulting fromthe particles, when viewed in the electron microscope, is characteristic andthe size is proportional to the size of the original particles. Hence, as wellas identification, a particle count and a particle-size distribution of eachspecies can be determined after an experimental calibration or proportionalityfactor for each has been determined. A discrepancy in the previouslypublished figure for the solubility of 4-amino-4’-chlorodiphenyl sulphatehas been found by Bengtsson,16 who gives the reason for the discrepancywhich resulted in the previous figure’s being low by a factor of 10.The acidicdissociation constant and solubility of the free arnine are also given, and itsuse as a reagent for the microanalytical determination of sulphate is discussedin the light of the new experimental results.The determination of small amounts of hydrogen cyanide in air, i.e.,down to 1 p.p.m. v/v, by a specific test depends for its success on thepreparation and storage of filter papers of proved performance.17 Thesepapers are of known weight and have total iron, ferrous iron, sulphate, andalkali contents between certain narrow limits. After passage of air con-taining hydrogen cyanide through the papers, they are immersed in sulphuricacid solution to yield Prussian-blue colours proportionate to the amountsof hydrogen cyanide absorbed.A method previously used for the determination of water in paper hasnow been adapted by S.Hill and A. G. Dobbs18 to its determination ingranulated sugar. The sample is weighed, sealed in a thin glass ampoule,which is then enclosed in a flask that can be evacuated. After evacuationof the flask the ampoule is broken and, by the grinding action of four steelballs, the sugar is ground to a specific surface of not less than 3500 sq. cm.per g. The water is now driven off by heating the flask and contents for14 P. B. B. Vosloo and D. Sampson, S. African Ind. Chemist., 1958, 12, 48.15 B. J. Tufts and J. P.Lodge, jun., Analyt. Chem., 1958, 30, 300.16 T. A. Bengtsson, Analyt. Chim. Acta, 1958, 18, 353.17 B. E. Dixon, G. C. Hands, and A. F. Bartlett, Analyst, 1958, 83, 199.18 S. Hill and A. G. Dobbs, zbid., p. 143HASLAM AND SQUIRRELL: GENERAL. 39315 hr. at 60°, the water being collected in a cold trap surrounded by ethanoland solid carbon dioxide. This condensed water is finally allowed toevaporate into a space of known volume; the pressure exerted in thisvolume is a measure of the water content. S. D. Gardiner and H. J. Keyte l9have investigated the problem on quite different lines. Their method,whether before or after grinding, i.e., whether for free or total moisture,involves reaction of the water with a chloroform solution of cobaltousbromide and the production of cobaltous bromide hydrate which is filteredoff.Its amount, determined by evaporation of known aliquot parts ofthe cobaltous bromide reagent before and after reaction with the sugar, is ameasure of the water absorbed by the sugar.In connection with the examination of salt-water and other swimming-pools, Johannesson 20 has developed useful methods of determination ofmonobromamine and monochloramine in water. In one method the wateris treated with a neutral o-tolidine reagent containing sodium hexameta-phosphate which acts as a combined buffer and sequestering agent. Themonobromamine produces a blue quinonoid derivative which is titratedwith ferrous ammonium sulphate solution. After this, iodide is added, andthe monochloramine produces another blue compound which is titratedwith ferrous ammonium sulphate solution.Alternatively, with a rotatingplatinum electrode and a saturated calomel electrode, the monobromaminecan be titrated amperometrically with phenylarsenoxide solution. Oncompletion of this titration, iodide is added and monochloramine nowtitrated with the same solution.Difficulties in the determination of phthalates in propellants containingnitro-aromatic compounds have been overcome by Nonvitz.21 In hismethod the phthalate ester is extracted with ether and the nitro-compoundsare reduced by boiling with a specified concentration and volume of aceticacid and zinc. The phthalate is then extracted from the acetic acid solutionby means of light petroleum, and determined volumetrically by a hydrolysisprocedure. The extent of any interference by other esters and compoundspresent in propellants has been tabulated.Guenther 22 has utilised methyl-magnesium iodide as Grignard reagent for the gasometric determinationof the silanol group in complex silanols and silicone resins. A solution ofthe sample is added in a specially designed apparatus to 2~-Grignard reagent,butyl ether being used as solvent and methane as the inert gas. Theadvantages of the method are simplicity and efficiency at room temperature.Other interesting tests have been developed for the determination ofalkyl- and aryl-chlorosilanes in air.23 These are based on the colorimetricreaction of the substituted silane with a solution of tetramethyldiamino-benzophenone in aniline.Phenyltrichlorosilane can be determined colori-metrically by reaction with anhydrous aluminium chloride and a benzenesolution of tetramethyldiaminobenzophenone. In addition, a usefulcryoscopic method which may find application in other ways has beenl9 S. D. Gardiner and H. J. Keyte, Analyst, 1958, 83, 150.2O J. K. Johannesson, ibid., p. 155.21 G. Norwitz. Analyt. Chim. Acta, 1958, 19, 216.22 F. 0. Guenther, Analyt. Chem., 1958, 30, 1118.23 F. D. Krivoruchko, J . Analyt. Chem. (U.S.S.R.), 1957, 12, 245394 ANALYTICAL CHEMISTRY.detailed by Mullen for the determination of the y-isomer in the impure" benzene hexachloride " isolated from various dip-washes. Portions ofthe isolated impure benzene hexachloride are dissolved both in benzene andin a pure sample of the y-isomer.The depression of the solvent freezing pointis measured in each case, and from the results obtained the proportion ofy-isomer in the original sample can be calculated. The use of the benzenesolvent allows the effective molecular weight of the impure extract to bedetermined. Interesting methods have been developed by Paul 25 in thebiochemical field for the determination of the major constituents of samplesof tissue of about 1 mg. dry weight. Lipids are extracted with fat solvents,and the amount in the extract determined by dichromate oxidation. Ex-traction with cold N-sulphuric acid now yields a second fraction; carbo-hydrates are determined in this fraction by the anthrone test.Extractionwith hot N-perchloric acid solution yields a third fraction containing morecarbohydrate and the nucleic acids. The carbohydrate is determined by theanthrone method, and the total nucleic acids by measurement of the ultra-violet absorption a t 268 mp. Deoxyribonucleic acid is determined inde-pendently in this fraction by the indole test and hence ribonucleic acid isobtained by difference. Protein nitrogen is determined in the residue fromall the above extractions by a modified Kjeldahl procedure. Also two newmethods for the oxidative determination of blood alcohol are described byKirk et a1.26 The first method uses a modified form of dichromate oxidationof the alcohol after steam-distillation from a folded filter paper impregnatedwith acid salts to retain any basic components in the blood sample.Aldehydes and ketones and any organic acids are retained by washingthrough an alkaline mercuric oxide suspension.After oxidation, the excessof dichromate is determined by the conventional procedures. The secondmethod utilises a diffusion technique and oxidation of the diffused alcoholby an enzyme preparation containing liver alcohol dehydrogenase andco-enzyme 1. The absorbance of the enzyme oxidation product is measuredat 340 mp. The apparatus used is extremely simple.3. QUALITATIVE AND QUANTITATIVE INORGANIC AND ORGANIC ANALYSISQualitative.-Clark has summarised much of his work on toluene-3,4-dithiol as a general analytical reagent in qualitative analysis.27 The reagentcan be used for many of the purposes for which hydrogen sulphide has beenused in the past.It has certain advantages; e.g., it can be added inregulated amounts and the complexes which it forms with various metallicions are often highly coloured and moreover they coagulate readily. Thereactions can often be made highly selective by appropriate choice of operat-ing conditions. Objections to the reagent on the grounds of its instabilitymay be overcome by generation of the substance when required from itsdiacetyl derivative or its zinc salt. As a result of further work, Clark2824 J. D. Mullen, Analyt. Chim. Acta, 1958, 18, 189.B6 J. Paul, Analyst, 1958, 83, 37.26 P. L. Kirk, A. Gibor, and K. P. Parker, Analyt. Chem., 1958, 30, 1418.27 R.E. D. Clark, Analyst, 1958, 83, 396.28 Idem, ibid., p. 103HASLAM AND SQUIRRELL : QUALITATIVE AND QUANTITATIVE. 396has drawn attention to other very useful properties of diacetyldithiol. Itis a very useful reagent for the coagulation of Group I1 sulphides and forsulphur in acid solution. In addition, it catalyses the reduction of arsenatesand molybdates in the presence of hydrogen sulphide and acts as a precipitantfor tungsten(v1) in dilute acid solution in the absence of silicate.Useful specific reagents and spot tests have been described with applic-ation to inorganic analysis. Goldstein and Stark-Mayor z9 have definedconditions under which glyoxal bis-(2-hydroxyanil) can be used as a specifictest reagent for calcium; the test is particularly applicable to the detectionof calcium oxide in admixture with other colourless oxides and to the detec-tion of calcium in pharmaceutical products.According to Xavier’s ~ o r k , ~ Othere appears to be considerable doubt as to whether 2-mercaptoquinolinewill find application for quantitative work. On the contrary, it is likelythat it will prove to be a sensitive spot reagent for copper and palladium,with which it yields orange-yellow colours.Extending their work on the use of naphthalene derivatives in organicanalysis, Anderson, Garnett, and Lock31 have described the use of2-hydroxy-6-nit ronaphthalene-8-sulphonic (7-hydroxy-3-nitronaphthalene-l-sulphonic) acid as a new fluorimetric reagent for the detection of tin in thepresence of many other ions.When tin is present in the spot of the testsolution treated with a O-lyo solution of the ammonium salt of the reagent adark purple colour showing faint fluorescence is observed. After spray-ing with 15~-ammonium hydroxide solution, however, an intense bluefluorescence is produced. Other ions give coloured spots which are notfluorescent; 10-8 g. of stannous tin can be detected under the conditions ofthe test.Feigl, Goldstein, and Rosel132 have put forward a useful test for thedetection of trace amounts of chloride in fine chemicals. The test substanceis heated with chromic-sulphuric acid. mixture, and the vapours evolvedare passed through 4,4’-bisdimethylaminothiobenzophenone paper, withwhich free chlorine gives a blue coloration.The test will detect silverchloride in the presence of the ferrocyanide, ferricyanide, thiocyanate, andcyanide of silver. Also on the detection of anions, Coldwell and McLean 33have shown that under the influence of short ultraviolet radiation diphenyl-amine and nitrate salts react photochemically to form a yellow product.This reaction has been used in a sensitive and specific spot test for nitrate ion,which is carried out on a filter-paper support. It is possible to detect1 pg. of NO,- in 0-01 ml. of solution by this test.In the field of organic qualitative analysis, Soloway and Rosen 34 haveclassified organic compounds according to their effectiveness in promoting orinhibiting chelation between ferric chloride and n-propyl gallate. It isreported that this classification, used in conjunction with others based on29 D.Goldstein and C. Stark-Mayor, Analyt. Chim. Ada, 1958, 19, 437.30 J. Xavier, 2. analyt. Chem., 1958, 163, 182.31 J. R. A. Anderson, J. L. Garnett, and L. C . Lock, Analyt. Chim. Ada, 1958, 19,32 F. Feigl, D. Goldstein, and R. A. Rosell, 2. analyt. Chem., 1957, 158, 421.33 73. B. Coldwell and S. R. McLean, Canad. J . Chem., 1958, 36, 652.34 S. Soloway and P. Rosen, Analyt. Chem., 1957, 29, 1820.256396 ANALYTICAL CHEMISTRY.solubility, elementary analysis, and reaction to acid-base indicators, willpositively identify many classes of compound and will distinguish somealiphatic compounds from their aromatic counterparts. KolSek and hisco-workers 35 have shown that P-dimethylaminobenzaldehyde appears tohave particular advantages for the characterisations of primary aromaticamines.The Schiff's bases produced by the reaction of the aromatic aminewith the reagent are readily isolated and possess high and sharp meltingTwo tests devised by Trofimenko and Sease= have made it possiblereadily to differentiate nitriles, N-unsubstituted amides, and N-mono- or-di-substituted amides. In the first test the sample is decomposed byheating with soda-lime. Any ammonia or amine liberated is detected byvola t ilisat ion into met hanolic copper sulphat e solution, characteristicreactions being observed. In the second test account is taken of the form-ation of organomercury compounds by those amides having a hydrogenattached to the amide nitrogen.In the presence of ignited calcium oxideo-aminophenol and glyoxal unite to give the intensely red inner complexcalcium salt of a Schiff's base. This reaction 37 forms the basis of new spottest reactions for o-aminophenol and glyoxal.have shown that 2-diphenylacetylindane-l,3-dionel-hydrazone is potentially a most valuable reagent for identification andcharacterisation of carbonyl compounds, giving crystalline derivatives withwell-defined melting points and showing strong fluorescence in solutionand in the solid state. It is suggested that this fluorescence might be ofuse for the identification of carbonyl compounds by paper or columnchromatography. The azines are easily prepared by refluxing the aldehydeor ketone with the reagent in a solvent containing an acid catalyst.Eventhe long-chain alkyl carbonyl compounds form crystalline derivatives asdistinct from the oils usually obtained with 2,4-dinitrophenylhydrazine., Quantitative (Gravimetric) .-Development of gravimetric methods hasbeen confined largely to inorganic analysis. For example, Riley39 hasshorn that water and carbon dioxide may be determined simultaneouslyand speedily in rocks and minerals. The sample is heated to 1100-1200° ina silica tube in a stream of nitrogen. Oxides of nitrogen are removed fromthe products of heating by passage over heated copper wire at 700". Thetube containing the copper wire also contains silver pumice for the removal ofsmall amounts of sulphur compounds. The water is determined gravi-metrically by absorption on magnesium perchlorate. After removal ofwater, excessive amounts of sulphur compounds are removed by passagethrough a bubbler containing chromium trioxide in phosphoric acid beforegravimetric absorption of carbon dioxide on soda-asbestos.A macro-procedure previously developed for the macro-gravimetricdetermination of beryllium has now been adapted for the correspondingmicro-determination.The beryllium is precipitated and weighed aspoints.Braun and Moshera5 J. KolSek, N. Novak, and M. Perpar, 2. analyt. Chern., 1957, 159, 113.a6 s. Trofimenko and J. W. Sease, Analyt. Chem., 1958, 30, 1432.87 F. Feigl and D. Goldstein, 2. analyt. Chem., 1958, 163, 30.$8 R. A. Braun and W. A. Mosher, J . Amer. Chem.SOL, 1958, 80, 3048.39 J. p. Riley, Analyst, 1958, 83, 42HASLAM AND SQUIRRELL QUALITATIVE AND QUANTITATIVE. 397[Co ( NH3)6] [ (H,O) ,Be,( CO,) 2( OH),] ,3H,O and the factor for conversion toBe is 0-041. The procedure40 is particularly useful for the determinationof beryllium in the-presence of iron, aluminium, and magnesium, which areheld in solution by means of the disodium salt of ethylenediaminetetra-acetic acid. 5-Hydroxy-4-azaphenanthrene has been described byJakubke 41 as a suitable reagent for the gravimetric determination of cupriccopper. The copper is precipitated as a chelate from alkaline tartratesolutions by an alcoholic solution of the reagent; chromium(III), iron(III),aluminium(m), bismuth(m), lead(@, and zinc(11) do not interfere.J. E.Banks,42 in continuing his work on the gravimetric uses of benzene-phosphinic acid, has put forward a micro-method for the determinationof iron. As little as 2 mg. of ferric iron can be precipitated quantitativelyas ferric benzenephosphinate. An advantage of the micro-method, apartfrom speed, is that errors due to solubility of the precipitate in wash-liquorsare much reduced. New information has been given by Ramana Rao*on the oxine complex of molybdenum. He has shown that under deiinedconditions the precipitate is not formed at pH values higher than 2.24.In the estimation of other metals by the oxine method interference due tomolybdenum may be avoided by using a high pH value; no other externalcomplexing agent is necessary.Pietsch 44 has made a critical examination of the precipitation of uraniumwith cacodylic acid in the presence of many other metals.Uranium may bedetermined gravimetrically by this test ; glass fdters are much more suitablefor the filtration of the precipitate than are filter-papers. A thorough studyof the experimental conditions for the quantitative precipitation of phos-phorus as 3-oxine-l-phospho-12-molybdate has been made by Gottschalk 45in the range 10-0.5 mg. of P and also in the range 1 4 . 0 5 mg. of P. Hedeals with the relative correction which has to be applied in different casesbecause of solubility losses and also with the determination of phosphorusin FeIII EDTA-masked solutions as well as in certain unmasked solutionsof bivalent cations.Quantitative (Volumetric) .-Conventional volumetric analytical methodshave as usual been very numerous and we are able to review but a few.Newprimary standards are always of interest and sodium hydrogen diglycollatehas been suggested by Keyworth and Hahn 46 as a new primary standard foralkalimetric work. The salt is easily prepared, recrystallised, and dried togive a material which assays at 100.OO~o & 0.05%. It is soluble in waterand gives easily detectable end-points when titrated with phenolphthaleinas indicator.Many new indicators have been described, and ion-exchange resinsincorporating Thymol Blue, Bromocresol Green, and phenolphthaleinindicators have been used successfully by Miller 47 in acid-base titrations.40 Th. I. Pirtea and G.Mihail, 2. analyt. Chem., 1958, 159, 205.4 1 D. Jakubke, ibid., 160, 6.42 J. E. Banks, Analyt. Chim. Acta, 1958, 19, 331.43 D. V. Ramana Rao, ibid., 1957, 17, 538.J1 R. Pietsch, 2. analyt. Chem., 1958, 159, 343.G. Gottschalk, ibid., p. 257.16 D. A. Keyworth and R. B. Hahn, Analyt. Chem., 1958, 30, 1343.1' W. E. Miller, iFid., p. 1462398 ANALYTICAL CHEMISTRY.The advantages of the resin indicators are that they are in a separate phasefrom the solution and thus do not contaminate it for further work and arerecoverable for re-use. The resin may also be loaded to the limit of itsion-exchange capacity with indicator, resulting in a highly concentrated“ point ” indicator. The indicator PAR (4-2’-pyridylazoresorcinol) 48appears to have particular advantages in the titration of tungstates withstandard lead solution.The titration is carried out at boiling temperaturein the presence of hexamethylenetetramine as buffering agent ; the titrantmust be added at a definite slow rate.Copper phthalocyaninesulphonic acid 49 has proved a particularly usefulindicator for titrations of dilute uranium(1v) solutions at room temperaturewith standard ceric sulphate solution in the presence of sulphuric and phos-phoric acids. The indicator correction is quite low and the principle ofthe test may be extended to the corresponding titrations of molybdenum(v)and arsenic@) solutions. This versatiie indicator 50351 has also been usedwith success for the cerimetric titration of quinol and in the pennanganatetitration of iron(II), uranium(Iv), and potassium ferrocyanide.New indicators have also been used for complexometric titrations andwe include these together with a few of the many papers describing more ofthe growing uses of EDTA in analytical chemistry. In an address to theSociety for Andytical Chemistry on Recent Deselopments in Chelatometry,R.Pfibi152 draws attention to the effect of pH on complex formation byEDTA and to new indicators for complexometric titration such asXylenol Orange, Methylthymol Blue, Thymolphthalenone and Fluorexone,Glycine-Thymol Blue and Glycine-Cresol Red as well as certain azo-dyes.As a result of the introduction of these indicators there are new possibilitiesin the reagents used to screen certain cations in complexometric work.Eriochrome Blue SE 53 in the presence of a green screening agent, Acid GreenG, has been shown to be a useful indicator for the titration, at high pHvalues, of zinc, cadmium, magnesium, lead, nickel, and manganese withEDTA.Many interesting possibilities have been opened up by the discovery 5pthat several bivalent metals, such as lead, zinc, calcium, and magnesium, canbe determined potentiometrically by an EDTA procedure.Excess ofEDTA is added to the solution and this excess, often in the presence ofother complexing agents, is back-titrated in the pH region 8-12 by meansof a standard solution of mercuric nitrate. A silver electrode amalgamatedwith mercury is used as indicator electrode, and a calomel electrode asreference.The principle has been extended to the corresponding determin-ation of thorium 55 and to the determination of small amounts of aluminiumand manganese.56 The aluminium (or manganese) is complexed with excess48 R. Puschel, E. Lassner, and R. Scharf, 2. analyt. Chem., 1958, 163, 344.49 T. P. Sastri and G. G. Rao, ibid., p. 1.50 I d e m , ibid., p. 263.51 I d e m , ibid., p. 266.62 R. Piibil, Analyst, 1958, 83, 188.5s A. A. Abd. El Raheem and Abdel-Aziz M. Amin, 2. anaZyt. Chem., 1958,163, 340.54 H. Khalifa, R. Patzak, and G. Doppler, ibid., 161, 264.55 H. Khalifa, ibid., p. 401.18 I d e m , ibid., 1958, 163, 81HASLAM AND SQUIRRELL : QUALITATIVE AND QUANTITATIVE. 399of EDTA, and this excess is back-titrated a t the appropriate alkaline pHwith standard mercuric nitrate solution, a silver-mercury amalgam electrodebeing used as indicator electrode.In determining chromium(II1) in chromicacid solutions, Weiner and Ney5' have shown that it is necessary first toremove chromate. This is precipitated and removed as mercurous chromate,and excess of mercurous reagent is precipitated and removed as chloride.The tervalent chromium is now complexed by boiling with excess of EDTAbefore back-titration of the excess with standard nickel solution, murexidebeing used as indicator.The metal content of salts of organic acids commonly used as paintdriers can be conveniently determined by a direct and simple methoddue to Lucchesi and H i ~ - n . ~ ~ The salt is dissolved in alcohol-benzenesolution, and the metal chelated with EDTA without prior decom-position of the salt.The excess of EDTA is then back-titrated withzinc chloride to the Eriochrome Black T end-point. Some of the metalsto which the method is applicable are calcium, cobalt, zinc, lead, andmanganese.General inorganic volumetric methods include that for the determinationof aluminium in bauxite and other aluminous materials, which can becarried out rapidly and precisely by a method described by Watts.59 Potass-ium fluoride is added to the aluminium solution at pH 10, and the hydroxidewhich is thus liberated is titrated with standard acid back to the original pH.The titration can be carried out in the presence of any iron and titaniumprecipitated as hydroxides, and interfering calcium can be precipitated asoxalate.Picolinic acid and quinaldinic acid are very effective reagents for theseparation of palladium from other ions.Majumdar and Sen Gupta60761have shown that the palladium in such precipitates can be readily determinedby volumetric procedures in which either (a) the precipitate is dissolved inpotassium cyanide, and the excess of cyanide determined by back-titrationwith standard silver solution in ammoniacal media with potassium iodidesolution as indicator, or (b) the precipitate is dissolved in potassium nickelcyanide, liberating an equivalent amount of nickel, which is titrated withthe sodium salt of ethylenediaminetetra-acetic acid, murexide being used asindicator.In the reduction of uranium(v1) solutions by passage through zincreductors, not all the uranium is recovered from the reductor in theuranium(1v) condition; this is the state which is desirable if the amountpresent is to be determined by titration with oxidising agent.Rao andRao,62 however, have shown that if the uranium(v1) solution is treatedwith diethyl ether and sulphuric acid, then exposed to bright sunlight orthe light from a high-pressure mercury vapour lamp, quantitative con-version only into uranium(1v) takes place.57 R. Weiner and E. Ney, 2. Analyt. Chem., 1958, 161, 432.58 C. A. Lucchesi and C. F. Hirn, Analyt. Chem., 1958, 30, 1877.59 H. L. Watts, ibid.. p. 967.60 A. I<. Majumdar and J. G. Sen Gupta, 2. analyt. Chem., 1958, 161, 179.61 Idem, ibid., p.181.62 V. Panduranga Rao and G. Gapala Rao, ibid., 160, 190400 ANALYTICAL CHEMISTRY.The analysis of metal vanadium oxygen compounds similar to 2MgO,VO,,where the state of oxidation of the vanadium may not correspond exactly to2.00 presents certain difficulties. These are solvede3 by the addition of aknown amount of standard permanganate to a solution of the sample, ie.,sufficient to convert the vanadium a t least above the oxidation stage V*+.The excess of permanganate is now back-titrated to the quadrivalent vanad-ium stage with standard ferrous sulphate, sodium N-methyldiphenylamine-+-sulphonate being used as indicator. The difference between the per-manganate and the ferrous sulphate titres gives a measure of the state ofoxidation of the vanadium. A modification of the procedure enables totalvanadium to be determined.LangM has shown that small proportions of borate can be determinedin aqueous extracts of certain fertilisers, by a rapid method which does notsuffer interference from either carbonates or phosphates.The methodinvolves titration of an acid extract to pH 6, followed by addition of mannitoland re-titration to pH 6. Szekeres and BakAcs-Polgar 65 have put forward arapid procedure for the determination of alkali bicarbonates in the presenceof alkali carbonates. The carbonate is first precipitated as barium carbonateby the addition of sodium or potassium chloride and barium chloride. Thebicarbonate in the suspension is now titrated with sodium hydroxidesolution, with phenolphthalein as indicator.A rather complicated methodhas been devised for the determination of chloride in connection with theexamination of de-caff einised coffee for traces of chlorinated hydrocarbonsolvent.66 This draws attention to the fact that much better methods areavailable for the determination of traces of chloride of which the authorsappear to be unaware. These authors precipitate and isolate the chlorideas silver chloride, then convert this into metallic silver with alkali andformaldehyde. The silver is dissolved in acid and determined by titrationwith dithizone solution.Archer 67 has developed methods for the titration of iodides, bromides,and chlorides which are based on titration of acetone solutions of the halidewith standard silver nitrate (usually in n-propyl alcohol solution), dithizonebeing used as end-point indicator. Modifications of the method enablecyanide and silver to be determined. The methods may also be adapted tothe determination of halogens in organic compounds after preliminarybreakdown by either the Carius procedure or the Schoniger oxygen com-bustion method.A useful paper by Sarson 68 describes how non-aqueoustitrations afford an accurate and rapid method of analysis of explosives.Organic and inorganic nitrates are first separated by hot isobutyl methylketone extraction, before titration with perchloric acid or tetrabutylammon-ium hydroxide to indicator end-points or by automatic titration. Variationsin titrant, indicator, and solvent permit differential determination of variousinorganic and organic nitrates.63 B.Reuter and J. Siewert, 2. analyt. Chem., 1958, 162, 175.64 K. Lang, ibid., 163, 241.65 L. Szekeres and E. Bakhcs-Polgar, ibid., 159, 414.66 H. Suter and H. Hadorn, ibid., 160, 335.6 7 E. E. Archer, Analyst, 1958, 83, 571.6 8 R. P. Sarson, Annlyt. Chein., 1958, 80, 932HASLAM AND SQUIRRELL QUALITATIVE AND QUANTITATIVE. 401Phosphorus in condensed sodium phosphates 69 can be readily deter-mined by a method in which the polyphosphate is hydrolysed extremelyrapidly with concentrated hydrochloric acid, and the resulting ortho-phosphate precipitated as quinine phosphomolybdate before titration withstandard alkali using an indicator end-point. Alternatively, the titrationis carried out automatically.A method, which may well have otherapplications, has been presented by Seidman 70 for the determination ofsmall amounts (as low as O . O O l ~ o v/v) of sulphur trioxide in stack gasescontaining up to 0.3% v/v of sulphur dioxide. The sulphur trioxide isabsorbed quantitatively in 80% propan-2-01 which prevents oxidation ofsulphur dioxide. The sulphate is then titrated with standard bariumchloride solution, Thorin being used as indicator. Ammonia and nitrogenoxides do not interfere. Krause and Busch71 have been interested inthiosulphate and the tri-, tetra-, and penta-thionates. They have describeda convenient method of oxidising these compounds to sulphate with a hotsolution of cerium(1v) in perchloric acid.The excess of cerium is then back-titrated with sodium oxalate, nitro-ferroin being used as indicator. Halideand dithionate interfere.We turn now to volumetric tests as applied to organic analysis. Aninteresting new test 72 has been developed for the determination of thechlorine content of some organo-chlorosilanes. The sample, in ether solution,is titrated with a solution of ammonium thiocyanate in acetone. Ammoniumchloride is produced as a result of the reaction, and the end-point in thetitration is recognised by the production of a persistent red colour due toferric thiocyanate, ferric chloride being added a t an appropriate point in thetitration. Grover and Mehrotra 73774 have paid particular attention to thepotentialities of alkaline hypobromite solution as a volumetric oxidisingagent in connection with the determination of ammonia (oxidised to nitrogenand carbon dioxide) and urea and thiourea (oxidised to sulphuric acid andnitrogen).The reagent oxidises both propan-2-01 and acetone quantitativelyto acetic acid and carbon dioxide; under controlled conditions, chromic acidoxidises propan-2-01 to acetone only. Hence, by procedures employingboth the oxidising agents, mixtures of propan-2-01 and acetone can bereadily analysed.A simple method for the determination of the equivalent weights ofphosphorus and sulphur esters and of alkyl halides described by Baldwinand Higgins 75 is based on the following principles. The ethanolamine saltof the ester or halide is first formed by refluxing the sample with the base.The solution of this salt is then passed down a column containing cation-exchange resin in the hydrogen form, the cation of the salt being exchanged forhydrogen; the free acid in the effluent from the column can be titrated, andthe equivalent of the original ester or alkyl halide calculated from the result.69505 172S.Zechner and V. Reppestam, 2. analyt. Chem., 1958, 163, 423.E. B. Seidman, Analyt. Chem., 1958, 30, 1680.R. A. Krause and D. H. Busch, ibid., p. 1817.T. Takiguchi, Analyst, 1958, 83, 482.73 K. C. Grover and R. C. Mehrotra, 2. analyt. Chem., 1958, 160, 267.54 Idem, ibid., p. 274.I . ’ W. H. Raldwin and C. E. Higgiiis, .47ro/j*t. Chew?., 1958, 30, 446. - 402 ANALYTICAL CHEMISTRY.Schulek and Maros76 have presented an iodometric method for thedetermination of methanesulphonic acid derivatives.The method, whichis particularly useful for medicinal products since it can be applied in thepresence of antipyrine and pyramidone, is based on decomposition of thesulphonic acid with potassium cyanide in moderately alkaline medium togive sulphite and glycollic acid. The sulphite is then titrated in acid mediawith standard iodine solution in the presence of the other degradationproducts and any excess of hydrogen cyanide.Under appropriate conditions 77 polyethylene oxides have been shown toreact with barium chloride and sodium tetraphenylborate to yield insolublesubstances of definite compositions, which may be used for the detectionand determination of the polyethylene compounds.The insoluble productsare readily analysed. On treatment with mercuric chloride solutionproportionate amounts of hydrochloric and boric acids are liberated whichmay then be determined by conventional procedures. In a short communic-ation Sant and Sankar Das78 have described a method for the assay oftributyl phosphate in organic solvents. The tributyl phosphate is convertedinto inorganic orthophosphate by fusion with sodium hydroxide at a lowtemperature. The orthophosphate is then titrated with standard bismuthylperchlorate, saturated 1,2-di[allyl(thiocarbamoyl)] hydrazine in chloroformbeing used as an extraction indicator.It is known that determinations of nitrogen in certain high polymersand copolymers, e.g., polyacrylonitrile , copolymers of vinyl chloride andacrylonitrile, polyvinylpyrollidone, and polyvinylcarbazole, may yieldlow results if conventional Kjeldahl procedures are used.Skoda andS c h ~ r y , ~ ~ however, have shown that if a digestion medium containing coppersulphate, mercuric oxide, sodium sulphate, and mercury is used in thedigestion, and thiosulphate added to the alkali required in the neutralisationof the digestion product, then accurate figures for nitrogen are obtained.Whitehurst and Johnsonso have developed a method for the chemicaldetermination of aliphatic nitriles. The solution of the nitrile is treatedwith hydrogen peroxide and a known amount of standard potassiumhydroxide solution. The solution is then concentrated, one mole of basebeing consumed for each mole of nitrile in the overall reaction of hydrolysingit to the salt of the corresponding acid.The excess of alkali is then deter-mined in the conventional way, and hence the concentration of nitrilecalculated.Huhn and Jenckel *l have shown that maleic acid-maleic anhydridemixtures may be analysed by a method involving (a) titration with aqueousalkali, in which both substances react, and (b) titration with sodium meth-oxide, in which the acid uses twice as much titrant as the anhydride. Theprinciple of the method has been applied to the analysis of mixed poly-merisates of maleic acid and anhydride containing large amounts of styrene.76 E. Schulek and L. Maros, Analyt. Chim. Acta, 1958, 19, 4.77 R.Neu, Fette u. Seifen, 1957, 59, 823.78 U. A. Sant and H. Sankar Das, Analyt. Chim. Acta, 1958, 19, 202.70 W. Skoda and J . Schury, 2. analyt. Chem., 1958, 162, 259.80 D. H. Whitehurst and J. B. Johnson, Analyt. Chem., 1958, 30, 1332.81 H. Huhn and E. Jenckel, 2. analyt. Chem., 1958, 163, 427HASLAM AND SQUIRRELL: PHYSICAL METHODS. 403Finally, Hennart and Merlin 82 prefer to use anhydrous propionic acidin place of acetic acid for the direct titration of quaternary ammoniumcompounds with perchloric acid in non-aqueous media. The titrationmethod, which is particularly useful for the assay of commercial products,can be carried out potentiometrically or by using Metanil Yellow; thisindicator is found to give its colour change exactly at the point of maximuminflection in the titration curve.4.PHYSICAL METHODSElectrical.-The papers presented at the Symposium on Modem Electro-chemical Methods of Analysis held in Paris, July, 1957, have been printedin the January/February issue of Analytzca Chimica A~ta.~3 In consequence,this volume will be of more than ordinary interest to those analysts usingelectrical methods in any form. Many branches of the subject are covered,and these include Chronopotentiometry, Amperometric titrations, Potentio-met ry, Coulomet ry, Polarography, and High-frequency t itrat ions, Paperson adsorption kinetics and electrode processes and on newer methods ofelectrolytic conductivity measurement are also given, as well as a contro-versial feature on the desirability of adopting the coulomb as a universalstandard in place of the many chemical standards at present used.Under the heading of potentiometric methods the following work meritsattention. Based on studies made on some compounds as sensibilisatorsin the photolysis of silver halides, G.Asensi Mora 84 has described a rathernovel method of titration of halides in solution. Sodium thiosulphate isadded to the halide solution which is then brought to a pH of 6-6.5, andthe titration carried out with silver nitrate solution. A t the equivalencepoint, when a slight excess of silver ions is present, these ions react withthe thiosulphate to give Ag2S,03 which in turn decomposes to give riseto hydrogen ions: S,O,Ag, + H20 Ag,S + S 0 2 - + 2H+. A t theequivalence point there is thus a sharp change in pH which can be followedby using a pH meter and appropriate electrodes.The method has givenexcellent results for the titration of N/10- and N/100-solutions of all threehalides and thiocyanate solutions. Ivanova and Kovalenko 85 have shownthat calcium and magnesium, when present together in solution, can bedetermined with considerable accuracy by titration with potassium phos-phate solution and use of a zinc phosphate electrode.dead-stop ” andpotentiometric titration methods for the determination of primary aromaticamines with nitrous acid. They find that the potentiometric method ismore widely applicable and is less subject to interference than the dead-stopprocedure. A platinum-calomel electrode pair is used, and in titrations ofsulphanilamide and o-anisidine comparable precision is obtained whethercommercial automatic titrimeters or normal manual titrations are used.In the organic field, Butt and Stagg 86 have compared82 C.Hennart and E. Merlin, Chim. analyt, 1958, 40, 167.83 Analyt. Chim. Acta, 1958, 18, 1-184.84 G. Asensi Mora, Anales real SOC. espafi. FCs. QuCm., 1957, 53, B, 697.85 2. I. Ivanova and P. N. Kovalenko, J . Analyt. Chem. (U.S.S.R.), 1957, 12, 179.86 L. T. Butt and H. E. Stagg, Analyt. Chim. Ada, 1958, 19, 208404 ANALYTICAL CHEMISTRY.The titration of weak bases such as aniline in aqueous solution is alsopossible by the method of Critchfield and Johnson,87 who find that reasonableend-points can be obtained by making the solution 6-SM with respect to aneutral salt before titration.Indicators can be used for bases of ionisationconstant down to 1 x and potentiometric methods are applicableto bases with ionisation constants as low as 1 x Differential titrationsare also possible. Schouteden and Herbots88 are interested in polymersobtained by treating polyacrylamidoxime with hydroxylamine in sulphatesolution. The determination of free ammonia and hydroxylamine in suchpolymeric systems containing a large percentage of hydroxamic acid sidegroups is not possible without first removing the polymer. They thus firstprecipitate the polymer as its Fe(1rr) complex, and after removal of excessof Fe@) ions as Fe(OH), by addition of sodium hydroxide solution, titratethe mixed bases potentiometrically.Reasonable differential titration curvesare obtained by using 0.1N-hydrochloric acid as titrant and a conventionalpH meter with a glass-calomel electrode system.An important contribution to the work on titrations in non-aqueousmedia has been made by Yakubik et aZ.,89 who have described how, by usinga glass-silver electrode pair and the proper choice of solvent and titrant,potentiometric titration curves with sharp voltage peaks at the end-pointcan be obtained. The shape of the titration curves is similar to the firstderivative curves from ordinary potentiometric titrations. ConventionalS-shaped curves are obtained with very weak acids, and differential titrationsare possible by using sodium methoxide as titrant and neutralised pyridineas the titration solvent.Cundiff and his co-workers in three papers90-92 have added to theinformation about the use of the versatile titrant tetrabutylammoniumhydroxide.In the first they describe the use of the titrant for the deter-mination of the " equivalent " of 2 : 4-dinitrophenylhydrazones of aldehydesand ketones, which may be titrated as weak acids in pyridine solution.Isomerisation of the carbonyl compound does not affect the neutralisationequivalent, and the figure obtained is most useful in the characterisation ofthe derivative and parent aldehyde or ketone. The second paper givesmore information on the effect of solvent in the non-aqueous titration ofstrong acids, and pyridine is recommended for this purpose.This work iscontinued in the third paper which described how an impurity in the titrant,which can give rise to an error in the titration of strong acids or mixturescontaining strong acids, can be easily removed by passage over an ion-exchange column Amberlite IRA 400 (OH). This treatment makes thetitrant effective for titrating all types of acid but is not necessary for usewith weak or very weak acids alone. Harlow and Wyld 93 have shown thatinflections obtained in potentiometric titrations of very weak acids in acetone87 F. E. Critchfield and J. B. Johnson, Analyt. Chem., 1958, 27, 1247.88 F. Schouteden and J. Herbots, Makromol. Chem., 1958, 27, 256.89 M. G. Yakubik, L. W. Safranski, and J. Mitchell, A~zalyt. Chem., 1958, 30, 1741.90 A.J. Sensabaugh, R. H. Cundiff, and P. C. Markunas, ibid., p. 1445.91 R. H. Cundiff and P. C. Markunas, zbid., p . 1447.92 Idem, ibid., p . 1450.93 c;. .A. IlarIow and C. E. A. Wyld, ibid., p. 73HASLAM AND SQUIRRELL PHYSICAL METHODS. 405and pyridine are greatly affected by the acidity of the titrant solvent.They show that tetra-n-butylammonium titrants prepared in propan-2-01are superior for this work to titrants made in ethanol, methanol, or water.A rapid and simple procedure for the titration of amides or bases withperchloric acid in acetic acid or dioxan has been developed by Wimer.94The solvent used for the a i d e is acetic anhydride, and the end-point of thetitration is readily detected by means of a modified calomel-glass electrodepair in this medium.Coulometric methods have been extended, and an amperometric methodof end-point detection has been employed by Kennedy and Lingane 95 intheir method for the coulometric titration of uranium(v1) with electro-generated titanous ion at a platinum cathode in the presence of ferrous ionas catalyst.Even mixtures of vanadium(v) and uranium(v1) can be titratedowing to the fact that the reduction of V5+ occurs at a more oxidisingpotential than that of V4+ and U5+. Two amperometric end-points areobserved, corresponding to the reduction of V5+ to V4+ and to thesimultaneous reduction of V4+ to V3+ and U6+ to U4+. A measure ofthe titre due to uranium is thus obtained by subtracting twice the time tothe first end-point from the total time to the second end-point.A coulometric procedure has been developed by Arcandg6 by whichsmall quantities (18-240 pg.) of bromate can be determined with con-siderable accuracy.The bromate is allowed to react with bromide in acidsolution and the bromine liberated is titrated with electrically generatedcuprous copper. The method employs a dual platinum electrode indicatorsystem as an amperometric end-point detector. Calibration titrations are,however, necessary, and for accurate results the time for the calibrationtitrations should be within 10% of the time required for the sample titration.Thiourea 97 can also be determined coulometrically by using a mercury anodein a ground electrolyte containing sulphuric acid and potassium sulphate.The electrically generated mercury ions react with the thiourea to producea complex ion (Hg[CS(NH2)2]2)2’.In-a method due to Alfonsi 98 the best conditions for the successivedeposition of copper, lead, tin, and antimony by controlled-potentialelectrolysis from the same solution are described.The copper is depositedfirst at a cathode potential of 0.35-0.4 v against saturated calomel electrode(S.C.E.) from a solution at pH 5 containing ammonia, succinic acid, andhydrazine hydrochloride. The pH is now adjusted to 54-56, and the leaddeposited at a cathode potential of 0.6-0.65 v against S.C.E. The tin isnow deposited under specified conditions from an acidified solution, andfinally the antimony from the boiling solution at a cathode potential of0.45 v against S.C.E. The determinations are not affected by moderateamounts of Ni, Zn, Mn, Fe, Al.This method as applied to the analysisof brasses and bronzes has been described in a later paper.9994 D. C. Wimer, Analyt. Chem., 1958, 30, 77.95 J. H. Kennedy and J. J . Lingane, Analyt. Chim. Ada, 1958, 18, 240.96 G. M. Arcand, ibid., 19, 267.97 H. L. Kies and G. J. van Weezel, 2. analyl. Chem., 1958, 161, 348.9s B. Alfonsi, Analyt. Chim. Acta, 1958, 19, 276.99 Idem, ibid., p . 389406 ANALYTICAL CHEMISTRY.A procedure which promises wider applicability has been described byBaker loo for the rapid estimation of hydrofluoric acid in fuming nitric acid.The method is based on measurement of the spontaneous current ofelectrolysis of the diluted sample between an aluminium anode and aplatinum cathode in a polythene beaker and an amount of 0.6% vlv hydro-fluoric acid can be accurately and rapidly determined.Po1arography.-Advances in polarographic methods of analysis through-out the year have been largely due to the production of new and moresensitive instruments.Sawyer and his co-workers lol have developed anew polarograph. It utilises an x - y recorder for the direct measurementof the electrode potential or applied voltage, whilst correction for theIR drop in the cell is made by use of a third electrode. By this arrange-ment the authors have greatly reduced the sources of instrumental errors,and the accuracy of the direct reading of potentials is limited only by theaccuracy of the recorder itself.A new polarographic instrument whichautomatically corrects for the variation of the potential of the indicatorelectrode has been designed by S. Oka.lo2 With this instrument, whichincorporates an electronic servo-amplifier and a means of applying anauxiliary potential, any potential drop resulting from internal and externalresistances is continuously compensated for, thus eliminating the needfor a preliminary measurement of the potential drop as is required formanual correction.Kolthoff et aZ.lo3 have recommended the use of the rotating droppingmercury electrode for the analysis of solutions containing one or moreelectroactive species at concentrations of less than 10-4~. This recommend-ation is made after studies of the reproducibility of measurement of theresidual and limiting currents at the rotating dropping mercury electrodeat these low concentrations.Polyacrylamide is suggested as the idealmaximum-suppressor for use with the electrode since it is retained at themercury surface over the whole potential range and does not (unlike gelatine)combine with heavy metals.has shown that thallium, iron, and copper can readily bedetermined in high-purity cadmium metal. Thallium and iron are firstextracted from a solution of the chlorides of the elements in the sample bymeans of diethyl ether in 8N-hydrochloric acid, and are then determinedpolarographically. Copper is determined in a separate portion of the sampleby an extractive titration with a carbon tetrachloride solution of sodiumdiethyldithiocarbamate.The result of this test can also be confirmedpolarographically. E. Polecek lo5 has utilised the principle of oscillographicpolarography in the development of a method for the direct analysis ofsubstances in solid colloidal gels. Tin may be readily determined in food-stuffs by wet digestion with nitric acid, sulphuric acid, and hydrogenR. Carson100 B. B. Baker, Analyt. Chem., 1958, 30, 1085.101 D. T. Sawyer, R. L. Pecsok, and K. K. Jensen, ibid., p. 481.102 S . Oka, ibid., p. 1635.103 I. M. Kolthoff, Y. Okinaka, and T. Fuginaga, Analyt. Chim. Ada, 1958, 18, 295,104 R. Carson, Analyst, 1958, 83, 472.lo5 E. Polecek, 2. analyt. Chem., 1958, 162, I HASLAM AND SQUIRRELL : PHYSICAL METHODS.407peroxide. The reaction products are dissolved, and the tin in the solutionis determined, again by an oscillopolarographic testA most useful three-step polarographic method has been presented bySandler and Chung lo7 for the quantitative determination of hydrogenperoxide, formaldehyde, and acetaldehyde in solutions. The diffusioncurrent-concentration relationship for each pure substance is the onlycalibration datum required ; lithium chloride supporting electrolytes areused. Hydrogen peroxide is determined in an acid buffer solution in whichaldehydes do not interfere. Formaldehyde is determined in alkaline mediato which titanium tetrachloride is added to eliminate interference due tohydrogen peroxide, and acetaldehyde is estimated in similar solutions in thepresence of dimedone which reacts with and prevents interference from theformaldehyde.Garn and Gilroy have determined the maleic anhydridecontent of polyesters by carrying out a hydrolysis with aqueous potassiumhydroxide in the presence of chloroform or benzene. The aqueous layer isthen acidified, and the maleic acid determined polarographically.In the field of amperometric titration it has been demonstrated that itis possible to determine small amounts of gold even in the presence ofselenium , tellurium, and palladium by titration with quin01.l~~ The method,also suitable for the determination of gold in ores and sludges, uses a rotatingplatinum micro-electrode at a potential of 1 v with respect to the referencecalomel half cell.The titration is carried out in 2~-sulphuric at a tem-perature of 60". The sensitivity of the amperometric titration of fluoridewith thorium solution has been greatly increased in a method due toHarris,llO who uses a rotating palladium electrode. Large amounts ofchloride, nitrate, sulphate, perchlorate, borate , calcium, and magnesium aresaid to interfere slightly but small amounts of aluminium and phosphate(more than 1 p.p.m.) interfere seriously. As little as 20 pg. of fluoride per100 ml. of solution can be satisfactorily titrated.Polaro-voltric titrations (i.e. , titrations in which the end-point is detectedby measurement of the difference between an adjustable reference voltageand the polarisation voltage set up between two platinum micro-electrodesin the solution) have been applied most successfully by Walisch and Ash-worth 111 to the titration of highly dilute halide solution with silver nitratesolution.Even at the high dilution of 0.004~ an accuracy of &0.2% isclaimed for the method.Chromatography.-Under this heading some of the interesting applica-tions of paper and column chromatography including ion-exchange methodsare first described; this is followed by a section on gas-liquid and gas-solidchromatographic methods.A method of ascending paper chromatography, which has the advantageof concentrating rather than spreading the developed spots or streaks, has106 2. Malkas, 2. Lebensm.-Untersuch., 1957, 106, 257.107 S. Sandler and Yu-Ho Chung, Analyt. Chem., 1958, 30, 1252.108 P.D. Garn and H. M. Gilroy, ibzd., p. 1663.109 L. S. Reishakhrit and W. S. Sukhobokova, J . Analyt. Chem. (U.S.S.R.), 1967,110 W. E. Harris, Analyt. Chem., 1958, 30, 1000.111 W. Walisch and M. R. F. Ashworth, Analyt. Chim. Ada, 1958, 18, 632.12, 145408 ANALYTICAL CHEMISTRY.been described by Osawa.l12 The method uses a circular paper cut so thatit can be folded round to form a cone. If a continuous line of test solution isapplied to the paper, then the resulting developed chromatogram appears asa series of arcs concentrated towards the apex of the paper. The authorhas used the method for the separation and identification of green plant pig-ments. E. Blasius and W. Gottling 113 have contributed a rather interestingpaper on the paper chromatographic separation of the common cations.Preliminary chemical tests divide the cations into small groups which arethen separated on circular paper.The eluted substances, present in zones ofcharacteristic RF value, are distinguished by appropriate spray reagents. Itwould appear that condensed phosphates are used as tenderising agentsin the meat industry. Their analysis is of importance, and Schormuller andWurdig 114 describe paper-chromatographic methods for the separation ofmono- to hepta- and cyclotri- and cyclotetra-phosphates in commercial pre-parations. The various polyphosphates are separated as concentric zones,and hexametaphosphate is used as standard. Information is given on thebehaviour of the various phosphates with different eluting media.Duffield 115 has developed some interesting methods for the determinationof certain trace metals in crops such as wheat, barley, and oats, althoughthe tests have obvious applications in other fields.The sample is first ashedunder controlled conditions, and the boron determined in an aqueous extractof the ash. This involves the reaction of the borate with curcumin and thepaper chromatographic preparation and purification of the blue-green alkalireaction product of this borate-curcumin derivative. Copper, cobalt, andnickel, in the acid extract of the ash, are separated on paper with an ethylmethyl ketone-hydrochloric acid solvent ; rubeanic acid is employed forcolour development. Zinc is determined in the acid extract of the ash bycomplexing interfering elements with sodium thiosulphate, then determiningthe zinc by a mixed colour method involving the extraction of the zincdithizone complex at pH 5. Molybdenum is determined on the acid extractof the ash by a procedure involving paper-chromatographic separation frominterfering elements and dithiol colour development.Manganese is deter-mined on a similar acid extract by direct application of the periodateoxidation test .Organic applications include a novel approach which has been madeto the detection and approximate determination of traces of acrylonitrile.116As little as 1 pg. of acrylonitrile can be detected. Advantage is taken of thefact that acrylonitrile reacts with thiourea in the presence of hydrochloricacid to yield 2-cyanoethylisothiuronium chloride.This substance isseparated from the reaction products by paper chromatography, thendetected on the paper by spraying with ammoniacal silver solution.Extending their work on centrifugally accelerated paper chromatography,McDonald et aL1l7 have described the fundamental factors involved in112 Y . Osawa, Nature, 1957, 180, 705.119 E. Blasius and W. Gottling, 2. analyt. Chem.. 1958, 182, 423.114 J. Schormuller and G. Wiirdig, 2. Lebensm.-Untersuch., 1958, 107, 415.115 W. D. Duffield, Analyst, 1958, 83, 503.116 J. M. Stephek and V. M. tern&, ibid., p. 345.117 H. J. McDonald, L. R. McKendall, and E. W. Bermes, J . Chvomatog., 1958,1, 259HASLAM AND SQUIRRELL: PHYSICAL METHODS. 400obtaining good separation and reproducible chromatograms by the method.They have studied the variables which influence RF values, and havefractionated mixtures of Bromophenol Blue, Methyl Orange, and MethylRed as well as solutions of leucine, methionine, and glycine.In view of the increasing interest being taken in antioxidants, the paperby Ter Heide 11* on the paper-chromatographic separation of antioxidantsshould be noted.It is concerned with those antioxidants such as gallicacid, ethyl, n-propyl, n-butyl, n-pentyl, n-octyl, n-decyl, and n-dodecylgallate, 4-methoxy-2- and 3-t-butylphenol, and 4-methyl-2,6-di-t-butyl-phenol which now find their way into fats and oils. Useful information isprovided about the extraction of the antioxidant from the product.Thepaper chromatography is conventional; the developing reagent consists of aferric salt-potassium ferricyanide spray. The phenolic antioxidant reducesthe ferric salt and blue ferrous ferricyanide is produced a t appropriate pointsin the chromatogram.A semi-quantitative ascending-paper chromatographic method ofdetermining salicin has been devised by P e r ~ s o n . ~ ~ ~ The developing solventis butanol-acetic acid-water (4 : 1 : 5) and 2~-sulphuric acid is used as sprayreagent. The sprayed chromatogram is heated in a thermo-desiccator, andthe salicin appears as distinct pink spots. Organic peroxides have beendetected, identified, and estimated in small amounts by Abraham et aZ.,120using a chromatographic method. For the higher-molecular-weight com-pounds a procedure including reversed-phase chromatography on Silicone-treated paper is used, with a developing solvent of chloroform-ethanol-water and a spray reagent of acidified ferrous thiocyanate. For peroxidesof low molecular weight, which are too volatile for paper-chromatographicmethods, a gas-liquid chromatographic procedure is described.The chromatographic separation of the volatile fatty acids has receivedconsiderable attention.Osteux and his co-workers 121 have isolated theacids by steam-distillation, and after preliminary removal of cations andconcentration by passage over a sulphated ion-exchange resin, have convertedthe mixed acids into their morpholine salts. These salts are then chromato-graphed on paper by using a developing solvent containing butanol, cyclo-hexane, propylene glycol, ammonia, morpholine, and water for the straight-chain series, and containing benzyl alcohol saturated with 1.5N-ammoniafor the iso-series.The method has been used in connection with the deter-mination of the fermentation type or pattern of anaerobic bacteria.We turn now to the extensive use of ion-exchange methods. The stronglybasic anion-exchange resin Amberlite IRA 400, after conversion into theascorbate form, is used by Korkisch and Farag 1229123 in two methods usefulfor the analysis of steels and minerals. The first paper, dealing with thebehaviour of ascorbate complexes of vanadium, molybdenum, and tungsten,118 R. Ter Heide, Fette u. Seifeen, 1958, 60, 360.11s A.Persson, J . Chromatog., 1958, 1, 269.120 M. H. Abraham, A. G. Davies, D. €3. Llewellyn, and E. M. Thain, Analyt. chinz.121 R. Osteux, J. Guillaume, and J. Laturaze, J . Chrmatog., 1957, 1, 70.122 J. Korkisch and A. Farag, Microchim. Ada, 1958, 646.123 Idem, ibid., p. 659.Acta, 1957, 17, 499410 ANALYTICAL CHEMISTRY.describes how at pH 4 those complexes having a negative charge are absorbedon the resin. The vanadium can now be quantitatively eluted with 0 . 1 ~ -hydrochloric acid, together with only a very small proportion of tungstenand no molybdenum. Other components of steel, zliz., iron, chromium, andmanganese, do not interfere with this method of vanadium separation.The second paper, dealing with the photometric determination of smallamounts of titanium as the titanium-ascorbic acid complex, uses theexchange resin as a means of concentration of the titanium, at the same timeremoving it from such ions as iron, chromium, and nickel which interfere inthe direct photometric determination.The titanium concentrated on theresin is eluted with N-hydrochloric acid prior to photometric determination.Lederer 124 has extended his work on chromatography, using paperimpregnated with ion-exchange resins, to the separation of such systems asCu-Cd, Mn-Fe-UO,++, Th-Ce-Fe, Al-Zr, and Ti-Fe-Al. The filter paperused is impregnated with colloidal Dowex 50 resin, and the chromatogramdeveloped with various concentrations of hydrochloric acid. An advantageof the method is that the concentration of the acid required to give aseparation can be predicted from the equation xpH = R, + constant, wherex = valency of the cation and R, = log[(l/RF) - 13.Considerable attention has been paid to the correct conditions for theseparation of uranium from thorium.Tomic and others 125 utilise the factthat uranium in 6-5~-hydrochloric acid solution forms a negatively chargedcomplex which is absorbed on an Amberlite IRA 400 resin (in the chlorideform), whilst thorium is not absorbed. The uranium is subsequentlyeluted with 1N-hydrochloric acid and then determined by an improved formof fluorimetric procedure. A useful method126 has been developed for thedetermination of uranium in phosphates, coal ashes, and bauxites in whichthe uranium as its negatively charged ~ ( I v ) chloride complex, in ~N-HCI-ascorbic acid solution, is separated from interfering substances by passagethrough an anion-exchange column containing Amberlite IRA 400 (Cl-) .The uranium is then eluted with 1N-hydrochloric acid solution, and deter-mined in the eluate either by the polarographic method involving the catalyticnitrate wave or by the method which takes advantage of the fluorescenceof the sodium carbonat e-po t assium carbonat e-sodium fluoride fusionproduct.The use of ion-exchange resin to concentrate fluoride and free the samplefrom interfering ions has allowed Nielsen 12’ to develop a method fordetermining microgram quantities of fluoride at quite high dilution.Thefluoride ion concentrated on the resin column is eluted with increasingconcentrations of sodium acetate and determined colorimetrically withCyanine R, zirconyl nitrate, and hydrochloric acid, density measurementsbeing taken at 527.5 mp.IgUchi1Bs129 has studied the distribution co-efficients of the di-, tri-, tetra-, and penta-thionates between anion-exchange124 M. Lederer, J. Chromatog., 1958, 1, 314.126 E. Tomic, I. M. Ladenbauer, and M. Pollak, 2. analyt. Chem., 1958, 161, 28.126 J. Korkisch, A. Farag, and F. Heckt, ibid., p. 92.127 H. M. Nielsen, Analyt. Chem., 1958, 30, 1009.128 A. Iguchi, Bull. Chem. SOC. Japan, 1958, 31, 597.129 Idem, ibid., p. 600HASLAM AND SQUIRRELL : PIIYSICAL METIIODS. 41 1resins and hydrochloric acid solution, and also the distribution coefficientsof sulphate, thiosulphate, sulphite, and sulphide between the resin andalkaline nitrate solutions.As a result of these studies methods have beenworked out for the separation of mixtures of these compounds by ion-exchange chromatography.Ion-exchange chromatography has also found use in organic analysis.In an extension of the work on salting-out chromatography, Rieman andhis co-worker 130 have developed a new technique of solubilisation chromato-graphy by which mixtures of some four phenols and six aliphatic alcohols(C,-C,) have been separated. The procedure involves the elution fromion-exchange resin by means of aqueous solutions of acetic acid. The effectof the extent of cross-linking in the exchange resins, and the rate of flow andconcentration of the eluent on the behaviour of the compounds have beenreported, A second paperf3l describes the separation of ketones by usingaqueous solutions of acetic acid or lower alcohols as eluents.By this methodseven ketones were separated, the highest being undecan-2-one. A newmethod for the determination of the saponification number of oils and esterswhich yields more precise information about the esters under test has beendeveloped by Swann, Zahner, and M i l r ~ e r . ~ ~ ~ After saponification, thehydrolysis products are passed through a cation exchanger in the hydrogenform. The eluate contains the free acids and these are titrated directly bya potentiometric method, differential curves often being obtained when estermixtures are under test.The rapid increase in the use of gas-liquid and gas-solid chromatographicmethods has necessitated some degree of standardisation of units andtechnical terms used.In connection with this, Johnson and Stross 133 havereported the findings of a study group of the A.S.T.M. Committee onPetroleum Products and Lubricants. Terms relating to technique,apparatus, reagents, conditions of determination, and reporting results aredefined. Jowes and Kieselbachl= have suggested a revised set of unitsbased on kinetics rather than equilibrium data, which they indicate will givemore convenient recording and interpretation of results and facilitate amore exact control of quantitative separation processes. An alternativemethod due to Ambrose and his collaborators135 suggests that the datashould be recorded in terms of the relationship between elution temperatureand the partition coefficient or specific retention volume of the substance.By this method a quantitative approach to solvent composition is possible.The complexity of the detection instruments for use in gas-chromato-graphy apparatus has been a valid criticism of the method.Davis andHoward 136 have partly overcome this difficulty by designing a simple andeasily constructed instrument which employs a single thermistor as thedetector. The detector is placed near the exit tube of the column and forms130 J. Sherma and Wm. Rieman, tert., Analyt. Chim. Acta, 1958, 18, 214.131 Idem, ibid., 1958, 19, 134.132 W. B. Swann, R. J. Zahner, and 0.I. Milner, Analyt. Chem., 1958, 30, 1830.133 H. W. Johnson and F. H. Stross, ibid., p. 1688.134 W. L. Jowes and R. Kieselbach, ibid., p. 1590.135 D. A. Ambrose, A. I. B. Keulemans, and J. H. Purnell, ibid., p. 1582.136 A. D. Davis and G. A. Howard. J . Appl. Chem., 1958, 8, 183412 ANALYTICAL CHEMISTRY.one arm of a Wheatstone bridge driven through a potentiometer by a battery.A 10 mv full-scale deflection recorder is used. The authors describe thenecessary precautions and conditions for use of a thermistor detector, andthe overall sensitivity and response to differing materials are outlined.J. E. Lovelock137 has utilised the unique ionisation properties of argon asthe basis of a stable detector for use with gas-chromatographic apparatus.The theory of the operation of the device depends on the ionisation, bycollision with excited argon atoms, of molecules of the test substance inthe vapour phase.The response of the detector is similar for differentmolecular species of organic compounds and over a considerable range thisresponse is also linear.The problem of selecting the liquid substrate for use in gas-liquidchromatographic columns has been clarified by Tenney,l38 who has studied18 liquid substrates with respect to their selectivity towards various typesof hydrocarbon and oxygenated compound. His results permit the com-parison of retention characteristics between compound types at variousboiling point levels and show dipropionitriles to have the highest selectivityand squalane to be the best non-selective substrate for hydrocarbons.With regard to selectivity, too, Eggertson and Knight 139 have shown that,in the chromatography of hydrocarbon gases, adsorption on the surface ofthe support medium may be controlled to provide a wide naphthalene-paraffin selectivity by adjusting the amount of the solvent.With smallamounts of supported solvent they find paraffins are retarded, whereasnaphthalenes are retarded on a liquid-partition type packing.By designing a column capable of operating at temperatures up to 4W0,Ogilvie and his co-workers 140 have been able further to explore the realmsof high-temperature gas-phase chromatography. The column utilisedasphaltines as stationary liquid phase and bare wire thermal-conductivityfilaments in the conductivity detector cell.Higher-molecular-weightseparations were made, and the method was used for the determination ofn-paraffin distribution in waxes. Dupire and Botquin,la using a stationaryliquid phase of Silicone grease on a powdered Silocel C22 firebrick supportand helium as carrier gas in a column working between 220" and 250", haveseparated the heavy tar oils, naphthalenic oils, wash oils, and anthracenicoils. Excellent quantitative results have been obtained on complexmixtures in a high-temperature apparatus of their own design. Nogare andSafranski 14% have described a relatively simple high-temperature gas-chromatographic apparatus capable of resolution and estimation of high-boiling organic mixtures , e.g.J polyethylene glycols, phthalate esters, andSilicone oils.High-vacuum Silicone grease and commercial linear poly-ethylene are used as partition media a t temperatures of 150-355' in con-junction with platinum-filament thermal conductivity detectors operated at10-100" higher than the column temperatures. Excellent resolution is187 J. E. Lovelock, J . Cht'omatog., 1958, 1, 35.188 H. M. Tenney, Analyt. Chem., 1958, 30, 2.180 F. T. Eggertson and H. s. Knight, &id., p. 16.140 J. L. Ogilvie, M. C. Simmons, and G. P. Hinds, jun., ibid., p. 26.141 F. Dupire and G. Botquin, Andyt. Chim. Ada, 1958, 18, 282.142 s. D. Nogare and L. W. Safranski, Analyt. Chem., 1958, 30, 894HASLAM AND SQUIRRELL: PHYSICAL METHODS. 413obtained on relatively short columns.A second apparatus,l& suitable forthe analysis of impurities in organic mixtures at the parts per million level,utilises an amplifier to increase the signal from thermistor detectors. Theconditions necessary for obtaining low noise and drift are described, togetherwith examples of some specific applications of the method, e.g., the determin-ation of p.p.m. of propan-2-01 in benzene, benzene in toluene, cyclohexanein toluene, methanol in water, and other impurities in cyclohexane andtoluene.The growing general importance of gas chromatography in analysis isshown by the fact that a particular part of the Zeitschrift fGr analytischeChemie is devoted almost wholly to the subject. There are papers which dealwith the gas-chromatographic examination of such diverse chemicals asethylene,lU ~yclopentadiene,~~~ mono- and di- hydric phenols,146 and phenolsbefore and after rnethy1ati0n.l~' Work on the examination of perfumerymaterials is interesting because it brings out the possibility of examiningsuch substances by gas-chromatographic test both before and after treat-ment with chemical reagents; the purpose of the latter treatment is toremove such classes of substance as ketones, etc.,l& from the mixtures.Avery comprehensive paper deals with the choice of stationary phase in gas-chromatographic analysis; 149 detailed information is presented about thebehaviour of 120 substances on various stationary phases. Other con-tributions are concerned with the comparative merits of infrared andgas-chromatographic methods of examination of toxic solvent mixtures,lMwith the quantitative evaluation of separated chromatographic fractions,l51with the automatic registration of gas volumes following gas-chromato-graphic ~eparation,l5~ and with the substitution of a combination of gaschromatography and gravimetry for Podbielniak di~tillati0n.l~~ Using new-type packings consisting of heterocyclic amines on a Celite support,Zlatkis 154 has been able to resolve mixtures of the hexane isomers.Suchstationary liquid phases as squalane, isoquinoline, and quinoline-brucinemixture have been used with helium as the carrier gas. With the lattersystem such multicomponent mixtures as cyclohexane, methylcyclohexane,2,4-dime th ylpent ane, c yclopen t ane , 2-me t hylpen t ane,2,3-dimethylbutane, n-pentane, and isopentane have been analysed, withthe column temperature at 250".Studies in the composition of azeotropeshave been aided by the gas-chromatographic method of Haskin and hisc o - ~ o r k e r s , ~ ~ ~ who have described analytical methods for the study of a3-met h ylpent ane,143 C. E. Bennett, S. D. Nogare, and L. W. Safranski, Analyt. Chem., 1958, 30, 898.144 G. Nodop, 2. analyt. Chem., 1958, 164, 120.145 E. A. M. Dahmen and J. D. Van der Laarse, ibid., p. 37.146 J. JanAk and R. Komers, ibid., p. 69.147 G. Bergmann and G. Jentzsch, ibid., p. 10.148 E. Bayer, G. Kupfer, and Karl-Heinz Reuther, ibid., p. 1.149 G. Raupp, ibid., p. 135.150 E. G. Hoffmann, ibid., p. 182.151 G.Schomburg, ibid., p. 147.162 J. JanAk and K. Tesarik, ibid., p. 62.153 U. Schwenk and E. Weber, ibid., p. 169.154 A. Zlatkis, Analyt. Chem., 1958, 30, 332.J. F. Haskin, G. W. Warren, L. J. Priestley, jun., and V. Yarborough, ibid.,I). 217414 ANALYTICAL CHEMISTRY.number of ternary and quaternary mixtures including the systems acetone-met hanol-me t h yl acetate ; chlorofonn-propan-2-ol-et h yl methyl ketone ;acetonitrile-ethanol-triethylamine-water; and di-n-butyl ether-butan-1-ol-n-butyl acetate-water. The reproducibility of the methods ranged fromk0.5 to &4.4% of the figure found for the specific compositions studied.Haslam and Jeffs 156 have described methods to determine the nature ofmixed solvents in plastic adhesives and it is probable that these tests willhave other applications. The method of isolation of the solvent is described,as well as its preliminary gas-liquid chromatographic separation using adinonyl phthalate stationary phase.The subsequent examination of thesolvent on a polar (tritolyl phosphate) column and a non-polar (paraffinwax) column is outlined, as well as infrared and chemical tests which may becarried out on the separated products.Gas-liquid chromatography has also been used by Irvine and Mitchell 157for the separation of tar acids; some 28 phenolic compounds were identifiedby retention-time data and infrared spectroscopy. Amberg 1% has measuredthe relative elution times of a number of organic sulphur compounds fromglass columns containing tritolyl phosphate supported on 35-80-meshJohns Manville C-22 firebrick.The elution gas used is nitrogen at columntemperatures of 8A101°. The injection system consists of a heatedchamber at the column head inside which are two 60-mesh stainless-steelgauzes from which the liquid can be flashed on to the column after injectionthrough a self-sealing cap in the usual way. Both detector and referenceelements of the thermal conductivity monitor are placed at the column out-let with liquid nitrogen traps interposed between the two. These traps aredesigned to fit directly on to a mass-spectrophotometer, facilitating theready determination of the mass spectra of each fraction.Haslam, Hamilton, and Jeffs159 have developed a novel method ofdetermination of polyethyl esters in methyl methacrylate copolymers.Thealkoxyl groups in the polymer are converted into the corresponding iodides,which are purified and collected in n-heptane. After addition of methylenedichloride and ethylidene dichloride as markers, the mixture of methyl andethyl iodides, markers, and n-heptane is subjected to gas-liquid chrom-atography in order to obtain the proportions of the two iodides and hencethe amounts of the corresponding polyesters.A useful method of gas analysis has been developed by Timms, Konrath,and Chirnside for the determination of trace impurities such as hydrogen,argon, oxygen, nitrogen, methane, and carbon monoxide in carbon dioxidewhich may be used as coolant gas for nuclear reactors. In this method,carbon dioxide is removed by soda-lime, and moisture by magnesiumperchlorate. If required, oxygen is taken out by means of De-oxo catalyst.The separation of the various impurities is then effected on a column ofmolecular sieves.By appropriate use of hydrogen as carrier gas on the onehand and argon on the other hand, it is possible to obtain chromatograms156 J. Haslam and A. R. Jeffs, Analyst, 1958, 83, 465.157 L. Irvine and T. J. Mitchell, J . AppZ. Chem., 1958, 8, 425.158 C. H. Amberg, Canad. J . Chem., 1958, 36, 690.159 J. Haslam, J. B. Hamilton, and A. R. Jeffs, Analyst, 1958, 83, 66.160 D. G. Timms, H. J. Konrath, and R. C. Chirnside, ibid., p. 600HASLAM AND SQUIRRELL: PHYSICAL METHODS. 415from which the concentrations of the various impurities in carbon dioxidecan readily be deduced.R.N. Smith and his co-workers161 have found that, by using a gas-chromatography column consisting of two layers of silica gel separated byiodine pentoxide, they are able to separate and determine the components ofmixtures of (a) nitrogen, nitrous oxide, nitric oxide, and carbon monoxideor (b) nitrogen, nitric oxide, carbon monoxide, and carbon dioxide in10 minutes. The column at elevated temperatures oxidises carbon monoxideto carbon dioxide and nitric oxide to nitrogen dioxide.Absorption Spectroscopy (Inorganic) .-The use of absorption spectroscopyin analytical chemistry increases year by year and it was a noticeable featureof 1958 that very many excellent methods were developed based onabsorptiometric measurements in the visible region of the spectrum.Improvements in instruments have facilitated rapid methods of analysis ofcomplex mixtures. For example, differential spectrophotometric methodshave been extended by Banks and his co-workers,162 who have applied theprinciple to the analysis of neodymium-erbium, praseodymium-erbium, andpraseodymium-neodymium-samarium mixtures, and in addition havediscussed an accurate method for the measurement of small molarabsorptivities.Freund and Holbrook,l@ however, by application of theAlizarin Red differential method of Manning and have developeda method for the determination of zirconium in the presence of hafnium.The test is applied to a fixed sum of the elements as their oxides.A useful method of calibration or checking the wavelength scale of aspectrophotometer without the use of the conventional standard filters orsolutions has been described by Parthasarathy and Sar~ghi.l~~ They utilisethe so-called isosbestic point of indicator solutions.This is the point atwhich all transmittancy curves of an indicator solution intersect irrespectiveof pH. Thus, acidic and basic solutions made from equal amounts of thesame stock indicator solution are placed in matched cells and the differencein their absorbance is measured a t different wavelengths. The differencein transmittancy is zero at one specific wavelength only, e.g., 469 mp forMethyl Orange. Provided pure indicators are used, the method is free fromthe usual sources of calibration error. The isosbestic point of severalindicators is given.Bohnstedt and Budenz 166 have put forward a semimicro-method ofdetermination of arsenic, especially in steel and iron, etc.The materialis dissolved in perchloric acid and nitric acid, and the arsenic distilled astrichloride after treatment of the solution with ferrous sulphate, potassiumbromide, and hydrochloric acid. The neutralised distillate is treated with amolybdenum-blue reagent , and the resulting colour measured at 900 mp.The same authors have developed a photometric semimicro-method ofdetermination of sulphur in iron and steel, etc., based on the determination161 R. N. Smith, J. Swinehart, and D. G. Lesnini, Analyt. Chem., 1958, 30, 1217.162 C. V. Banks, J. L. Spooner, and J.W. O’LaughIin, ibid., p. 458.163 H. Freund and W. F. Holbrook, ibid., p. 462.164 D. L. Manning and J. C. White, ibid., 1955, 27, 1389.165 N. V. Parthasarathy and I. Sanghi, Nature, 1958, 182, 44.166 U. Bohnstedt and R. Budenz, 2. analyt. Chem., 1957, 159, 95, 102416 ANALYTICAL CHEMISTRY.of sulphur dioxide in combustion products arising from the iron, etc. Thecolourless N-sulphinic acid of Fuchsinleucosulphonic acid formed by actionof sulphur dioxide on Fuchsin, yields with formaldehyde a red product whichis measured at 580 mp.In a series of papers 1679168 Spier and Strickland have described methodsfor the determination of boron. In the first, those methods based on thereaction of boric acid and curcumin reagent to form red compounds, rubro-curcumin and rosocyanin, are detailed.The first two methods, suitablefor the ranges 2-15 and 0.5-4 pg. of boron respectively, are based on theevaporation of the boron distillate in the presence of sodium hydroxide andglycerol and removal of the bulk of the glycerol before ignition of theresidue at a dull red heat. This residue is now evaporated under carefullycontrolled conditions with a curcumin-oxalic acid reagent, and the ab-sorbancy of the rubrocurcumin thus formed is measured in ethyl alcoholsolution at 555 mp. In the third, more sensitive method the residues of theignition are treated with water and curcumin solution and neutralised withacetic acid before re-evaporation to dryness. This residue is treated witho-chlorophenol reagent under controlled conditions, and the absorbancy ofthe rosocyanin thus formed measured in aqueous alcohol solution a t 550 and630 mp.The principles of the methods of separation of boron by distillationfollowed by evaporation of the distillate are also described.Lilie1G9 has taken advantage of the fact that bismuth reacts withthionalid to develop a useful method for the determination of very smallamounts of bismuth in lead and tin. The method for lead involvespreliminary precipitation of the bulk of the lead as chloride, followed byfiltration and extraction of the bismuth from the filtrate as its chloroform-soluble thionalid complex. This is decomposed with sulphuric acid solution,and the bismuth in the acid extract determined by means of its colorimetricreaction with iodide in the presence of sulphite.Bismuth in tin is deter-mined by similar principles.The requirements of industry and medicine for the determination ofsmaller and smaller quantities of copper have long made necessary a moresensitive reagent for this element. Turkington and Tracey 170 have foundthat 1,5-diphenylcarbohydrazide is such a reagent, for it gives an extremelysensitive colour reaction with copper in basic solutions, which obeys Beer’slaw over the range 0.01-0.26 pg. of copper per 3 ml. of final coloured solution.The molar absorbance index of the complex is 158,800 at 495 mp comparedwith 12,700 for the copper diethyldithiocarbamate complex in pentyl alcoholmeasured at 440 my. The rate of the reaction is dependent on pH and inconsequence it is carried out in a buffered solution.An interesting method has been developed by Riedel 171 for the determin-ation of copper in nickel cathodes.From the solution of the sample, all thecopper and some nickel are brought together by extraction withcarbon tetrachloride and cadmium diethyldithiocarbamate. Treatment167 G. S. Spier and J. D. H. Strickland, Analyt. Chim. Acla, 1958, 18, 231.168 Idem, ibid., p. 523.169 H. Lifie, 2. analyt. Chem., 1958, 159, 196.150 R. W. Turkington and F. M. Tracey, Analyt. Chem., 1958, 30, 1699.171 I<. Riedel, 2. anal-vt. Chem., 1957, 159, 25HASLAM AND SQUIRRELL: PHYSICAL METHODS. 417with mercury salt now yields a new aqueous phase containing all the copperand some nickel.Subsequent complexing with ethylenediaminetetra-acetic acid followed by addition of excess of magnesium and cadmiumdiethyldithiocarbamate yields a mixture from which the copper can beextracted quite cleanly as its coloured diethyldithiocarbamate complex byusing carbon tetrachloride as solvent. 2-Furoyltrifluoroacetone forms anintense green chelate with copper which is quantitatively precipitated fromaqueous solution but can be extracted into organic solvents to yield a solutionwhich absorbs strongly at 660 mp. This property is the basis of a methoddescribed by Berg and Day 172 for the rapid and precise determination of aslittle as 1 mg. of copper at dilutions down to 5 p.p.m. Fe, Ce, V, and Ni arethe only serious interferences, since the other metal chelates are colourless,not extracted, or do not absorb at 660 mp.Ziegler 173 has described amethod by which appreciable proportions of copper can be determined byreaction with aminoacetic acid, in the presence of citrate buffer, a t a pH of2 - 9 4 . 1 . The test is fairly specific for copper and the colour of the reactionproduct is measured a t 735 mp.Blundy 174 and Simpson and Blundy 175 have developed interestingsolvent-extraction methods. In the first method, which is concerned withthe determination of chromium in the presence of iron, nickel, uranium, andcopper, the chromium is first oxidised to chromate with ammonium hexa-nitratocerate in hot acid solution. This chromate is then extracted withisobutyl methyl ketone from a solution M in hydrochloric acid.Extractionof the extract with water yields a solution of the chromate, which is thendetermined by the diphenylcarbazide reaction. Blundy and Simpson dealwith the determination of nickel in solutions containing uranium, thorium,copper, iron, and chromium. The nickel is precipitated as its 4-methylcyclo-hexane-l,2-dione complex in the presence of tartaric and thioglycollic acids.The nickel complex is extracted with toluene before absorptiometricmeasurement.A novel use of EDTA has been presented by Lott and Cheng176 in asimple method for the determination of iron in clay and limestone. TheEDTA serves two purposes : it prevents interference due to cations, and alsoproduces a colour with iron in the solution of the limestone sample in thepresence of hydrogen peroxide.The absorbance due to this colour ismeasured at 520 mp. Very good use of the comparatively new reagentfor iron, viz., bathophenanthroline (4 : 7-diphenyl-1 : 10-phenanthroline)previously described by Smith et aZ.,177 has been made by Booth andE ~ e t t . ~ ~ ~ They describe a method for the determination of iron (2-100 p.p.m.) in bismuth in which the iron in the hydrochloric acid solution ofthe sample is first reduced to the ferrous condition by means of stannouschloride. The reagent is then added, followed by a mixture of disodium172 E. W. Berg and M. C. Day, Analyt. Chim. Ada, 1958, 18, 578.173 M. Ziegler, 2. analyt. Chem., 1958, 183, 197.174 P. D. Blundy, Analyst, 1958, 83, 555.175 P.D. Blundy and M. P. Simpson, ibid., p. 558.176 P. F. Lott and K. L. Cheng, Analyt. Chem., 1957, 29, 1777.17' G. F. Smith, W. H. McCurdy, jun., and H. Diehl, Analyst, 1952, 77, 418.1 7 ~ E. Booth and J. W. Evett, ibid., 1958, 83, 80.REP-VOL. LV 41 8 ANALYTICAL CHEMISTRY.ethylenediaminetetra-acetate and sodium citrate. The ferrous complex isextracted with n-hexyl alcohol, and the optical density of the extractmeasured at 533 mp.A new colorimetric reagent for ferric ion in nitric acid solutions (0.3-2.5111) has been described by Holdaway and Willan~.l7~ The reagent,tris-(o-hydroxypheny1)phosphine oxide, gives a violet-red complex withferric ions which obeys Beer’s law over the concentration range 0-02-0.1 mg. of Fe3+ per ml. The only cation known to interfere in thedetermination of iron by this reagent is Ce4+ which should make its useattractive to many analysts.Underwoodlao has described the use of anew reagent, ethylenedi-(o-hydroxyphenylacetic acid), for the spectrophoto-metric determination of iron. The reagent forms a stable red complex withferric iron over a wide pH range and has been used for the determination ofiron in aluminium alloys. Interferences from other elements are no morenumerous than in other methods.Very small amounts of mercury in ores can be determined by a procedureput forward by Michal et aZ.18l After isolation of the mercury by distillationand removal of any free chlorine produced in the distillation by means offerrous sulphate solution, the mercury solution is brought to pH 5.Abenzene solution of mercupral is now added. Mercupral which possessesa strong yellow colour in benzene solution is produced by reaction of tetra-ethylthiuram disulphide with cupric salts. On reaction with mercurycompounds a diminution in yellow colour is produced and this reduction incolour is measured at 4 2 0 4 3 0 mp.An ingenious method has been devised by Sutcliffe and P e a k 182 for therapid determination of nickel in copper-nickel alloys. The alloy is dissolvedin nitric acid-phosphoric acid and, after evaporation and treatment withhydrogen peroxide to avoid the interference of manganese, the absorbancedue to nickel is measured at 3950 A. Another measurement is made a t4900 if to permit a background correction to be made.Cobalt(I1) is theonly cation which interferes in a useful and sensitive colorimetric methodfor the determination of nickel(rr) and copper(I1) described by Jonassenet aZ.183 The complexing reagent employed is disodium ethvl bis-(5-tetrazoly1azo)acetate trihydrate which with nickel gives a product whichabsorbs at 335 and 505 mp, and with copper a complex absorbing a t 268,300, and 535 mp.A new colorimetric method for the determination of small amounts ofselenium in the presence of sulphuric acid has been presented by Danzukaand Ueno.184 The reagent used is 3,3’-diaminobenzidine and interferenceof the sulphate ion is eliminated by the addition of a large excess ofammonium chloride. As little as 5 pg. of selenium per g.of sulphuric acidcan be determined. The problem of the determination of small amounts ofM. J. Holdaway and J. L. Willans, Analyt. Chim. Ada, 1058, 18, 376.I8O A. L. Underwood, Analyt. Chem., 1958, 30, 44.181 J. Michal, B. Pavlikov%, and J. Zjrka, 2. analyt. Chem., 1958, 159, 321.183 H. B. Jonassen, V. C. Chamblin, V. L. Wagner, and R. A. Henry, Analyt. Chem.,IB4 T. Danzuka and K. Ueno, Analyt. Chem., 1958, 30, 1370.G. R. Sutcliffe and D. M. Peake, Analyst, 1958, 83, 122.1958, 30, 1660HASLAM AND SQUIRRELL: PHYSICAL METHODS. 41 9silicon, i.e., down to O.OOOl%, in high-purity iron is one of considerablediffic~1ty.l~~ Chemically, it has been shown that the total silicon can bedetermined by procedures depending on production of a molybdosilicatecomplex and reduction to molybdenum-blue. The latter is measuredabsorptiometrically.The results obtained have now been checked by radio-activation of the iron samples. To the solution of the activated sample isadded silicon carrier. The silicon is isolated as purified silicic acid beforecounting.A very sensitive test for silver, capable of detecting 0.5 pg. per ml.and particularly effective in the presence of copper, has been put forwardby Ciuhandu and Giuran .l86 Copper solution and sodium sulphamido-benzoate are first added to the silver solution, followed by a known amountof sodium carbonate and sodium hydroxide solution and filtration. Thefiltrate retains a definite amount of copper. On reduction of this filtratewith carbon monoxide, a silver sol is produced suitable for photometricmeasurement. 5,7-Di-bromo-8-hydroxyquinoline lS7 appears to be a very satisfactory reagent forthe volumetric determination of tin.At pH 1.0 tin forms a complex withthe reagent which is soluble in isobutyl alcohol to yield a yellow solution,which is measured at 410 mp.Everest and Martin 188 have sought to develop a method for the deter-mination of thorium in medium- and low-grade ores. The method is onein which only one separation is required and which has an absorptiometricfinish in which small amounts of impurities (particularly zirconium) presentin the eluate can be tolerated. Their method involves the separation of thethorium from gross amounts of other elements by elution on a cellulose-alumina column.The thorium is determined by a selective absorptiometricfinish with APANS [1-(o-arsonophenylazo)-2-hydroxynaphthalene-3,6-di-sulphonic acid], mesotartaric acid being used as a masking agent for zirconium.A sensitive extraction method has been put forward by Ziegler et aZ.lB9 for thedetermination of small amounts of titanium in the presence of comparativelylarge amounts of foreign ions, e.g., ferric ions, which must first be reduced.The method depends on the production of the tributylammonium-titanium-sulphosalicylate complex and its extraction by chloroform. The complexis measured at 400 nip. G. Eckert and E. Bauersachs 190 have shown thattungsten, present to the extent of 3 4 % in cathodic nickel, can be deter-mined quite satisfactorily.The method involves a pre-reduction of thesample solution with stannous chloride before completion of the reductionto a definite valency state with titanous chloride. The reduced solution isthen treated with thiocyanate to yield a colour suitable for absorptiometricmeasurement.In a short communication Shibata and Matsumae 191 have outlined a newThe copper in the filtrate stabilises the silver sol.lS5 H. G. Short and A. I. Williams, Analyst, 1958, 83, 624.lE6 G. Ciuhandu and V. Giuran, 2. aizalyt. Chem., 1958, 159, 250.E. Ruf, 2. analyt. Chem., 1958, 162, 9.D. A. Everest and J. V. Martin, Analyst, 1957, 72, 807.M. Ziegler, 0. Glemser, and A. V. Baeckmann, 2. analyt. Chem., 1958, 160, 324.loo G. Ecltert and E. Bauersachs, ibid., 163, 161.ls1 S.Shibata and T. Matsumae, Bull. Chem. SOC. Japan, 1958, 31, 377420 ANALYTICAL CHEMISTRY.colorimetric method for the determination of micro-amounts of uranium.The method is based on the formation of the stable blue uranium complexof neo-thorone (o-arsonophenylazochromotropic acid) in aqueous solutionat pH 6-0. At 600 mp the molar extinction coefficient of this complexis about 25,000. Clinch and Guy lg2 have made considerable improvementsin the thiocyanate method for the determination of this element, and theirprocedure is well suited to the determination of uranium in low-grade oresand in thorium oxide. The uranium is extracted from a solution containingEDTA a t a pH of 3-5-3.9 with a 32.5% v/v solution of tributyl phosphatein carbon tetrachloride.The colour of the extract solution is measuredat 350 mp.A useful method for the determination of vanadium has been putforward by Cozzi and Raspi.lg3 The evaporated vanadium solution isdissolved in acid hydrogen peroxide solution and made anhydrous by theaddition of the reagents (viz., hydrochloric acid, glacial acetic acid, andmethyl salicylate) in solution in acetic anhydride. Under these conditionsa violet-blue colour is formed (Imx. = 565 mp) and as little as 1 pg. ofvanadium can be quantitatively detected. 4-Chlororesorcinol has beenused by Stewart and Bartlet lg4 as a useful reagent for the colorimetricdetermination of zinc. The zinc is separated from interfering cations byextraction as zinc diethyldithiocarbamate with chloroform under controlledconditions and then re-extracted into hydrochloric acid solution.Thezinc4-chlororesorcinol complex is now formed, and the colour which ismeasured at 640 mp obeys Beer’s law over the convenient range 0-1-5p.p.m. of zinc.Colorimetric procedures for the determination of anions include a rapidspectrophotometric method for the determination of microgram quantitiesof chloride in sweat and blood serum which has been described by Gerlachand Frazier.lg5 The prepared sample solution is brought to pH 3.3-3.5 bythe addition of nitric acid and the chloride is caused to react with an excessof mercury(I1) reagent. The excess of mercury(I1) ions is now determinedcolorimetrically by measurement, at 520 mp, of the co-ordination compoundformed by reaction with diphenylcarbazone at pH 3.2, Beer’s law beingobeyed within a concentration range of 0-240 pg.of chloride per 100 ml.Cyanide and thiocyanate in small amounts can often be determinedcolorimetrically by Aldridge’s method. By means of bromine-water thecyanide and thiocyanate are converted into cyanogen bromide which thengives a colorimetric reaction with a reagent containing pyridine and benzidine.Wagner lg6 has shown, in making tests on effluents containing large propor-tions of sulphide, that it is necessary to make an independent preliminarydetermination of sulphide and then to add a calculated excess of bromine inpotassium bromide before proceeding with the colorimetric reaction for totalcyanide plus thiocyanate.The cyanide is determined independently afterdistillation from acid solution.ls2 J. Clinch and M. J. Guy, Analyst, 1957, 82, 800.19s D. Cozzi and G. Raspi, Analyt. Chim. Ada, 1957, 17, 590.ls4 J. A. Stewart and J. C. Bartlet, Analyt. Chem., 1958, 30, 404.195 J. L. Gerlach and R. G. Frazier, ibid., p. 1142.196 F. Wagner, 2. analyt. Chem., 1958, 162, 106HASLAM AND SQUIRRELL: PHYSICAL METHODS. 421A rather novel procedure for the colorimetric determination of fluorideion has been described by Yasuda and Lambert.lg7 They use the classicalthorium-Alizarin Red S reagent, but support it on previously preparedsquares of filter paper. The solution under test is freed from interfering ionsby passage through an ion-exchange resin, and the pH adjusted to 3.5 priorto treatment with silver nitrate at boiling temperature and readjustment ofpH to 7.0.The prepared reagent paper is now dropped into an aliquotpart of the solution, whereupon anion exchange occurs on the paper andcolour proportional to the fluoride present is liberated into the solution. Thiscolour is measured in the usual way a t 520 mp.In the colorimetric determination of phosphate the reduction of molybdo-phosphate by stannous chloride to a molybdenum-blue colour, has notproved entirely satisfactory in the hands of some investigators. Whensolid ascorbic acid is used,198 and the reduction carried out at boiling point,the process is apparently quite satisfactory. The colours obtained are stableat room temperature and may be measured either visually or instrumentally.Ruf lg9 has made a comprehensive study of the pH and other conditionswhich are required in the determination of both silicic and phosphoric acidsby photometric procedures involving the use of ammonium molybdate andsodium molybdate.He draws attention to the error which may arise if thesilicic acid is present in solution in polymeric forms.Increasing use is being made of ultraviolet absorptiometric methodsfor the determination of such elements as iron and lead.determine iron in high-purity bismuth by solution of the sample in nitricacid, then conversion into the chlorides and measurement at 390 mpof a solution of the chlorides in constant-boiling hydrochloric acid. Atthis wavelength the ferric iron absorbs and the interference of bismuth isnegligible.The infrared spectra of the metal anthranilates such as thoseof CuZ+, Ni2+, Zn2+, Cd2+, Fez+, and Mn2+ are often very characteristic andparticularly so in the region 9-10 t/-. With binary mixtures of theseanthranilates of known weight, it is often possible to determine the propor-tions of the individual aiithranilates by measurement of the infrared spectraa t appropriate wavelengths.201 Mutschin and Maennchen 202 have made athorough study of the infrared spectra of diphosphoric acid, mono- anddi-sodium diphosphate, trisodium diphosphate as both mono- and mono-hydrate, tetrasodium diphosphate both anhydrous and as decahydrate, andtetrapotassium diphosphate in sodium chloride, potassium bromide, andczesium bromide solutions.Corbridge and Tromans 203 have indicated that the greater resolutionand increased definition of the diffraction lines recorded with a Guiniertype X-ray focusing camera, compared with those obtained with the usualDebye-Scherer type camera, offer several advantages for the identificationHigh and PlacitolQ7 S.K. Yasuda and J. L. Lambert, Analyt. Chem., 1958, 30, 1485.lQ8 D. N. Fogg and N. T. Wilkinson, Analyst, 1958, 83, 406.lQS E. Ruf, 2. analyt. Chem., 1958, 161, 1.J. H. High and P. J. Placito, Analyst, 1958, 83, 522.201 R. Neeb, 2. analyt. Chem., 1958, 161, 161.202 A. Mutschin and K. Maennchen, ibid., 160, 81.209 D. E. C . Corbridge and F. R. Tromans, Analyt. Chem., 1958, 30, 1101422 ANALYTICAL CHEMISTRY.and analysis of certain compounds, for example, phosphates.They presentthe X-ray data thus obtained for 60 crystalline sodium phosphates.Absorption Spectroscopy (Organic) .-Organic applications of absorptionspectroscopy have been numerous and two methods which, in conjunctionwith one another, can be used for the quantitative differentiation of tertiaryamines, amine salts, and/or quaternary amines have been put forward bySass et aZ.204 The first method is based on the reaction of amines withaconitic anhydride in toluene solutions at 100" to produce a coloured complexmeasured at 500 mp. The second procedure is similar but uses chloranilas reagent to produce a green colour measured at 610 mp. The sensitivitiesof the two methods are 3 pg.and 50 pg. per ml., respectively.A most useful procedure for the determination of small amounts ofthiourea or nitrite has been developed by Hutchinson and Boltz206 basedon the reaction of the two compounds to form thiocyanic acid. An excessof the one reagent is added to the solution containing the other being deter-mined, and the thiocyanic acid thus produced is measured by the ferricthiocyanate complex colour reaction. Use of the stable free radical a-di-phenyl-p-picrylhydrazyl has been made by Blois 2M for the determinationof antioxidants. The free radical in solution shows a deep violet colourwith an absorption band at 517 mp. This absorption and colour is due toan odd unpaired electron in the structure of the radical, and when thiselectron becomes paired owing to reaction with an antioxidant, there is acorresponding decrease in colour of the solution.This is the basis of themethod of antioxidant determination.The well-known reaction of esters of carboxylic acids with hydroxyl-amine hydrochloride giving hydroxamates which yield coloured reactionproducts with ferric salts has been studied in some detail by W. Pi1z.207An agreed procedure has been developed for the examination of esters ofacetic acid, i.e., with strict attention to the conditions required for thereaction with hydroxylamine, addition of the buffer, and colour development.Carbon monoxide can be determined in air by its reaction with the silvercompound of 9-sulphamidobenzoic acid in alkaline solution , with whichit yields a coloured silver solution.20* The range of the test has beenextended to cover 0.001-2%.Up to 06%, the extraction of the solutionis measured a t 420 mp, but above this concentration the measurement ismade at 610 mp.The iodoform reaction has been utilised by Takayama2OQ in a usefulmethod for the determination of small amounts of acetone in methylmethacrylate monomer. The acetone-containing sample is added to ahypoiodite solution, the colour of excess of iodine is discharged with thio-sulphate, and the iodoform formed extracted into chloroform solution.The absorbance due to iodoform is measured as 347 mp, and by referenceto an appropriate calibration curve the acetone content of the sample is204 S.Sass, J . J. Kaufman, A. A. Cardenas,and J.J.Martin, Analyt. Chem., 1958,30,529.205 K. Hutchinson and D. F. Boltz, ibid., p. 54.206 M. S. Blois, Nature, 1958, 181, 1199.207 W. Pilz, 2. analyt. Chem., 1958, 162, 81.208 G. Ciuhandu, ibid., 1958, 161, 345.2os Y. Takayama and F. Tokiwa, Bull. Chem. SOC. Jafian, 1958, 31, 369HASLAM AND SQUIRRELL: PHYSICAL METHODS. 423calcuIated. As little as O - O O l ~ o of acetone in methyl methacrylate monomercan be determined by this method.The availability of high-resolution spectrophotometers covering thenear infrared region has provided the analyst with a new tool. A paperby Goddu 210 has evaluated this region for the determination of unsaturationin organic compounds. He describes how both terminal methylene groupsand cis-double bonds can be selectively determined in a wide variety ofmaterials, and explains that other types of unsaturation do not interferewith terminal methylene determination. The bands at about 1.62 and2.10 p are used for determining terminal unsaturation and a band at 2-14p for cis-unsaturation.The potassium bromide disc technique has been adapted to suit micro-amounts of compounds which are soluble in organic solvents but insolublein water.211 Freeze drying is used exclusively for the preparation of thesample suspension in potassium bromide and precautions are taken to avoidcontamination of the mixture; the die for casting the disc is lubricatedwith graphite.The reproducibility of the discs prepared in this way issuch that the method is suitable for the quantitative infrared micro-analysisof binary mixtures of complex organic compounds.A semi-quantitativemethod for the determination of t-butyl groups has been prepared by Boogand I<unstJ212 mainly, they suggest, for the elucidation of the structure ofpure compounds. It appears that an infrared band near 1250 cm.-l andRaman bands near 1250 cm.-l and 745 cm.-l should be regarded as character-istic of the neopentyl group, while a Raman band near 925 cm.-l occurs withall types of t-butyl group. The sum of the intensities of the three Ramanbands can be plotted against concentration of t-butyl groups and gives areasonable straight-line relationship.Bartlet and Mahon 213 have described a useful application of differentialinfrared spectroscopy for the identification of vegetable oils and for thedetection of adulteration of one oil with another. The differential technique,for example, picks out the small differences in the spectra of olive and rape-seed oils, particularly in the 1200-900 cm.-l region, and a 10% contaminationof rape-seed oil in olive oil is readily detected.The effect of refining on thedifferential spectra of the oils has been studied and the principle of themethod of analysis extended to many vegetable oils, animal fats, andhydrogenated fats. The effect of the unsaturation in the oils on thisdifferential spectrum is also discussed.The infrared analysis of plasticisers obtained from plastic materials isfacilitated by a method due to Cachia et d 2 1 4 in which the mixed plasticisersare separated by column chromatography.The column is packed withsilica gel and Celite and the separations are achieved by elution with solventsincluding carbon tetrachloride-isopropyl ether and benzene-isopropyl210 R. F. Goddu, A~zalyt. Chem., 1957, 29, 1790.211 H. P. Schwarz, R. C. Childs, L. Dreisbach, S. V. Mastrangelo, and A. Kleschick,212 W. Boog and E. D. Kunst, Speclrochirn. Acta, 1957 Suppl., p. 568.*13 J. G. Bartlet and J. H. Mahon, J . Assoc. Ofic. Agric. Chern., 1958, 41, 450.214 M. Cachia, D. W. Southwart, and W. H. T. Davison, J . AppL Chem., 1958, 8,Appl. Spectroscopy, 1958, 12, 35.291424 ANALYTICAL CHEMISTRY.ether in different ratios. The component plasticisers are recovered byevaporation of the eluted fractions.Emission Spectroscopy .-The speed and direct applicability of spectro-graphic and flame photometric procedures have been fully exploited during1958.Fassel and Gordon 215 have devised an emission spectrophotometricmethod for the determination of oxygen in titanium, the precision of whichis said to be comparable to the vacuum fusion or bromination procedure.The method is based on direct-current excitation of a special electrodeassembly which provides a molten platinum bath after the arc is initiated.The oxygen in the titanium is rapidly liberated from this bath into an argonatmosphere, and the intensity ratio of the line pair (0) 7771 &(A) 7891 A isrelated to the oxygen content of the titanium sample. A new method hasbeen presented by Rusanov and Khitrov 216 for the spectrographic analysisof powder samples.The sample is carried in a stream of air and thusintroduced evenly into the discharge zone of a horizontal carbon or metallicA.C. arc which is initiated by a high-frequency discharge. The procedureis speedy and gives an increase in sensitivity and reproducibility. Electrodereactions are prevented, as well as fractional evaporation of the differentelements. Thus, because the radiation remains effectively constant themethod offers great possibilities for direct photoelectric measurement of theline intensities.An attempt has been made to devise rapid methods of analysis whichwill provide information about the mineral matter associated withAlthough alkalis are determined by a flame-photometric procedure, andsilica often by a gravimetric process, the main feature of the test is itsspectrographic nature.An appropriate solution of the coal ash or the wetdigestion product of a coal is treated with a chromium solution as internalstandard. Portions of this solution are absorbed by porous graphiteelectrodes and spectra are produced by the use of an uncontrolled condensed-spark discharge. Examination of the relative intensities of the appropriatelines in the spectrum enables one to obtain information about the content ofmajor elements such as aluminium, iron, titanium, calcium, magnesium, andmanganese very quickly; copper, strontium, and barium may also bedetermined.A paper by Fabian and Bode 218 may be of far-reaching importance inflame-photometric work.It is concerned with the flame-photometricdetermination of copper (a) in aqueous solution and (b) when the copper ispresent as a complex dissolved in an organic solvent immiscible with water.The copper complexes dealt with are those with sodium diethyldithio-carbarnate, cupferron, 8-hydroxyquinoline, and salicylaldoxime. With theappropriate solvent it is often possible to increase the sensitivity of theflame-photometric test very considerably as compared with the result inaqueous media. The interference of aluminium, silicic acid, and phosphoricacid in the flame-photometric determination of calcium has been largely215 V. A. Fassel and W. A. Gordon, Analyt. Chem., 1958, 30, 179.216 A. K. Rusanov and V.G. Khitrov, Spectrochim. Acta, 1958, 10, 404.217 K. Dixon, Analyst, 1958, 83, 362.H. Bode and H. Fabian, 2. analyt. Chem., 1958, 162, 328HASLAM AND SQUIRFCELL. 425overcome by the addition to the test solution of an amount of strontium 219which bears some relationship to the amount of calcium being sought, beforeflame-photometric examination. Fornwalt 220 has studied the effect of theconcentration of methanol upon the flame emission of the 352.5 mp nickelline in a 1 : 1 methanol-water solution, and also the effect of nickel on theflame emission of boron in 1 : 1 methanol-water solution for the oxide bandsystem at 518 mp. As a result, a rapid flame-photometric method for theestimation of nickel and boron in nickel-plating solution has been worked out.The interference of anions in the flame-photometric determination ofmetals is a serious one.Yofi: and Finkelstein z21 have overcome thisdifficulty due to phosphate and sulphate in the determination of calcium,in which these anions cause a fall in intensity. The addition of lanthanumor iron to the sample, it is postulated, replaces the calcium present as phos-phate or sulphate, allowing it to ionise to give the full intensity. Withoutthis addition, the calcium does not form ions from phosphate or sulphatesolutions even at the high temperature of the flame.The close similarity between the mass spectra of alcohols and hydro-carbons makes the determination of one in the presence of the otherimpracticable by mass-spectrographic methods. Langer and his co-workers 222 have overcome this difficulty by treating the mixture of hydro-carbon and alcohol with hexamethyldisilazane, which converts the alcoholinto its methylsilyl derivative, leaving the hydrocarbon unchanged.Thisderivative can be readily determined without prior separation, even in thepresence of water which is converted into hexamethyldisiloxane and has itsown distinctive mass-spectral peak. The authors suggest that the methodmight be useful for the determination of glycols, amines, and phenols.Atomic absorption spectroscopy is likely to be very useful for thedetermination of magnesium in plant materials, soil extracts, etc., as hasbeen shown by J. E. Allan; 223 Le., it will enable magnesium tests to becarried out with the same ease and reliability as the flame-photometricdetermination of sodium, potassium, and calcium.The principle of thetest is that a lamp is used as a light source which emits the line spectra ofmagnesium. The sample is sprayed into a flame alongside so as to providea reproducible and clearly defined cloud of atoms, and the light absorbed atthe wavelength of the resonance line by the unexcited atoms of magnesiumis measured.Micro-analysis.-The micro-determination of chlorine has received muchattention throughout the year. In a paper by Seligson et al. 224 an electro-metric method for the rapid titration of chloride in serum and other biologicalfluids is described. This method, which appears accurate, precise, andspeedy, utilises a silver-silver amalgam electrode pair and titration withsilver nitrate solution is carried out in sulphuric acid media, delivery of219 W.Schuhknecht and H. Schinkel, 2. Analyt. Chem., 1955, 162, 266.220 D. E. Fornwalt, Analyt. Chim. Acta, 1957, 17, 597.221 J. Yofi: and R. Finkelstein, Analyt. Chim. Acta, 1958, 19, 166.232 S. H. Langer, R. A. Friedel, I. Wender, and A. G. Sharkey, jun., Analyt. Chem.,223 J . E. Allan, Analyst, 1958, 83, 466.224 D. Seligson, G. J. McCormick, and K. Sleeman, CEinicaZ Chem., 1958, 4, 159.1958, 30, 1353426 ANALYTICAL CHEMISTRY.titrant being made from a Scholander micro-burette. A special electrodecircuit is described incorporating bucking potential controls by which theindicating galvanometer can be adjusted to read zero in the presence of aknown excess of chloride and 100 in the presence of a similar excess of silverions; with this means of standardisation the end-point occurs at a readingof 50 irrespective of small peculiarities in different electrode pairs.J.H. Cannon,225 drawing largely from the previous work of Kirk (1950),Schwartz (1933), Malmstadt and Fett (1954), and Baledel and Malmstadt(1952), has described an automatic method for chloride titration by whichas little as 10 pg. of chloride can be titrated with a precision of h0.2-0.3 pg. in a total titration time of 2 minutes. This apparatus consistsessentially of a silver-silver chloride electrode system and a horizontalcapillary burette fitted with automatic stopping devices controlled by anelectronic end-point detector.The titration is carried out in sulphuric acidsolution.For the micro-determination of chlorine in organic compounds also con-taining nitrogen and sulphur, Makineni and his co-workers 226 prefer to usea direct titration method which can be applied to wet or dry combustion pro-ducts. The halide is precipitated with silver, and the precipitate washed freefrom acids and from excess of silver solution before reaction with ion-exchangeresin Amberlite IRA 120 (H) at 70". The hydrochloric acid thus liberatedis titrated with standard mercuric nitrate solution, with use of diphenyl-carbazone indicator, or with standard silver nitrate solution by a conducto-metric procedure. A. R. Panicker and N. G. Banerjee227 have developedan ingenious method for the rapid determination of carbon and hydrogen inhighly volatile combustible organic liquids such as benzene and toluene.The sample is weighed out accurately in a three-bulb Pyrex-glass capillarywith open capillary end, then volatilised from this in a stream of nitrogenbefore combustion in added oxygen. Any residue in the sample tube isfinally combusted completely at an elevated temperature in a stream ofoxygen.Co,O, has been used by VeEera, Snobl, and Synek2= as a com-bustion catalyst for the micro-determination of carbon and hydrogen inorganic compounds; they report this oxide to be superior to all othercatalysts previously used, and it has the advantages of universalapplicability, simplicity, reliability, and speed.Results obtained by usingthis catalyst have been shown to have an accuracy equal to that obtainedby other methods.The method of preparation and typical analyses of a new multi-purpose microchemical standard have been described by Smith.229 Thecompound, 5-chloro-P-hydroxy-3-methoxybenzylisothiuronium phosphate,contains seven elements in reasonable quantities and in addition has amethoxyl group.Feigl and his co-workers have continued their work on spot-test analysis.They have shown that chloramine-T can be readily detected in the presence225 J. H. Cannon, J . Assoc. Ofic. Agric. Chenzists, 1958, 41, 428.226 S. Ma.kineni, W. McCorkindale, and A. C. Syme, J. AppZ. Chew., 1958, 8, 310.227 A. R. Panicker and N. G. Banerjee, Analyst, 1958, 83, 296.228 M.Teeera, D. Snobl, and L. Synek, Microchim. Acta, 1958, 9.229 W. H. Smith, Analyt. Chem., 1958, 30, 149HASLAM AND SQUIRRELL. 427of hypo~hlorite.~~~ The hypochlorite is first converted into chloride byreaction with hydrogen peroxide and zinc chloride solution. An alcoholicsolution of thio-Michler's ketone is now added; in the presence of chlor-amine-.r a blue oxidation product is produced. Very dilute solutions ofchloramine-T are without oxidative action on tetrabase ("""-tetra-methyldiaminodiphenylmethane) In the presence of iodide ions, how-ever, the base is catalytically oxidised to a blue product. This test is usefulin the detection of very small amounts of iodide in various waters and inorganic substances. In the biochemical field the same workers have alsodeveloped new spot tests for vitamin B, 232 and for ephedrine.2BA test which can be regarded as specific for orthophosphate has beenbased by Robinson and West 234 on the reaction of orthophosphate witho-dianisidine molybdate in acid solution, followed by addition of hydrazinehydrate.In this spot test a brown precipitate is fornied upon the additionof o-dianisidine molybdate if orthophosphate is present. The conversionof this brown precipitate into a blue colour when hydrazine hydrate is addedserves to confirm the presence of phosphate. Although specific for phos-phate, certain ions, particularly sulphide, complicate the interpretation ofthe results.Radio-chemical Methods-On the expanding subject of radio-chemicalanalysis, a critical survey of the factors which influence the accuracy ofthe determination of potassium in solution by the Geiger-Muller rapidradiological test has been made by H.Dresia and R. B e ~ k m a n n . ~ ~ ~ Trace-impurity determination by radio-chemical methods has been used byThompson et aLB6 in the examination of ultra-pure silicon; 29 elementshave been determined by using the combined technique of y-spectrometryand radiochemical separation with @-counting of the fractions. Theimpurities are classified into long-lived and short-lived elements accordingto the half lives produced from neutron activations, and the method is statedto be applicable to determinations in the parts per lo9 range.In a method for the determination of uranium dioxide in stainless steel,Silverman et aZ.237 have used a direct fluorescent X-ray procedure on aperchloric acid solution of the sample.The intensity of the radiation ismeasured with a scintillation counter, and strontium is used as an internalstandard. The determination suffers no interference from large amounts ofiron, chromium, and nickel, and reasonable accuracy is obtained in a four-pair count which can be completed in 12 minutes.Apparatus.-The development of new instruments and improved designof apparatus has continued to provide the analyst with the means for morerapid and accurate methods of test. Examples of a few of these develop-ments are described below.The nature of these complications is described.230 F. Feigl and R. A. Rosell, 2. analyt. Chem., 1958, 159, 335.231 F. Feigl and E. Jungreis, ibid., 1958, 161, 87.232 Idem, Clinica Chim. Acta, 1958, 3, 399.233 F. Feigl and E. Silva, J . Amer. Pharm. Assoc. (Sci. Edn.), 1958, 4'4, 460.234 J. W. Robinson and P. W. West, Microchim. J., 1957, 1, 93.236 H. Dresia and R. Beckmann, 2. analyt. Chem., 1957, 159, 1.2313 B. A. Thompson, 13. Ri. Strause, andM. B. Leboeuf, Analyt. Chem., 1958,30,1023.237 L. Silverman, W. W. Houk, and L. Moudy, ibid., 1057, 29, 1762428 ANALYTICAL CHEMISTRY.An apparatus, simple in design yet efficient and adaptable in use, hasbeen made by Strain 238 for the continuous electrochromatography of variousinorganic ions. The chromatographic medium is a tapered sheet of soft,thick, filter-paper supported on a block of modified polystyrene foam.A loose curtain of polyethylene film serves to cut evaporation losses ofsolvent.P U C & - , ~ ~ in connection with experiments on the electrophoresis ofhalogen complexes of many metals, has designed a heavy-current and high-tension electrophoresis apparatus. This apparatus, which is for use withfilter-paper, avoids the difficulties that occur in electrophoresis in baseelectrolytes that are concentrated and show greater electrical conductivity.The zones migrate along the paper strips with constant speed, and con-sequently mobility measurements can be made. With the apparatus themobility dependence of the chloro-complexes of Hg(n), Cd(II), Pb(II), andCU(II) on concentration and volume of solution used have been investigated.The application of paper electrophoresis in the field is now possible by usingthe portable apparatus described by Marini-Bettolo and Cochfr~goni.~~~The apparatus, including all accessories, is only 35 x 27 x 16 cm. in size,weighs 6.650 kg., and is powered by a 360-volt dry battery generator adjust-able by means of a potentiometer. It has been used in preliminary fieldanalysis of curarising alkaloids in plants and curares.Paulik, Paulik, and Erdey 241 have constructed a new type of apparatuswhich they described as a “ derivatograph.” With this apparatus they areable to carry out differential thermoanalysis, thermogravimetric, anddifferential thermogravimetric investigations simultaneously, and moreoverregister the results automatically. Application of the procedure enables theauthors to make very interesting observations on the mineral composition of,e.g., bauxite. The same authors a2 have extended the method to the gravi-metric micro-distillation of quite small amounts of mixed liquids. In thisway, they are able to determine the composition of such mixtures as benzene,ethyl alcohol, and water. A further aid to those chemists concerned withthermal analysis has been described by Williams et aLN3 in the form of aversatile differential thermal analysis apparatus. A constant reproducibletemperature gradient is established in a heavy-walled metal tube. Thesample holder containing the base and differential thermocouples is pulledin either direction through this tube at speeds required to produce thedesired temperature change.Miller and Deford 244 have made a useful contribution in the design of asimple apparatus for the quantitative hydrogenation of unsaturated com-pounds, using electrically generated hydrogen. The current used to producethe hydrogen by automatic electrolysis is measured on an electroniccoulometer which also serves as a recorder. The great advantage of the238 H. H. Strain, Analyt. Chem., 1958, 30, 228.239 2. Puck-, Analyt. Chim. Acta, 1957, 17, 476.240 G. B. hlarini-Bettolo and J. S. Cochfrugoni, J . Chromatog., 1958, 1, 182.241 F. Paulik, J. Paulik, and L. Erdey, 2. analyt. Chem., 1955, 160, 241.242 Idem, ibid., p. 321.243 D. D. Williams, R. D. Barefoot, and R. R. Miller, Analyt. Chem., 1958, 30, 492.244 J. W. Miller and D. D. Deford, Analyt. Chem., 1958, 30, 295HASLAM AND SQUIRRELL. 429method is that the accuracy appears independent of the size of sampleavailable for test.Those concerned with the evaluation and characterisation of smallamounts of materials by means of their melt viscosities will be interested ina simple melt viscometer designed by Small 2G which allows viscosities fromlo4 up to at least lo9 poises to be measured on 0.1 g. of material or less.Details of construction, use, and calibration are given.Finally, an electronic spectro-analyser has been developed by theFederal Telecommunication Lab0rat0rie.s.~~~ This instrument combines arecording device which encodes the spectral data on magnetic tape or paperand a reference " library " of infrared data on possible constituents, alsorecorded in numerical form on digital tape. The instrument also incorporatesa high-speed digital computer to calculate and record the analysis directly.J. HASLAM.D. C. M. SQUIRRELL.245 P. A. Small, J . Polymer Sci., 1958, 28, 223.240 Chem. Eng. News, 1958, 36, 62