摘要:

May, 19511 EXAMINATION OF DETERGENT PREPARATIONS 279 Analytical Methods Committee REPORT PREPARED BY THE SOAPLESS DETERGENTS SUB-COMMITTEE Examination of Detergent Preparations THE Analytical Methods Committee has received from its Sub-Committee on Soapless Detergents the following information based on its work to explore the advisability of standard methods for these preparations, and its publication has been duly authorised. The members of the Sub-committee were: W. H. Simmons (Chairman), H. E. Cox, C. G. Daubney, S. R. Epton, P. J. C. Haywood, A. MacArthur, K. A. Williams, D. C. Garratt (Honorary Secretary), and valuable assistance was given by W. J. Dwerryhouse and L. E. George of the Ministry of Food, Oils and Fats Division. The Sub-committee was formed primarily to consider whether in view of the Soap Substitutes (Labelling and Prices) Order, 1943 (S.R.& O., 1943, No. 638), standard methods of analysis of soapless detergents were possible or necessary. Under the Soap Substitutes Order certain standards were laid down by the Ministry of Food for the formulation of these products, preliminary to the granting of a licence to label them for retail sale. Briefly these standards consisted of a specified minimum of active detergent and a maximum percentage of alkali and for the pH value when the product was claimed to be suitable for washing woollens, silks and other delicate fabrics. The definition of “active detergent’’ is capable of very wide interpretation, but general agreement was reached that it referred to the water-soluble organic compounds capable of reducing the surface tension of water, in the molecule of which a hydrophilic group or groups is attached to a hydrophobic alkyl or alkyl-aryl chain.For quantitative purposes the descrip- tion should refer to the anhydrous organic portion of the product, together with inorganic radicles chemically combined with it. The Sub-committee do not wish to put this description forward as a definition, since to do so would be of little value unless the definition could be correlated to a performance test-and agreement on a specification for this is a matter of considerable difficulty. The various types of modern detergents generally commercially available in this country have been listed. This list is in no way complete but it should be useful to illustrate the difficulties that were manifested in the attempt to collate the analytical data available as a preliminary to investigating a method applicable as a standard procedure to all products.PRODUCTS CONTAINING FATTY ACID RESIDUES- Igepon A type . . .. . . R-CO-O-C,H4-S0,Na (I.G.) Igepon T type . . .. . . R-CO-N(CH,)-C,H,-S0,Na (I.G.) .. . . R-CO-N(CH,)-CH,-C0,Na (I.G.) Medialan A type . . - Lissapol LS . . .. .. R-CO-NH<?CH, (1;C.I. Ltd.) S0,Na Breeze (active constituent) . . R-CO-NH-C.,H4-O-CH,S0,Na (Levers, U.S.A.)280 EXAMINATION OF DETERGENT PREPARATIONS [Vol. 76 The fatty acid residues used in the foregoing types are varied, from around C,, upwards, depending on the particular use for which the detergent is intended. SULPHATED PRIMARY ALCOHOLS- These are products of the general type R-0-SO,Na, where R generally ranges from C,, to C18.Commercial examples are Lissapol C. (I.C.I.), Gardinols (I.G.), Blansols, Empicols L and LQ (Marchon Products Ltd.) and Dreft: (Hedleys). The triethanolamine salts also appear as shampoos, e.g., in Drene. SULPHATED SECONDARY ALCOHOLS- These are products of the type R-CH(0--SO,Na)-R’. Teepol (Shell Chemicals Ltd.) and Comprox (Irano Products Ltd.) are example:;. Comprox has also been known as Iranopol and By-Prox. ALKANE SULPHONATES- General formula: R-CH(S0,Na)-R’. Examples are MP.189 (Du Pont) and Mersolats (I.G.). ALKYL ARYL SULPHONATES- For example : “-Q :SO,Na and similar products based on alkylated naphthalene sulphonates. Commercial examples are Perminal BX and Dispersols L and T (I.C.I.) and Nansa (Marchon Products Ltd.), some of the Nacconols (National Aniline 81 Chemical Co.), Santomerses (Monsanto), Igepon NA and sulphonated Oronite.Petroleum sulphonates are generally of this type. ESTERS AND ETHERS- Polyglyceryl stearates and ethylene oxide condensates of fatty alcohols, phenols, amides, etc., containing, for instance, a long polyethylene glycol chain. Lubrol W and Lissapol N (I.C.I.), Igepals (I.G.), Stergene (Domestos Ltcl.) and some of the Tritons (Rohm & Haas) are commercial examples. SULPHOSUCCINIC ESTERS- For example, sodium dioctyl sulphosuccinate C,H,,-OOC.CH.SO,Na I C,H,,-OOC.CH, as commercially represented by the Aerosols OT, etc. (American Cyanamid). SULPHATED VEGETABLE OILS, FATTY ACID ESTIERS, ETC.- For example : as commercially represented by the long-established Turkey Red Oils (sulphated castor oil) and Calsolene Oil HS (I.C.I.).All the foregoing types are anionic surface active agents, except those of the ester and ether type, which are classed as non-ionic agents. In addition there are the cationic detergents. These are quaternary ammonium com- pounds, such as cetyl trimethylammonium bromide. Cetavlon, Lissolamine A, Cirrasol OD (I.C.I.) and cetyl pyridinium bromide, Fixanol C (I.C.I.), are commercially available types. All these various types of surface active agents, with possibly the present exception of the cationic types, may appear either singly or admixed with each other and also mixed with other detergent “boosters,” solubilising agents, etc.May, 19511 EXAMINATION OF DETERGENT PREPARATIONS 281 With the rescinding of the Soap Substitutes Order in November, 1949, the need for standard methods of analysis became less urgent. This was fortunate, as the Sub-committee had reached the conclusion that, with our present knowledge and the diversity of materials grouped together under this class of product, a method of quantitative determination suitable for all was impracticable.In fact, it is doubtful whether much progress can be made in devising accurate methods of analysis unless each product is considered as an individual rather than as one of a type. Hence the Sub-committee consider that it is unable to recom- mend any standard method for the analysis of these detergents. Throughout the work the Sub-committee has had unstinted assistance from those of its members who are connected with commercial firms manufacturing particular products (Shell Refining & Marketing Co., Ltd., Monsanto Chemicals Ltd.and Imperial Chemical Industries Ltd., Dyestuffs Division), also from Mr. C. G. Daubney of the Department of the Government Chemist and from the Ministry of Food, Oils and Fats Division. It was obvious that this analytical information would be of value to analysts if published collectively. But it must be emphasised that a particular quantitative method is only recommended for a specified product, although it may be found that many of the methods are of wider or even of general applicability for products of the same class. QUALITATIVE TESTS A qualitative examination to ascertain the presence of one or more of the under-noted classes of substances is a primary necessity in dealing with a detergent of unknown com- position.True soap (fatty or resin acids with alkali, ammonium or ethanolamine base). Alkali carbonate, phosphate or silicates ; abrasives and fillers. Synthetic organic detergents of the sulphated or sulphonated type. Sulphated glycerides. Polyethylene glycol derivatives. Quaternary bases. Methyl and carboxymethyl cellulose. Naphthenic acid soaps. To this end a few simple tests can be applied; the following procedure will give useful information. A portion of the sample is taken up in water and any insoluble matter removed by filtration or centrifuging to give the test solution. (i) Some of the test solution is boiled.The formation of a gel may indicate cellulose ethers, see (vii); separation of an oil may indicate polyethylene glycols. (ii) A portion of the test solution is tested for alkalinity and, if alkaline, made acid to methyl red. The separation of fatty acids indicates soap (note that silica may also separate here). If, after removal of any fatty acid, the solution still froths on shaking there is pre- sumptive evidence of a synthetic detergent. (iii) If a synthetic detergent is indicated, chloroform and a few drops of methylene blue solution are added and the mixture is shaken. If the chloroform is coloured blue, a sulphated or sulphonated product is indicated. Absence of colour in the chloroform suggests a poly- ethylene glycol or cationic product or both.(iv) If a polyethylene glycol product is indicated, 5 ml of a 1 per cent. solution of the glycol in water (Le., test solution) are treated with 5 ml of thiocyanate reagent (174 g of ammonium thiocyanate and 2-8g of cobalt nitrate per litre) at room temperature. After 2 hours the colour of the liquid (not of any precipitate) should still be blue-violet if a poly- ethylene glycol is present. Vigorous shaking is necessary. ( v ) A fresh portion of the test solution is boiled under a reflux condenser with N sodium hydroxide for half an hour. An increase in the fatty acid liberated on acidification compared with that from the untreated material indicates a sulphated glyceride. One is shaken with chloroform and methylene blue, the other with chloroform and bromophenol blue.Quaternary bases cause the bromophenol blue to colour the chloroform but the methylene blue remains in the aqueous phase. For this test a strong electrolyte must be present. (vii) Methyl cellulose and carboxymethyl cellulose can sometimes be detected in pre- parations by their typical fibre-like structure under the microscope. To distinguish between (vi) Two further portions of the test solution are taken.282 EXAMINATION OF DETERGENT PREPARATIONS [Vol. 76 them, the following tests can be applied to suitably purified aqueous extracts after concentra- tion under reduced pressure or at a low temperature (not exceeding 50° C). The reagents are added to the concentrate. Fehling’s copper Aluminium sulphate, Substance Boiling reagent 10 per cent. Methyl cellulose .. . . Coagulation No reaction No reaction Carboxymethyl cellulose . . No reaction Heavy, pale blue ppt. Gelatinous white ppt. A portion of the aqueous extract is evaporated to dryness on a watch glass. Saturated lithium chloride solution is added and mixed intimately with the film of residue. Addition of 0.01 N iodine solution produces a red colour with methyl cellulose and a blue colour with carboxyl cellulose. There is little information in the literature o n the qualitative examination of detergent preparations, but the schemes of analysis set out by Linsenmeyerl and Van der Hoeve,2 although they have not been tried out fully, will be found useful and might well form the basis for further work. QUANTITATIVE METHODS OF ANALYSIS Although quantitative methods are generally given in outline only, the Sub-committee decided to republish in extenso those met hods that have been used satisfactorily for certain surface active materials in the laboratories of its various members.These methods appear in the appropriate places in the text. TOTAL ORGANIC MATTER- (a) A given weight of sample is extracted by ‘boiling with alcohol under a reflux condenser or by means of a Soxhlet extractor. If inorganic chloride is present in the sample, some may be extracted by the solvent. It should be determined in the usual way and its weight deducted. (b) A given weight of sample is dissolved in dilute acid and extracted with ether in a Werner - Schmidt tube. Other solvents that dissolve less water, e g . , methyl isobutyl ketone, can be used.The extract is evaporated, the residue freed from traces of moisture by warming with a little acetone, dried at as low a temperature as possible and weighed. Re-extraction of the dry residue with dry ether is sometimes advisable. This method gives a measure of the total organic matter as sodium salt or as free acid and includes neutral bodies. See also, “Official and Tentative Methods of the American Oil Chemists Society,” Section F.3 SOAP - DETERGENT MIXTURES- Berkowitz4 describes a method based on a difference figure, whereby the percentage of synthetic detergent = per cent. alcohol-soluble matter - (per cent. soda soap plus per cent. fatty matter plus per cent. NaCl in alcohol-soluble matter). ANIONIC DETERGENTS- A given weight of the organic matter extracted from the original sample is dissolved in water and a suitable aliquot titrated with standard 0-004 M cetyl pyridinium bromide with methylene blue as indicator. The following method is described by Epton5- Ten millilitres of an approximately 0.005 Ad solution of the anion-active material is transferred by means of a pipette to a 250-ml stoppered reagent bottle.To this is added 25 ml of a solution containing 0.003 per cent. of :methylene blue (B.P. quality), 1.2 per cent. of concentrated sulphuric acid, 5-0 per cent. of sodium sulphate (anhydrous) and then 15 ml of chloroform. The bottle is shaken with just sufficient force to ensure that the phases mix thoroughly. At this stage the upper layer is a pale blue and the lower layer dark blue.A solution of cetyl pyridinium bromide containing about 2 g per litre of purified material is added about 2 ml at a time with intermittent shaking. When the colour of the upper layer begins to deepen the rate of addition is reduced. The end-point is reached when both layers, viewed in reflected light, are the same colour.May, 19511 EXAMINATION OF DETERGENT PREPARATIONS 283 The molar concentration of the cetyl pyridinium bromide solution is determined as f 0110 ws- Fifty millilitres of the solution are transferred by means of a pipette to a beaker and 25 ml of 0.01 M potassium dichromate solution are added. This precipitates insoluble cetyl pyridinium dichromate. The solution is heated to 90" C to coagulate the precipitate and then filtered through a No. 40 Whatman paper.The excess of dichromate in the filtrate is determined iodimetrically. Another method of determination is due to Barr, Oliver and StubbingsJs who use the following reagents and procedure- Reagents- (a) Bromophenol blue indicator solution, a 0.04 per cent. solution in 20 per cent. aqueous alcohol. (b) Chloroform B.P. (c) Cetyl trimethylammonium bromide solution, approximately 0.001 M . ( d ) Sodium oleyl sulphate (the reference sample). Procedure-Weigh accurately such a quantity of the sodium oleyl sulphate, reagent ( d ) , as will contain about 0.4g of active agent, dissolve in about 200ml of distilled water at 50" to 60" C (not above 60" C) and add a few drops of M sodium carbonate solution until the solution is faintly alkaline to Brilliant Yellow indicator paper.Cool the solution to 20" C, transfer to a measuring flask, dilute to 1 litre and mix well. Transfer 25.0 ml of this reference solution by means of a pipette to a glass-stoppered bottle and add 100 ml of water, 50 ml of chloroform and 5 drops of brornophenol blue indicator solution. Titrate the mixture with the cetyl trimethylammonium bromide solution, reagent (c), shaking after each addition of titrant. In the early stages of the titration, the chloroform emulsifies in the aqueous phase, but ready separation into two layers occurs as the titration proceeds, particularly as the end-point is approached. About 1 minute is allowed to elapse between successive additions of titrant, added in 0.1-ml increments towards the end-point, which is taken as the point at which the first indication of blue colour appears in the chloroform layer.The blue colour intensifies with further additions of cetyl trimethylammonium bromide solution. The cation active solution (c), thus standardised against the reference sample, is used for the titration of other anion active compounds by an exactly similar titration procedure, using as far as is practicable an approximately 0.001 M solution of each substance to be titrated. S CHINDLE R SEPARATION- This method is described in "Sulphated Oils and Allied Products," by Burton and Robertshaw,' and has been widely used for the separation of the components of detergent preparations. The various fractions recovered are- (1) Free fatty acids from soap, which can be weighed and titrated to check the equivalent weight.(2) Sulphated compounds. These tend to darken when the solvent, carbon tetra- chloride, is removed. (3) Highly sulphated compounds and sulphonic acids. This fraction can be titrated as desired. (4) Higher alcohols, mineral oil and other neutral fatty bodies. It should be noted that the method fails for sulphated glycerides, which do not dissolve Further, cellulose ethers upset the method, for they stabilise in carbon tetrachloride. emulsions and prevent the proper separation of the initial aqueous and solvent layers. SULPHATED BODIES- These are regarded as substances that can be hydrolysed by heating under a reflux condenser with N hydrochloric acid for several. hours, after which the liberated sulphuric acid is titrated. If it is difficult to see the end-point, the hydrolysate can be extracted with ether before titration.The ether-soluble hydrolysis product can be weighed as a check. TotaZ alkalinity-Weigh out accurately from 10 to 15 g of the sample into a litre beaker, make into a thin cream with water, dilute to 600 ml and titrate with 0.1 N hydrochloric acid and not more than 4 drops of the mixed indicator (see below). Let alkalinity expressed as milligrams of potassium hydroxide per gram of sample = A.284 EXAMINATION OF DETERGENT PREPARATIONS [Vol. 76 Concentration of organic sd$hates-Weigh out accurately about 15 g of the prepared sample in a 500-ml Geissler flask and add 50 ml of N hydrochloric acid. Heat gently under a reflux air-condenser, until the layer of oil covering the aqueous layer is clear and frothing has entirely ceased; this may take from 4 to 24 hours.During the heating, the flask should be supported by a clamp, about a half to three-quarters of an inch above an asbestos gauze with the flame carefully regulated so that the liquid is just kept gently boiling. If boiling is vigorous there is a danger of loss of hydrochloric acid, with consequent vitiation of the results. A piece of congo red paper loosely inserted in the upper end of the air condenser will indicate whether or not volatile mineral acid is being lost during the hydrolysis. Allow the flask to cool and wash down the condenser with a little ether and then a little distilled water. Add 25ml of ether (free from acid or alkali) to dissolve the fatty layer; then add 50 ml of neutral saturated sodium chloride solution and mix the contents of the flask by swirling.Add not more than 4 drops of mixed indicator and titrate with N sodium hydroxide solution. Near the end-point care must be taken to allow the ethereal layer to separate after each addition of alkali, as the dark colour of the ethereal solution tends to mask the colour change of the indicator. It is preferable to separate and wash the ether layer and then to titrate the combined aqueous extracts. Let the titre = B ml of N sodium hydroxide. Titrate 50 ml of N hydrochloric acid with N sodium hydroxide solution and 4 drops of Let the titre = C ml of N sodium hydroxide. The increase in acidity after hydrolysis = 13 - C ml of N sodium hydroxide. This is equivalent to (B - C) x 56*l/wt.taken = D milligrams of potassium hydroxide per gram of sample. (A + D)/56.1 x 36 = per cent. sodium alkyl sulphate (m.wt. 360). The molecular weight used in calculating the result is empirical and is based on the results of analysis. Mixed indicator-Methyl orange and xylene cyanol FF. (a) Xylene cyanol FF (Disulphine Blue FFS), 1.4 g, dissolved in 250 ml of 50 per cent. alcohol. (b) Methyl orange, 1 g, dissolved in 250 ml of 50 per cent. alcohol. Note-A mixture of solutions (a) and (b) tends to deteriorate on keeping. This mixed indicator is to be preferred to methyl orange alone, which is not very sensitive mixed indicator as before. For Lissapol C, m.wt. = 360, for 'Teepol, m.wt. = approximately 320. Two drops of (a) and 2 drops of (b) are used in each titration. in artificial light.SULPHONATED BODIES- A method suitable for Igepon type detergentss is based on the precipitation of the sul- phonate with benzidine. The insoluble portion is filtered off and dissolved in alcohol, and aliquots are taken for titration with alkali and for direct weighing. The method is unsuitable for sulphated bodies since it involves boiling with hydrochloric acid. A sirnilar method by Marron and SchifferlP is based on the fact that alkyl aryl sul- phonates react in aqueous solution with an mine salt of a mineral acid, $-tohidine hydro- chloride, to produce an amine salt that can be extracted with carbon tetrachloride. The carbon tetrachloride extract is mixed with neutral alcohol and titrated with 0.1 N sodium hydroxide.On titration the compound is broken up into P-toluidine and the sulphonic acid, which reacts with the alkali. It is essential, however, that the molecular weight of the alkyl aryl sulphonate is known. This molecular weight can be ascertained by carrying out several determinations on a sample of known composition. The weakly basic amine does not interfere. METHOD- Thirty-four grams of the salt, after recrystallisa- tion from industrial methylated spirit, are dissolved in distilled water and made up to 1 litre. Further recrystallisation must be carried out if the solution is not clear, or if the titration blank is too high. Dissolve 0-250 g of m-cresol purple in 6.5 ml of 0.1 N sodium hydroxide. Rinse into a 50-ml volumetric flask with Reagent+-Toluidine hydrochloride.Indicator-A 0.5 per cent. solution of m-cresol purple.May, 19511 EXAMINATION OF DETERGENT PREPARATIONS 285 sufficient water to make a total volume of 25 ml and then dilute to 50 ml with industrial methylated spirit. Procedure-Weigh out by difference from a weighing bottle 2 to 3 g of the detergent into a 250-ml separating funnel. Add 100 ml of p-toluidine hydrochloride solution, stopper the funnel and shake well. Continue with alternate shaking and settling until all the solid dissolves. Add 50 ml of carbon tetrachloride and shake well. Allow to stand until there is complete separation of the phases. Run off the lower layer into a 500-ml iodine flask. Make further extractions with 25 and 10 ml respectively of carbon tetrachloride. To the combined extracts add 100 ml of neutral industrial methylated spirit and 6 drops of m-cresol purple indicator.Titrate with 0.1 N sodium hydroxide with vigorous shaking between the additions. At the end-point, the grey colour of the emulsion takes on a blue or lavender tint. The two phases will separate if allowed to stand and a reddish purple colour in the upper layer denotes the end-point. Carry out a blank determination of the p-toluidine hydrochloride solution. Subtract any value, which should not exceed 0.3 ml of 0.1 N sodium hydroxide, found for the blank from the value obtained when the sample is present. The percentage of alkyl aryl sodium sulphonate contained in the sample is given by- Volume (ml) of 0.1 N NaOH required for the titration x F x M Weight of sample taken x 100 where F is the factor for the 0.1 N sodium hydroxide and M is the molecular weight of the alkyl aryl sodium sulphonate. NON-IONIC DETERGENTS- The method given by Oliver and PrestonlO has been found satisfactory.The factor used for converting the weight of the precipitate to weight of detergent must, however, depend on information obtained by experiment. Reagents-The reagents required are : (a) Detergent solution of known concentration. (b) Hydrochloric acid solution, 1 volume of the concentrated acid diluted to 4 volumes with distilled water. (c) Barium chloride, l o g of BaC1,.2H20 dissolved in 1OOml of distilled water. (d) Phosphomolybdic acid (B.D.H.), 10 g of P20,.20M00,.51H20 dissolved in 100 ml of distilled water. Sulphates, if present, interfere with the determination and must be removed by treatment with barium chloride solution before the following procedure.Procedure-An aliquot part of solution (a), containing a known amount of detergent, usually not more than 100 mg, is placed in a 250-ml beaker, to which are added, in the following order, 5 ml each of hydrochloric acid (solution b), barium chloride (solution c) and phospho- molybdic acid (solution d ) ; the contents of the beaker are then diluted to 150 ml with distilled water. The yellowish-green precipitate formed is flocculated by heating the mixture to boiling-point; the beaker is then covered and allowed to stand overnight (18 hours). The precipitate is filtered through a tared No. 4 sintered-glass crucible that has been previously heated for 15 minutes at 100” C and cooled in a desiccator, washed with a minimum of 100 ml of distilled water and dried at 100” C to constant weight. After two 1-hour periods, no substantial change in weight should be found. Solutions of the detergent under test are then precipitated in the same way. This procedure establishes the ratio of the weight of the complex to the weight of the detergent for the particular detergent preparation under test.The ammonium cobalt thiocyanate test mentioned above (qualitative test iv) as described by H. Gnammll and by Van der Hoeve2 is said to be available as a quantitative method. CATIONIC DETERGENTS-QUATERNARY BASES- These can be determined by a reversal of the methylene blue titration method. A sample of Teepol is titrated against 0.004 M cetyl pyridinium bromide (see “Anionic Detergents”) and this standardised Teepol used to titrate the quaternary base under test.NAPHTHENIC ACID SOAP- This can be determined as described in “Standard Methods for Testing Petroleum and its Products”12 by separating the copper salt, in presence of excess copper sulphate , and dissolving it in benzene. Polyethylene glycol compounds.286 EXAMINATION OF DETERGENT PREPARATIONS p o l . 76 NEUTRAL ELECTROLYTE AND ALKALINE ELECTROLYTE- Since the type and quantity of these materials vary enormously in a given product from time to time, it is impossible to give general methods for their analysis. The presence of the detergent usually interferes seriously with any precipitation method involving aqueous solution, because of its dispersing action on the precipitate.It is therefore usually necessary to remove, extract or destroy the active material before estimating inorganic salts. Alkalinity as hydroxide, carbonate or bicarbonate may be estimated by electrometric titration or by the indicator method. In the latter method indicators must be selected with care, as the active material tends to alter the apparent pH value of certain indicators. For the estimation of sodium sulphate, which is a very common constituent of detergent preparations, the following method can be recommended. It will be noted that alkalinity is neutralised by sulphuric acid before the determination is begun. Sufficient sample to yield approximately 0.2 g of sodium sulphate as final precipitate is weighed to the nearest milligram into a 400-ml beaker.Five millilitres of water are added and thoroughly mixed with the sample by stirring with a glass rod. One drop of bromo- phenol blue indicator is added and the solution is titrated with 0.5 N sulphuric acid until exactly neutral. The solution is rapidly raised almost to boiling-point by heating over a gas burner, but is not actually boiled. (This process must not take longer than 1 minute.) One drop of phenolphthalein indicator is added and then 1 to 3 drops of 0-5 N sodium hydroxide solution, which will make the test solution just alkaline to phenolphthalein. Two hundred millilitres of neutral ethyl alcohol are added and the mixture is boiled gently for 30 minutes, with the beaker covered by a clock-glass during this period. The mixture is cooled to about 50” C and the precipitated sulphate is removed by filtration through the tared sintered-glass crucible and washed with 100 ml of warm neutral ethyl alcohol. The crucible is then dried repeatedly in the oven, cooled in a desiccator and weighed to the nearest milli- gram until it reaches constant weight. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. REFERENCES Linsenmeyer, K., Melliard Textilber, 1940, 21, 468. Van der Hoeve, J. A., Rec. Trav. Chirn. Pays-Bas., 1948, 67, 649. The American Oil Chemists Society, “Official and Tentative Methods,” Section F. Berkowitz, D., and Bernstein, R., Ind. Eng. Chew., Anal. Ed., 1944, 16, 239. Epton, S. R., Trans. Farad. SOC., 1948, 44, 226. Barr, T., Oliver, J., and Stubbings, W. V., J . .TOG. Chew. Ind., 1948, 67, 45. Burton, D., and Robertshaw, G. F., “Sulphatetl Oils and Allied Products-Their Chemistry and Analysis,” A. Harvey, London, 1939, p. 81. Sheraeff, D. A., American Dyestuffs Reporter, 1947, 36, 313. Marron, T. U., and Schifferli, J., Ind. Eng. Chew., Anal. Ed., 1946, 18, 49. Oliver, J.; and Preston, C., Nature, 1949, 164, 242. Gnamm, H., “Die Ltjsungs und Weichmachungsmittel,” Stuttgart, 1941, p. 330. Institute of Petroleum, “Standard Methods for Testing Petroleum and its Products,” 1.P.-87/44.

ISSN:0003-2654    

DOI:10.1039/AN9517600279

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

May, 19511 DE WHALLEY, ALBON AND GROSS 287 Applications of Paper Chromatographic Methods in the Sugar and Allied Industries It is shown that separation of sugars of industrial interest can be made rapidly and quantitatively by the paper chromatographic procedure first used by S. M. Partridge of the Low Temperature Research Station, Cambridge, whose advice and assistance at the beginning paved the way for the later investigations. The procedure adopted and the construction of the apparatus is described. The chromatogram is sprayed with various reagents that give colour reactions with the sugars. The sugars are identified by comparison with controls of pure sugar run on the same-piece of paper. By means of paper-chromatography it is possible to demonstrate the Lobry de Bruyn conversion and show the equilibrium concentrations of glucose, fructose and mannose; the presence in the conversion products of a fourth sugar, allulose, is also made evident.The heat degradation products of fructose have been separated and, by co-operation with Zerban and Sattler, were shown to be identical with the non-fermentables of cane molasses; some “compounds” were shown to be mixtures. Raffinose can be estimated in raw beet sugar with precision and certainty for the first time. The degree of purity of sugars such as raffinose has been determined; freedom from traces of sucrose could previously only be inferred, as the exact physical constants of the sugar were, and still are, open to conjecture. Mixtures of starch hydrolysis products, such as those present in beer, can be separated and identified quantitatively.At intermediate stages of the inversion of sucrose by invertase the appearance of at least one synthesised tri-saccharide has been noticed. Chromatographic methods enable the organic and inorganic non-sugars, including colouring matters, t o be separated, and make their examination a relatively simple matter; much still remains to be done in the identification of the separated products. LIMITATIONS OF THE TRADITIONAL CONTROL INSTRUMENTS ALTHOUGH scientific control of process operations in the sugar industry has been progressively improved during the last twenty-five years, the greater part of this control is still very empirical. The most important instruments used in a control laboratory include the polarimeter, refractometer, Brix hydrometer, pH meter, conductivity ash meter (which indicates the amount of salts present), colorimeters and narrow-waveband light absorption instruments.All of these have been and will remain invaluable to the industry, but they are not sufficient to solve many of the problems that occur daily. The optical rotation provides an exact measure of the concentration of pure sugar solutions, but in the sugar industry most of the samples examined are far from pure, as they contain mixtures of sugars together with non-sugars, some of which possess optical activity that varies with concentration or pH; measurement of polarisation gives only a commercial approximation to the sucrose content. The use of lead acetate defecants in solution or powder form confuses the result still further.Even with refined white granulated sugar1 the sucrose content must be indirectly calculated by difference from 100 of the separately determined percentages of moisture, ash and reducing sugars, because the polarimeter is limited to an accuracy of 0-05, or with more recent instruments 0.02, and this is sufficient only for commercial purposes. The Lane and Eynon method for determining reducing sugars, which has received international acceptance as a standard method, gives, with sugar-house products, only the total reducing value expressed as invert sugar, which is rarely, if ever, found in any samples from juice to finished sugars, syrups or molasses. The reducing effect of large quantities of sucrose with a low content of reducing sugars vitiates the accuracy of the method;288 [Voi.76 some of the reducing power may not even be due to sugars.2 Tables based on the refractive index of sucrose solutions give values for the total solids whose accuracy is limited by the content of non-sugars. The Brix hydrometer at 20” C gives a true value for total solids for pure sucrose solutions, but with solutions of lower purity it gives values that are correct only for density purposes. The determined pH at 20” olr 25” C does not necessarily correspond to the pH found at the factory working temperature, nor is there any precise relation between the two figures owing to the unknown dissociatiori of the non-sugars. Abnormal discrepancies between ash determined by conductimetric and gravimetric methods depend on the composition of the non-sugars.A figure for colour content, whether found by iise of Lovibond glasses or by light absorp- tion at 5 6 0 0 ~ , gives no information about the composition of the colouring materials, nor does it give information about their selective ads.orption during carbonatation or by bone charcoal. DE WHALLEY, ALBON AND GROSS: APPLICATIONS OF PAPER NEW TOOLS FOR INDUSTRIAL RESEARCH How then are we able to investigate problems that, owing to the presence of non-sugars in relatively low concentration, cannot be solved with the instruments available ? The most satisfactory procedure is to make use of chromatography in one or other of its several forms. Chromatography, sometimes with the use of ion-exchange agents, wavelength spectrophotometers, flame spectrometers and recording polarographs, will go far to assist research in the sugar and allied industries.GENERAL POSSIBILITIES OF CHROMATOGRAPHY It is appropriate at this stage to quote from Dr. R. C. Hockett’s foreword to the scientific: report on Chromatography of Sugars and Related Substances3 published in 1948- “The close chemical similarity of the various sugars to one another has always made the problem of sugar analysis a fundamentally difficult one. The progress of biochemistry, physiology, botany and food technology has ‘been impeded for many years by the lack of quick and accurate methods by which the various carbohydrates might be differentiated from one another and estimated quantitativeky. The recent developments of chromato- graphic adsorption methods of analysis have provided investigators in many fields with an exceedingly powerful and useful tool for separation and purification.The application of such methods to the sugars has required ingenuity and painstaking work but has attained a notable degree of success. Although they are not generally useful as quick routine procedures, they do provide means of separating sharply the constituents of many mixtures composed of substances that are chemically similar. They have also the extremely important merit of isolating these constituents in tangible and usually crystalline form so that the identification is not a matter of surmise or deduction. Qualitatively they meet the ultimate specification and are capable also, in conjunction with more traditional methods, of providing quantitative information.” This paper is concerned with chromatography.METHODS- Some progress has been made since these words were written. Chromatography has been described as a specialised type of adsorption. Solutions are run through a bed or column of adsorbent and., with subsequent percolation of the proper solvent or mixed solvent down the column, the least strongly adsorbed compounds move down the column faster than the more strongly adsorbed compounds. This results in a series of bands or zones that can be separated by mechanical extrusion of the adsorbent in fractions or alternatively can be fractionally eluted by the solvent. A practical example of adsorption in the sugar refinery is provided by bone charcoal contained in cisterns.Colour and non- sugars and even sugars themselves are adsorbed, and the sugars are desorbed together with some colour and non-sugars when the charcoal is “sweetened off” or washed prior to regenera- tion of the charcoal. As bone charcoal is a black adsorbent, no zones are visible in the earlier stages of running, although some of the coloiired zones would be visible if the adsorbent were colourless. Even colourless adsorbents may fail to show zones if the adsorbed compounds are colourless, but sometimes they can be made to fluoresce in ultra-violet light and at otherMay, 19511 CHROMATOGRA4PHIC METHODS I N SUGAR AND ALLIED INDUSTRIES 289 times they can be made visible by addition of mixed dyestuffs that have individual adsorption characteristics resembling those of the main compounds being separated..For standard column chromatography, a narrow glass tube with a constriction at the lower end with a perforated disc, sintered-glass disc or a cotton wool plug just above the constriction can be used when the adsorbents are to be eluted. When the adsorbent is to be extruded mechanically, a straight tube of glass or plastic is selected; it should have a slight taper to assist extrusion. The tube is attached by means of a ground joint to a lower portion containing a perforated cap and ending in a constriction. The adsorbent used must be selected in accordance with the required characteristics. Many different substances have been used; they include carbon, alumina, silica gel, kieselguhr, magnesia, clays of different types, sucrose, starch and cellulose powder. Generally the particle size of the adsorbent should not be larger than would pass a 200- mesh screen, and the material should be freed from very fine particles that hinder flow.The packing of the tube is an art acquired only by experience. Flow may be facilitated by use of a hydrostatic head or by gentle suction. The first advances in the separation of sugars came with the discovery of suitable coloured sugar derivatives. However, these methods did not commend themselves to the sugar industry because the yields of the derivatives were not quantitative, nor were the preparations entirely free from products of side reactions; also one simple sugar might yield several derivatives. After flow-through or elution chromatography the eluates of colourless sugars could be examined by physical means, e.g., by measurement of the refractive index, or by chemical means, e.g., by oxidation with periodate or Fehling’s solution.The method was tedious and empirical, however. The interest of the sugar industry was aroused when Partridge: who used the method elaborated by Consden, Gordon and Martin6 for separating the amino-acids of wool hydro- lysates, separated simple sugars by what is now known as paper chromatography. Goppelsroder, in the nineteenth century, was the first to use sheets or strips offilter-paper for separation, and he found it extremely useful. Consden, Gordon and Martin revived this dormant technique and placed it on a more scientific basis; it has since been applied more generally to other problems, including that of the separation of sugars, by Partridge and his colleagues at the Low Temperature Research Station, Cambridge.With the helpful advice and guidance of Dr. Partridge we were able to make an easy start, although we were forewarned by Dr. Bate-Smith that in many of the investigations planned we should have to work out our own salvation. The technique of paper chromatography lies in placing a drop of the test solution on a filter-paper strip at a point near the top end; the paper is then hung from a trough containing a suitable solvent or mixture of solvents into which the top end of the paper dips. The trough and the paper strip are enclosed in a chromatographic cabinet, a lagged air-tight compartment in which the atmosphere is saturated with the vapours of water and the solvents.As the solvent is drawn into the paper by capillary forces and starts flowing vertically over the strip it extracts the individual constituents of the spot at different rates, carries them down the strip to form a series of invisible spots each containing a separate constituent, and so effects the desired separation. The strip is removed from the solvent trough and dried, and then sprayed with suitable reagents to make the spots visible; coloured compounds that are already visible do not need to be sprayed. The resultant strip is termed a chromato- gram. THEORY OF THE PROCESS- To explain the mechanism, it is assumed that the water in the paper (cellulose) forms a static liquid phase and the solvent a mobile liquid phase.Hence the paper takes no part in the actual process and acts only as an inert support for the water. From the moment the solvent flows over the original spot of test solution there is a distribution or partition of the constituents of the mixture between the water and the solvent, the solvent carrying more and more away until the extraction is complete. Every constituent is extracted according to its partition coefficient, which is a constant characteristic of each compound. The differences in the partition coefficient account for faster or slower movement and hence for the ultimate separation of the constituents of a mixture. By dividing the distance each substance moves by the distance moved by the solvent front on the chromatogram, a constant,290 DE WHALLEY, ALBON AND GROSS: APPLICATIONS OF PAPER [Vol.76 the RF value, is obtained; it is characteristic of the compound in a given solvent under standard conditions, is easily reproducible and can be used for the identification of the compound. As the partition coefficients vary with each solvernt, the selection of a suitable solvent for various mixtures of compounds is very important. The solvents are used saturated with water to help to attain phase equilibrium and also to increase the solvent power. Mixtures of several solvents are often more efficient, and the addition of small quantities of acidic or basic compounds has also been found useful. The partition ratio is proportional to the solubilities of the compounds in the solvent concerned; a knowledge of the solubilities is therefore helpful in the selection of solvents.Indeed, because the partition principle was assumed to be the basis of the method, the process was called “paper partition chromato- graphy.” However, it is not possible always to exclude the action of adsorption on the paper, as various workers have reported more or less strong adsorption effects when running paper chromatograms. I t is sometimes difficult to decide what part is played by partition and whether partition is still the only governing factor. It can, however, be safely assumed that both partition and adsorption, and possibly some other factors of minor importance, are responsible for the functioning of paper chromatography. Paper is very convenient as an adsorbent because it is easily stored in a small space, is ready for instant use, is inexpensive and can be photographed for permanent record.A large number of different samples can be run in parallel under identical conditions. Minute quantities down t o a microgram can be analysed, which helps to conserve precious compounds. The paper strips can be of various sizes, or large sheets can be used, as is necessary with two- dimensional chromatograms when the paper is developed in two directions with two different solvents selected according to the requirements of the substances to be separated. The one serious limitation to the use of paper in chromatography is the very small amount of material that can be placed upon a single sheet of paper. However, the small volume of solution applied to the spot can be dried .and a further volume superimposed upon it, so doubling the quantity of material to be separated, although if the quantity is greatly increased, clean separation is hindered.An alternative method of increasing the yield is by making a series of sample drops across the paper and, after separation, collecting them together by water elution; in this way quantities can be increased up to tenfold. For larger quantities recourse must be made to clolumn separation with powdered cellulose as the adsorbent and the appropriate solvent. Extensive work has been carried out by this method by Jones, Hirst et aL6 at Birmingham and later at Bristol and Edinburgh. The completeness of separation of the fractions can be tested by subjecting a drop of each fraction to the paper technique.A very minor constituent can sometimes be concentrated on a column and from the still impure mixture so prepared the required constituent can be separated cleanly on paper for identification or quantitative measurement. DETAILS OF THE TECHNIQUE- For the identification of the separated sugars a supply of pure sugars is required, as it is necessary to run solutions of these on the paper as controls, If a micro-technique is used, only minute quantj ties of samples and controls are required. The drops applied to the paper are measured with a micro-pipette, an Agla micrometer syringe or, for routine purposes, carefully calibrated capillary tubes. For quantitative work exact measurement of the volume is essential.The areas of the spots of reacted sugars are compared with those of the controls for quantitative estimation. The filter-paper must be carefully selected, as different grades are required for different purposes. Even with one grade of paper, because no two sheets of paper are exactly identical in thickness or density, it is essential to make control tests on the same sheet as the test sample. Trouble has been caused by merely following directions to use a particular grade of paper, when the results obtained have conflicted with those of other workers. The suppliers request that when a particular grade of filter-paper is ordered it should be stipulated that it is to be used for chromatographic work; such supplies are vetted to ensure that the tolerance in weight and thickness is much smaller than for paper supplied for less critical analytical work.Other requirements that, unless fulfilled, are almost certain to cause trouble to the beginner are as follows- (1) The sample spots applied to the paper must be air-dried before the strip is transferred to the chromatographic cabinet.May, 19511 CHROMATOGRAPHIC METHODS I N SUGAR AND ALLIED INDUSTRIES 291 (2) The solvent or mixed solvents must be saturated with water before use by shaking the two liquids together in a separating funnel; the solvent phase must be free from fine droplets of water. This phase is used in the trough and the aqueous phase is placed in the bottom of the cabinet. (3) The atmosphere of the cabinet should be in equilibrium with respect to themixed vapours before the paper is dipped in the trough and suspended in the cabinet, and in consequence it is advisable to prepare the cabinet in advance. (4) The paper should nearly reach the liquid in the base of the cabinet but must not touch it.(5) The cabinet should be lagged to prevent or slow down temperature changes, as increase or decrease of temperature will alter RF values. Small changes are immaterial as control tests are likewise affected, but large changes of temperature may speed operations and run the spot off the paper. (6) When the separation of mixtures is not satisfactory the cause may be insufficient time in the chromatographic cabinet. If increased time produces no improvement, the solvents may require modification. As the RF value is increased, the separation in a given time first improves, but then becomes less satisfactory because the spots spread.EXPERIMENTAL APPARATUS- Figs. 1, 2 and 3 show the chromatographic cabinet and constructional details; the cabinets are made inexpensively from glass accumulator boxes by cutting off and grinding the base of one and fitting it above another, the joint being made with adhesive tape. The trough is of glass or polythene; the latter is more easily worked by making use of hot nitrogen welding. L -Ground Surface Glass Slat Solvent Trough Glass Support Water Layer Fig. 2 . Chromatographic cabinet Trough u Slass S'upport Fig. 3. and paper Trough assembly, showing solvent After removal from the chromatographic cabinet, the paper is suspended in the drying cabinet, as shown in the photograph, Fig.5. This cabinet has been constructed in our workshops from angle-iron, which forms a framework that is fitted with asbestos-cement sheet sides and a hinged top. The base of the cabinet is a perforated plate under which is fitted an electric resistance heater. Through the heater and the perforated bottom plate a current of air is drawn by suction from the fan situated on one side at the top of the cabinet; the air exhausts through a duct to the fume cupboard-a necessary precaution, as the solvent vapours, especially when pyridine or collidine are used, are unpleasant and may be toxic. The paper must not be overheated or it will become discoloured; uniform heating, not entirely satisfactorily attained with this particular design, is most desirable, The function of the cabinet is simply to remove water and solvent.292 DE WHALLEY, ALBON AND GROSS: APPLICATIONS OF PAPER [Vol.76 If colourless substances are being separated, it is necessary to spray the paper with a suitable reagent, and this is done as shown in Fig. 4 by means of an atomising spray bottle actuated by compressed air. It is essential to spray the paper evenly and completely. During this operation the paper is suspended in a fume cupboard. When sprayed, the paper is returned to the drying cabinet for drying and sometimes for further heating to complete the reaction of the reagent; Fig. 5 shows the sprayed chromatogram in the drying cabinet. REAGENTS- External heat for reaction is applied when the paper has been sprayed with a-naphthol.Molisch's reagent, as ordinarily used in testing for :sugar in refinery or factory waste waters, relies on the addition of concentrated sulphuric acid to the water sample to which an alcoholic solution of a-naphthol has been previously added. The heat generated serves two purposes; one is to hydrolyse sucrose, the other is to enable the breakdown product of fructose, methyl hydroxy furfural, to react with the phenol to give the characteristic colour reaction. As such a reagent as sulphuric acid must not be used on a paper chromatogram, the alcoholic solution is acidified with phosphoric acid and heat is applied externally. The a-naphthol- phosphoric acid reagent serves to indicate not only ketoses, but also di-saccharides containing ketoses, e.g., sucrose or raffinose, which are completely hydrolysed by the phosphoric acid.It gives no reaction for aldoses or compound aldose sugars such as maltose. An alcoholic solution of ammoniacal silver nitrate containing ca.ustic soda serves to indicate all reducing sugars and reacts slightly with non-reducing sugars. An alternative reagent for sugars is aniline clxalate, and the chromatogram prepared in this way shows spectacular fluorescence in ultra-violet light. When it is required to show aldoses and ketoses on the same chromatogram, an alcoholic solution of naphtho-resorcinol containing phosphoric acid can be used ; this produces a pink spot with ketoses and a green spot with aldoses. It is evident from the foregoing that the reagent has to be selected to suit the particular mixture of sugars. Many other reagents have been tried, but those described have given the most successful results.Sometimes the reagent gives a fugitive colour reaction, e.g., that with ct-naphthol, and then a photographic copy must be made within fifteen minutes if a record is to be kept. With colouring matters separated from raw sugar solutions or intermediate products, the coloured spots are visible; however, some are pale in colour and some are present in sufficiently low concentration to be invisible. Examination in ultra- violet light will enable them to be seen clearly, and exposure to ammonia gas will intensify all the coloured spots. USES OF PAPER CHROMATOGRAPHY Paper chromatography has been applied by the authors to many problems, some of which have been readily solved; others have been partly solved and much information has been gained, but further work is required for their complete solution.The separation of sucrose, glucose, fructose and other simple sugars by the technique of Partridge is shown in Table I. TABLE I CONDITIONS FOR SEPARATION OF SUGARS Paper used . . .. .. . . . . 'Whatman No. 1 Solvent . . .. .. .. . . . . :Ethyl acetate, 2 parts by volume Pyridine, A.R., 1 part by volume 'Water, 2 parts by volume Time in cabinet . . .. .. . . . . .16 hours a t 17" C Time and temperature of drying . . . . Q hour a t 90" C Reagent used . . .. .. . . . . Naphtho-resorcinol and phosphoric acid Time and temperature of heating . . . . 10 minutes at 90" C In the chromatogram, Fig. 6, the individual sugars, each of different RF value, are shown to have travelled different distances downwards.In addition, a mixture of the sugars has been separated into its constituents. Several practical applications of this procedure have been useful. In one, a sample of supposed mannose was quickly shown to contain no mannose but only sucrose with traces of glucose and fructose (invert sugar) and a patch of colour; it was in fact a low purity granulated sugar of unknown origin.May, 19511 CHROMATOGRAPHIC METHODS I N SUGAR AND ALLIED INDUSTRIES 293 A sample of Dutch fondant was found to be composed of sucrose and high conversion glucose, with only a trace of invert sugar formed by hydrolysis in manufacture. Sugar confectionery containing a mixture of sugars can easily be examined by this procedure and such constituents as lactose, sucrose, maltose, glucose and fructose may be shown to be present. Once the approximate amounts of the components of these mixtures are known, it is possible to make a quantitative assessment by polarimetric and reducing methods and simple calculation.LOBRY DE BRUYN CONVERSION OR EFFECT OF ALKALI ON HEXOSES- The re-arrangement of simple sugars in aqueous alkaline solution can be clearly demon- strated. Glucose, fructose and mannose in saturated lime-water solutions are shown in Figs. 7 and 8 t o have formed equilibrium mixtures of all three sugars from each individual sugar through the common enolic form. With the chromatogram sprayed with the reagent aniline oxalate, glucose and mannose are shown in Fig, 7.With cc-naphthol reagent, Fig. 8 shows fructose and, in addition, a spot that is almost certainly d-allulose, the epimer of fructose, but as yet we have no specimen of d-allulose as a control. The demonstration of this conversion is not as academic as it may seem, as it is connected with the problems of alteration or loss of reducing sugars in the operations of liming, carbonatation and charring. Samples of brewer's caramel produced by the action of heat and ammonia on sugars such as sucrose, invert sugars, glucose and hydro1 all showed a quantity of fructose among the residual sugars present. DESTRUCTION OF SUGARS DURING THE HEATING OF RAW CANE SYRUP- An investigation was being made into the relative advantages of heating syrup by steam coil or by hot-water coil through equal temperature ranges in the same time, the ,upper temperature limit being maintained for twelve hours.In the normal estimation of total sugars after inversion, which were expressed as the proportion of invert sugar in the total TABLE I1 CONDITIONS FOR SEPARATION OF FRUCTOSE DECOMPOSITION PRODUCTS Paper used .. . . . . .. . . Whatman No. 1 Solvent . . .. .. .. .. . . Ethyl acetate, 2 parts by volume Pyridine, A.R., 1 part by volume Water, 2 parts by volume Time in cabinet . . .. .. I . . . 24 hours a t 17" C Time and temperature of drying . . . . 1 hour a t 90" C Reagent used . . . . . . .. .. Naphthol Time and temperature of heating , . . . 10 minutes a t 90" C solids, the errors of estimation were found to be of the same order as the expected differences.A chromatographic examination of destruction products showed, as might be expected, that some were present in the original sample. A number of compounds could be separated but not identified. It was then realised that these compounds, some still possessing optical rotation or reducing power, resembled one class of the non-fermentables of cane molasses, that produced by destruction of fructose, into which Zerban and Sattler were carrying out an investigation. By their procedure,' pure fructose was partly destroyed by long heating of its solution and fermentation of the residual fructose; the unfennentable syrup that remained was chromatographed and showed many similar spots to those found with raw cane syrup. Chemical fractionation of the residue by the procedure of Zerban and Sattler followed by chromatographic treatment of the separated fractions showed that some that had previously been considered to be simple were really more complex.Collaborative work with the New York Sugar Trade Laboratory has enabled some of the compounds to be identified. Minute samples for examination have been flown from the U.S.A. to England and chromatograms have been flown back. As a result the joint knowledge of all the colla- borators has been much advanced. A full paper on the subject by Zerban, Sattler and their various collaborators is being prepared for publication in the near future. With the conditions shown in Table 11, a chromatogram of some of the fructose decom- position products has been prepared.294 DE WHALLEY, ALBON AND GROSS: APPLICATIONS OF PAPER [Vol.76 The three dimers have been isolated as described from the mixture, but only two of the pure dimers have been identified on the chromatogram; we were unable to carry out the identification as we had no pure specimens, but Zerban and Sattler were able to do so. In the chromatogram, Fig. 9, the compounds visible include (a) the three dimers of fructose, the di-hetero levulosans, (b) a compound that is probably d-allulose and (c) the monomeric anhydride of fructose. In following the procedure of Zerban and Sattler in fermentation of residual fructose, it was found by chromatographic examination that all fructose had been removed by one instead of by two fermentations. Any attempt at quantitative measurement of destruction of fructose by heating had to be postponed to allow work to be concentrated on the estimation of raffinose in beet sugar.PURITY OF SUGARS- Pure sugars are used for controls and now that more exact methods are available for testing for impurities, the purity of purchased sugars should not be assumed. A sample of pure fructose was found to contain traces of the breakdown products already described, and a laboratory reagent maltose contained dextrins. A sample of purified raffinose was submitted to us for estimation of traces of sucrose. It was part of a larger quantity intended for use in preparation of International tables of refractive indices and no other means existed for determination of small quantities of sucrose. 'The physical constants that were available had probably been determined on sugars of even lesser purity.Direct chromatographic tests on 400 pg of the raffinose sample in a 10 per cent. solution did not reveal any sucrose spot, but this only meant that the sucrose, if present, was less than 0.5 per cent. of the total. Larger quantities, limited by the solubility of raffinose, of 800 and 1200 pg still did not reveal a sucrose spot; this placed the content of sucrose below 0.25 and 0.16 per cent. respectively, although under such concentrated conditions the separation could not be perfect as overloading would occur. A series of spots, each containing 1OOOpg of the raffinose sample, were chromatographed on one sheet of thicker paper (to prevent overloading) and the portion where the sucrose should have appeared was cut off, extracted with water, concentrated in ztacuo and :re-chromatographed.The result was the appearance of a sucrose spot indicating that the original sample contained 0.15 per cent. of sucrose. The test had to be carried out in an atmosphere free from traces of sucrose, a requirement not easily met in a sugar laboratory. When the result was reported, the bulk was once recrystallised and another sample submitted. This was tested in a similar manner and showed less than 0.02 per cent. of sucrose, whkh was the limit of the test, The bulk of the raffinose was then recrystallised twice more a s an added safeguard before being used for its intended purpose. RAFFINOSE IN RAW BEET SUGARS- Our work on the subject of raffinose in raw beet sugars has already been reported,*yg but as the subject is of some importance and interest, a brief account is included in this paper.By the use of paper chromatography, raffinose in raw beet sugars can be detected and estimated with precision and certainty for the first time. The classical two-enzyme method is useless for determining small quantities in raw products and may be unreliable even with amounts of the order of 0.5 per cent. Table I11 shows the conditions of the separation. TABLE I11 CONDITIONS FOR RAFFINOSE SEPARATION Paper used . . . . .. .. . . Whatman No. 1 Solvent . . .. .. .. .. . . w-Butanol, 5 parts by volume I'yridine, A.R., 3 parts by volume Water, 3 parts by volume Eienzene, 1 part by volume Time in cabinet . . .. .. .. . . 24 hours at 16°C Time and temperature of drying .. . . 1 hour a t 90°C Reagent used . . .. .. .. . . a-Naphthol Time and temperature of heating . . . . 10 minutes at 90' C A chromatogram of a number of British raw beet sugars with controls of raffinose is The ash, or more correctly the salts present in raw beet sugars, hinders, shown in Fig. 10.Fig. 1. Chromatographic cabinets, lagged and unlagged Fig. 4. Spraying the reagent Fig. 6. Heating the sprayed paper in the drying cabinetFig. 6. Chromatogram of sucrose, glucose and fructose Fig. 8. Lobry de Bruyn conversion: glucose, mannose and fructose Fig. 7. Lobry de Bruyn conversion; glucose and mannose Fig. 9. Chromatogram of fructose decom- position productsFig. 10. Chromatogram of raffinose separation from raw- beet sugars Fig.12. Chromatogram of raffinose in U.S. white beet sugars from baryta-treated Steffenised molasses and in residual molasses Fig. 11. Chromatogram of raffinose in U.K. beet molasses Fig. 13. Chromatogram of raffinose hydrolvsed by enzvmesFig. 11. Sucrose partl!. hydrolysed b!- invertase Fig. 15. Sugars in beer Fig. 16. Colouring matters in raw sugar solutionMay, 19511 CHROMATOGRAPHIC METHODS IN SUGAR AND ALLIED INDUSTRIES 295 to a slight degree, the movement of the raffinose, and to compensate for this in the controls the pure raffinose is dissolved in a solution of raw cane sugar that has been tested and shown to be completely free from raffinose. The spots now known to be raffinose were at first identified with caution as it was considered that they might possibly be a non-sugar of RF value equal to raffinose, but this was disproved by the following procedure.A number of equal volumes of raw sugar solution were chromatographed together on a paper sheet and the portion of the paper containing spots corresponding to raffinose was cut out and extracted with water. After concentration in VLZCUUO and precipitation by acetone characteristic needle- shaped crystals were deposited; they were indistinguishable from those precipitated from pure raffinose. In other experiments the spots were extracted and concentrated to give a 2 per cent. solution of the supposed raffinose. On inversion with Wallerstein's invertase this solution gave fructose and melibiose, and on inversion with invertase containing Waller- stein's melibiase it yielded fructose, glucose and galactose. The presence of these sugars in the expected yields was demonstrated chromatographically, as also was the complete inversion of the supposed raffinose.All the samples of raw beet sugars received for refining at our two London refineries during the 1949 to 1950 campaign have been tested for raffinose by paper chromatography with the results summarised in Table IV. No sample was free from raffinose. The total number of samples from both refineries was 429; they contained an average of 0.28 per cent. of raffinose hydrate. Factory Refinery P- 1 2 3 4 Miscellaneous 1 3 5 Miscellaneous Refilzery T- TABLE IV RAFFINOSE IN RAW BEET SUGARS Raffinose hydrate No. of r- A -, samples Max., Min., Average, % % % 90 0.48 Trace 0.22 77 0.60 0.15 0.32 42 0.40 Trace 0.21 34 0.50 0.05 0.23 8 0.50 0.20 0.33 22 0.50 0.30 0.39 12 0.45 0.15 0.32 139 0.48 0.02 0-28 5 0.28 0.10 0.24 Average ash, % 0.96 2.12 0.69 0.56 0.79 0.98 0.68 0.76 0.81 As 1 per cent.of raffinose polarises 1-52', the amount of fictitious sucrose shown by polarisation is 0.42 per cent. This is equivalent to a thousand tons of fictitious sucrose for which payment has been made. The figures of apparent sucrose losses in refining require an adjustment of sucrose input because, although the raffinose appears in an output item, viz., the molasses, the method of analysis takes only partial account of this. The value of the apparent sucrose in molasses is considerably lower than that in the major output item, viz., refined sugar.RAFFINOSE IN FACTORY BEET MOLASSES- The effect of the high ratio of salts to sucrose makes direct chromatography of a molasses solution unsatisfactory. If, hQwever, the salts are removed by passing the solation through a mixed bed of cation and anion removal material, e.g., Zeocarb and De-acidite B (Pennutit), it is found that a clean chromatographic separation of raffinose can be made, as shown in Fig. 11. They were obtained through the courtesy of the British Sugar Corporation to assist in our search for a supply of molasses containing a high proportion of raffinose for culturing melibiase yeasts. The results of these tests showed that raffinose hydrate was present in amounts of 1.1 to 2.4 per cent. A larger supply of one of these samples showed 2 per cent. of raffinose by the paper test and 2.1 per cent.by the two-enzyme method. A number of samples of factory beet molasses has been tested.296 [Vol. 76 RAFFINOSE IN WHITE SUGAR RECOVERED FROM MOLASSES- No processing of molasses for recovery of sucrose is practised in the United Kingdom, but various countries in Europe use the lime, strontia or baryta processes. The Great Western Sugar Company of Denver, Colorado, uses the Steffen lime process in about half of their beet sugar factories in order to get a double dose of the lime for juice purification and for the recovery of sugar. When the raffinose in the final molasses from the Steffen process accumulates to the extent of 5 per cent. or thereabouts it is termed Steffen discard. This amounts to 40,000 to 50,000 tons per year and is further processed in the barium process plant at Johnstown, Colorado, which works for eight to ten months of the year entirely on molasses. A considerable elimination of the raffinose takes place in the mother liquors from the barium saccharate precipitation, but some goes through and builds up in the final molasses to about 20 per cent.The total quantity of this molasses is very small, but it is reported that several thousand pounds of high-purity crystalline raffinose have been recovered from it. The refined white granulated sugar crystals produced at Johnstownlo are distinctive in appearance, being rather long and needle shaped in comparison with normal sucrose crystals. This variation is attributed to crystallisation in presence of raffinose.Fig. 12 is a chromatogram of the sugar and the molasses. On the molasses chromatogram there appeared an unknown spot, afterwards identified; this is referred to later in the paper. The raffinose hydrate in the sugar was estimated at 0.8 per cent. by paper chromatography and as 0.9 per cent. by the two-enzyme method. DE WHALLEY, ALBON AND GROSS: APPLICATIONS OF PAPER The analysis showed- % Polarisa t ion .. * . .. .. .. .. 100.18 Sucrose . . .. .. .. .. .. .. 98-88 Moisture . . .. . . .. .. .. .. 0-05 Ash . . .. .. .. .. * . .. .. 0-027 Organic non-sugars . . .. .. .. .. 0.143 Raffinose hydrate . . .. .. .. . . .. 0.90 100*00 In the molasses the proportion of raffinose as shown by paper tests was about 16 per cent. and as shown by the two-enzyme method 15.5 per cent.The analysis was as follows- % Raffinose . . .. .. .. .. .. .. 15.48 Sucrose . . .. .. . . .. .. .. 55.84 Ash . . .. .. .. .. .. .. .. 2.67 Moisture . . .. .. .. .. .. .. 23.4 Organic non-sugars . . .. .. .. .. 2.61 100.0 From these figures it can be deduced that the shape of the sugar crystals is due to the fact that they are mixed crystals and that the effelct on the shape has not merely been caused by crystallisation in presence of raffinose. It is now known that raffinose is widely distributed in the vegetable kingdom and that it is generally found together with the tetra-saccharide stachyose. Raffinose is not sweet, but it is in all probability assimilated and utilised in human metabolism like most other naturally occurring sugars.Molasses cattle foods may contain up to 1 per cent. or more of raffinose. RAFFINOSE IN NORMAL WHITE BEET SUGARS- quality is low, the amount, although very small, is measurable. the following curious analysis- If white beet sugars are of good quality they contain very little raffinose, but if the One such sample showed % Polarisation .. .. .. .. .. .. 99.93 Ash . . .. .. .. .. .. . . .. 0.030 Reducing sugars . . .. .. .. .. .. 0.014 Moisture . . .. .. . . * . .. .. 0.048 Organic non-sugars . . .. .. .. .. ? 100.022May, 19511 CHROMATOGRAPHIC METHODS I N SUGAR AND ALLIED INDUSTRIES 297 The total, 100.022 per cent., indicates that the percentage of organic non-sugars, found by subtracting the figures for the other items from 100, is either absent or negative.The former is unlikely with an ash figure of 0.03 per cent., and the latter is impossible. A chromatogram showed the presence of 0.15 per cent. of raffinose hydrate, which enabled the polarisation to be corrected and the analysis re-arranged as follows- 0 1 70 Corrected polarisation (sucrose) . . .. . . . . 99.705 Raffinose hydrate . . .. . . .. . . . . 0.15 Ash . . .. .. .. . . .. . . . . 0-030 Reducing sugars . . .. . . .. . . . . 0.014 Moisture (8 hour at 105" C)* . . .. . . . . 0.048 Unestimated or organic non-sugar . . . . .. 0.053 100~000 * The loss of water from raffinose hydrate under these conditions is negligible. There is no doubt that the raffinose present is optically active and is affecting the polarisation reading to the extent of 1.5 per cent. for every 1.0 per cent.of raffinose present. HYDROLYSIS OF SUGARS WITH ENZYMES- Mention has already been made of the use of the enzyme invertase and the mixed enzymes invertase and melibiase to prove the identity and estimate the quantity of raffinose in raw beet sugar. Fig. 13 is a chromatogram of raffinose after treatment with the enzymes. With invertase the raffinose has yielded fructose and melibiose, and with the mixed enzymes the raffinose has completely hydrolysed to the three constituent hexoses, fructose, glucose and galactose. Another worker had informed us that he had experienced difficulty in the separation of lactose and melibiose; hydrolysis of either with acid yielded the same two sugars, glucose and galactose. Our experience showed that a satisfactory separation could be achieved with a suitable ethyl acetate - pyridine - water solvent.If the hydrolysis was made with melibiase instead of with acid, lactose was unaffected, whereas melibiose was broken up and a very clearly defined separation was obtained. ENZYME PURITY- trace of melibiase was also present. separation of the sugars this could not have been detected. A sample of invertase was found to hydrolyse melibiose slightly, so showing that a I t is almost certain that without chromatographic HYDROLYSIS OF SUCROSE BY INVERTASE- During the examination of the activity of melibiase in culture yeasts, the progress of hydrolysis was followed by chromatography of minute samples. A preparatory study was made of the hydrolysis of sucrose by invertase. In the early stages of the hydrolysis the chromatogram (Fig.14) showed much unchanged sucrose and some glucose and fructose. In addition to these there is an unknown spot above the sucrose spot. Our colleague, P. Blanchard, who was engaged on this study, was interested in the nature of this compound, the quantity of which decreased as hydrolysis proceeded and finally disappeared when hydrolysis was complete. The portion of the paper containing this spot of unknown material was cut out, extracted, and re-chromatographed after complete hydrolysis with invertase. From the position of the spot on the original chromatogram and the pro- portion of the hexoses produced by hydrolysis the material appears to be a new tri-saccharide composed of two fructose and one glucose units. This compound is the cause of the unknown spot in Fig.12. Further investigation of this reaction revealed the presence of yet another compound, which is thought to be another tri-saccharide of slightly different structure. These new compounds have been separated by adsorption chromatography on activated charcoal and small quantities have been crystallised. Bacon and Edelman, working with a similar tech- nique at the University of Sheffield, have also observed this effect with invertase preparations. This complication could be due to the reactive fructofuranose first liberated by the hydrolysis of sucrose linking up with some unchanged sucrose. It may be catalysed by invertase or by some other enzyme occurring in invertase preparations. No similar effect has yet been observed during the hydrolysis of sucrose by acids.298 [Vol.76 BEER- Small amounts of sugar remain after the normal course of fermentation, and for the determination of these we suggested paper chromatography. A trial gave satisfactory results. In Fig. 15 dextrins appear at the top of the chromatogram as a string of compounds. The sugars maltose, glucose and fructose are also clearly separated in that order. The investigations were not pursued, but they indicate the possibilities of the method for the separation of the hydrolysis products of starch. DE WHALLEY, ALBON AND GROSS: APPLICATIONS O F PAPER COLOURING MATTER- The study of the colouring matter present in raw materials and also of that formed during refining processes is made difficult by the presence of a multitude of coloured compounds of several types.Spectrophotometric absorption curves over the range of the visible spectrum show no characteristic peaks to assist identification. It was thought that if separation into components could be effected, more information could be obtained either chemically or spectrophotometrically from each isolated component. The colour in a solution of a raw cane sugar was isolated by column chromatography and the concentrated colour was re-chromatographed on paper with a solvent consisting of a mixture of butanol, acetic acid and benzene. The resultant chromatogram when viewed in ultra-violet light showed a number of strongly fluorescent, well-separated spots in addition to those visible by normal light, as shown in Fig.16. Further investigations showed that sufficient quantities for the identification of the :separated compounds could be obtained by the use of adsorption columns packed with acdsorbents such as activated alumina, or magnesia. According to the adsorbent used the coloured compounds can be divided into several components by the formation of distinct bands or zones that can be separated by extrusion or elution. It is now possible to treat each component or group of coloured com- pounds separately and to obtain information specific for the particular component ; this represents an obvious advance over previous investigations in this field. WORK OF OTHER INVESTIGATORS Some mention should be made of the valuable contributions to the subject by other investigators.Brown and Dahlbergll have separated various nitrogenous compounds from Steffen waste waters and then applied paper chromatography to the identification of the various amino-acids. They have identified aspar tic acid, glutamic acid, serine, glycine, alanine, tyrosine, valine and leucines in both protein-free beet juice and in Steffen waste. Pratt and Wiggins12 have adsorbed all the amino-acids present in cane juice on a column of Zeocarb 215 and eluted the column with 2 N hydrochloric acid. The concentrated eluate was submitted to paper chromatography, first with a mixed solvent of collidine and lutidine and then in a second dimension with phenol as solvent. They identified the following amino- acids: aspartic acid, glutamic acid, serine, glycine, alanine (and possible glutamine), amino- butyric acid, lysine, valine and leucine or an isomer.Binkley and Wolfrom13 used chromatographic fractionation with various concentrations of ethanol on columns of clay and obtained four. crude fractions, characterised by their copper reduction values before and after hydrolysis. Re-fractionation of two of these crude fractions followed by the extraction of sections yielded crystalline sucrose, d-glucose and .meso-inositol; these were also isolated, toget her with d-fructose and d-mannitol, as crystalline acetates by acetylation and subsequent chromatographic separation. &Glucose is present in excess of d-fructose; reducing substances other than hexoses were present in the reducing fraction. Application of similar procedures to a residue from the fermentation of molasses led to the isolation of crystalline sucrose and d-mannitol and their crystalline acetates, and also crystalline acetates of d-glucose, d-fructose, d-erythritol, d-arabitol and an unidentified substance.We desire to thank those of our colleagues who have given us a great deal of time and help, and in particular D. Harrison for the preparation of the photographs and F. Carman for his advice and guidance on these. We also thank W. Underwood and A. Graves for their skill and ingenuity in constructing, at low cost, the apparatus required. Finally, we express our thanks to the Directors of Tate and Lyle Limited for permission to publish the results of the work carried out in their Research Laboratories.May, 19511 CHROMATOGRAPHIC METHODS IN SUGAR AND ALLIED INDUSTRIES 299 REFERENCES 1.2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. de Whalley, H. C. S., “Proceedings of the Tenth Session of the International Commission for Uniform Methods of Sugar Analysis,” Supplement to Int. Sug. J., 1950, p. 31; Int. Sug. J., 1949, 52, 223. Zerban, F. W., and Sattler, L., Sugar, 1947, 42, 44. Binkley, W. W., and Wolfrom, M. L., “Chromatography of Sugars and Related Substances,” Sugar Research Foundation Scientific Report Series No. 10, 1948. Partridge, S. M., Nature, 1946, 158, 270; Biochem. J., 1948, 42, 238. Consden, R., Gordon, A. H., and Martin, A. J. P., Biochern. J . , 1944, 38, 224. Jones, J. K., Hirst, E. L., and colleagues, Disc. Farad. SOC., 1949, 7, 268. Sattler, L., and Zerban, F. W., Ind. Eng.Chem., 1945, 37, 1133. de Whalley, H. C. S., Int. Sug. J., 1950, 52, 127-129, 151-152 and 267. Albon, N., and Gross, D., Analyst, 1950, 75, 454. Van Hook, A., “Sugar,” Ronald Press Co., New York, 1949, pp. 73, 74. Brown, R. J., and Dahlberg, H. W., Abstracts, 117th Meeting Amer. Chem. SOC., March and April, Pratt, 0. E., and Wiggins, L. F., Proceedings of the 1949 Meeting of B.W.I. Sugar Technologists, Binkley, W. W., and Wolfrom, M. L., J . Amer. Chem. Soc., 1950, 72, 4778 1950, p. 2Q. pp. 29-38. TATE AND LYLE LIMITED RESEARCH LABORATORY RAVENSBOURNE, WESTERHAM ROAD KESTON, KENT DISCUSSION MR. L. EYNON congratulated the authors on their very important contribution to sugar analysis. In particular, they had solved the problem of the estimation of raffinose in raw beet sugars.Before the 1914-18 war, we had imported large quantities of raw beet sugar from the Continent and a con- siderable proportion of Continental sugar was so-called “process” sugar, extracted from beet molasses by the lime or strontia extraction method. This sugar contained considerable amounts of raffinose; with the methods of estimating raffinose available at that time it was not possible to estimate less than 0-3 to 0.5 per cent. with certainty. With the chromatographic method developed by the authors it was now possible to estimate as little as 0.05 per cent. Mr. Eynon opened the discussion by asking whether any attempt had been made to determine the melassigenic factor of raffinose. MR. DE WHALLEY replied that not much was known about the melassigenic factor of raffinose.A. Herzfeld, in 1910, stated that joint work with Foerster a t an earlier date had shown i t to be negligibly small. DR. J. R. NICHOLLS said that from a scientific point of view the paper was a highly valuable con- tribution. It showed that minor constituents could now be separated and determined to a high degree of accuracy and that the presence of hitherto unsuspected constituents could be demonstrated. The method showed great promise for the separation of such vague constituents as the colouring matters, and the identification of the separated components should be greatly facilitated. DR. A. CARRUTHERS said that as a representative of the Corporation that produces sugar from beet, he offered congratulations to Mr. de Whalley and his colleagues on their work.He referred to the degree of accuracy pertaining to the estimation of raffinose in raw beet sugar and asked whether the results of comparisons between the test material (raw sugar) and the same test material with added known quantities of pure raffinose would differ from the results found by comparing the test material with mixtures of raw cane sugar and known amounts of raffinose. The addition of raffinose to the sample under test had been tried and i t had been found that raw cane sugar containing no raffinose was a more suitable adjunct to the pure raffinose to balance the slowing of the separation by the salts present. MR. H. E. MONK enquired about the possibility of using these methods in the examination of honey for the presence of invert sugar made either with acid or enzymes.MR. DE WHALLEY replied that they had not examined honey. If invert sugar had been made by invertase and the sucrose was not completely inverted, some of the newly discovered tri-saccharides would be present. These might normally be present in honey. Hydroxymethylfurfural was formed either by heating or on inversion with acid. MISS E. I. BEECHING asked the lecturer if he had tried precipitation of the protein before making chromatographs. MR. DE WHALLEY replied that removal of fat and protein, if present, was desirable, but that in his laboratory no work had been done on this subject as neither were present in their sugar products. DR. J . H. HAMENCE said that there were two questions concerning technique that he would like t o ask the authors.First, the authors had shown paper chromatography to be an extremely valuable analytical tool for the differentiation and estimation of the different sugars present in mixtures, but from the description given by MI-. de Whalley, it took a t least 18 hours before results were obtained. Was it MR. DE WHALLEY replied that there was no reason to expect a difference in the results.300 TINSLEY, TAYLOR AND MOORE: DETERMINATION OF CARBON DIOXIDE [VOl. 76 possible t o shorten this time in order to obtain a qualitative indication of the sugars present in a mixture? Secondly, was it essential that the form of apparatus used by the authors be strictly adhered to? For a number of years they had worked in their laboratory with strips of filter-paper between two sheets of glass plate, following the technique described by Rutter. Was it possible to use this technique for the separation of sugars qualitatively, instead of using the rather more elaborate technique described by Mr. de Whalley? MR. DE WHALLEY said that separation was normally achieved overnight and did not consume operator's time. MR. N. ALBON said it was possible in qualitative work and simple investigations to obtain separations more quickly by variation of the solvent mixtures by inclusion of more water in the solvent phase or by use of faster paper. The spots were less well defined and the method was unsatisfactory for quantitative work. The apparatus described sounded more elaborate than i t actually was. It could be further simplified for short-travel separations designed for qualitative use. MR. E. M. LEARMONTH enquired whether it was necessary to have control spots a t both edges and to keep a stock of pure sugars. MR. ALBON said that a photometric method was possible and there were a number of references in the literature. However, when only a few samples had to be tested, controls need only be on one edge. MR. DE WHALLEY pointed out that the quantities of stock sugar required were very small as only microgram amounts were used 'for controls. DR. E. H. CALLOW asked what were the minimum amounts that could be estimated. He also asked whether the authors had tried the use of a higher temperature. MR. ALBON replied that the minimum amount required depended on the particular sugar; it could be less than 6 pg or even as little as 1 pg. The use of a higher temperature appeared to be a useful modifica- tion, but the design of the apparatus would become more complicated. MR. R. F. MILTON asked if two-dimensional paper chromatography was useful for separating closely related sugars. MR. ALBON replied that it had on occasion been used for rarely occurring constituents, but that they had found it better to use one-dimensional chromatography and suitably modify the solvent as required. Various types of apparatus could be used to give similar results. The glass-plate method described by Dr. Hamence had not been tried. Was there any method of assaying photometrically? A large number of test samples could be run on one sheet of paper.

ISSN:0003-2654    

DOI:10.1039/AN9517600287

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

300 TINSLEY, TAYLOR AND MOORE: DETERMINATION OF CARBON DIOXIDE [Vol. 76 The Determination of Carbon Dioxide Derived from Carbonates in Agricultural and Biological Materials BY J. TINSLEY, T. G. TAYLOR AND J. H. MOORE Methods for the estimation of carbonate are reviewed. A simple titration procedure for the determination of carbonate in a wide variety of materials is described. Decomposition by acid and absorption of carbon dioxide by alkali are accomplished under reduced pressure within one reaction vessel. Amounts of carbon dioxide ranging from 1 to 80 mg can be determined to within k1 per cent. Part 1. Introductory Review METHODS for determining carbon dioxide released by the action of acids on carbonates comprise three main groups according to the way the measurement is made.GRAVJMETRIC METHODS- In these the carbon dioxide may be weighed either (i) directly, by the gain in weight of an absorbent, or (ii) indirectly, by the loss in weight of the system. For the most accurate results a direct procedure is usually employed in which the gas evolved from the reaction vessel is dried and purified by passing through a suitable absorption train, Formerly the carbon dioxide was usually absorbed in potash bulbs, but now it is more convenient to use solid absorbents such as soda-lime, soda-lime - asbestos (Carbosorb) or sodium hydroxide - asbestos (Ascarite). Provided that other gaseous products are not derived from, e.g., nitrite, sulphite or sulphide, accurate However, for many purposes an indirect method is less tedious to operate.May, 19511 DERIVED FROM CARBONATES IN AGRICULTURAL MATERIALS 301 results are secured if due precaution is taken to prevent loss of water from the system.Several designs of apparatus have been used but probably the best known is the Schrotter tube. An apparatus of simpler construction designed by Scott and Jewelll was recommended by Scott and Furman.2 The reaction vessel consists of an ordinary conical flask in which is placed a small volume of acid solution. The test material, contained in a squat flat- bottomed tube, is carefully introduced, and the flask is then closed with a rubber stopper carrying a calcium chloride exit tube and an inlet tube dipping below the surface of the acid. After weighing, the assembled apparatus is tilted to overturn the small tube and gently shaken to mix the contents.A slow stream of dried air is passed through the apparatus for 30 minutes before the final weighing is performed. A very similar type of apparatus was used by Erickson, Li and Gieseking3 for the deter- mination of carbonate in soils. A small tube was fused to the inside of the reaction flask to hold 5ml of a concentrated solution of trichloroacetic acid and the reaction was allowed to proceed for 12 hours after mixing. GASOMETRIC METHODS- In these the carbon dioxide is released and measured in a closed system either (i) volu- metrically, at known temperature and pressure , or (ii) manometrically, at known temperature and volume. The form of apparatus depends somewhat on the type of material being analysed. For solids such as powdered limestone or soil, Collins* developed a volumetric apparatus from an earlier design by Scheibler.A full description of the Collins calcimeter and its operation is given in Wright’s textbooks of agricultural and soil analysis.5~~ Briefly, it consists of a gas- measuring unit immersed in a water jacket and connected by rubber tubing to a reaction vessel. For this a 150-ml bottle or flask is used to contain the test sample together with a small horn tube holding diluted hydrochloric acid (1 + 2). This arrangement is the reverse of that used by Scott and Jewell, but the same result is achieved by tilting the flask to discharge the acid on to the carbonate material. The maximum volume of carbon dioxide that can be handled in the apparatus at room temperature and pressure is less than 50m1, corresponding to about 0.2 g of calcium carbonate.If carefully calibrated, the apparatus gives reliable results, except with small quantities of carbon dioxide. An alternative form of apparatus for soils and other solid materials was described by Singh and MAth~r.~ This was adapted from a design used originally for biological fluids, embodying a manometric tube filled with Brodie’s solution. Burns and Hendersons employed the Barcroft manometric apparatus for the determination of carbonate in bone tissue. However, for measuring carbon dioxide and other gases evolved from biochemical solutions, the well-known Van Slyke apparatus has attained great popularity. The original tube described by Van Slykeg was calibrated for volumetric measurement of the gas over mercury at atmospheric pressure.Later, Van Slyke and NeilllO introduced the manometric version for measuring the pressure of the gas when reduced to a standard volume of 0-5 or 2.0ml. Greater accuracy was possible with this apparatus, a full description of which is given in the textbooks of Peters and Van Slykell and of Hawk, Oser and Summerson.12 Robinson13 described a modification of the original Van Slyke tube in which powdered solids were admitted through an opening fitted with a ground-glass stopper. Bowes and Murray,14 in studies on the composition of teeth, used a Van Slyke - Neill tube having a stopcock with a large bore of 7mm diameter, through which the powdered material was introduced from a weighing funnel. Such arrangements have not persisted, because it is more generally convenient to use the separate reaction vessel devised by Van Slyke, Page and Kirk15 for connection to the standard Van Slyke - Neill tube. These workers first used it for the micro-determination of carbon in organic compounds by a wet combustion procedure, which was later modified by Van Slyke and Folch.lG The carbon dioxide is absorbed in a 0.5 N solution of sodium >hydroxide containing 0.5 M hydrazine, to reduce any halogens, and then regenerated with 2 N lactic acid solution.MacFadyenl7 used the same apparatus for deter- mining carbon dioxide released from amino-acids by the ninhydrin reaction but, in order to reduce the solubility of carbon dioxide, the alkali and lactic acid solution were almost saturated with sodium chloride.For the micro-estimation of carbonate in bone, Sobel, Rockenmacher and Kramer18 found that it was necessary to use a water jacket around the connecting arm to prevent fumes passing into the Van Slyke - Neill tube when the bone sample was boiled with 3 N hydrochloric acid in the reaction vessel. The great utility of the Van Slyke - Neill302 TINSLEY, TAYLOR AND MOORE: DETEIRMINATION O F CARBON DIOXIDE [VOl. 76 apparatus with attached reaction vessel is further illustrated by a method for the determination of inorganic and organic carbon in the same soil. sample that has recently been devised by Bremner.19 Rapid manometric devices that involve no ad.justment in volume of the gas are available for carbonate solutions and materials such as limestones.For use when moderate accuracy suffices, Piper20 has described a modification of Passon's method that is used for soils a t the Waite Institute in Australia. Dixon and Williarns21 developed a stainless steel vessel fitted with a false base forming two compartments. These hold the carbonate material separate from a 20 per cent. solution of phosphoric acid until mixing is performed by tilting. TITRIMETRIC METHODS- In these the carbon dioxide is absorbed in a standard alkali solution, usually barium or sodium hydroxide. Methods differ in points of detail according to the form of the reaction vessel, the means of absorption and the mode of titration. The reaction vessel may consist simply of a flask with provision for the inlet of acid and the outlet of carbon dioxide to the absorption vessel.In the method described by Amos22 for the determination of carbonate in soils, the sample, containing not more than 0 6 g of calcium carbonate, was boiled with a solution of approximately 2 N hydrochloric acid. The carbon dioxide passed through a condenser into a Reiset absorption tower containing 0.5 N sodium hydroxide solution. When absorption was complete, the solution was titrated with standard hydrochloric acid in two stages, according to the method used by Brown and Escomb.23 The first stage was titrated with phenolphthalein as indicator and the second stage with methyl orange. The difference between these two titration values represents bicarbonate. However, such a method is not employed a t present for soils, because boiling with hydro- chloric acid often gives high results, as was emphasised by the Report of the Organic Carbon Committee to the Third International Congress olf Soil Science.24 The high results are due to decomposition of organic matter, especially uronides, and also to the formation of chlorine by oxidation when manganese oxides are present.These errors can to a great extent be avoided either by carrying out the reaction at :room temperature or by boiling at a low temperature under reduced pressure with hydrochloric acid containing a reducing agent such as ferrous or stannous chloride. The 1owe:r the temperature and the lower the con- centration of acid the less is the danger of decarhoxylation of the organic matter, but when magnesian limestone granules are present in the sample, they often dissolve very slowly in dilute acid at room temperature.In the tentative A.O.A.C. method26 for soils, the sample, ground to pass a 60-mesh sieve, is treated with approximately N hydrochloric acid solution containing 5 per cent. of stannous chloride. The reaction flask is mounted on a reciprocating shaker and connected by rubber tubing either to an absorption train or to a tower containing 0.5 N sodium hydroxide solution, depending on whether the determination is to be made gravimetrically or titri- metrically. For titration, the procedure of Winkler26 is followed, the alkali being first washed out of the tower into a 500-ml graduated flask, 10 ml~ of a neutral 25 per cent. barium chloride solution being then added to precipitate barium carbonate.After dilution and mixing, the flask is set aside for 4 hours before the excess of hydroxide is titrated with standard acid, phenolphthalein being used as indicator. For soils containing resistant carbonates, the reaction vessel is heated with a condenser inserted before the absorption vessel. It should also be noted that Fraps2' examined this method for the determination of carbonate in feeding stuffs and found it necessary to use stronger hydrochloric acid (diluted 1 + 1) for the complete decomposition of bone meal. In place of this rather lengthy procedure, Schollenberger's method has been widely used in America for soils and limestone materials. Decomposition is effected by boiling with acid under reduced pressure at about 30" C; although the method has undergone minor modifications since it was first described in 1930,28 the essential features remain un- changed in the latest (1945) desc~iption.~~ The sample, containing not more than the equivalent of 0.25 g of calcium carbonate, is placed in a 200-ml reaction flask, which is con- nected through an upright condenser with a one-litre absorption flask.The system is evacuated to about 2 cm of mercury, and then 50 ml of 0.2 N barium hydroxide solution are admitted to the absorption flask. Dilute hydrochloric acid containing ferrous or stannousMay, 19511 DERIVED FROM CARBONATES IN AGRICULTURAL MATERIALS 303 chloride is sucked into the reaction vessel and the contents are heated to maintain a steady rate of boiling at about 30" C. Generally the reaction is completed within a few minutes, after which the inlet to the reaction flask is opened to admit a slow stream of air, free from carbon dioxide.The barium hydroxide solution must be agitated by shaking the apparatus to break the film of carbonate and so ensure complete absorption of carbon dioxide before the flask is disconnected. Finally, the excess of hydroxide is titrated with 0.1 N hydrochloric acid, thymolphthalein or phenolphthalein being used as indicator. The volume of acid solution in the reaction vessel should be about 100m1, but the strength is varied from a dilution of 1 in 25 for ordinary use to P in 10 for use with resistant carbonates. Normally the determination can be completed within an hour. A quicker method is that described by Shaw and MacIntire30; in this, steam distillation is used to ensure rapid decomposition.The reaction flask is fitted with a stopper carrying a steam inlet tube and a tap funnel for introducing acid. A Liebig condenser forms the exit tube, which leads through a gas-scrubbing bottle to the absorption vessel. This consists of a 500-ml suction flask containing dilute sodium hydroxide solution, and is provided with a special reservoir tube that can be moved vertically in the neck of the flask while main- taining an airtight joint with a rubber sleeve. Inorganic carbonate materials are decomposed by introducing approximately N hydrochloric or perchloric acid solution, which is diluted to a concentration of about 0-4 N within the reaction vessel. It is necessary to pass steam for about 2 minutes and, if 1 minute is allowed for shaking the absorption vessel, the whole operation of decomposition and collection is completed within 5 minutes.The absorbent is then titrated with 0.1 N hydrochloric acid, either directly to the bicarbonate stage at pH 8.4, with phenolphthalein as indicator, or after the addition of barium chloride solution. For soil samples, the same procedure is followed except that the acid introduced into the reaction vessel contains 5 per cent. w/v of stannous chloride (SnC12.2H20). Even with this reducing agent and short digestion period, the error due to carbon dioxide released from the soil organic matter is slightly greater than by Schollenberger's method because of the higher temperature. In all these titrimetric methods so far described, each piece of apparatus, besides requiring individual attention during operation, needs to be heated for decomposition to take place.Hutchinson and MacLennanFl however, devised a method in which the carbonate was decom- posed in vacuo a t room temperature, and a slight modification of their method was recom- mended by Piperz0 for accurate determinations. The reaction vessel is connected through a simple spray trap to a larger absorption vessel, containing 0.1 N sodium hydroxide, After assembly the apparatus is evacuated as completely as possible before admitting approximately N hydrochloric acid into the reaction vessel from a tap funnel. For soil samples this acid may contain 3 per cent. w/v of ferrous chloride (FeC1,.4H20).The apparatus is shaken intermittently for 20 minutes before admitting a slow stream of air, free from carbon dioxide, through the tap funnel. The shaking is repeated to complete the absorption of carbon dioxide, and then barium chloride solution is added to the absorbent before titration with 0.1 N hydrochloric acid. If the end-point is taken as the stage when thymolphthalein just becomes colourless, it may be checked by addition of phenolphthalein, which should then give a pink colour that requires only another one or two drops to discharge it. A very similar method was used by Williams32 for the determination of carbonate in calcareous soils before the extraction of exchangeable bases. An excess of 0.5 N acetic acid was used in place of hydrochloric acid and the system was allowed to stand overnight under reduced pressure.Instead of sodium hydroxide, 0.2 N barium hydroxide solution was used as absorbent. In order to avoid individual manipulation during the absorption process, several workers have endeavoured to simplify this form of apparatus still further by carrying out decomposition and absorption within one closed vessel under reduced pressure. Van Slyke= devised the apparatus illustrated in Fig. 1 for the determination of carbonate in powdered bone and other materials. The powdered material is weighed into the bottom of a suitable tube about 20 mm in diameter, and this is placed inside a 250-ml suction flask containing a measured volume of 0.1 N barium hydroxide. The stopper carrying the tap funnel is inserted and the flask evacuated to less than 50 mm of mercury through the side tube, which is then closed with a screw clip.An excess of approximately N hydrochloric acid, usually about 5 ml, is dropped cautiously on the carbonate material from the tap funnel. When the first reaction has304 TINSLEY, TAYLOR AND MOORE: DETERMINATION OF CARBON DIOXIDE [VOl. ‘76 subsided, the flask is shaken with a rotary moveinent for 3 minutes and then set aside for 5 hours, for bone or other resistant material, with occasional shaking to break the carbonate film. Afterwards the vacuum is broken and, according to the original method, the barium carbonate is filtered on a Gooch crucible before the excess of hydroxide is titrated with 0.1 N hydrochloric acid, phenolphthalein being used as indicator.The same equipment and pro- cedure was used by Robinsonf3 for soils, limestones and marls. Materials containing much Fig. 1. Van Slyke’s apparatus Fig. 2. Simplified apparatus carbonate should be covered with water in the tube, a drop of octyl alcohol being added to minimise frothing when the acid is admitted. For soils containing little carbonate he suggested placing the sample in the flask and the barium hydroxide in the tube. Independently, Hepburnx devised an almost identical form of apparatus and used approximately 3 N hydrochloric acid for decomposition and 50 ml of 0.1 N barium hydroxide solution as absorbent. For resistant materials the flask was allowed to stand overnight and the excess of hydroxide was then titrated directly in the flask with 0.1 N oxalic acid solution, phenolphthalein being used as indicator. The acciiracy claimed for the method was &0*5 per cent.The same principle underlies the well-known m icro-diffusion method of analysis developed by C0nway.3~ Part 2. Proposed Method PRELIMINARY IXVE STIGATIONS The authors required a convenient method for the determination of carbonate in amounts equivalent to 1 to 100 mg of carbon dioxide in such diverse materials as bones, fresh intestinal contents from experimental animals, feeding stuffs, soils and liming materials. For some materials the yields of carbon dioxide are minute, yet most of the foregoing methods are not designed primarily for dealing with small amounts. A Van Slyke - Neil1 apparatus was not available, and it did not appear feasible to use Conway’s micro-method because of the difficulty of drawing small representative samples from such heterogeneous materials.Therefore recourse was made to the Van Slyke type of appxatus shown in Fig. 1. Preliminary trials were made with measured volumes of a standard sodium carbonate solution placed in the tube with the absorbent in the flask, and then with solution and absorbent interchanged. Initially, 0.2 N barium hydroxide solution was used as absorbent and the excess was titrated with hydrochloric acid, a mixed indicator solution of thymol- phthalein and phenolphthalein being used, as suggested by Conway. The surface of theMay, 19511 DERIVED FROM CARBONATES I N AGRICULTURAL MATERIALS 305 absorbent became coated with a film of barium carbonate and, although this could be broken by repeatedly shaking the flask, in the narrow tube it quickly reformed and prevented complete absorption.Attempts were made to avoid this caking by adding various reagents to the absorbent, but neither synthetic surface active agents nor glycols proved successful. Of the various alcohols tried only ethyl alcohol gave any promise. By adding an equal volume of rectified spirit to the 0.1 N barium hydroxide solution just before evacuation, caking was prevented, but this procedure could not be adopted because absorption of carbon dioxide was not quantitative in the presence of alcohol. In view of this difficulty and because it seemed desirable to retain the absorbent in a tube for convenience of titration, it was decided to use sodium hydroxide instead.Winkler’s procedure was first used for titration, barium chloride solution being added to precipitate carbonate, but results were variable with standard carbonate solutions and it was discontinued. It appears difficult to ensure the right conditions for precipitation of the normal barium carbonate, although greater success might possibly have been secured with strontium chloride in place of barium chloride, as used by Benedetti-Pichler, Cefola and Waldman.36 Instead, titration of the absorbent in two stages was investigated and finally adopted, as described later. This gave satisfactory results when a good vacuum was maintained in the flask overnight, but all too frequently air leaked in. Eventually the simpler form of apparatus illustrated by Fig.2 was devised. This arrangement embodies features from several of the methods described previously( the absorbent, carbonate material and acid all being placed within the flask before evacuation. The method has proved very successful and is described in detail below. METHOD APPARATUS FOR RELEASE AND ABSORPTION OF CARBON DIOXIDE- For quantities of carboPz dioxide zlp to 10 mg-A convenient number of thick-walled 200-ml conical Pyrex flasks are selected for uniform neck dimensions to allow a long and a short tube to be placed within each flask (Fig. 2). These are supplied with well-fitting one-holed rubber stoppers (size M or N) carrying a short length of capillary tubing to which is joined about 8 to 10 cm of 2-mm bore rubber pressure tubing with a screw clip attached.Two sets of flat-bottomed specimen tubes are used, each of 19mm (2 inch) diameter; the larger, in which is placed the absorbent, is 80mm long and the smaller is cut to 60mm long. For quantities of carbon dioxide exceeding 10 nzg-For larger quantities of carbon dioxide, 500-ml suction flasks are convenient, the rubber tubing being joined to the side-arm and the neck closed with a solid rubber stopper. Appropriate sizes for tubes are 25 mm (1 inch) in diameter, the larger being 127 mm (5 inches) long and the smaller either 51 mm (2 inches) or 76 mm (3 inches) long. REAGENTS- Distilled water-Boil vigorously for 20 minutes to free from carbon dioxide and protect with a soda-lime guard tube. Sodium hydroxide stock solution-Prepare a carbonate-free stock solution containing 1 g per ml of water.Allow to stand until clear and then siphon off the supernatant liquid from the carbonate residue. About 10.5 rnl of this solution diluted to 1 litre is used as a 0-2 N solution. Sodium hydroxide solution for adding to intestinal contents-Dilute the stock solution with carbon dioxide-free water to give an approximately N solution. Sodium hydroxide solution for absorption-Dilute the stock solution to 0.05 N , 0.1 N or 0.2 N , according to the amount of carbon dioxide to be determined. These solutions need not be accurately standardised, but must be protected from carbon dioxide by soda-lime guard tubes. Hydrochloric acid-Prepare a solution for titration by dilution of A.R. concentrated acid (approximately 11 N ) with carbon dioxide-free distilled water to a strength equivalent to the alkali absorbent.Acid solutions for decomposition-(i) Perchloric acid, 20 per cent. w/w, approximately 3 N , for soils, etc. (ii) Hydrochloric acid or trichloracetic acid, approximately N , for bone tissue, etc. Thymol blue indicator-Prepare a 0.1 per cent. solution by triturating 1 g of thymol blue with 21.5ml of 0.1 N sodium hydroxide and dilute to 1 litre with water. This solution must be accurately standardised.306 TINSLEY, TAYLOR AND MOORE: DETERMINATION OF CARBON DIOXIDE [Vd. 76 Screened methyZ mange indicator-Prepare a solution containing 0.2 per cent. of methyl orange and 0.28 per cent. of xylene cyanol FF in 50 per cent. ethanol. Standard sodiwn carbonate solution-Prepare from A.R. anhydrous sodium carbonate heated in an electric muBe at 270" C for half an hour. Dissolve 9.6355 g in 1 litre of water, to give a solution equivalent to 4 mg of carbon dioxide per millilitre. This solution is further diluted as desired. These solutions readily absorb atmospheric carbon dioxide and must be protected with soda-lime guard tubes. Bufer solutions-(i) 0.05 M potassium hydrogen phthalate adjusted to pH 3.8 with hydrochloric acid. (ii) 0-05M borax adjusted to pH 8.2 and 8.3 with hydrochloric acid. PROCEDURE FOR RELEASE AND ABSORPTION OF CARBON DIOXIDE- use. All glassware is thoroughly cleaned in chromic acid mixture, washed and dried before For 1 to 10 mg of carbon dioxide in small flasks- (i) If no more than about 2 g of dry powdered material is to be taken, weigh the sample into the short tube and insert it in the flask into which 10 to 25 ml of decomposition acid has been measured with a pipette.For this o:peration hold the flask at an inclined angle in one hand and gently slide the tube into an upright position with the aid of the index finger of the other hand. Finally, carefully introduce the larger tube, containing 5 ml of 0.05 N or 0.1 N sodium hydroxide solution for absorption, and then insert the rubber stopper in the neck of the flask after wetting it with distilled water to obtain a good seal. To minimise errors due to absorption of carbon dioxide from the air in the flask, immediately connect the rubber tube to a suction line operated by a vacuum pump or water jet and evacuate the flask until small bubbles of gas arise in the absorbent.At this point, corresponding to a pressure of about 6 cm of mercury, close the screw clip and disconnect the suction line. Tilt the flask to overturn the smaller tube and allow the contents to react with the acid. Ensure thorough mixing by gentle swirling and set the flask aside overnight for 18 hours to allow complete absorption. The sodium hydroxide solution should be measured into the larger tube just before evacuation; it is convenient to draw each aliquot from a flask in which the solution is protected from the atmosphere by a layer of petroleum ether. If possible, it is convenient for one worker to measure the sodium hydroxide solution while another evacuates the flasks, when a batch of analyses is being made. (ii) If the material is so bulky and low in carbonate content that a large sample must be taken, weigh it into the flask and place 5 ml of acid in the short tube.For dry materials, such as feeding stuffs, add 10 to 20 ml of carbon dioxide-free water to ensure thorough wetting and mixing with the acid, provided that there is no risk of losing carbon dioxide from the wetted material. Follow the remainder of the procedure as usual. With wet material, such as intestinal contents, no extra water is needed, but the operations should be performed without delay to avoid loss of carbon dioxide or gain from the atmosphere. Samples of intestinal contents that are liable to lose carbon dioxide during evacuation of the flasks should be made alkaline with 5 m l of approximately N sodium hydroxide added to about 5 g of wet material in the flask.After evacuation and mixing with 5 ml of 20 per cent. perchloric acid solution a sufficiently large excess of acid remains. For more than 10 mg of carbon dioxide- For soils and ground limestone samples 500-ml flasks are used, but essentially the same procedure is followed. Weigh from 0-5 to 5 g of soil into the flask, depending on the carbonate content, Place 10 ml of 20 per cent. perchloric acid solution in the smaller tube and 20 ml of 0.2 N sodium hydroxide in the larger one. Theoretically this amount allows for the absorption of 88mg of carbon dioxide to the sodium carbonate stage. For soils with a very low carbonate content more than 5 g can be taken and a larger volume of acid (15 or 20 ml) used to ensure thorough mixing.The use of reducing agents in the acid solution has not been examined, as it was considered that the error due to decom- position of organic matter would be very small under these conditions. Whichever procedure is followed, blank determinations on the reagents must be carried out in duplicate or, preferably, triplicate for each batch of analyses. After the flask has stood overnight, release the vacuum by unscrewing the clip on the inlet tube. Originally carbon dioxide-free air was admitted through an absorption tube, A ring of water placed around the stopper helps to maintain the seal.&fay, 19511 DERIVED FROM CARBONATES IN AGRICULTURAL MATERIALS 307 but this proved unnecessary provided the titration was performed immediately.A con- venient precaution is to add 1 or 2 ml of petroleum ether to the absorption tube as soon as the flask is opened. This serves to protect the solution from atmospheric contamination and to prevent the escape of carbon dioxide during the first stage of the titration. TITRATION PROCED u RE- To the small tubes containing 5 ml of absorbent, add 4 drops of thymol blue indicator and titrate the solution to the bicarbonate stage with hydrochloric acid of appropriate strength. It is convenient to use a'5-ml burette with a rubber tube and pinchcock connected to a capillary jet of sufficient length to reach to the bottom of the tube. Add the acid in small amounts until the indicator begins to change colour from blue to green. At this point make further additions very carefully, use the jet itself for stirring, and allow adequate time after each addition. Match the colour against two standard tubes of the same dimensions containing about the same volumes of the indicator and buffer solutions adjusted to pH 8-3 and 8-2 respectively.At least one minute should elapse between the addition of the last fractions of acid required to bring the colour to between those of the two standards. The standards can be used for some time provided that the tubes are kept well stoppered and are stored in the dark when not in use. If the end-point is overshot it may be quickly recovered by a small addition of carbonate-free sodium hydroxide solution. Record the volume of acid required for this first titration, and then add 2 drops of screened methyl orange indicator to the tube before continuing the titration rapidly to the second stage.For this end-point, match the tube with a tube of buffer solution at pH 3.8. We have found i t convenient to use a separate 2-ml or 5-ml burette for this second titration to avoid errors due to draining of the first burette on standing. Alternatively, a micro-burette of Rehberg or Conway type could be used with acid of suitable strength. For the larger tubes the same procedure is followed, except that 10 drops of thymol blue and 5 drops of screened methyl orange indicator are adequate for 20 ml of absorbent. It is best to agitate the solution with a glass rod with a flat circular disc at the end; care must be taken not to break the surface of the petroleum ether layer during the first stage of the titration.It is convenient to use 25-ml and 10-ml burettes and, if desired, these may be of a self-levelling type, connected with a reservoir. CALCULATION- The carbon dioxide content is represented by the second titration value. (i) (ii) NaOH + Na&O, + 2HC1+ 2NaC1+ NaHCO, + H,O (pH 8-3) NaHCO, + HCl 3 NaCl + H,CO, (pH 3.8) From equation (ii), 1 ml of N hydrochloric acid = 44 mg of carbon dioxide. A correction must be made for the blank on the apparatus and reagents. Hence- Carbon dioxide content = 444y - x) mg, where x = normality of acid y z = mean value for blank in millilitres. = millilitres of acid used in second titration If the volume of standard sodium hydroxide solution is carefully measured by pipette, the total volume of acid used for the titration will be the same each time within the limits of experimental error.For routine operation it is therefore unnecessary to perform the second stage of the titration once the mean values for the total titration and the blank have been determined. Alternatively, the acid used for the first part of the titration need not be recorded in routine work. However, it is safest to record both titrations because they afford a useful check on the accuracy of working, particularly if some acid should perchance be carried into the absorption tube as spray or splash. This we have found to happen only very occasionally during many determinations, but even so, provided an excess of alkali remains in the tube, the second titration is unaltered and the result is valid.DISCUSSION OF TITRATION PROCEDURE The procedure adopted is essentially that devised by WardeF and developed by Brown and Escombe.23 It has been fully described and discussed by Kolthoff and StengeF in their textbook of Volumetric Analysis. A careful examination of the method was made by308 TINSLEY, TAYLOR AND MOORE DETERMINATION OF CARBON DIOXIDE [VOl. 76 Benedetti-Pichler, Cefola and WaldmarP and also by Shaw and MacIntire.30 Since the accuracy of this method for the determination of carbonates depends so largely on the estima- tion of the carbon dioxide absorbed by sodium hydroxide solution it is worth considering the main features of the titration. TITRATION TO THE BICARBONATE STAGE- According to Kolthoff and Stenger the pH of sodium bicarbonate solution is 8.35 whether the solution is 0.1 N or 0.01 N, but other workers have accepted values as low as 8-2 for the equivalence point.Because the inflexion point of the titration curve is not very sharp, phenolphthalein and thymolphthalein are not very suitable indicators. Thymol blue is more satisfactory, but it is still necessary to employ a solution of the indicator at known pH for matching the end-point of the titration. Some workers have used sodium bicarbonate solution for this purpose, but it is not stable and readily loses carbon dioxide with a con- sequent rise in pH value. It is convenient to use two comparison tubes of borate buffer solution adjusted to pH 8.3 and 8.2 respectively. These show a distinct colour difference with thymol blue when viewed against a white background, the lower value being more TABLE I RECOVERY OF CARBON DIOXIDE FROM SODIUM CARBONATE SOLUTIONS Carbon Number of Standard Coefficient of dioxide, Recovery, determinations deviation Variation mg % (a) With sinall Jusks- 1 100.7 8 0.012 1.22 2 100.4 9 0.023 1-15 3 100.5 9 0.018 0.61 5 100.2 9 0.004 0.07 10 100.5 8 0.029 0.29 (b) With large $ush- 12.5 100.4 40 100.3 66 99.4 82.5 99.2 9 0.120 8 0.186 10 0.626 8 0.295 0.96 0.46 0.60 0.36 yellow than green.This contains 6 volumes of 0.1 per cent. thymol blue to 1 volume of 0.1 per cent. cresol red and changes colour sharply from violet a t pH 8.4 tcl blue a t 8.3 and rose a t 8.2. However, the presence of cresol red renders the end-point with methyl orange at the second stage of the titration somewhat less distinct.It is most important to realise that the reaction proceeds relatively slowly, especially as the end-point is approached and the addition of hydrochloric acid forms carbonic acid locally in the solution. The dehydration reaction, H,CO, -+ H,O + CO,, then proceeds much more readily than the formation of bicarb'onate, CO, -+ OH' --f HCO,'. Consequently the titration to the first stage should be performed slowly with gentle stirring to avoid local concentrations of acid. When the end-point is almost reached, about 1 minute should elapse between the last two additions. Benedetti-Picliler, Cefola and Waldman showed the nee? to prevent loss of carbon dioxide from the system and recommended that a stopperedflask be used as the titration vessel.In the present investigation good results were obtained in narrow titration tubes by having the surface s f the alkali solution protected with alayer of petroleum ether and directing the tip of the capillary jet to the bottom of the tube. Alternatively, the indicator mixture of Simpson39 may be used. TITRATION TO THE CARBONIC ACID STAGE- Originally, WardeP removed carbon dioxide by boiling the solution with an excess of acid, but Brown and E~combe,~ used methyl orange as indicator for the direct titration of the bicarbonate. The second inflexion point cd the titration curve occurs around pH 4, but it becomes less sharp as the concentration of carbonate is decreased and it is also influenced by the amount of carbon dioxide retained in solution. Cooper"lo demonstrated thatthe pH value of the equivalence point approached 5 at very great dilutions, such as occurin natural water.We have found for concentrations of the order of 0.1 N that screened methyl&fay, 19511 DERIVED FROM CARBONATES I N AGRICULTURAL MATERIALS 309 orange is satisfactory when used in conjunction with a comparison tube containing phthalate buffer solution adjusted to pH 3.8. Vigorous stirring to facilitate removal of carbon dioxide sharpens the end-point. RESULTS The accuracy of the method described was first established with standard solutions of sodium carbonate, and a summary of the results obtained is shown in Table I. Further, the method has been used for extensive analyses on various materials, and a comparison with other methods was made on samples of bone and soils.With the small flasks, replicate determinations were made of the carbon dioxide content of a standard sample of powdered bone kindly supplied by Dr. A. E. Sobel of the Jewish Hospital, Brooklyn, U.S.A. The results, given in Table 11, show close agreement with his value obtained with the Van Slyke - Neil1 apparatus. With the larger flasks, a group of ten soil samples, air-dried and passed through a 40-mesh sieve, were analysed for their carbonate content, and the results agreed closely with the values obtained gravimetrically with a Schrotter tube, as shown in Table 11. TABLE I1 DETERMINATION OF CARBONATE IN BONE AND SOILS (a) In Powdered bone, with small flasks and N hydrochloric acid for decomposition- Bone sample Weight, Carbon dioxide, 1 29.6 3.397 2 29.9 3.274 3 32.8 3.306 4 37.3 3.327 5 4 2 4 3.439 6 45.8 3,274 7 52.0 3.263 8 62.2 3-288 9 99.7 3.375 10 117.2 3.384 11 126-9 3.333 12 149.6 3.292 mg % Mean carbon dioxide content, 3.329 per cent.; Std.dev., 0.053. Coefficient of Variation, 0.456. By Van SZylie - NeilZ-Mean carbon dioxide content, 3.315 per cent. _+0-0167. (b) In soils, with largerjasks and 20 per cent. perchloric acid- Calcium carbonate A I > By titration By Schrotter Soil sample method, tube, Deviation, Yo O f % /O 1 0.00 0.002 - 0.002 2 0~010 0.013 - 0.003 3 0.020 0.016 + 0.004 4 0-020 0.018 + 0.002 5 1.19 1-17 + 0.08 7 2.24 2.23 + 0.01 8 7.08 7-06 + 0.02 9 32.50 32-22 + 0-28 10 46.80 46-63 +0.17 6 1.82 1-85 - 0.03 1. 2. 3. 4. 5. 6. 7. a. REFERENCES Scott, W.W., and Jewell, P. W., Ind. Eng. Chew., Anal. Ed., 1930, 2, 76. Scott, W. W., and Furman, N. H., “Standard Methods of Chemical Analysis,’’ 5th Edition, D. Van Erickson, A. E., Li, L. C., and Gieseking, J. E., Sod Sci., 1947, 63, 451. Collins, S. H., J . SOC. Chem. Ind., 1906, 25, 518. Wright, C. H., “Agricultural Analysis,” Murby, London, 1938. -, “Soil Analysis,’’ Second Edition, Murby, London, 1939. Singh, B. N., and Mathur, P. B., Soil Sci., 1936, 41, 433. Burns, C. M., and Henderson, N., Biochern. J., 1935, 29, 2385. Nostrand, New York, and The Technical Press, London, 1939.310 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. RYAN THE COLORIMETRIC DETERMINATION OF PALLADIUM [Vol. 70 Van Slyke, D. D., J . Biol. Chem., 1917, 30, 347. Van Slyke, D. D., and Neill, J. M., Ibid., 1924, 61, 527. Peters, J. P., and Van Slyke, D. D., “Quantitative and Clinical Chemistry, Part 11. Methods of Analysis,” Williams and Wilkins, Baltimore, 1932. Hawk, P. B., Oser, B. L., and Summerson, W. H., “Practical Physiological Chemistry,” 12th Edition, J. & A. Churchill Ltd., London, 194’7. Robinson, C. S., SoiE Sci., 1920, 10, 41. Bowes, J. H., and Murray, M. M., Biochem. J., 1935, 29, 2721. Van Slyke, D. D., Page, I. H., and Kirk, E., J‘. Biol. Chem., 1933, 102, 635. Van Slyke, D. D., and Folch, J., Ibid., 1940, 136, 509. MacFadyen, D. A., Ibid., 1942, 145, 387. Sobel, A. E., Rockenmacher, M., and Kramer, B., Ibid., 1944, 152, 265. Bremner, J. M., Analyst, 1949, 74, 492. Piper, C. S., “Soil and Plant Analysis,” University of Adelaide, 1942. Dixon, B. E., and Williams, R. A., Analyst, 1849, 74, 360. Amos, A., J. Agric. Sci., 1905, 1, 322. Brown, H. T., and Escombe, F., Phil. Trans., B’, 1900, 193, 223. Third International Congress of Soil Science, Transactions, 1935, 1 , 114. “Official and Tentative Methods of Analysis,” 6th Edition, Association of Official Agricultural Winkler, L. W., 2. anal. Ckenz., 1901, 40, 523. Fravs, G. S., J , Ass. Ofl. Agric. Claem., 1944, 27, 438. Schollenberger, C. J., Soil Sci., 1930, 30, 307. Shaw, W. M., and MacIntire, W. H., J. Ass. 05‘. Agric. Chenz., 1943, 26, 357. Hutchinson, H. B., and MacLennan, K., J. Agxic. Sci., 1914, 6, 323. Williams, R., Ibid., 1932, 22, 838. Van Slyke, D. D., J. Bid. Chem., 1918, 36, 351. Hepburn, J. R. I., Analyst, 1926, 51, 622. Conway, E. J ., “Microdiff usion Analysis and Volumetric Error,” Crosby, Lockwood & Son, London, Benedetti-Pichler, A. A., Cefola, M., and Waldman, B., Ind. Eng. Chem., Anal. Ed., 1939, 11, 327. Warder, R. B., Chem, News, 1881,43, 228. Kolthoff, I. M., and Stenger, V. h., “Volumetric Analysis, Vol. 11, Titration Methods,” 2nd Edition, Intencience Publishers Inc., New York, 1947. Simpson, S. G., I n d . Eng. Chenz., 1924, 16, 709. Cooper, S. S., I n d . Eng. Chem., Anal. Ed., 1941,, 13, 466. Chemists, Washington, D.C., 1945. -, Ibid., 1945, 59, 57. 1947. DEPARTMENT OF AGRICULTURAL CHEMISTRY THE UNIVERSITY READING October, 1950

ISSN:0003-2654    

DOI:10.1039/AN9517600300

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

310 The RYAN THE COLORIMETRIC DETERMINATION OF PALLADIUM Colorimetric Determination of Palladium [Vol. 78 with 2-Mercapto-4 : 5-dimethylthiazole BY D. E. RYAN A method is described for the determination of palladium based on the colour produced by its reaction with 2-mercapto-4 : 5-dimethylthiazole. Rhodium and gold in amounts up to 10 p g per ml do not interfere. Iridium (IrC16n) solutions are nearly completely decolorised ; large amounts of iridium, therefore, do not interfere. Platinum interferes, but an extraction procedure that permits small amounts of palladium to be separated quickly and easily from platinum, iridium and rhodium is described. WHILE investigating the use of 2-mercapto-4 :5-diniethylthiazole for the colorimetric deter- mination of rhodium1 it was found that palladium interfered by reacting to give a colour similar to that for rhodium, Further investigation has shown that this reaction can be used to determine palladium colorimetrically.It is practically instantaneous and is independent of the amount of reagent present provided that an excess is used. It is not sensitive to variations in the acidity of the solution or in the concentration of salts present. The reaction is selective in that platinum is the only platinum metal causing serious interference; this interference is easily eliminated, moreover, by an extraction procedure that allows small mounts of palladium to be separated from large quantities of the platinum metals.May, 19511 WITH 2-MERCAPTO-4 :&DIMETHY LTHIAZOLE 31 1 Maximum absorption, as with rhodium, occurs in the ultra-violet, but excellent results can be obtained in the visible region of the spectrum.A Spekker photo-electric absorptio- meter, with an Ilford No. 601 filter (maximum transmittancy at 430 mp), was used to obtain the data below. The colour developed in aqueous solution is stable for a time, but the palladium complex has a tendency to precipitate, particularly with increasing palladium concentration. Samples containing 2 pg of palladium per millilitre show less than 5 per cent. deviation in the values for optical density after 12 hours, whilst those containing 8 pg per ml are stable for less than 15 minutes. The difficulty is overcome by making the solution 20 per cent. v/v in ethyl alcohol; under these conditions the following optical densities were obtained for a solution containing 8.6 pg of palladium per millilitre- Time .. .. . . . . 5 min. 30 min. 1 hour 2 hours 20 hours Optical density . . .. . . 0.548 0.548 0,549 0-548 0.546 The amount of alcohol added should be closely controlled as the optical density decreases with increasing alcohol concentration ; solutions containing 8-6 pg of palladium per millilitre gave the following optical densities at various alcohol concentrations- Alcohol concentration, yo . . 0 5 10 20 30 Optical density . . .. . . 0.590 0.576 0.565 0.543 0.517 The maximum concentration that can be used with a 1.00-cm cell is 15 pg of palladium per millilitre, whilst a concentration of 0.5 pg per ml gives a minimum transmittancy of 92 per cent. METHOD REAGENTS- 2-Mercapto-4 :6-dimethyZthiaxoZe-The reagent solution was prepared as previously described.1 Standard palladium soZutiow--Palladium sponge was dissolved in aqua regia, the nitric acid removed by evaporating, to dryness several times with concentrated hydrochloric acid, and the residue finally dissolved in a solution containing 1 ml of concentrated hydrochloric acid per litre.This solution, standardised by the dimethylglyoxime method, contained 0.862 mg of palladium per millilitre. Suitable concentrations were prepared by diluting this standard with a solution containing 1 ml of concentrated hydrochloric acid per litre. PROCEDURE- Measure palladium samples into 100-ml volumetric flasks, add about 2 ml of concentrated hydrochloric acid and 20 ml of 95 per cent.ethyl alcohol, and dilute to about 70 ml. Add the reagent (1 ml is sufficient for each 5 pg of palladium present per millilitre) and dilute to the mark with distilled water. The optical density can be measured immediately. The optical densities for solutions under varying conditions are shown below. These results were obtained with 1-8 pg of palladium per millilitre and show that variations in the acid, reagent, or salt concentration have little effect on the sensitivity. Condition Optical density 100 ml of solution containing- 1 ml of hydrochloric acid .. . . . . .. . . 5 ml of hydrochloric acid . . . . . . .. . . 20 ml of hydrochloric acid . . .. . . . . . . 5 ml of hydrochloric acid + 6 ml of sulphuric acid 1 ml of reagent . . .. .. .. . . .. . . 6 ml of reagent .. .. .. .. . . .. .. . . . . 5 ml of hydrochloric acid + 3 ml of nitric acid (nitrous-free) 0.2 M ammonium chloride . . * . .. .. .. 0.133 0.129 0.132 0.132 0.130 0.134 0.133 0.132 Fig. 1 shows that the system deviates from Beer’s law. This deviation was found also in solutions in which no alcohol was present. COMPOSITION OF COMPLEX- neutral solution. complex dried in a desiccator over silica gel, was 27-9 per cent. palladium complex is (C,H,NSCS),Pd, which contains 27.0 per cent. of palladium. The palladium complex with 2-mercapto-4 :5-dimethylthiazole was precipitated from a The average value obtained for the palladium content, on analysis of the This suggests that the312 RYAN : THE COLORIMETRIC DETERMINATION OF PALLADIUM [Vol. 76 The reactions of various cations with 21-mercapto-4 :5-dimethylthiazole have been previously described.l Under the conditions of reaction with palladium, however, the following differences should be noted.Rhodium does not react and does not interfere except for the colour of the rhodium solution itself; this becomes troublesome above 10 pg of rhodium per millilitre. Iridium solutions are almost completely decolorised and palladium can be determined without interference in soluticlns containing 35 pg of iridium per millilitre. Gold is not precipitated from solutions containing 10 pg per ml and palladium can be deter- mined without interference at such gold concentrations; attempts to eliminate interference from gold at high concentrations by precipitating the gold complex in aqueous medium and filtering were unsuccessful owing to loss of palladium.Platinum interferes by giving a yellow colour or precipitate and must be absent. No interference was noted when 200pg per ml of the following ions were present in solutions containing 2 p g of palladium per ml: NO,', C2H,02', F', Cl', Br', 1', SOe", SO,", C20q", PO4'", ClO,', AsO;", K , Na', NH,', Pb", Sn", Sb"' and Ce"". Particularly note- worthy are the non-interferences of iodide and stannous tin, which have been used as drop reagents for detecting palladium. EFFECT OF VARIOUS IONS- Palladium Concentration, pg per ml Fig. 1 . Deviation from Beer's law Cyanide and carbonate must be absent as they inhibit the reaction; even 2 pg of cyanide per millilitre gives results that are low by 20 per cent.Nitrite interferes by reactingwith the reagent to give a greenish-yellow colour and must be absent. SEPARATION OF PALLADIUM FROM PLATINUM, RHODIUM AND IRIDIUM- Palladium can be separated from platinum by applying the hydrous oxide method of Gilchrist and Wichers2 to quantities of the order of 5 p g . This method was used to separate palladium before determining it colorimetrically with 2-mercapto-4 :5-dimethylthiazole and TABLE I DETERMINATION OF PALLADIUM WITH 2.-MERCAPTO-4 :&DIMETHYLTHIAZOLE Palladium taken, CLg Per 2-2 8.6 2.2 8.6 2.2 4.3 1-7 1.7 1.7 1.7 1.7 Palladium found, 2.2 8.7 2-3 8.6 2.2 4.4 1-8 1.7 1.7 1.8 1.9 CLg Per ml Other metal present, PLg Per ml 10 Au, as AuC1,' 36 Ir, as IrC1," 20 Ir 99 10 Kh, as RhCI,'" 10 Rh.* Hydrous oxide procedure used 20 Ir* 39 99 n 20 PI: Hydrous oxide procedure used to separate I't 10 Ir + 20Pt* n 59 99 10 Rh + 20 Ir + 20 Pt* 39 39 20Au 99 16 Rh 91 * Rhodium and iridium precipitate with the palladium; good results are obtained, however, on applying the hydrous oxide procedure to solutions containing the concentrations shown.May, 19611 WITH 2-MEHCAPTO-4 :&DIMETHYLTHIAZOLE 313 was primarily as that given by Yoe and Overholser.3 Since, however, palladium is determined in acid solution with this reagent there was no necessity to evaporate to dryness after dis- solving the oxide in hydrochloric acid.In obtaining the results shown in Table I, when applying the hydrous oxide procedure, it was necessary to boil the solutions vigorously for 16 minutes to eliminate the greenish-brown colour obtained on first dissolving the dioxides.The possibility of separating palladium from platinum, iridium and rhodium was suggested in a previous paper by Ryan4 in which the palladium was allowed to react with p-nitrosodi- phenylamine and the palladium complex extracted with an organic solvent. Results shown in Table I1 prove that small amounts of palladium can be separated in this manner. Procedure-Evaporate the platinum metal solution to dryness on a steam-bath and dissolve the residue in 10 ml of water containing one drop of concentrated hydrochloric acid. Add 3 to 4 ml of a 0.005 per cent. solution of P-nitrosodiphenylamine and, after 10 minutes, extract with 10 ml of ethyl acetate or chloroform. Repeat the reagent addition and extrac- tion and combine the two extracts.Evaporate to dryness, add 5 ml of concentrated sulphuric acid and heat until the organic matter is well charred before adding nitric acid to destroy organic matter completely. Since nitrous acid is liberated in this operation and this is a positive interference in the determination, it must be completely removed; this is ensured by fuming at least twice, washing down the sides of the beaker after each fuming. Cool, dilute and determine the palladium with 2-mercapto-4 :5-dimethylthiazole. In extracting the palladium complex, the colour of the organic layer is a good guide to the completeness of extraction; a yellow colour of reagent appearing in this layer rather than the red of the complex indicates complete extraction. It is advisable to let the aqueous layer that wets the walls of the separatory funnel drain down, after nearly all the aqueous layer has been drawn off, in order to remove the last few drops; since two extractions are made it is convenient to have two funnels available.Although both chloroform and ethyl acetate were used in this work, the latter is preferred as there is less tendency to form emulsions or for the complex to gather at the interface. TABLE I1 SEPARATION (BY EXTRACTION AS ITS P-NITROSODIPHENYLAMINE COMPLEX) The hydrous oxide procedure accomplishes a separation from platinum only. DETERMINATION OF PALLADIUM WITH 2-MERCAPTO-4 :B-DIMETHYLTHIAZOLE AFTER FROM PLATINUM, RHODIUM AND IRIDIUM Palladium taken, mg 0.172 0.172 0.085 0.085 0.085 0.034 0-018 0.018 0.018 0.018 0.022 0.022 0.007 0-007 Palladium found, mg 0.172 0.170 0.087 0.082 0.084 0.034 0.0 18 0.019 0.018 0.017 0.023 0-023 0.007 0.007 Other metal present, mg 4 Pt 10 Pt 6 Pt 5 Ir 6 Rh 10 Rh 4 Pt 5 Rh 5 Ir 5 P t + 5 Rh + 3 Ir 5 Pt + 5 Rh + 3 Ir 5 P t + 5 R h + 31r 4 Pt 5 R h + 6 I r The author thanks Mr. L. S. Theobald for his encouragement and help in this work. Sincere appreciation is expressed to Lord Beaverbrook for providing the scholarship, through the Beaverbrook Overseas Scholarship Fund , which enabled this work to be completed. REFERENCES 1. Ryan, D. E., Analyst, 1950, 75, 657. 2. Gilchrist, R., and Wichers, E., J . Amer. Chem. SOC., 1935, 57, 2565. 3. Yoe, J. H., and Overbolser, L. G., Ibid., 1939, 61, 2058. 4. Ryan, D. E., Analyst, 1951, 76, 167. DEPARTMENT OF INORGANIC AND PHYSICAL CHEMISTRY IMPERIAL COLLEGE OF SCIENCE AND TECHNOLOGY LONDON, S.W.7 October, 1950

ISSN:0003-2654    

DOI:10.1039/AN9517600310

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

314 NORWITZ: THE ANALYSIS OF MANGANESE BRONZE (Vol. 76 The Analysis of Manganese Bronze BY GEORGE NORWITZ A new method is proposed for the analysis of manganese bronzes. The sample is dissolved in a mixture of perchloric and nitric acids, hydrogen peroxide is added, and the copper and leaad electrolysed. Manganese, nickel, iron, aluminium and tin are determined on aliquot parts of the electrolyte. THE methods that have been p r o p o ~ e d ~ s ~ ~ ~ ~ * s ~ for the analysis of manganese bronze require many time-consuming operations such as filtrations and hydrogen sulphide precipitations. In this paper a method is proposed for the analysis of manganese bronze that entirely eliminates filtrations and hydrogen sulphide precipitations. The time required is one-quarter of that required for the most rapid of the older methods.It is based on the observation by the author that copper and lead can be separated by electrolysis from as much as 2 per cent. of tin by the use of a perchloric - nitric acid electrolyte. After the copper and lead have been separated by electrolysis, the manganese, nickel, iron, aluminium and tin are determined on aliquots of the electrolyte by means of conventional procedures. APPARATUS AND REAGENTS- A Klett - Summerson photo-electyic colorimeter. Platinum gauze cathodes-Height 50 mm, diameter 45 mm and platinum spiral anodes (with a spiral width of 10 mm). Benzoic acid solution-A 10 per cent. solution in methanol. Bufer solution-Mix 470 ml of ammonium hydroxide and 430 ml of glacial acetic acid. Cool to room temperature and add more acid 01: base as necessary to make the pH value lie between 5.25 and 5.35 when 1 volume is diluted to 20.Dilute to 1 litre. Gelatin solution-A 1 per cent. w/v solution. Aluminon reagent-Dissolve 0.3 g of aluminon in 200 ml of water and add 60 ml of benzoic acid solution (10 per cent.). Dilute to 300 ml, add 300 ml of buffer solution and 300 ml of gelatin solution and shake. Allow to stand for 3 days before using. The reagent will keep for a t least 2 months if stored in a dark place. PROCEDURE Copper and lead-Take 1 g of the sample in a 300-ml electrolytic beaker. Add, in the following order, 5 ml of water, 10 ml of perchloric acid and 10 ml of nitric acid. Allow the sample to dissolve without heating (this will take about a minute). Add 1 ml of 30 per cent.hydrogen peroxide to oxidise the lower oxides of nitrogen, mix and dilute to 19Oml with water. Electrolyse for copper and lead at 2 amperes per square decimeter for 75 to 90 minutes. Immerse the electrodes in water and in alcohol. Dry the cathodes at 110" C for 3 minutes, cool and weigh the deposit as metallic copper. Dry the anodes at 110" C for 15 minutes, cool and weigh the deposit as PbO,. Manganese-Dilute the electrolyte from the copper and lead determinations to 250 ml. Pipette 25 ml into a 100-ml graduated flask, add 5 ml of 85 per cent. phosphoric acid and 0.5 g of potassium periodate. Heat to boiling and boil gently for 15 minutes. Cool to room temperature and dilute to 100 ml. Measure the absorption at 540 millimicrons in a photo- electric absorptiometer (1-cm cell) that has been set to zero against water.Convert the readings to percentage of manganese by means of' a graph prepared from standard perman- ganate solution. Nickel-Transfer 10 ml of the test solution to a 100-ml graduated flask, add 5 ml of 10 per cent. citric acid solution, 5 ml of bromine water and 5ml of diluted ammonium hydroxide solution (1 + 1). Mix after the addition of each reagent. Add 3 ml of a 1 per cent. alcoholic solution of dimethylglyoxime and dilute to 100 ml with water. Measure the absorption at 540 millimicrons in a photo-electric absorptiometer (1-cm cell) that has been During the electrolysis stir the solution by suitable means. The factor for converting PbO, to Pb is 0.866.May, 19511 NORWITZ: THE ANALYSIS OF MANGANESE BRONZE 315 set to zero against water.Convert the readings to percentage of nickel by means of a graph prepared from standard nickel solution. Iron-Pipette 10 ml of the test solution into a 100-ml graduated flask and add 1 ml of 3 per cent. hydrogen peroxide. Add 25 ml of 10 per cent. ammonium thiocyanate solution and dilute to 1OOml with water. Measure the absorption a t 470 millimicrons in a photo- electric absorptiometer (l-cm cell) that has been set to zero against water. Convert the readings to percentage of iron by means of a graph prepared from standard iron solution. ALumini~m~1~-Pipette 25 ml of the test solution into a 200-ml graduated flask. Dilute to the mark and pipette a 5-ml aliquot into a 100-ml graduated flask. Add 15 & 0.5 ml of the aluminon reagent from a pipette.Do not allow the aliquot or the aluminon reagent to run down the neck of the flask. Heat on the steam-bath for 20 minutes in such a way that the steam comes in contact with the sides of the flask. Cool and dilute to 100ml. Measure the absorption at 540 millimicrons in a photo-electric absorptiometer (4-cm cell) that has been set to zero against a reagent blank. Read from a graph prepared from pure aluminium solution the percentage of aluminium represented by the combined aluminium and iron colours, and then make a correction-for the iron. The iron colour was found by the author to be 0.364 times as intense as the aluminium colour. The exact ratio should be determined by each laboratory by means of standard aluminium and iron solutions.Tin-Measure 150 ml of the test solution into a 500-ml Erlenmeyer flask. Add 20 ml of sulphuric acid and evaporate to fumes. Continue to fume strongly for a few minutes to drive off the perchloric acid. Allow to cool, add 150 ml of water and 45 ml of hydrochloric acid and reduce with lead in the usual way.8 Titrate with 0.01 N iodine solution from a short accurately -calibrated burette. RESULTS The results obtained by the author in analysing two representative manganese bronzes by the above method are shown in Table I. RESULTS Copper .. Lead . . .. Manganese . . Nickel . . .. Iron . . . . Aluminium . . Tin . . .. Copper . . Lead . . .. Manganese . . Nickel . . .. Iron . . .. Aluminium . . Tin . . .. FOR .. .. .. .. .. .. .. . . .. .. .. .. .. ..TABLE I TWO SAMPLES OF MANGANESE BRONZE .. . . .. .. r . .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . .. .. .. .. .. .. .. Pn sent,* 57.39 0.28 1.29 0.27 0.82 0.97 0.96 % 56-82 0.18 0.98 0.02 1.02 0.88 0.34 Sample 62b Found, % 57.40, 57.37, 57.39 0.29, 0.27, 0.28 1.28, 1.29, 1-29 0.27, 0.27, 0.26 0.82, 0.82, 0.81 0.98, 0.95, 0.96 0.98, 0.96, 0.95 Sample R 56-86, 56.82 0.18, 0.17 0.97, 0.98 0.02, 0.02 1.02, 1.01 0.86, 0.89 0.37, 0.34 * Values for 62b are National Bureau of Standards certified values; values for sample R were obtained by using A.S.T.M. umpire meth0ds.l NOTES- The samples must be dissolved without heating, otherwise some metastannic acid may precipitate. It is essential that 30 per cent. hydrogen peroxide be used to oxidise the lower oxides of nitrogen.The use of the equivalent amount of 3 per cent. hydrogen peroxide is not satisfactory, since this grade usually contains sufficient phosphate (added as a preservative) to interfere with the deposition of lead dioxide. When spiral anodes are used there is no appreciable contamination of the lead dioxide by manganese dioxide. The method described is designed for the analysis of the MnC type of manganese bronze. This is the type usually encountered in commerce. It has maximum permissible limits of 1.5 per cent. of manganese, 0.35 per cent. of lead and 1;O per cent. of tin.316 NOTES [Vol. 76 REFERENCES Ameritan Society for Testing Materials, “A.S.T.M. Methods of Chemical Analysis of Metals,” Babson, E. K., and Johnson, W. W., Ind. Eng. Chem., Anal. Ed., 1942, 18, 292. Jones, B., AnaZyst, 1933, 58, 11. Ravner, H., Ind. Eng. Chem., Anal. Ed., 1945, 17, 41. Scott, W. W., “Standard Methods of Chemical Analysis,” Vol. 11, D. Van Nostrand Co., New York, 1939, pp. 1352-1362. Carroll, J., Geld, I., and Norwitz, G., American Foundryman, 1949, 16, No. 5, 43. Craft, C. H., and Makepeace, G. R., I d . Eng. Chem., Anal. Ed., 1945, 17, 260. American Society for Testing Materials, “ A S .T.M. Methods of Chemical Analysis of Metals,” 1. 2. 3. 4. 6. 6. 7. 8. Philadelphia, Pa., 1946, pp. 197-212. Philadelphia, Pa., 1946, p. 202. 3353 RIDGE AVENUE PHILADELPHIA 32, P A . July, 1960 U.S.A.

ISSN:0003-2654    

DOI:10.1039/AN9517600314

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

316 NOTES [Vol. 76 Notes THE CHROMATOGRAPHIC ESTIMATION OF AQUEOUS SOLUTIONS OF ACIDS THE separation and estimation of the lower fatty acids is easily effected by partition chromato- graphy.l One slight difficulty is that acids in aqueous solution have to be transferred to a solvent before being placed on the column. It has been found possible to eliminate this step by mixing an aliquot of the aqueous solution with a small amount of silica gel and then adding this to the top of the column. PROCEDURE- with chloroform and packing it into the tube with a tightly-fitting glass piston. Prepare the column by adding dilute hydrochloric acid to silica gel,l forming it into a sludge Place a circle of 8 I z 0 "t 0 2 I 4 ---&--- 100 200 LJ Volume of eluate, ml Fig. 1. Titration curve of acidity of eluate filter-paper on the top of th6 column.Take a quantity of silica gel, about 6 to 10 per cent. of that used for the column, and add to it an aliquot of the aqueous solution so that the silica gel remains powdery after the addition. The aliquot should contain 0.001 equivalent of total acid. With a spatula, mix the solution and the silica gel thoroughly in a small beaker. Tap the powder on to a folded filter-paper and thence on to the colurnn. There should be just enough chloroform above the column to cover the surface after the extra silica gel has been added. Press the surface firmly with the glass piston, add three 1-ml portions of chloroform to wash in the acids and then develop the column in the usual way, titrating the eluate with 0.01 N sodium hydroxide solution.The method has been found to work well for the routine estimation of mixtures of butyric, propionic, acetic and formic acids. The result of a typical separation of these acids is shown in Fig. 1. The eluate from the column was titrated continuously with 0-01 N sodium hydroxide solution; a stream of air, free from carbon In this experiment a column containing 10 g of silica gel was used.May, 19511 NOTES 317 dioxide, was bubbled continuously through the mixture of eluate and water. The solvents used were: chloroform up to the point A, chloroform containing 5 per cent. of butanol by volume up to B, and benzene containing 10 per cent. of butanol by volume beyond B. The acids appear in the order butyric, propionic, acetic and formic.The quantitative results were- Taken in Acid Found, mixture, % % Butyric . . .. .. 28.1 28-4 Propionic . . .. .. 15.6 15-1 Acetic . . . . .. .. 39.1 39.5 Formic .. .. .. 17.2 17-1 I should like to thank the Directors of the Distillers Company Limited for permission to publish this note. REFERENCE 1. Neish, A. C., Canad. J . Kes., 1949, 27B, 6. T H E DISTILLERS COMPANY LTD. RESEARCH AND DEVELOPMENT DEPT. GREAT BURGH, EPSOM, SURREY W. S. WISE November, 1960 THE PHOTOMETRIC DETERMINATION OF COPPER I N ALUMINIUM ALLOYS WITH SODIUM DIETHYLDITHIOCARBAMATE WE have examined the method described by Williams1 for the photometric determination of copper in aluminium alloys with tetra-ethylene pentamine and, although we agree that sodium diethyl- dithiocarbamate may have certain disadvantages (e.g., its high sensitivity) when compared with tetra-ethylene pentamine, we wish to emphasise that, in spite of this, it can be and is used success- fully for the analytical control of alloying quantities of copper in a wide variety of normal industrial aluminium alloys.Moreover, in the important group of aluminium - copper - nickel alloys, interference caused by the presence of nickel is not insurmountable, as i t may be reduced to negligible proportions by suitable adjustment of the ammonia concentration in the final solution. Table I shows the typical accuracy attainable in routine copper determinations carried out over a period of several months on two normal aluminium alloys. Table I1 shows the composition of these alloys. The results are quite satisfactory for control purposes.TABLE I DETERMINATION OF COPPER I N ALUMINIUM ALLOYS BY SODIUM DIETHYLDITHIOCARBAMATE Standard Average of result routine Number of Standard Standard (volumetric), results, determinations deviation, deviation, % % 0 % 1-53 1-555 64 0-0274 1-76 1 *53 1.55 39 0.0225 1-45 2-02 2.015 98 0.0522 2.58 2-57 2.58 144 0.0399 1.55 Alloy HID55 (483) HID55 (483) HIDRR50 (461) HID55 (484) TABLE I1 COMPOSITION OF ALLOYS Alloy CU I Ni, Mg1 Fe, Si, Ti, Mn, Zn, % % Yo % % % % % HIDRR50 (461) . . 2-02 1.82 0.064 0.69 1-61 0-23 0.13 - HID66 (483) . . 1.53 1.43 1.31 0.40 1.52 0-20 0.31 0.055 HID55 (484) . . 2-57 0.92 0-78 0.88 0.92 0-15 0.50 0-10 In 1945, Phillips and Edwards2 published some results of their work on a method that used sodium diethyldithiocarbamate for the determination of copper in the presence of a wide variety of elements normally encountered in aluminium alloys.Their method was tested with particular reference to Y-alloy containing 3-5 to 4.59 per cent. of copper and 1.8 to 2.3 per cent. of nickel. With regard to nickel interference, they demonstrated that turbidity caused by this element could be avoided by ensuring a suitable concentration of ammonia in the final solution. Their318 NOTES [Vol. 76 method in practice covered ranges of from 0.2 to 5 per cent. for copper and 0.2 to 3 per cent. for nickel. Typical standard deviations were given for a Y-alloy and are quite satisfactory for most purposes. Stross3 has also-investigated the influence of large quantities of various elements likely to be present in aluminium alloys.With a 10-mg sample, the determination of from 0.5 to 2 per cent. of copper is not affected by the presence of 2 per cent. of nickel, and with a 5-mg sample 4 per cent. of nickel is without any appreciable influence on the determination of 4 per cent. of copper. Similarly, Steele and Russell4 have shown the efficacy of a suitable ammonia concentration in reducing the effect of nickel on the determination of copper in nickel-bearing steels and cast irons. By way of confirmation we have recently carried out a limited number of experiments with synthetic solutions, all of which contain aluminium, together with various concentrations of copper, nickel and iron. These solutions were prepared by dissolving suitable weights of the highest purity metal available in the minimum quantity of nitric - hydrochloric acid solution.After dilution to a convenient volume in a graduated flask, aliquots representing the amounts of the above metals contained in 0.1 g of alloy sample were transferred to beakers and evaporated almost to dryness in order to remove any excess of acid. The solutions were diluted with water and filtered into 200-ml graduated flasks. From the latter, IO-ml aliquots (representing 5 mg of the alloy sample) were then transferred by means of a pipette to 100-ml graduated flasks, to each of which was added 10 ml of 30 per cent. aqueous citric acid solution, 20 to 21 ml of fresh ammonia solution, sp.gr. 0.880, cooled to 20" C, 5 ml of fresh 1 per cent. aqueous gum arabic solution, 10 ml of fresh 0.1 per cent.aqueous solution of sodium diethyldithiocarbamate and sufficient water to make up to the 100-ml mark. The extinctions were measured on the Spekker absorptiometer in a 4-cm cell with Ilford violet 601 filters and a tungsten filament lamp. The measurements were carried out immediately after the solutions were coloured. With alloys, including calibration standards, 0-1 g of sample is dissolved in 10 ml of aqua regia (1 volume of hydrochloric acid + 1 volume of nitric acid) and evaporated almost to dryness. From this stage the procedure is the same as that outlined above. Blank determinations are always carried out. The results are shown in Table 111. TABLE I11 DETERMINATION OF COPPER IN SYNTHETIC SOLUTIONS Observed extinction ; Elements present in addition to ahminiurn average of two Copper, Nickel, Iron, determinations A r - separate % % % 1.81 1.81 1.81 2.94 2-94 2.94 2-94 4-08 4.08 4-08 - 2.0 2.0 2.0 2.0 3.0 2-0 2.0 - - - 0.529 - 0.535 1.08 0.531 - 0-826 - 0.824 1.08 0.828 0.54 0.841 - 1.089 - 1.105 1-08 1.103 Inspection of Table 111 shows that variations in extinction caused by the presence of nickel and iron are quite small, and the measurements are capable of giving sufficient accuracy for control purposes. Moreover, in our laboratories, these variations are still further reduced by calibration on actual standard alloys representative of the alloys under investigation. We wish to thank the Directors of High Duty Alloys Limited for permission to publish this note. REFERENCES 1. 2. 3. 4. CENTRAL RESEARCH LABORATORIES HIGH DUTY ALLOYS LTD. SLOUGH, BUCKS. Williams, L. H., Analyst, 1950, 75, 425. Phillips, D. F., and Edwards, L. L., Metal Ind., 1945, 66, 409. Stross, W., Metallurgia, 1945, 32, 267. Steele, S. D., and Russell, L., Analyst, 1949, :74, 105. E. C. MILLS S. E. HERMON November, 1960

ISSN:0003-2654    

DOI:10.1039/AN9517600316

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

May, 19511 APPARATUS 319 Apparatus PREPARATION OF GLASS-DISTILLED WATER GLASS-DISTILLED water is essential for the preparation of media for microbiological assays. The apparatus shown in Fig. 1 is suitable for the production of a continuous supply of glass-distilled water and can be constructed from apparatus common in any laboratory. Water distilled in this apparatus was used in the assay of vitamin B,, as described by Pritchard.l Fig. 1. Simple still for the continuous supply of glass-distilled water suitable for micro- biological assays The “boiler” consists of an ordinary distilling flask inclined at an angle of about 20” to the horizontal. The side-arm, after shortening if necessary, is bent in the form of a U-tube in a plane at right angles to the axis of the flask in such a fashion that the free limb of the U-tube is vertical when the flask is inclined downwards a t an angle of about 20”.The cylinder of a constant-level device can be made from 1Q or 2-inch diameter glass tubing or from a boiling tube, the bottom of which has been cut off. At one end it is fitted with a stout rubber bung, bored with two holes. One hole cames the bent side-arm of the flask, the other carries a wide-bore tube to lead water to waste. The height that this tube reaches into the cylinder determines the depth of the water in the distilling flask. The constant-level apparatus is completed by bending a length of tubing and attaching it by a rubber band so that one end enters the cylinder and the other is conveniently attached to the outlet of the condenser by a length of rubber tubing.The bore of the inlet should, of course, be less (about half) than that of the waste pipe. Steam is conveyed from the flask to a vertical condenser by a wide tube fitted into the flask and condenser by rubber bungs. The end of the conduit that is in the flask must be so bent that the open end is directed away from the freely boiling surface of the water to avoid contamina- tion from priming. The angle of this bend is determined by trial and is so made that the tube can be passed down the neck of the flask when the apparatus is assembled. It is more convenient to have an all- glass condenser, as thin rubber connections become rotted after the still has been working for some weeks. The condenser is of the ordinary single-surface type.320 MINISTRY OF FOOD [Vol. 76 The steam conduit and waste tubes are 7-mm bore tubing. The cylinder of the constant level device is made from a boiling tube of 3 cm diameter and the inlet to it is made from 4-mm bore tubing. In this laboratory the boiler is a distilling flask of 700-ml capacity. REFERENCE 1. 13, HAMILTON SQUARE BIRKENHEAD Pritchard, H., Analyst, 1951, 76, 155. D. R. WRAIGE February, 1951

ISSN:0003-2654    

DOI:10.1039/AN9517600319

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

320 MINISTRY OF FOOD [Vol. 76 Ministry of Food STATUTORY INSTRUMENTS* 1951-No. 314. The Meat Products and Canned Meat (Amendment) Order, 1951. Price !&I. This Order, which came into operation on March 4th, 1951, permits the manufacture and sale of pork sausages and beef sausages of less than the prescribed minimum meat content, i f the deficiency in meat i s compensated by the use of milk powder in the proportion of 6 parts of milk powder to 10 parts of meat. The actual meat content may not be reduced in this way below 55 per cent. for pork sausages apzd 40 per cent. for beef sausages. The Order amends the Meat Products and Canned Meat (Control and Maximum Prices) Order, 1948 (S.I., 1948, No. 1609; Analyst, 1948, 73, 341), as amended by S.I., 1949, Nos. ’182, 1303 and 2045 and S.I., 1950, No.1764 (Analyst, 1951, 76, 119), as follows- (a) by adding after the definition of “Meat product” in Article 1 thereof the following definition : “ ‘Milk powder’ means milk, partly skimmed milk or skimmed milk, buttermilk or whey which has been concentrated to the form of powder or solid by the removal of its water.” (b) by substituting for Article 3 thereof the following Article: “3. A person shall not by way of trade prepare or manufacture or sell or have in his possession for sale any description of specified food mentioned in column 1 of the First Schedule to this Order the meat content of which is less than the minimum meat content prescribed as respects that descrip- tion in column 2 of the said Schedule: * Obtainable from H.M. Stationery Office.Italics indicate changed wording.May, 19511 BRITISH STANDARDS INSTITUTION 321 Provided that- (i) any milk powder used in the manufacture of pork sausages, pork sausage meat and pork slicing sausage and beef sausages, beef sausage meat and beef slicing sausage shall be deemed to be equivalent to 6/3 of its own weight in meat for the purpose of assessing the meat content of any of the said products, if the quantity of such milk powder so used does not exceed 6 per cent. of the total weight of the product, and (ii) any fat of vegetable origin used in the manufacture of beef sausages, beef sausage meat, or beef slicing sausage shall be deemed to be meat for the purpose of assessing the meat content of any of those products, if the total quantity of such fat so used does not exceed 25 per cent. of the prescribed minimum meat content of the product.” CIRCULARS ME” 17/50 AND 5/51 Approved Oxidising and Preservative Agents Circular MF 17/50 (price ld.), dated September 19th, 1950, refers to Circular M F 11/50, dated June 27th 1950 (Analyst, 1950, 75, 604), and gives the names of two further products whose use for the cleansing of milk tankers, vessels or appliances has been approved by the Minister of Agriculture and the Minister of Food. Circular MF 6/51 (price 2d.), dated March lst, 1951, gives the names of three further Products similarly approved. The further list of products i s as follows- CLORFECT, UNIVPURI CHLORSOL, DUROS, GASCOIGNE-RED LABEL, SOLCHLOR.

ISSN:0003-2654    

DOI:10.1039/AN951760320b

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

May, 19511 B.S. 997 : 1951. B.S. 1365 : 1951. B.S. 1704 : 1951. BRITISH STANDARDS INSTITUTION British Standards Institution NEW SPECIFICATIONS* Sperm Oil. Price 2s. 6d. Short-range, Short-stem Thermometers. Price 2s. General Purpose Thermometers. Price 2s. DRAFT SPECIFICATIONS 321 A FEW copies of the following draft specifications, issued for comment only, are available to interested members of the Society, and may be obtained on application to the Secretary, Miss D. V. Wilson, 7-8, Idol Lane, London, E.C.3. Draft Specification prepared by Technical Committee LBC/2-Glassware for Medical and Bacterio- logical Use. CM(LBC) 9917-Draft B.S. for Petri Dishes (Revision of B.S. 611). Draft Specifications prepared by Technical Committee LBC/l-Volumetric, Mouldblown and Lamp- blown Glassware. CN(LBC) 36-Draft B.S. for Nessler Cylinders (Revision of B.S. 612). CM(LBC) 9886-Draft B.S. for Density Bottles (Revision of B.S. 733). Draft Specification prepared by Technical Committee DAC/G-Rennet, Annatto and other Colours for Dairying Purposes, Sub-committee DAC/6/2--Annatto and other Colours (drafting). CN(DAC) 80-Draft B.S. for Annatto for Dairying Products. Draft Specification prepared by Non-ferrous Metals Industry Standards Committee NFE/-, Sub- CM(NFE) 9831-Draft B.S. Method for the Determination of Zinc in Aluminium Alloys committee NFE/-/3-Samgling and Analysis of Aluminium and Aluminium Alloys. (Polarographic Method). * Obtainable from the British Standards Institution, Sales Department, 24, Victoria Street, London, S.W.1.

ISSN:0003-2654    

DOI:10.1039/AN9517600321

出版商:RSC    

年代:1951

数据来源: RSC

摘要:

322 BOOK REVIEWS [Vol. 76 Book Reviews THE NATIONAL FORMULARY. Ninth Edition. Pp. xl + 877. Washington : American Pharma- This book should not be confused with the British publication having a similar title, which is a compilation of prescriptions for use in connectionwith the National Health Service Act, 1946, whereas the book under review may conveniently be regarded as being approximately the American equivalent of our British Pharmaceutical Codex. Its purpose is to present a compilation of standards for widely used drugs and preparations that are not included in the United States Pharmacopoeia. The importance of the National Formulary has steadily increased since the appearance of the first edition in 1888 and by the terms of the Federal Food and Drugs Law of 1906 it was designated an official compendium along with the United States Pharmacopoeia, and this arrangement was confirmed by the Federal Food, Drug and Cosmetic Act of 1938.The body of the work comprises 662 pages devoted to monographs giving detailed standards for drugs and their preparations. These monographs are arranged in the same way as those of the United States Pharmacopoeia and, in general, include a description of the medicament, statement of solubility, tests for identification, examination for likely impurities and, where applicable, a procedure for assay. For nearly every substance there is a direction about packaging and storage. After the monographs there follows a compilation of reagents and preparations for use in the clinical laboratory and then an important section devoted to standards for reagents, for ingredients of preparations and for dyes used as biological stain.s.Together, these two portions of the work occupy 114 pages of small print. Many of the specifications are very exhaustive, particularly those applying to the dyestuffs. The next section of 98 pages is entitled general tests, processes and apparatus and defines numerous analytical operations, such as the test foir arsenic, the determination of boiling range, disintegration tests for tablets, standards for light transmission, the measurement of viscosity, microbiological methods for the assay of nicotinarriide and riboflavine and many more general and special analytical procedures. A further list of general reagents and volumetric solutions, two tables of permitted coal-tar dyes for colouring purposes, a short addendum and.an index occupying 57 pages complete this handsomely bound volume. From all these thousands of tests and analytical processes it would not be difficult to make suggestions for alternatives or improvements, the more easily by concentrating on one’s own specialities. It is sufficient to say that every page reveals the immense care that has been devoted to the production of this compendium. Much of the matter has, of necessity, been derived from previous editions or from earlier Pharmacopoeias of the United States; but, on the other hand, much is new and the whole is dovetailed together to produce a work of reference invaluable to all concerned with the examination of medicinal substances.ceutical Association. 1950. Price $8.00 (in U.S.A.) ; $8.75 (elsewhere). But no useful purpose would be served. N. L. ALLPORT DDT AND NEWER PERSISTENT INSECTICIDES. By T. F. WEST, D.Sc., Ph.D., F.R.I.C., and G. A. CAMPBELL, M.Sc., F.R.I.C. Second Edition.. Pp. xiv + 632. London: Chapman and Hall Ltd. 1950. Price 50s. The first edition of this book appeared in 1946 when DDT was still something of a novelty. Even so, it ran to some 300 pages in recording the large amount of research results published up to 1945. In the succeeding years an even greater volume of publications on the applications of DDT in the control of animal and plant pests has appeared. Hence it is not surprising that the authors have found it necessary to produce a revised and extended second edition, or that the new edition should run to some 500 pages devoted to DDT, with a further 7 0 pages devoted to newer chlorinated hydrocarbon insecticides.Inclusion of the latter has, incidentally, necessitated an appropriate change in the original title. Chapter I, describing the development of DDT, has been completely rewritten on the basis of more detailed information obtained from Switzerland since 1945. Similarly, advances in knowledge of the chemistry of DDT have necessitated considerable revision of Chapter 11, which now includes a valuable tabulation of the many analogues that have been synthesised and gives an indication of their relative insecticidal activity, with references. This chapter deals with assay methods Part One of the book, dealing with DDT, follows the pattern of the first edition.May, 19511 BOOK REVIEWS 323 for DDT and its estimation in residues on biological or other treated surfaces, and also gives much more information on the solubility of DDT than appeared in the first edition.It is to be regretted that this latter information refers in the main to the pure p-p’-isomer and that the solubilities are expressed as “grams per 100 ml” and “grams per 100 grams” of solvent. The ordinary user is likely to be much more interested in the solubility characteristics of technical DDT, and even then on a weight/volume percentage basis. The remaining chapters, while preserving much of the matter presented in the first edition, have been considerably extended by references to more recent work.Part Two deals, rather briefly, with the newer chlorinated hydrocarbon insecticides, chapters being devoted respectively to benzene hexachloride, chlordane, toxaphene and “various new insecticides. ” The authors are to be congratulated on the manner in which they have undertaken the task of surveying the vast literature on this subject. Wisely, they have not attempted to do more than report on all work that appears to be relevant. In fact, they specifically disclaim any effort to make, at this stage of development, a “critical rbsumb.” It does, however, seem unfortunate that a book published in 1950 should review the literature only up to 1946; although, through unavoidable delay in production of the book, opportunity has been taken to include, as an appendix, lists of references tabulated according to chapter headings up to 1948.Suffice it to say that this is a book that must be available to all those who, as chemists or biologists, are interested in insecticides. The book is well produced and bound, contains relatively few mis- prints, and includes 13 plates, most of which appeared in the first edition. It is quite impossible in a short review to single out items of particular interest. W. MITCHELL ORGANIC REAGENTS FOR ORGANIC ANALYSIS. By the Staff of the Research Laboratory of Hopkin and Williams Ltd. Second Edition. Pp. viii + 255. Published by Hopkin and Williams Ltd., Chadwell Heath, Essex. 1950. Price 12s. 6d. (plus 5d. postage) from the publishers. The first edition of this laboratory handbook to the use of organic reagents in the identification of organic compounds by means of derivatives with diagnostic melting-points has already received favourable notice in The Analyst (1946, 71, 503).The general plan of the second edition follows that of !he first, but the book has undergone enlargement by the inclusion of about half a dozen new reagents and many additions, revisions and corrections to the melting-point tables. This increase in subject-matter fulfils, in some measure, the hope expressed by the reviewer of the first edition. From the large number of reagents that are continually being proposed for the identification of organic groups and compounds by means of the melting-points of characteristic derivatives, a selection is made of those that are, in the opinion of the authors and for reasons stated, the most useful and readily obtainable; for these, the properties and methods of use are described and the melting-points of the derivatives listed. Bibliographies to the literature of the selected reagents are supplied and the properties of many other reagents that are, for various reasons, considered to be of less general importance or usefulness are critically reviewed. It forms a useful companion to the standard textbooks on practical organic chemistry. F. L. OKELL The book is well indexed, and bound in proofed cloth for bench use.

ISSN:0003-2654    

DOI:10.1039/AN9517600322

出版商:RSC    

年代:1951

数据来源: RSC