摘要:

August, 19611 BY LIQUID - LIQUID PARTITION CHROMATOGRAPHY 527 Thin-la y er Chromatography of 3,5=Dini trobenzoates (Central Institute f o v Nutvition and Food Reseavch T.N.O., Utvecht, The Nethevlands) The behaviour of some 3,5-dinitrobenzoates on chromatoplates has been studied; the RF values are compared with the RB values found by measuring the locations of the compounds against that of a standard substance. A procedure is described for preparing the dinitrobenzoates of hydroxyl com- pounds present in very dilute solutions. The method is suitable for detecting small amounts of these compounds in water and in the presence of large amounts of methanol or ethanol. Small amounts of the dinitrobenzoates can be purified, since the esters can easily be separated from the excess of unchanged reagent.IN a study of steam-volatile alcohols and phenols in foods, a simple and rapid method was required for separating and purifying certain 3,5-dinitrobenzoates prepared by reaction of the hydroxyl compounds with 3,5-dinitrobenzoyl chloride. Results obtained in the separation of 2,4-dinitrophenylhydrazonesl prompted us to investigate the behaviour of some dinitro- benzoates on chromatoplates. EXPERIMENTAL APPARATUS AND TECHNIQUE- Chromatoplates were prepared, as described previously,l by using Stahl's apparatus, * with silica gel G (obtainable from E. Merck A.G.) as adsorbent. The plates were activated before use by heating them at 110" to 120" C for 15 minutes. After the plates had been dowed to cool on a large plate of glass for 5 minutes, spots (usually 5 to 10 pl) of the solutions being investigated were placed on them, and the treated plates were put into a tank for development within 10 minutes after removal from the heated cabinet.The atmosphere in the tank was kept saturated with the vapour of the developing solvent by covering the walls of the tank with strips of filter-paper dipping into the solvent. FORMATION OF 3,5-DINITKOBENZOATES- A weighed amount of the alcohol or phenol was placed in a small round-bottomed flask, slightly more than the calculated amount of 3,5-dinitrobenzoyl chloride and then 10 ml of benzene and 0.1 ml of dry pyridine were added, and the contents of the flask were heated under reflux for 30 minutes to 1 hour. When cool, the reaction mixture was extracted with 26 ml of 0-1 N sulphuric acid, then with 25 ml of a 0-5 to 1 per cent.solution of sodium528 DHONT AND DE ROOY: THIN-LAYER CHROMATOGRAPHY [Vol. 86 carbonate and finally with water. The aqueous layers were discarded, and the benzene solution was evaporated to dryness under reduced pressure in a bath of boiling water. The residue was covered with a few millilitres of benzene, so that a saturated solution of the ester was obtained. Solutions of the esters in methanol were not stable; after some time, these produced a second spot on the chromatogram, in the position of that formed by the 3,5-dinitrobenzoate of methanol. The unchanged dinitrobenzoyl chloride is converted into dinitrobenzoic acid. However, it is not necessary to remove this acid by extraction with sodium carbonate solution, as it remains a t the start in the solvents used for developing the chromatograms. SOLVENT SYSTEMS- The solvent systems tested included a (1 + 1) mixture of benzene and light petroleum (boiling range 60" to 80" C), hexane containing 15 or 25 per cent.of ethyl acetate and toluene containing 10 per cent. of ethyl acetate. The results for the 3,5-dinitrobenzoates of normal Number of carbon atoms in chain Fig. 1. R, values for 3,5-dinitrobenzoates of aliphatic alcohols as a function of chain length: curve A, benzene - light petroleum mixture (1 + 1) as solvent; curve B, toluene containing 10 per cent. of ethyl acetate as solvent; curve C, hexane containing 15 per cent. of ethyl acetate as solvent aliphatic alcohols when these systems were used are summarised in Fig.1. The system giving the best results was the benzene - light petroleum mixture, and this was subsequently used throughout. COMPARISON BETWEEN R p AND Rig VALUES- During our work on the thin-layer chromatography of 2,4-dinitrophenylhydrazones, we found that RF values were dependent on the distance travelled by the solvent front, and we defined1 the R B value by the expression- Movement of substance from start, mm Movement of butter yellow from start, mm' R R = (Nicolaus3 also recently recommended the use of a standard substance for measuring the positions of the spots.) The R g and Rp values for the dinitrobenzoates of methanol and hexanol were measured under different conditions; two adsorbents were used, Merck's silica gel G and silica gel obtained from Mallinckrodt.The latter adsorbent was prepared by mixing 25 g of Mallinc- krodt's silica gel with 5 g of gypsum in a mortar and then adding 60 ml of distilled water. From this slurry, chromatoplates were prepared in the way described for silica gel G.l Table I shows the results of a series of experiments in which the RB and RF values for the 3,5-dinitrobenzoate of hexanol were compared. The largest difference between RF values was 30.1 per cent. and that between R B values was 18.6 per cent. The results for the 3,5-di- nitrobenzoate of methanol were essentially the same, the largest differences between RF and RB values being 50.0 and 14.3 per cent., respectively. For this reason, we use RB valuesAugust, 19611 OF 3,5-I)INITROBENZOATES 529 in preference to RF values for defining the positions of the compounds on the chromatograms.The RB values of the aliphatic alcohols can be seen from Fig. 1, and those for some other hydroxyl compounds are listed in Table 11. TABLE I R R ~ AND RF VALUES FOR THE 3,5-DINITROBENZOATE OF HEXANOL Distance of Standard Standard Time of solvent front Sumber deviation deviation activation, from start, of MeanRv of KF Mean RB of RB Silica gel used minutes cm cxperinients value value value value Merck . . . . 15 10 9 0.39 0.006 1.34 0.020 Mallinckrodt . . 15 10 6 0.405 0.012 1-28 0.023 Merck . . . . 30 9-5 6 0.26 0.01 1 1-27 0.016 Merck . . .. 5 14 6 0.284 0.002 1-19 0.014 Mallinckrodt . . 5 9.5 1 0 0.494 0.018 1-24 0.02 I Merck . . . . 60 6 ti 0.273 0.020 1-50 0.023 TABLE I1 Rg VALUES FOR VARIOUS ALCOHOLS AND PHENOLS Compound, as 3,6-dinitrobenzoate Rg value Compound, as 3,5-dinitrobenzoate HB value Phenol.. .. . . .. . . 0.76 Eugenol . . Thymol . . . . . . . . 1-42 Isoeugenol . . l-Naphthol . . . . . . . . 0.89 Citronellol . . 2-Naphthol . . . . . . . . 0-93 Geraniol . . o-Cresol . . . . . . . . 1.00 Terpineol . . m-Cresol . . . . . . . . 1.02 Furfuryl alcohol p-Cresol . . . . . . . . 1.05 Benzyl alcohol Maltol . . . . . . . . . . 0.55 .. . . . . 0.56 . . . . . . 1.69 . . . . . . 1.26 .. . . . . 1-41 . . . . . . 0-71 . . . . . . 0-76 . . . . . . 0.92 METHOD PREPARATION OF 3,5-I)INITROBENZOATES FROM DILUTE SOLUTIONS- As the hydroxyl compounds were always obtained as dilute solutions in water or ethanol, it was difficult to prepare the derivatives directly.Although some success was achieved by using Holley and Holley’s method,* in extremely dilute aqueous solutions this method was not very sensitive. It was especially difficult to prepare the dinitrobenzoates of phenolic compounds. A large excess of ethanol is often present in steam-distillates from foods; in such distillates the minor components are completely masked. However, results were satisfactory when these distillates were extracted two or three times with small amounts of pentane, methanol and ethanol then remaining almost completely in the aqueous layer; even small amounts of higher alcohols and phenols were easily extracted. The pentane extract was dried over anhydrous sodium sulphate for some hours, and then decanted into an Erlenmeyer flask containing 10 to 20 ml of benzene, 0-5 g of 3,5-dinitro- benzoyl chloride and 1 to 2 ml of dry pyridine.The mixture was heated under reflux for 30 minutes to 1 hour and, when cool, was extracted successively with 25-ml portions of 0.1 N sulphuric acid, a 5 per cent. solution of sodium carbonate and water. The organic layer was then evaporated to dryness under slightly reduced pressure in a bath of hot water. To the dry residue were added 2 to 4 ml of benzene, 5 to 10 pl of this saturated solution were placed on an activated plate, and the dinitrobenzoates were separated by development in (1 + 1) benzene - light petroleum mixture. The developed chromatogram was sprayed with a 1 per cent. solution of l-naphthylamine in ethanol; the benzoates then became visible as yellow to orange spots.DISCUSSION OF THE METHOD Fig. 2 shows two chromatograms obtained with mixtures of known composition; 3,5-dinitrobenzoic acid, which may be present in relatively large amounts, does not affect the separation, as it remains at the start when the benzene - light petroleum solvent is used. Since, for separation, RB values must differ by at least 0.3 to 0.4, the method is clearly unsuitable for separating the members of homologous series of alcohols or phenols. However, very small amounts of higher alcohols in ethanol can easily be detected. Small amounts530 ALLAN THE DETERMINATION OF ZINC IN AGRICULTURAL [Vol. 86 i i Fig. 2. Detection of octanol and thymol in presence of a large excess of ethanol: (a) chromatogram for mixture of 100 mg of octanol and 100 ml of 10 per cent. ethanol; (b) chromatogram for mixture of 1 litre of water, 1 ml of ethanol and 5 mg of thymol. A = 3,5-dinitrobenzoic acid; B = ethanol; C = octanol; D = thymol of the 3,5-dinitrobenzoates can even be separated from large amounts of the unchanged acid for melting-point determinations, saponification, etc. The esters can be hydrolysed, and the original hydroxyl compounds can then be subjected to other chromatographic methods or to chemical procedures. The locations of the spots are most suitably measured by comparison with a standard substance, as RF values are influenced by many factors in this kind of adsorption chromatography . REFERENCES 1. 2. 3. 4. Dhont, J . H., and de Rooy, C., Analyst, 1961, 86, 74. Stahl, E., Pharmazie, 1956, 11, 633. Nicolaus, B. J. R., J . Chromatograplzy, 1960, 4, 384. Holley, A. D., and Holley, R. W., A~zal. Chem., 1952, 24, 216. Received March 13th, 1961

ISSN:0003-2654    

DOI:10.1039/AN9618600527

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

530 ALLAN THE DETERMINATION OF ZINC IN AGRICULTURAL [Vol. 86 The Determination of Zinc in Agricultural Materials by Atomic-absorption Spectrophotometry BY J. E. ALLAN (Defiartment of Agriculture, Rukuhia Soil Research Station, Hamilton, New Zealand) An investigation of various aspects of the determination of zinc in agri- cultural materials by atomic-absorption spectrophotometry is described. Satisfactory recoveries for zinc were obtained from fertilisers, soils, soil extracts and plants. The method is rapid and accurate, and with the appara- tus used the limit of sensitivity is 0.025 p.p.m. of zinc. Coefficients of variation of 2 to 3 per cent. are obtained over the range 0.3 to 8.0 p.p.m. of zinc. THE use of atomic-absorption spectrophotometry for determining zinc in plant material was described by David,l who found that no interference was caused by the other elements present in plants and considered that the technique was superior to chemical and polarographicAugust, 19611 MATERIALS BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY 531 methods for determining this element.Gidley and Jones2 have reported the successful application of atomic-absorption spectrophotometry to the determination of zinc in a variety of metallurgical materials. They investigated many elements for interference effects and found that only silicon caused depression of absorption. In this laboratory, atomic absorption has been successfully used for the past 3 years for the routine determination of zinc in a variety of agricultural materials, and the results are presented in this paper.MEASUREMENT OF ZINC ABSORPTION APPARATUS AND METHODS- The general arrangement of the apparatus and the method of measurement were described previously for the determination of magne~ium.~ Two types of hollow-cathode lamp have been used. One (kindly supplied by Mr. A. Walsh, C.S.I.R.O., Melbourne, Australia) had a zinc cathode and was operated at a current of about 10 to 12 mA; the other (obtained from Hilger & Watts Ltd., London) had a brass cathode and was operated at about 40 to 45 mA. Concentration of zinc, p.p.m. Fig. 1. Absorption for zinc measured at 2138.6 A : curve A, in aqueous solution; curve B, in 40 per cent. acetone; curve C, in isobutyl methyl ketone With both types of lamp, a “warming-up” period of 30 to 60 minutes was necessary before maximum intensity was attained.Both lamps gave results similar in sensitivity and repro- ducibility, but that having the brass cathode had the longer life. The operating currents used were the minimum necessary to ensure that full-scale readings were attainable, and, with both lamps, reduction of current to below the values mentioned above led to a rapid decrease in intensity. The burner used provided a flame 12 cm in length and has been described by Clinton*; the atomiser was of the Lundeggrdh type, modified to aspirate solutions directly from a beaker. The zinc line at 2138.6 A was used for all measurements, this being the only line in the zinc spectrum that gives measurable absorption. SENSITIVITY- The sensitivity attainable with this equipment when aqueous solutions were sprayed into an air - acetylene flame is shown by curve A in Fig.1. Linearity and sensitivity depend to some extent on the particular lamp used, and this curve is an approximate average of results obtained over 3 years, which in the extremes gave transmissions of 20 per cent. for 3 and 7 p.p.m. of zinc. The sensitivity is about four times that reported by Gidley and Jones and about ten times that reported by David, and, although most of this increase in sensitivity is due to the longer flame used, some contribution may have come from the atomiser. When a coal-gas flame is used, the sensitivity is increased by about 20 per cent., but there is no increase with a fuel-rich flame, indicating that the dissociation of zinc into atoms is virtually complete.532 ALLAN : THE DETERMINATION OF ZINC I N AGRICULTURAL [Vol.86 The effect of spraying an organic rather than an aqueous solution has been discussed in detail el~ewhere.~ Curves B and C in Fig. 1 were plotted from the results obtained with solutions made up in 40 per cent. acetone and isobutyl methyl ketone, respectively. Am- monium pyrrollidine dithiocarbamate was used to form a complex with the zinc before extraction into the latter solvent, as this complex is readily soluble in esters and ketones, which are particularly suitable for spraying into flames. To a 10-ml portion of each of a series of sodium acetate - hydrochloric acid solutions, containing from 1 to 20 pg of zinc, were added 1 ml of a 1 per cent. aqueous solution of amonium pyrrollidine dithiocarbamate and 10ml of isobutyl methyl ketone.The mixture was shaken, and, after the layers had separated, the zinc content of each layer was determined. The percentages of zinc extracted into the organic layer from solutions of pH 1.5, 2.1 and 2.5 to 5-0 were, respectively, 67, 95 and 100. Somewhat surprisingly, it was found that the organic solvents used (isobutyl methyl ketone, ethyl pentyl ketone and ethyl acetate) frequently contained traces of zinc; however, these solvents could readily be purified by washing with 2 per cent. hydrochloric acid. REFRODUCIBILITY- Reproducibility was determined by repeatedly measuring the absorptions of a series of solutions containing 0.4, 1.6, 4.0 and 8-0 p.p.m. of zinc. At these concentrations, the coefficients of variation of the apparent zinc content, calculated from twenty-one sets of measurements, were 3.1, 2.5, 1-8 and 3-0 per cent., respectively.With adequate control over hollow-cathode current, photomultiplier voltage, air pressure and flow of gas, reproducibility depends largely on the stability of the hollow-cathode lamp. In my experience, the above results are fairly average ; on occasions, better have been obtained, and, with some lamps, worse. That Gidley and Jones obtained a coefficient of variation of 3.5 per cent. after integration for 30 seconds reflects the fact that at present not all hollow- cathode lamps are equally good. INTERFERENCE- David’s observation1 that none of the elements present in plant digests interferes with the determination of zinc has been confirmed.Likewise, to judge from the results in Table I1 (p. 533), none of the elements present in soil or fertiliser solutions causes any interference, provided always that the total concentration of salt and acid is not sufficiently great to alter the physical properties of the solution to an extent such that atomisation is affected. The U S E OF ORGANIC SOLVENTS- The effect of pH on the extraction has been investigated. Zinc present, p.p.m. 0.0 0.3 0.6 1.2 3-0 6.0 TABLE I EFFECTS OF DIFFERENT ACIDS ON ZINC ABSORPTION Galvanometer reading for solution in- hydrochloric sulphuric nitric perchloric acid acid acid acid 88.3 88.5 88.5 87.0 78.5 79.0 79.0 79-0 64.0 64.5 64.7 64.8 35.5 35.5 35.5 34.5 18.5 18.5 18.5 18.3 100 100 100 100 plant-digest solutions used contained about 2ml of 72 per cent. perchloric acid in 20ml of solution; this was sufficient to cause results for zinc to be low by some 5 to 7 per cent.unless standards containing approximately the same amount of acid were used. The effects of various acids on measurement of the zinc absorption were tested by using a series of solutions containing up to 6 p.p.m. of zinc, prepared in 0.5 N hydrochloric, sulphuric, nitric and perchloric acids. For these experiments the apparatus was set to give 100 per cent, transmission when distilled water was sprayed into the flame. The results obtained with an air - acetylene flame are shown in Table I, from which it can be seen that no acid itself caused any absorption and that, within the experimental error of reading the galvanometer, zinc absorption was the same in the four acids.Similar results were obtained with air - coal-gas and air - propane flames.August, 19611 MATERIALS BY ATOMIC-ABSORPTION SPECTROPHOTOMETRY 533 The observation by Gidley and Jones2 that an absorption band was produced when halogen acids were sprayed into the flame and interfered with measurement of zinc absorption led to experiments in which it was established that the presence of hydrochloric acid caused no difference in the absorption at 2138 A under a wide variety of flame conditions and in presence of a number of compounds that could conceivably occur in flames in some circumstances. It was concluded that Gidley and Jones’s results were caused by contamination from a brass burner giving rise to the zinc line at 2138 A and the copper lines at 2165, 2178 and 2183 A, which have been shown6 to absorb strongly.This conclusion was reached by Gidley and Jones in a Note7 that appeared after this paper had been submitted. ANALYTICAL APPLICATIONS The methods used and the results obtained for the determination of zinc in various agricultural materials are briefly described below. FERTILISERS- A suitable weight of sample was boiled with 3 N hydrochloric acid, the solution was evaporated to dryness, the residue was dissolved in 0.5 N hydrochloric acid, and this solution was diluted to 100ml. Recoveries and results for different weights of sample are shown in Table 11. TABLE I1 ZINC FOUND IN FERTILISERS, SOIL AND SOIL EXTRACT Amount of Zinc found per Sample sample taken Zinc added, g of sample, tLg Pg 1.0 g 0.25 g 0.25 g 0.25 g 1.0 R Fertiliser A* .. r 0.26-g 0.25 g 0.25 g Fertiliser Bt . . - :i2 } Average 293 - 100 692 (Expected 693) 200 1084 (Expected 1093) - 2:; } Average 487 100 892 (Expected 887) 200 1280 (Expected 1288) - Soil . . . . 4-0 g - 2.0 g - 1.0 g - 1.0 g 100 188 (Expected 190) 1.0 g 200 295 (Expected 290) ‘J”:> Average 90 90 K: } Average 10-5 10 ml undiluted - 5 ml diluted to 10 ml 2.5 ml diluted to 10 ml - 10.8 5 ml diluted to 10 ml 6-25 11-4 (Expected 11.5) - Soil extract . . * Superphosphate. t Proprietary fertiliser consisting essentially of lime, superphosphate, limonite and small amounts of various salts, including zinc sulphate. Low result probably due to decrease in efficiency of atomiser. SOILS- Various amounts of soil were digested with nitric and perchloric acids, silica was removed with hydrofluoric acid, and the solution was evaporated to dryness; the residue was dissolved in 5 per cent.perchloric acid, and the solution was diluted to 50ml. Results for soils are also shown in Table 11. The low result for the most concentrated solution was probably caused by a decrease in the efficiency of the atomiser owing to the high salt content. A similar effect occurred when these solutions were analysed for copper by the same method. For most soils, 1- to 2-g are sufficient. SOIL EXTRACTS- A 2-g sample of a peat soil was shaken with 20 ml of a 1 per cent. aqueous solution of disodium ethylenediaminetetra-acetate (a commonly used extractant for “available” zinc) for 2 hours, and the mixture was then filtered.Various aliquots of the filtrate were diluted to 10 ml, and the solutions were analysed for zinc; standards made up with 0.1 N hydrochloric acid were used. The results are also shown in Table 11.534 ALLAN [Vol. 86 PLANTS- Triplicate samples of various sizes were digested with 20 ml of a nitric - perchloric acid mixture (17 ml of concentrated nitric acid and 3 rnl of 72 per cent. perchloric acid). After the solutions had cleared, each was concentrated to about 2 ml, transferred to a 20-ml cali- brated flask, diluted to the mark with water and filtered. The filtrates were analysed for zinc, standards made up in 10 per cent. perchloric acid being used. The results are shown in Table 111; each represents a single determination, and sample heterogeneity and chance contamination may have contributed to the variation between replicates.TABLE I11 ZINC FOUND IN APPLE AND GRASS Weight of Zinc Sample sample taken, added, Zinc found per g of oven-dried sample, g Pg Pg 1.0 - 2.8, 3.1, 2.9 (mean 2-93) Average 3.00 1.0 0.5 - 42.0, 39.0, 40.0 (mean 40.3) - 39.8, 40.5, 38.8 (mean 39.7) Apple . . . .{ 2.0 - 3-1, 3.1, 3-0 (mean 3.07) } Grass . . 4.8 7.7, 7.4, 7.9 (mean 7.7; expected 7.8) 39.5, 40.5, 39.5 (mean 39.8) - 37.5 76-5, 78.0, 78.3 (mean 75.6; expected 77.4) 1.0 CONCLUSION The determination of zinc by atomic-absorption spectrophotometry compares more than favourably with other methods in simplicity, speed, reproducibility and accuracy. In a recent review of chemical methods for determining trace amounts of zinc, Margerum and Santacanas stated that their preferred method had an operating range of 3.3 (coefficient of variation 6-1 per cent.) to 32.7 pg of zinc (coefficient of variation 1.93 per cent.).The determination of zinc by neutron-activation analysis has also been describedgJ0; with this technique the ultimate sensitivity was 0.04 pg of zinc, and reliable determinations were possible on samples containing 0.3 pg of zinc. In the proposed method, the limit of sensitivity (99 per cent. transmission) for an aqueous solution is 0.025 p.p.m. of zinc, which, when 2 ml of solution are used for a determination, amounts to 0.05 pg, and determinations of reasonable accuracy (coefficient of variation of, say, 5 per cent.) can be carried out on a solution containing 0.2 p.p.m. of zinc, i.e., 0.4 pg per 2 ml. When it is convenient to extract the zinc into an organic solvent, these figures are decreased to about one-fifth. I thank Miss V. 0. Mamie Wright for assistance with the experimental and analytical work. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. David, D. J., Analyst, 1958, 83, 655. Gidley, J. A. F., and Jones, J. T., Ibid., 1960, 85, 249. Allan, J. E., Ibid., 1958, 83, 466. Clinton, 0. E., Spectrochim. Acta, 1960, 16, 985. Allan, J . E., Ibid., 1961, 17, 467. - , Ibid., 1961, 17, 459. Gidley, J. A. F., and Jones, J. T., Analyst, 1961, 86, 271. Margerum, D. W., and Santacana, F., Anal. Chem., 1960, 32, 115'7. Bowen, H. J. M., Int. J , AfipI. Radiation and Isotopes, 1959, 4, 214. Banks, T. E., Tupper, R., White, E. M. A., and Wormall, A., Ibid., 1959, 4, 221. Received December 28th, 1960

ISSN:0003-2654    

DOI:10.1039/AN9618600530

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

August, 19611 SHIMI, NOUR EL DEIN AND IMAM 535 Micro-quan ti tative Analysis by the Zone - Strip Technique BY IBRAHIM R. SHIMI, (Biochemistry Department, Faculty of Science, Ain Shams [Jnnivevsity, A bbassiah, Cairo, Egypt) MOUSTAFA S . NOUR EL DEIN (Biology Department, High Training College, Heliofiolis , Cairo, Egypt) AND GAMAL M. IMAM (National Research Centre, Caim, Egypt) The application of solution in a zone across a narrow strip of chromato- graphy paper restricts the width of the loaded area on the developed strip. The length of the developed zone is proportional to the logarithm of the zone content, but this relationship holds only below a certain maximum concentration, depending on the nature of the compound. Calibration graphs were plotted from the results for analytical-reagent grade compounds.The graphs obtained for several compounds were adequate for assaying solutions containing 1 mg of the compound per ml. The technique may be of particular value for analysing biological fluids. THE use of paper chromatography for micro-quantitative analysis has not yet attained a high degree of accuracy. Fisher, Parsons and Morisonl reported that the area of a round or ovoid spot increased with the logarithm of the spot content. For ovoid regular spots, the length is proportional to the logarithm of the spot content.2 Brown and Marsh3 increased the sensitivity of evaluating low concentrations of light-absorbing materials on paper-strip chromatograms by using a special attachment designed for use with a monochromator incorporating a stabilised light source.However, round or ovoid spots occur infrequently on developed chromatograms, and, because of this, we applied solutions in zones 2 mm wide near the tops of strips 0.5 cm wide. When these strips were developed, the width of a loaded area was restricted to 0.5 cm and its length was influenced by the content of the zone; graphs of length of zone against logarithm of zone content were linear. EXPERIMENTAL The pure standard antibiotics used were generously offered by Pfizer & Co. Inc., U.S.A., Leo Pharmaceutical Products Trading Ltd., Denmark, Imperial Chemical Industries Ltd. and the Medical Research Council (International Standard penicillins), England. The rest of the compounds used were of analytical-reagent grade and were obtained from the British Drug Houses Ltd.or from E. Merck A.G., Dannstadt, Germany. CHROMATOGRAPHY OF PENICILLINS AND TETRACYCLINES- For assaying the different penicillins (G, V, K, X, F and dihydro-F), sheets of Whatman No. 1 chromatography paper were soaked in a 30 per cent. solution of phosphate buffer (pH 6.2) and then cut longitudinally into strips 0.5 cm wide. Results were more accurate when narrower strips were used, but such strips were difficult to manipulate. At 8 cm from the end of each strip, two pencil lines were drawn 2 mm apart; the area between the two lines was found to absorb 0.00075ml of liquid when a fine capillary tube was passed over it. Diethyl ether saturated with water was used as mobile phase, and saturation of the atmosphere in the chromatographic tank with ether and water vapours resulted in improved resolution of the different penicillins. The mobile phase was allowed to descend through the strips for 40cm, and the strips were then removed, dried and subjected to any of the six treatments described below.(i) They were sprayed with an aqueous 2 per cent. solution of ferric chloride. (ii) They were sprayed with a 1 per cent. solution of iodine and then with a solution of starch.536 (iii) (4 (4 (4 The solution I -0 0 SHIMI, NOUR EL DEIN AND IMAM MICRO-QUANTITATIVE [Vol. 86 They were sprayed with the iodine solution and then subjected to ultra-violet light. The strips after treatment (i) were washed in a stream of tap-water until free from excess of the ferric salt, dried and then sprayed with a 20 per cent.solution of potassium ferrocyanide or with a 2 per cent. solution of sodium thiocyanate. They were sprayed with a 1 per cent. solution of potassium pennanganate. They were sprayed with an ammoniacal solution of silver nitrate and then set aside for 6 hours at 60" C. strips used for assaying the different tetracyclines were soaked in a 20 per cent. of phosphate buffer (pH 3.5) and dried in air, and the solutions were applied to 2*oI: 2-mm zones: the strips were then developed with ethyl acetate or aqueous butyi >lcohol. Penici G Penici F Direction of solvent travel Penicil'i+ F I (a) (6) Fig. 1. Reproduction of strips loaded with penicillins G and F and treated with (a) solutions of iodine, ferric chloride and potassium ferrocyanide and (b) solutions of iodine and starch.The left-hand strip of each pair is undeveloped The developed strips were sprayed with a 1 per cent. solution of alkaline potassium pennan- ganate or, preferably, a 4 per cent. solution of sodium hydroxide and then subjected to ultra-violet light. For comparison, antibiotics were also assayed by biological methods. The procedure used for the penicillins was a slightly modified version of that described by Goodall and Levi.4 The lengths of the zones on developed strips were measured with a travelling microscope. CHROMATOGRAPHY OF OTHER COMPOUNDS- Sugars, amino acids, ascorbic and other organic acids and adrenalin were assayed on unbuffered strips of paper, with a butyl alcohol - acetic acid - water mixture (4 : 1 : 5) as developing solvent, The loaded zones were made visible by spraying them with ethanolic 5 P M 01 J Log (weight of penicillin K, pg) Fig. 3.Graph showing relationship between length and logarithm of content of zone containing penicillin KFig. 2. Dcvclopcd strips for penicillins after treatment with solutions of iodine and starch [To face Doge 536August, 19611 ANALYSIS BY THE ZONE - STRIP TECHNIQUE 537 solutions of ninhydrin (for amino acids) or p-bromophenol (for other organic acids). For sugars, Harris and MacWilliamJs5 diphenylamine - aniline - phosphoric acid - water reagent was used. Aqueous solutions of 2,6-dichlorophenol endophenol and ferric chloride were used to define the areas containing ascorbic acid and adrenalin, respectively. RESULTS AND DISCUSSION OF THE METHOD The different reagents applied to the strips containing penicillins induced marked colour contrasts (see Figs.1 and 2), and results were in good agreement with those found by bioassay. Table I shows the effects produced by the various reagents. The relationship between length and logarithm of content of zone for penicillin K is shown in Fig. 3. We favour treatment of the penicillin-loaded strips with iodine solution, ferric chloride solution and then ultra-violet light; this gives an obvious colour contrast, and the zones have sharp edges. Application of ferrocyanide solution preserves the colour contrast for several months. This procedure can be successfully applied to solutions of penicillin-producing moulds, even in presence of some natural products, eg., corn-steep liquor or molasses.2 20 - 1.0 0. 1.0 20 Log (weight of acid, pg) Fig. 4. Curve A, tryptophan: curve B, cystine; curve C, glutamic acid Log (weight of acid, pg) Fig. 6. Curve A, succinic acid; curve B, gluconic acid 0 5 I -5 2 5 Log.(weight of compound, pg) Fig. 8. B, ascorbic acid Curve A, adrenalin; curve log (weight of acid, pg) Fig. 5. Curve A, aconitic acid; curve B, citric acid; curve C, malic acid ,Log (weight of sugar, pg), Fig. 7. Curve A, fructose; curve B, maltose; curve C, sucrose; curve D, raffinose Log( weight of tetracydine, yg) Curve A, chlortetracycline; curve B, Fig. 9. oxytetracycline; curve C, tetracycline Figs. 4 to 9. Graphs showing relationshipsibetween lengths and contents of zones containing various compounds538 SHIMI, NOUR E L DEIN AND IMAM [Vol.86 TABLE I COLOURS PRODUCED WITH PENICILLINS BY DIFFERENT REAGENTS Treatment No. * Colour of unloaded area White or pale buff (ii) (iii) Shiny white Dark shiny violet White or pale blue White or pale red Colour of penicillin-loaded zone Buff, with hard texture White Dark blue Deep blue Deep red White Violet White Brown or black (4 (i4 t ( i v ) :: (4 (4 * See pp. 535-536. t Sprayed with potassium ferrocyanide solution. :: Sprayed with sodium thiocyanate solution. The colours produced when developed strips containing other compounds are sprayed with different reagents are listed in Table 11, and the relationships between length and logarithm of content of zone for these compounds are shown in Figs. 4 to 9. Compound Amino acids . .Organic acids Sucrose Raffinose Ascorbic acid Adrenalin . . Tetracyclines . . TABLE I1 COLOURS PRODUCED WITH VARIOUS COMPOUNDS Colour of Colour of Spray reagent loaded zone unloaded area .. Ninhydrin Pink White or pale red .. p-Bromophenol Yellow lBlue or violet Brown Brown Grey or blue . . Diphenylamine, aniline a,nd . . 2,6-Dichlorophenol endophenol White .. Ferric chloride Dark green changing Pale buff to brown . . Sodium hydroxide Yellow White or violet in lWhit" Pink or pale violet { phosphoric acid ultra-violet light The linear graphs relating length of zone to logarithm of zone content are adequate for determining concentrations of about 1 mg of the compounds per ml of solution. The maximum errors calculated when these graphs were used for assaying penicillins, tetracyclines, organic acids, amino acids, sugars, adrenalin and ascorbic acid were 2.2, 1.6, 1.1, 1.7, 2.8, 1.3 and 2.4 per cent., respectively. For each of the compounds tested, there is a maximum weight below which the relation- ship between length of zone and logarithm of zone content is linear; above this level, such a relationship does not exist. Within a narrow range of concentrations linear graphs relating length of zone and zone content could be obtained. Such graphs were excellent for assaying the compounds tested, but were of limited application. Further work on the various factors influencing this technique and its applications is in progress. REFERENCES 1. 2. 3. 4. 5. Fisher, R. B., Parsons, D. S., and Morison, G. A., Nature, 1948, 161, 764. Fowler, H. D., Ibid., 1951, 168, 1123. Brown, J. A., and Marsh, M. M., Anal. Chem., 1953, 25, 1865. Goodall, R. R., and Levi, A. A., Analyst, 1947, 72, 277. Harris, G., and MacWilliam, I. C., Chem. & Ind., 1954, 249. Received Febrziary 6th, 1961

ISSN:0003-2654    

DOI:10.1039/AN9618600535

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

August, 19611 NOTES 539 Notes THE MICRO-DETERMINATION OF CARBON AND HYDROGE-N I N FLUORINE-CONTAINING ORGANIC COMPOUNDS WHEN fluorine-containing compounds are analysed for their carbon and hydrogen contents, special precautions must be taken to avoid errors arising from the presence of fluoride ion. Low values can be obtained for the carbon contents of fluorocarbons because they resist decomposition, owing to the stability of the fluorine - carbon bond. Generally, results are high because some silicon tetrafluoride is formed by interaction between the fluoride ions and the silica of the com- bustion tube; this tetrafluoride is retained by the absorbent in the carbon dioxide absorption tube. These observations are related to slow-combustion methods, in which combustion is carried out with oxygen flowing a t 4 to 10 ml per minute and the products of decomposition pass through layers of copper oxide or are in contact with a catalytic surface and silver in the hot zone of the combustion tube, or both. The constituents of the products of oxidation of fluorine-containing compounds are not known exactly, but it is reasonable to assume that elemental fluorine is the initial product.This assumption is based on the fact that the theoretical amount of elemental chlorine is obtained when a chlorine-containing compound is decomposed by combustion in a fast stream of oxygen in an empty tube.1 Similarly, bromine is the main product from a bromine-containing compound, and certainly only iodine is obtained from combustion of an iodine-containing substance. Some hydrofluoric acid must also be produced to account for the formation of silicon tetrafluoride.This reaction can be quantitative, as shown by Milner2 and Clark? who have used the reaction to determine fluorine by absorption of the terafluoride in water and subsequent titration of the hydrofluoric acid produced. Under normal conditions of combustion, some of the silicon tetra- fluoride is decomposed and silica is deposited in the combustion tube, as found by Belcher and G~ulden,~ who have studied the determination of carbon and hydrogen in fluorine-containing compounds. The elemental fluorine and its hydride will then be retained by the hot silver forming part of the combustion-tube filling. Belcher and Coulden4 retained the silicon tetrafluoride by means of sodium fluoride heated t o 270" C, the reagent being packed in the "beak" end of the combustion tube.The hot zone contained a roll of platinum foil maintained a t 800" C. Granulated magnesium oxide was used by Throckmortoh and H ~ t t o n , ~ who replaced part of the copper oxide packing at the front end of the tube with a 35- to 40-mm layer of the reagent; the temperature of the hot zone was 775" C. Fluorocarbons and chlorofluorocarbons were analysed without trouble, whereas Belcher and Goulden found it necessary to mix samples of fluorocarbons with a known weight of a hydrogen- containing organic compound, e.g., benzoic acid. The carbon content of the sample was calculated by deducting the theoretical amount of carbon derived from the weight of the additive.Mag- nesium oxide has also been recommended by McCoy and Bastin,6 who placed a 50-mm layer of the reagent a t each end of the copper oxide layer, which was heated at 900" C. Rush, Cruikshank and modes,' however, found that the magnesium oxide became inactive after a few determinations, and this they attributed to prolonged heating or exhaustion of the reagent. These workers recommended the use of a magnesium - aluminium compound, SMgO.Al,O,, maintained at 960" C, together with lead sesquioxide a t 450" C. Short layers of silver gauze were distributed throughout the combustion zone, where the temperature was about 450" C. Lead dioxide at 180" C was also included in the tube filling to retain oxides of nitrogen arising from the combustion of nitrogen- containing compounds.It appears from the observations of several workers that, for fluorocarbons, little of the sample is decomposed during the initial stages of the combustion. Most of the sample is vaporised and passes virtually unchanged into the hot zone, where complete or partial decomposition occurs, depending on the thermal stability of the compound. The extent of decomposition is also de- pendent on the temperature of the hot zone and, equally, on the efficiency of the tube packing (either in its oxidative capacity or catalytic effect) or on its affinity for fluorine ions. McCoy and Bastin,6 for example, found that values for carbon were less accurate (-10.5 per cent.), but Macdonalds found that, when a platinum catalyst was used in place of the copper oxide, results had the required accuracy (within &to-3 per cent.).The presence of water, even in trace amounts, has a pronounced effect on the course of combustion of fluorocarbons. Freier, Nippoldt, Olson and Weibleng have extended the silicon tetrafluoride method of determining fluorine to include540 NOTES [Vol. 86 the determination of carbon. For this, they used a tube packed with alternate layers of quartz chips and platinum gauze, which was maintained at about 1200OC; moist oxygen was passed through the tube at 25 ml per minute. Application of Belcher and Ingram'slO rapid-combustion method has only recently been extended to the determination of fluorine-containing compounds. Woodll has shown that fluoro- organic compounds (fluorocarbons were not tested) could be analysed within the accepted limits of accuracy by using the metbod developed by Belcher and Goulden4 (absorption of the silicon tetra- fluoride by means of sodium fluoride).The reagent was placed in a tube attached to the exit of the baffle-chamber combustion tube and heated to 270" C. The combustion tube contained the customary plug of quartz wool and roll of silver gauze. Some years ago, I had to determine the carbon contents of some specimens of polytetra- fluoroethylene. The determinations were carried out by the rapid-combustion rnethod,lO and it was found that the values for carbon were within the accepted limits of accuracy, and the hydrogen contents, which should have been negative, were not more than 0.1 per cent. No special pre- cautions were taken to remove fluorine, other than by the roll of silver gauze normally present in the exit part of the combustion tube. Rush, Cruikshank and Rhodes' reported that when polytetrafluoromethane was analysed by the conventional combustion procedure, correct values for carbon were obtained, suggesting that the fluorine was completely retained by the silver packing.In view of these separate observations, further experiments have been carried out to determine the fate of the halogen when other fluorine-containing compounds are analysed by the rapid-combustion method. It has not been possible to study the efficiency of the rapid method for fluorocarbons, owing to the lack of suitable test compounds of the type known to cause trouble. It is hoped to continue the experiments when such test compounds become available. EXPERIMENTAL The combustion experiments were carried out with a modified version of Belcher and Ingram's rapid-combustion unit.The baffle-chamber combustion tube was mounted horizontally, and the combustion heater was a small electrically heated furnace, 65mm in length, which was moved forward by a motor-driven mechanism. The speed of travel was controlled by means of coarse and fine resistors in the motor circuit, and manual movement of the furance was simulated by controlling the motor current with an energy regulator (Sunvic TYC) . This arrangement permitted a wide range of combustion times. In this work, the speed of the heater was adjusted to give a 5-mm movement in 5 seconds, and the stationary time was 22.5 seconds; the sample boat was 40mm from the main furnace. The total time of combustion was 8 minutes, and, when the heater had reached the main furnace, it was left in position for a further 7 minutes, during which combustion was completed and the products formed were swept out.The flow of oxygen was 50 ml per minute. In the various experiments described below, the inlet of the baffle chamber contained the customary plug of quartz wool, and the chamber was maintained at 900" f 20° C . The water and carbon dioxide were collected in Flaschentrager absorption tubes packed with suitable reagents ; when nitrogen-containing compounds were examined, an absorberlo containing manganese dioxide was connected between the two absorption tubes to retain the oxides of nitrogen formed.RETENTION OF FLUORINE BY SILVER- Weighed samples (3 to 4 mg) of trifluoromethylbenzoic acid were combusted in the apparatus, a roll of silver gauze, 160 mm long, being inserted in the exit tube of the baffle chamber. Part of this roll (about 70 mm) was heated to 550" f 10" C; the remainder reached to the "beak" of the exit tube in the normal way. Typical results are shown in columns A of Table I and indicate that, possibly, some silicon tetrafluoride was formed during the combustion and retained in the carbon dioxide absorption tube. However, when the nitrogen oxides absorber was connected between the absorption tubes, the values for carbon were within the limits of error, as shown. A small white deposit was also formed in the inlet connecting tube of the water-absorption tube; this was presumably silicon dioxide arising from decomposition of the tetrafluoride compound.The weight of the deposit was always less than 0.05 mg. Assuming that the theoretical amount of carbon dioxide had been obtained and that the excessive weight was related to silicon tetrafluoride, calculation showed that about 50 per cent. of the available fluorine had been collected as that compound. Confirmation of the formation of silicon tetrafluoride was obtained in another experiment. The products of combustion were dried by passage through a water-absorption tube attached toAugust, 19611 NOTES 541 the exit of the combustion tube and then passed through a bead absorberlz containing about 5 ml of water. was titrated with 0.01 N sodium hydroxide.absorber at the point where the gas first came into contact with the moistened beads. of alkali consumed was equivalent to approximately 60 per cent. of the available fluorine. RETENTION OF FLUORINE BY MAGNESIUM OXIDE- Weighed samples (3 to 4 mg) of trifluoromethylbenzoic acid were analysed as described above, but the silver in the exit tube was replaced by a layer of 14-mesh granules of pure magnesium oxide. The layer of reagent (about 60 mm long) was positioned between 10-mm wads of silver wool in the part of the tube heated by the furnace. The remaining space up to the "beak" was occupied by a roll of silver gauze. The oxide was heated to 650" C in some tests, but this temperature was increased to 750' C in others. The results of these experiments are shown in columns B of Table I, together with some values found for trifluoroacetanilide, the nitrogen oxides absorber being included in the system.The results for m-trifluoromethylbenzoic acid show that no silicon tetrafluoride was carried over into the carbon dioxide absorption tube. There was no deposit in the inlet connecting tube of the water-absorption tube. Further, the reagent was efficient in retaining the fluorine compound at the lower temperature of 550' C. RETENTION OF SILICON TETRAFLUORIDE BY MANGANESE DIOXIDE- In this series of experiments, only silver was used to retain the fluorine combustion products, but, as the compound being examined was trifluoroacetanilide, the absorber containing manganese dioxide was connected between the water and carbon dioxide absorption tubes.Samples weighing 3 to 4 mg were taken for the analyses, and some results are shown in columns A of Table I. In distinct contrast to the previous analyses of m-trifluoromethylbenzoic acid, when only silver was used to retain fluorine, the values for carbon had the required accuracy (within k0.3 per cent. of the theoretical figure). In these experiments, the white deposit was found in the inlet of the water-absorption tube. It was thus established that manganese dioxide at room temperature retains silicon tetra- fluoride quantitatively, possibly according to the reaction- A further study of this absorption phenomenon is being carried out to determine the mechanism of the reaction. The hydrofluoric acid formed according to the reactiong- SiF, + 2H,O -+ 4HF + SiO, A white band of silica was deposited in the bead The amount MnO, + SiF, --+ MnF, + SiO, RECOMMENDED TECHNIQUE FOR ANALYSING FLUORINE-CONTAINING COMPOUNDS The values found for carbon and hydrogen indicated, at least for the two substances examined, that silicon tetrafluoride could be quantitatively retained within the combustion tube by means TABLE I CARBON AND HYDROGEN CONTENTS FOUND UNDER VARIOUS CONDITIONS The values for hydrogen content listed under A were determined after deposited silica had been removed A B C I&& Carbon Hydrogen Carbon Hydrogen Carbon Hydrogen content content content content content content Compound tested found, found, found, found, found, found, % % % :0 % % 01 55.60 2-77 50.42 2.82* 50.70 2.80 57.90 2.68 50.37 2-57" 50.50 2.52 50.57 2-53t 50.72 2-85: 50.73 2.52 50.54 2-61? 50.55 2-60: 50.46 2.68 m-Trifluorometh ylbenzoic acid (carbon content, 50.54% ; hydrogen content, 2.65%) Trifluoroacetanilide (carbon content 50+30~0 ; hydrogen content, 3.20%) 50.50 3-22 50.84 3.15* 50-7 1 3.35 50.66 3.12 50.83 3.20* 50-73 3.13 50.93 3.25 50-95 3-23* 50-83 3.18 - - 50.65 3-35: 50.84 3-22 - - 50.99 3-14: - - * Magnesium oxide maintained a t 550" C.t Absorber containing manganese dioxide inserted between absorption tubes. Magnesium oxide maintained a t 750" C.642 NOTES [Vol. 86 of magnesium oxide when the rate of flow of oxygen was 50 ml per minute. Manganese dioxide can also serve as an external absorbent for the tetrafluoride at room temperature, but then a slight error in the values for hydrogen can be expected because of the small amount of silica collected in the inlet tube of the water-absorption tube.The deposit can be easily removed before the absorption tube is weighed, but traces of the solid may be swept into the absorber, although no evidence of this was obtained. Instead of the magnesium oxide being packed directly into the combustion tube, i t was considered that it would be more convenient to place the reagent inside the roll of silver gauze, so that it could readily be changed when necessary without dismantling the apparatus. This simplified procedure was examined and found to provide an alternative to the packed-tube tech- nique without a decrease in accuracy. A cylinder was made from a strip of 60-mesh silver gauze 160 mm long and about 40 mm wide, and this cylinder fitted snugly into the exit tube of the baffle chamber.A second strip of the gauze, 85 mm in length, was rolled round a piece of thick silver wire, which was bent into a hook to allow the prepared cylinder to be withdrawn from the tube. This roll was inserted in one half of the cylinder so that the hook projected outside. The remaining space in the cylinder was packed with 14-mesh granules of pure magnesium oxide, and the open end was pinched together to hold the reagent in the cylinder. The prepared roll was inserted in the combustion tube and conditioned in a current of oxygen at its working temperature of 550” C until the blank values were satisfactory. This modified filling has proved reliable in the analysis of fluorine-, chlorine-, bromine-, iodine- and sulphur- containing compounds, but has not yet been used for the analysis of fluorocarbons.Some typical results for carbon and hydrogen are shown in columns C of Table I. REFERENCES 1. 2. 3. Clark, H. S., Ibid., 1951, 23, 659. 4. 5. 6. 7. 8. Ingram, G., Mikrochim. Acta, 1953, 71. Milner, 0. I., Anal. Chem., 1950, 22, 315. Belcher, R., and Goulden, R., Mikrochem. Mikrochim. Ada, 1951, 36/37, 679. Throckmorton, W. H., and Hutton, G. H., Anal. Chem., 1952, 24, 2003. McCoy, R. N., and Bastin, E. L., Ibid., 1956, 28, 1776. Rush, C. A., Cruikshank, S. S., and Rhodes, E. J. H., Mikrochim. Acta, 1956, 859. Macdonald, A. M. G., in Wilson, C. L., and Wilson, D. W., Editors, “Comprehensive Analytical Chemistry, Volume IB.Classical Analysis,” Elsevier Publishing Co., Amsterdam, London, New York and Princeton, 1960, p. 550. Freier, H. E., Nip oldt, B. W., Olson, P. B., and Weiblen, D. G., Anal. Chem., 1955, 27, 146. 10. Belcher, R., and ggram, G., Anal. Chim. Acta, 1950, 4, 118. 11. Wood, P. R., Analyst, 1960, 85, 764. 12. Belcher, R., and Ingram, G., Anal. Chim. Acta, 1952, 7, 319. COURTAULDS LIMITED, RESEARCH LABORATORY 9. G. INGRAM MAIDENHEAD, BERKS. Received March 22nd, 1961 DETECTION OF PYRAMIDON AND ANTIPYRIN WHEN PRESENT TOGETHER THIS Note describes two tests for detecting pyramidon (amidopyrin ; 4-dimethylamin0-2~3-di- methyl-l-phenylpyrazol-&one) when present together with antipyrin (phenazone ; 2, S-dimethyl- l-phenylpyrazol-5-0ne)~ sS v4 ; ceric sulphate, platinic chloride and mercurous nitrate solutions are used.REAGENTS- METHOD All materials should be of the highest grade of purity obtainable. Ceric sulphate solution, 2 per cent.-Repared from the hydrated salt. Platinic chloride solution, 10 per cent. Mercurous nitrate solution, saturated. Pyramidon - antipyrin solutions-A series of aqueous solutions containing various concentra- tions of both compounds. TEST A- To about 1 ml of the pyramidon - antipyrin solution in an 80-mm x 8-mm test-tube are added 1 or 2 drops of the ceric sulphate solution; a violet colour is produced in the presence of pyramidon. By this reaction, 0.012 mg of pyramidon per ml of solution can be detected in theAugust, 19611 NOTES 543 presence of any amount of antipyrin that does not react with the ceric sulphate.(If ceric sulphate is present in excess, antipyrin produces a red-orange colour when heated.) After the reaction with pyramidon, 5 to 6 drops of the mercurous nitrate solution are added to the contents of the test-tube. If antipyrin is present, a grey or black spongy sediment is formed immediately or after a short time; this precipitate, when heated with concentrated nitric acid, produces a red colour. By this reaction, about 0.03 mg of antipyrin can be detected in 1 ml of solution containing any amount of pyramidon. It was found that the sensitivity of the test for antipyrin could be increased by using a mixed reagent consisting of 2 or 3 drops of the 10 per cent. solution of platinic chloride added to 100 ml of the ceric sulphate solution.When this reagent was used, a precipitate was formed in the presence of 0415mg of antipyrin per ml of test solution after the mercurous nitrate solution had been added. A similar result was achieved with a mixed reagent consisting of about 0.5 ml of a 1 per cent. solution of auric chloride added to 100 ml of the ceric sulphate solution. TEST B- To about 1 ml of the pyramidon - antipyrin solution in a small test-tube is added 1 drop of the platinic chloride solution; if pyramidon is present, a violet-blue colour appears. The colour is formed immediately with solutions containing about 0.02 mg of pyramidon per ml, but more slowly (or immediately after being heated) when the concentration of pyramidon is about 0.01 mg per ml. However, when 5 or 6 drops of the mercurous nitrate solution are added, a grey or black sediment appears when antipyrin is present, even at the level of 0412 mg per ml of solution.REFERENCES Antipyrin produces no colour. 1. 2. 3. 4. Bourcet, K., Angew. Chem., 1906, 19, 390. Escaich, J., 2. anal. Chem., 1921, 54, 432. Patein, J., Chew. Ztg., 1905, 45, 222. Ribdre, P., J . Pharm. Chim., 1930, 13, 444. GENERAL CHEMISTRY DEPARTMENT OLSZTYN, POLAND SCHOOL OF AGRICULTURE WIKTOR WAWRZYCZEK Received February 6th, 1961 DIRECT TITRATION OF HYDROLYSABLE SULPHUR IN ORGANIC COMPOUNDS As previously described, numerous sulphur compounds react with o-hydroxymercuribenzoic acid in alkaline solution to form soluble colourless sulphides having the structure RHgSHgR ; this reaction can be used for determining hydrolysable sulphur, and a procedure based on determining the unconsumed excess of o-hydroxymercuribenzoic acid after heating with the sample has been suggested.This Note describes the direct titration of hydrolysable sulphur with o-hydroxymercuribenzoic acid solution at room temperature, which has the advantages of being more selective and rapid. The procedure can be successfully used for the rapid determination of several sulphur compounds in the presence of each other, depending on their different rates of de-sulphuration. When dithiocarbamates are titrated, the isothiocyanates formed undergo slow hydrolysis and so interfere with the determination. However, this can be avoided by removing the isothio- cyanates from the aqueous solution with an organic solvent.Compounds such as 8-aminoethyl- and 8-hydroxyethylisothiocyanate undergo rapid transformation to the corresponding cyclic compounds, which are resistant to further de-sulphuration and do not interfere. METHOD PROCEDURE FOR PHENYL- OR ETHYLMONOTHIOCARBAMATE, DITHIOCARBAMATE, 8-AMINOETHYL- OR p-HYDROXYETHYLDITHIOCARBAMATE, DIPHENYLTHIOUREA AND RUBEANIC ACID- Add to the sample 5 ml of N sodium hydroxide, dilute to between 30 and 50 ml with water (or with methanol if the sample is insoluble in water), and titrate with 0.001 to 0.05 N o-hydroxy- mercuribenzoic acid. Thiofluorescein or dithizone can be used as indicator; with the former, the blue colour disappears at the end-point, and with dithizone the yellow colour changes to purple. The titration can be carried out a t 20" to 25' C, but a temperature of 30" to 40" C is preferable.544 NOTES [Vol.86 PROCEDURE FOR ETHYL-, PHENYL- OR BENZYLDITHIOCARBAMATE- Add to the sample 5 ml of N sodium hydroxide, dilute with water to 50 ml, add 5 to 20 ml of toluene, and titrate at 20" C with 0.001 to 0.05 N o-hydroxymercuribenzoic acid. Use thio- fluorescein as indicator, shake well after each addition of titrant, and take the end-point as being when the blue colour disappears for at least 30 seconds. Note that thiourea, ethylxanthate, o-phenylenethiourea, mercaptobenzothiazole, thiosulphate, thiocyanate and sulphite interfere not at all, or only slightly, with either titration. Recovery of the compounds tested was between 98-5 and 100 per cent., assuming that de-sulphuration was 50 per cent.for the dithiocarbamates and complete for the other compounds. REFERENCE 1. Wronski, M., AIzaE. Chem., 1960,32, 133; Chem. Anal., 1960, 5, 101. DEPARTMENT OF CHEMICAL TECHNOLOGY UNIVERSITY OF k6Di, POLAND MIECZYSLAW WRONSKI Received January 26th, 1961 ENZYMIC HYDROLYSIS OF PHOSPHOLIPIDS AS A MEANS OF DETERMINING EGG I N FOODS WE have already outlinedl a method of determining egg in food based on the hydrolysis of the phospholipids by lecithinase D followed by determination of the liberated choline. Originally, we prepared the enzyme concentrate from cabbage, but more recently we have had an opportunity of testing two samples of dried lecithinase D (one prepared from cabbage and the other from carrot) kindly supplied to us by C. F. Boehringer G.m.b.H., Mannheim, Germany (English agents, Messrs. Courtin and Warner Ltd., Lewes, Sussex).These samples represented enzyme concen- trates that the suppliers propose to make available commercially. TABLE I CHOLINE CONTENTS FOUND IN EGG AND ICE-CREAM Sample Enzyme A r \ Total Amount used choline Source per determination, found, mg 043 Boehringer (from carrot) Boehringer (from cabbage) 30 30 dried whole egg . . .. . . Boehringer (from cabbage) 30 * Choline content of dry matter. 7 Extracted and dried as described previouslv.' Cabbaget 200 Dried whole egg . . . . Ice-cream No. 1 (without egg) . . Boehringer (from cabbage) Ice-cream No. 1 plus 2 per cent. of 1-51" 1-62" 1-61* 1-63" 0.125 0.466 The results are shown in Table I, from which it is evident that a suitable proportion of either Boehringer lecithinase is 30 mg per determination.For the ice-cream, the choline derived from egg is, by difference, 0.341 per cent.; this corresponds to 2.1 per cent. of dried whole egg, whereas 2 per cent. was added. REFERENCE 1. KENSINGTON, LONDON, W.14 Casson, C. B., and Griffin, F. J., Analyst, 1959, 84, 281. THE LABORATORIES, J. LYONS & Co. LTD. F. J. GRIFFIN C. B. CASSON Received April 12th, 1961 RAPID EXTRACTION AND SPECTROPHOTOIVIETRIC DETERMINATION OF THE reagent N-benzoyl-N-phenylhydroxylamine, recommended by Shomel as superior to cup- ferron, has been widely used in analysis; it is an excellent reagent for the spectrophotometric determination of vanadium.2 p3 94 Tandon and Bhattacharyya5 studied many N-acyl-substituted N-arylhydroxylamine derivatives to obtain more information on the factors influencing the VANADIUM WITH N-CINNAMOYL-N-PHENYLHYDROXYLAMINEAugust , 19611 NOTES 545 selectivity and sensitivity of their reactions with metal ions.These workers recommended S-cinnamoyl-N-phenylhydroxylamine (I) as a highly specific spot-test reagent for vanadium, ~ ~ , H = c H - c = o I and this Note describes its use for the rapid spectrophotometric determination of milligram amounts of vanadium. N-Cinnamoyl-N-phenylhydroxylamine has all the useful features of N-benzoyl-N-phenyl- hydroxylamine as a reagent for vanadium, but is superior in sensitivity of reaction. The molecular extinction coefficients of the violet complexes of N-cinnamoyl- and N-benzoyl-N-phenylhydroxyl- amine with vanadium are 6300 & 50 and 4650 f 50, respectively, at their wavelengths of maximum absorption (calculated on the basis of vanadium).METHOD REAGENTS- N-Cinnamoyl-N-PhenyZhydroxyZamine soZution-The reagent as prepared in the laboratory5 consists of pale-green crystals, melting-point 162" to 163" C, and is stable to heat, light and air. A 0.1 per cent. w/v solution in ethanol-free chloroform was used for the extractions; this solution, stored in a dark bottle, is stable for several days. Ammonium vanadate solution, aqueous-The vanadium concentration was determined volu- metrically with potassium permanganate solution. PROCEDURE- Ensure that the vanadium in the sample solution is in the quinquivalent state by treating it with a few drops of a dilute solution of potassium permanganate until a faint pink colour persists.Transfer an aliquot of the treated sample solution, containing not more than 0.12 mg of vanadium, to a separating funnel, and add distilled water and hydrochloric acid until the volume is about 25ml and the acidity is between 2.7 and 7 . 5 ~ . Add 8 to 10ml of N-cinnamoyl-N-phenyl- hydroxylamine solution, shake the funnel vigorously, allow the layers to separate, and collect the chloroform layer in a 50-ml beaker containing about 1-5g of anhydrous sodium sulphate. Wash the aqueous layer twice with 5-ml portions of chloroform to remove any residual violet colour, and add the washings to the contents of the beaker. Decant the violet solution from the beaker to a 25-ml calibrated flask, wash the adhering colour from the sodium sulphate crystals with small portions of chloroform, combine the washings with the main solution, and dilute to the mark.Measure the optical density against chloroform at 640 mp in matched 1-cm silica or Corex cells with a Unicam SP500 spectrophotometer. Calculate the amount of vanadium corre- sponding to the optical density by reference to a calibration graph. The 0.1 per cent. w/v solution of N-cinnamoyl-N-phenylhydroxylamine has practically no absorption at 540mp, so that the use of chloroform in the reference cell is satisfactory. All glassware must be free from ethanol. &LOUR REACTION- A solution of N-cinnamoyl-N-phenylhydroxylamine in chloroform reacts with quinquivalent vanadium in solutions 2 to 10 N in hydrochloric acid and produces violet-coloured extracts.The absorption spectrum of such an extract has a broad band at 530 to 550 mp, the sides of the band being symmetrical. The chloroform used must be free from ethanol, otherwise the absorption spectrum of the complex is affe~ted.~ Many other organic solvents, e.g. , carbon tetrachloride, benzene, ethyl acetate and diethyl ether, could be used to extract the violet complex from the aqueous phase, but none of these was suitable for quantitative work, owing to the unfavourable distribution ratios for both reagent and complex. There is no reaction between N-cinnamoyl-N-phenylhydroxylamine and quadrivalent vanadium. ACIDITY- be between 2.7 and 7.5 N ; most of our measurements were made at an acidity of about 4 N. DISCUSSION OF THE METHOD For maximum colour development the concentration of acid in the aqueous phase should Only546 NOTES [Vol.86 hydrochloric acid was suitable for adjusting the acidity, but the presence of other acids could be tolerated, provided that their concentration in the aqueous phase was less than 1 N. STABILITY OF COLOUR- for a few days if stored in a cool dark place. decreased by about 4 per cent, Colour is extracted into the chloroform layer in about 2 minutes, and the extracts are stable The optical densities of extracts stored for 1 week CONCENTRATION OF REAGENT- The optimum concentration of the reagent solution is 0.1 per cent. w/v, and maximum development of colour takes place when the molar ratio of vanadium to reagent is 1 to 10. In practice, approximately 70 mg of reagent were used for each 1 mg of vanadium, but larger excesses of reagent can be tolerated.ADHERENCE TO BEER’S LAW- 10 p.p.m. is approximately 1.5 to 5 p.p.m. EFFECTS OF OTHER IONS- The effects of various ions on the determination of vanadium with N-cinnamoyl-N-phenyl- hydroxylamine were almost the same as those found when N-benzo yl-N-phenylhydroxylamine was used.3 Results obtained in the presence of ions commonly encountered in determinations of vanadium are shown in Table I ; the ions were added as solutions prepared from analytical-reagent grade salts by West’s proced~re.~ There was interference from the ions Ti4+ and MOB+, and the permissible limit of the former ion in the aqueous phase was 20 p.p.m. for each 1 p.p.m. of vanadium. Much higher concentrations of Mos+ can be tolerated if several successive extractions are made with the reagent solution.The ions listed below did not interfere when the weight ratio of each to vanadium was 260 to 1- AP+, Ba2+, Ca 2+ , Cd2+, Ce4+, Co2+, C9+, Cu2+, Fe3+, Hg2+, Mg2+, Mn2+, Ni2+, Sr2+, Th4+ UO 2+, W042-, Zn2+, 2++, acetate, borate, citrate, nitrate, perchlorate, phthalate, phosphate, sulphate and tartrate. Higher weight ratios were studied for a few ions and were tolerable, The order in which the reactants are mixed is not critical. The coloured system obeys Beer’s law at 540 mp for concentrations of vanadium from 0-5 to The optimum range for determining vanadium, based on Sandell’s recommendations,a TABLE I EFFECTS OF VARIOUS IONS The concentration of vanadium present in each test was 0.093 mg per 25 ml Ion present None .. .. .. AP+ added as’il(NO,), . - Co2+ added as Co(NO,), . . Cr3+ added as Cr(NO,), . . Cu2+ added as Cu(NO,), . . Fe3+ added as Fe(NO,), . . [Mo0,l2- added as (NH,),MoO, WO,2- added as Na2W04 Zr4+ added as Zr(N03), Ti4+ added as TiOCl, .. . . . . Amount of ion added, mg - .. .. 25 .. 25 ::} 25 .. d . 15 .. 5 .. 20 .. 25 Optical density at 640 mp 0.460 _+ 0.005 0.470 0.465 0.460 0.400 0.455 0.460 Amount of vanadium recovered, mg 0.093 0.095 0.094 0.093 0.087 0.091 0.081 0.092 0.093 * Normal extraction procedure. t Two further extractions with 5-ml portions of reagent solution. PRECISION- The average optical density found for twelve replicate samples, each containing 0.093 mg of vanadium in a final volume of 25 ml, was 0.460; the mean deviation of these results was 0.005. The method is rapid, gives reproducible results and is not affected by wide variations in factors such as temperature, ionic strength and volume of the aqueous phase. The sensitivity of the reaction, calculated on the basis defined by Sandell,6 is 0.008 pg of vanadium per sq. cm at 640 m p .August, 19611 APPARATUS 547 We thank Principal Umadas Mukherjee and Dr. Sameer Bose for the generous provision of laboratory facilities. We also thank Dr. S. C. Bhattacharyya, Assistant Director, National Chemical Laboratory, Poona, for helpful discussion and interest in the work. REFERENCES 1. Shome, S. C., Analyst, 1950, 75, 27. 2. Priyadarshini, U., and Tandon, S. G., Chem. & Id., 1960, 931. 3. - - , Anal. Chem., 1961, 33, 435. 4. Rya;, D. E., AnaZyst, 1960, 85, 569. 5. Tandon, S. G., and Bhattacharyya, S. C., Anal. Chim. Acta, in the press. 6. Sandell, E. B. , “Colorimetric Determination of Traces of Metals,” Third Edition, Interscience 7. West, P. W., J . Chem. Educ., 1941, 18, 528. Publishers Inc., New York, 1959, pp. 83, 97 and 925. DEPARTMENT OF CHEMISTRY MAHAKOSHAL MAHAVIDYALAYA JABALPUR UNIVERSITY JABALPUR, INDIA MISS U. PRIYADARSHINI S. G. TANDON Received January 30tk, 1961

ISSN:0003-2654    

DOI:10.1039/AN9618600539

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

August, 19611 APPARATUS 547 Apparatus APPARATUS FOR THE PREPARATION OF STANDARD GAS MIXTURES PROBABLY the greatest drawback to the gas-chromatographic analysis of mixtures of permanent gases is the need for calibrating the chromatograph. Calibration can be carried out very simply and rapidly if standard mixtures of gases contained under pressure in cylinders are available; however, such mixtures of certified composition are expensive and not readily available in most laboratories. We have found that an extremely simple apparatus incorporating calibrated glass syringes and a rubber football bladder can be used to prepare mixtures containing known amounts of permanent gases rapidly and with sufficient accuracy for routine work. DESCRIPTION AND USE O F APPARATUS The apparatus is shown diagrammatically in Fig.1. The 10- and 30-ml syringes are normal hypodermic syringes calibrated in 1- and 5-ml divisions, respectively. The 100- and 250-ml syringes are readily constructed from glass and are made gas-tight by the use of O-ring seals at Fig. 1. (100 ml)- Diagram of apparatus for preparing standard mixtures of gases548 APPARATUS [Vol. 86 the ends of the plungers ; they are calibrated in 25- and 50-ml divisions, respectively, Connections are by means of glass capillary tubing. To prepare a gas mixture of known composition, the bladder is first evacuated via taps B and C; a suitable vacuum pump can be used or the plunger of the 250-ml syringe can be withdrawn several times, with manipulation of taps B and C. The latter method normally gives adequate evacuation of the bladder.The connecting lines are then flushed out via taps A and C (with tap B closed) with the pure gas to be introduced, and the volume of gas required is drawn into the appropriate syringe. Taps A and C are then closed, tap B is opened, and the desired volume of gas is transferred to the bladder. The procedure is repeated for as many components as are required. Very small amounts of a component may be injected directly into the bladder through the self-sealing rubber cap with a hypodermic syringe. The gases are finally mixed by manipulation of the bladder, The total volume of gas introduced should be such that significant pressure is not produced within the bladder. Samples for injection into the chromatograph can be withdrawn through the serum cap or, if sampling valves of the types described by Harvey and Chalkleyl and Timms, Konrath and Chirnside2 are used, the bladder can be detached and affixed to the sample inlet of the valve.It should be mentioned, naturally, that the material of the bladder must be completely inert, chemically and physically, to the gases used. TABLE I RESULTS FOUND FOR VARIOUS MIXTURES Concentration Component range, yo v/v Equation* 0 to 1.75 H,, % = 0.108h - 0.006 * ' ' ' { 6 t o 40 H,, % = @172h - 0.32 Hydrogen , . Carbon dioxide . . .. 15 to 36 Cog, % = 0.276h + 0.34 Carbon monoxide ,. 5 to 15 CO, yo =; 0.168h - 0.01 Methane . . .. * . 1 to 6 CH,, yo = 0.161h - 0.18 Oxygen . . .. .. Oto20 0 2 , 3 0.967h - 0.05 Nitrogen . . * . .. 0 to 1.5 N,, yo = 0,042h - 0.001 Standard No.of error? mixtures o-03~0 of H, 8 0.3% of CO, 0.3y0 of CO 0.12% of CH, 0.2% of H, 7 o.3y0 Of 0, 8 0.03% of N, 11 * I n these equations, h is the height, in millimetres, of the appropriate peak. 7 Calculated from the expressiond' (y - y')z, in which y is the concentration present (% v/v), n - 1 y' is the concentration calculated from the equation (% v/v) and n is the number of mixtures used. RESULTS The precision attained in preparing gas mixtures of known composition for calibration purposes is shown by the results in Table I; 1 ml of each mixture was injected into a Fisher Model 25 Gas Partitioner, the peak height for each component was measured, and the calibration equation for each component was then calculated. The standard errors shown provide a good indication of the precision with which the mixtures were prepared, as the instrument gave reproducible results for several samples of the same mixture, and the response of the instrument was known to be linear over the small ranges of concentration used.REFERENCES 1. 2. Harvey, D., and Chalkley, D. E., Fuel, 1955, 34, 191. Timms, D. G., Konrath, H. J., and Chirnside, R. C., Analyst, 1958, 83, 600. AFRICAN EXPLOSIVES AND CHEMICAL INDUSTRIES LTD. RESEARCH DEPARTMENT, P.O. NORTHRAND TRANSVAAL, SOUTH AFRICA J. LACY K. G. WOOLMINGTON Received March 6th, 1961 A SIMPLE AND INEXPENSIVE FRACTION COLLECTOR FOR CHROMATOGRAPHY THE fraction collector described was built from readily available materials and components at a cost of approximately f15 and was assembled in a working time of 18 hours.Up to 100 fractions can be collected, and simple alternative devices for controlling collection by timed intervals or by volume are incorporated.August, 19611 APPARATUS 549 Construction was carried out in three stages, the first of which was the assembly of the motor and power pack. The motor used was a 12-volt, 1-amp ratchet motor obtained from 2 and I Aero Services Ltd., 14 South Wharf Road, Paddington, London, W.2, which required only minor modifications. The spindle for the ratchet wheel was removed and cut off A transverse cut was made across the cut end of the threaded piece to take a screwdriver. The upper guide hole for the spindle was drilled out to & inch, and a No. 3 BA bolt 2 inches long was used as a spindle to carry the turn-table for con- nection to the ratchet wheel.The hollow stem of the ratchet-wheel assembly was threaded to take the No. 3 BA bolt. The ratchet wheel was then replaced, the shortened spindle being used as lower bearing, and the motor was bolted to a heavy wooden base. The power pack was assembled from a 12-volt battery-charger transformer and a full-wave rectifier giving an output of 12 volts 1 amp. The ratchet-wheel assembly was already fitted with four “make-and-break” contacts operated by arms set a t 90’ intervals (see Fig. l), and these were used to stop the fraction collector after 25, 50, 75 or 100 fractions had been collected. These were carried out as follows. inch from the threaded end. 6 Make-and-- break switches ‘m actuated \ / \ by solenoid W Pawl Fig.1. Diagram of ratchet-wheel assembly showing “make-and-break” contacts The second stage was the construction of the turn-table. The diameter chosen for the tubes was Q inch, so that 3-inch x Q-inch or 5-inch x Q-inch tubes could be used. The centre-to-centre distance allowed for rimmed tubes Q inch in diameter was 16 mm, which gave 25.5 cm as the radius from spindle to centre of tube; tubes of larger diameter would require a correspondingly larger turn-table. The turn-table (total radius 28 cm) consisted of an upper disc of 18 s.w.g. aluminium bolted to a *-inch thick disc of hardboard for rigidity. After the positions of the tube centres had been marked on the aluminium disc the holes were drilled; the turn-table was then fixed on the 2-inch No.3 BA bolt, which was screwed into the stem of the ratchet-wheel assembly. A friction brake acting on the rim of the turn-table was used to damp the movement. The time control was based on a synchronous electric-clock movement. A “wiping” terminal was attached to, but insulated by plastic sleeving from, the second hand and had a radius of 4 cm; the minute and hour hands were not used and were not attached to the movement. The current was carried by means of an overhead strip of aluminium to the centre where continuous contact was made with the “wiping” terminal. The “wiping” terminal was made to contact with copper rivets, set at 15, 30, 45 and 60 seconds, as it rotated. The rivets were inch in diameter and were sloped on the approach side and sharply cut off on the reverse side.Three Perspex shields were made to cover the rivets so that, by selective insulation, intervals of 16, 30 or 60 seconds could be obtained. The volume-control device (see Fig. 2) was a balanced siphon, the movement of which was made to operate a simple switch consisting of a platinum wire, attached to the balance arm, making contact with a pool of mercury in a tube. The movement was restricted to + inch at the extremity. The balawe arm, made of aluminium rod, was used as one terminal and the pool Finally, the control devices were made.550 1 1 I t Clock-motor 112 volts switch I I amp I I f APPARATUS [Vol. 86 Q, \ 1 T xx I Siph I -1 1 I 1 I I solenoid Fig. 2. Diagram of volume- and time-control-devices of mercury as the other. Contact of the switch was made when the siphon was empty, and the balance was set to break contact when 0.5 g of eluate had been collected.This setting does not cause overheating of the motor solenoid coil during operation. The mercury switch and siphon assembly are removed when the time controller is in use. In the last position for collection is fitted a bottomless tube, so that excess of eluate can be collected in a beaker placed under the turn- t able. PATHOLOGY DEPARTMENT WEST WALES GENERAL HOSPITAL CARMARTHEN A. CLARK Received March 16th, 1961 A SIMPLE LOCK FOR USE ON A HIGH-VACUUM APPARATUS THE determination of gases in metals by the vacuum fusion technique requires the introduction of the material to be analysed into a high-vacuum apparatus. Samples can be either loaded into an apparatus before outgassing or introduced during a run via a vacuum lock or, in certain special instances, by means of a mercury lift. The first of these approaches has been widely used, and samples can be loaded into a suitably designed manifold or “tree” and released in turn by raising retaining plungers magnetically.“Trees” of this type have been described by Booth, Bryant and Parker1 and Swann and Williams2; Gregory, Mapper and Woodward3 have given details of a manifold containing hinged trip-buckets operated by solenoids. However, this method of introduction lacks flexibility, and it is often desirable when analysing material having anAugust, 19611 APPARATUS 551 Medium kn u rli ng 2 Grooves to take O-rings Fig. 1 . Construction of lock (all dimensions are in inches) unknown content of gas to be able to adjust the weight taken after an initial analysis has been carried out.Further, if a “tree” is used, the samples will be stored under vacuum for several hours while the crucible is being outgassed, and for certain types of material this may be un- desirable. A suitable vacuum lock can overcome these limitations, as, when such a device is used, samples can be rapidly introduced into the apparatus during an analysis. Although a small vacuum lock has been described recently by Still,4 it was thought that there was a need for a lock of simple design, which could be easily fabricated. DESIGN OF LOCK The lock shown in Figs. 1 and 2 is made of brass or stainless steel and consists of a hollow cylinder, grooved to take two pairs of O-rings (OS12; inside diameter Q inch, outside diameter Fig.2. Assembled lock inch), through which slides a smoothly polished shaft. The O-rings, which are lubricated with Apiezon L grease, form vacuum seals between the shaft and the cylinder. When the shaft is pushed into the cylinder, the air trapped in the small well containing the sample is pumped rapidly away through the apparatus, and, after a satisfactory working pressure has been attained, rotation of the shaft through 180” allows the sample to drop through the taper joint and thence via a funnel into the crucible.552 BOOK REVIEWS [Vol. 86 PERFORMANCE Although a small amount of gas at atmospheric pressure is introduced into the system when the shaft is slid into the dropping position, it has been found that the apparatus “pumps down’’ after 4 to 5 minutes, attaining the original blank rate. Several of these locks have been made and have proved to be extremely satisfactory during operation. Their use permits the design of a vacuum fusion apparatus to be simplified, as both the metal forming the bath and the samples can be introduced simply; further, the “dead” volume of the furnace assembly is considerably decreased. This may be advantageous when a high rate of transfer of evolved gases from the crucible is important, as in the determination of oxygen in beryllium. I thank Mr. W. R. Johnson, Chemical Inspectorate, War Office, for his valuable help in the construction of this lock. REFERENCES 1. 2. 3. 4. Booth, E., Bryant, F. J., and Parker, A., Analyst, 1957, 82, 52. Swann, D. A., and Williams, D. A., Ibid., 1958, 83, 113. Gregory, J. N., Mapper, D., and Woodward, J. A., Ibid., 1953, 78, 414. Still, J. E., Special Report No. 68, The Iron and Steel Institute, London, 1960. U.K. ATOMIC ENERGY AUTHORITY RESEARCH GROUP WOOLWICH OUTSTATION, WOOLWICH, S.E.18 A. PARKER Received April 13th, 1961

ISSN:0003-2654    

DOI:10.1039/AN9618600547

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

552 BOOK REVIEWS [Vol. 86 Book Reviews THE COMPOSITION OF FOODS. By R. A. MCCANCE, C.B.E., M.D., Ph.D., F.R.C.P., F.R.S., and E. M. WIDDOWSON, D.Sc. Medical Research Council Special Report Series No. 297 (Third revised edition of Special Report No. 235). Pp. viii + 252. London: Her Majesty’s Stationery Office. 1960. Price 30s. Having, scientifically speaking, grown up with a book as with “McCance and Widdowson” and confronted with a new edition, one has got about an equal chance of being blind to its faults and unappreciative of its virtues. Which of the two looms the larger probably depends on the mood of the reviewer at the “moment of truth,” which is doubtless the moment when he first puts pen to paper or finger to key. So I have perhaps been wise to hesitate for some weeks before venturing to express in print any opinion on this, the third revised edition of the Medical Research Council’s Special Report No.235, now renumbered as shown above. The pause has, I hope, given time for enthusiasm and wonder to be replaced by critical evaluation and a sense of historical perspective. But it is indeed difficult, if not impossible, to recapture the mood in which we welcomed the first paper- bound edition in the grim year 1940. The AnaZyst itself (1940, 65, 458) records the birth of the Special Report in a purely formal (unsigned) summary, which opens with the reminder that the Report “supplements, but does not supersede, the three Reports . . . previously issued by the Council.” A little research- shows that these three early Reports, the scaffolding amid which the present noble edifice was built, appeared as long ago as 1929, 1933 and 1936.It is more difficult to think back to those still earlier pioneer days when our knowledge about the detailed nutrient analysis of foods was about on a par with our knowledge about nutrient requirements and both were so elementary as to be an embarrassment to all concerned. It is more due to the work of Professor McCance, Dr. Widdowson and their succession of colleagues at Cambridge than to that of any other worker or group of workers, living or dead, that to-day our knowledge of the former, even when only to be given as a mean or modal value within a pretty wide range, has outstripped our information about the latter, which we can still express as no more than a figure, and an average one a t that, for the mythical “normal” man, woman or child. Still, it’s good to find even an islet of terraJirma to sustain the increasing pressure of the questing nutritionist.To describe in detail how the 252 pages of the new (cloth-bound) Report No. 297 differ from the 156 pages of the previous (1946 and cloth-bound) Report No. 235, and they in turn fromAugust, 19611 BOOK REVIEWS 653 the earlier (1942 and paper-bound) third impression of the original Report No. 235, would surely be a work of supererogation. All those who have to-day a dog’s-eared copy-and all extant copies must assuredly be in that condition-of any earlier edition will inevitably order and im- mediately use the new one. And many new arrivals at the camps of food technology, nutrition, dietetics and curative and preventive medicine will meanwhile have joined their ranks.All will require the book, so that none of them will require words of commendation from any reviewer. Even though its price be five times that of the 1946 and 7& times that of the 1942 edition, and whatever number of it may have been printed, which I am unable to decipher from the Stationery Office’s coding, I’ll lay all the equipment of a modern kitchen to a single antiscorbutic orange that supplies will soon be exhausted and another impression called for. Scientists had better get in before bibliophiles reduce stocks in the hope of having unearthed another collector’s piece. Anyhow, out of sympathy and astonished admiration for the years of perspicacious, painstaking and immaculate work that the authors have had to put into this steadily growing mass of essential data, I trust that they will not be called on to supply another wholly new edition for some years to come.Meanwhile, it should suffice to say that Part I of the new Report adheres to the admirable pattern and conventions adopted before, but that it now includes two further Parts-a short one on the amino acid make-up of the proteins in all important animal and vegetable foods containing significant quantities and a longer one on the vitamin contents of all our main foods. The former covers 18 amino acids, including the eight-or perhaps it is ten-“essential” ones: the latter gives figures, when appropriate, for carotene and the fat-soluble vitamins A and D and for all the main water-soluble vitamins of the “B group,’’ including vitamin B,,, which they do not call by its official name of cobalamin, and ascorbic acid.And I am glad to add, though not in the least surprised to find, that they steadfastly refuse to call riboflavin “vitamin B,,” which it isn’t, or nicotinic acid “niacin,” which it never should have been. Perhaps a slightly querulous question may here be addressed to the Publication Section of the M.R.C.-or should it perhaps be to H.M. Stationery Office? When can the last two letters of the word “gramme” be finally, formally and officially dropped in scientific British literature ? Even before we all go over to the metric system, this could save an appreciable amount of space. As heretofore, all that the authors have to say in their General Introduction, and now what they add in the special Introductions to the three Parts of the book, along with their numerous obiter dicta, help to make this Report something far more than might have been expected from its original purpose as a catalogue of analytical results presented as nutrient “data.” It is indeed a book without peer or parallel, especially for those working in relevant fields in the United Kingdom and the Commonwealth, but also elsewhere. Few, if any, of them can do without it for a day unless they are prepared to limit their efforts by a self-inflicted handicap.A. L. BACHARACH HANDBUCH FUR DAS EISENH~TTENLABORATORIUM. Band I. DIE UNTERSUCHUNG DER NICHT- METALLISCHEN STOFFE. Edited by the Chemikerausschuss des Vereins Deutscher Eisen- huttenleute.Second Edition. Pp. xviii + 322. Dusseldorf : Verlag Stahleisen M.B.H. 1960. Price DM43. The Chemist’s Committee of the Vereins Deutscher Eisenhuttenleute, which corresponds roughly to the Methods of Analysis Committee of B.I.S.R.A., but seems to have somewhat wider terms of reference, has published a four-volume work intended to cover the entire analytical needs of an integrated iron and steel works. This is the second edition of volume I and deals with the analysis of non-metallic materials. After an introduction explaining the purpose and scope of the work, there follow sections on the analysis of ores, phosphates, slag-forming materials, flue-dust and blast furnace potassium cyanide, slags, cements, refractories, solid fuels, gases, tar, pitch and benzol, fuel oil and lubricants, insulating oil and water.There is no section on sampling in the second edition, as this is now dealt with in volume 111, “Probenahme.” Although classical methods form the basis of the work, use is made, when appropriate, of instrumental methods, e.g., photometry, polarography, and potentiometry, but, as explained in the introduction, in order to save space no directions are given on the selection and use of particular instruments. Colorimetric methods indeed are pushed to the limit ; the molybdenum-blue method is taken to over 20 per cent. of silica, and the 1,lO-phenanthroline method to 60 per cent. of iron, at which level its error is stated to be f0.20 per cent. Alternative methods are given for most determinations, and the time for an analysis, as well as the probable error, is stated.The latter554 BOOK REVIEWS [Vol. 86 is taken as being approximately twice the standard deviation. Here, the German Committee are somewhat ahead of B.I.S.R.A., who have only just begun to think of reproducibility in statistical terms. In the section on ores, the methods given are general ones, and it may be for this reason that, e g . , the clumsy mercurous nitrate method is retained for determining tungsten. The use of titanous chloride to hasten the reduckion of tungsten in its colorimetric determination (British Standard 1121 : Part 32 : 1954) seems not yet to have spread to the Continent, In view of the evident belief of the Committee in potentiometric methods, recommended for chromium, vanadium, antimony and arsenic, it is rather surprising to find no mention of such a method for manganese. Among the procedures given for refractories are a number taking as long as 60 hours.It is here that one feels that more use could have been made of rapid procedures, such as those pub- lished by our own Ceramic Research Association. Neither here nor elsewhere is any mention made of spectrographic methods, although the volume “Probenahme” deals with sampling for spectrographic analysis. In the section on fuels, oils and greases, where methods tend to be more empirical, extensive reference is made to the methods of the German Institute of Standards (DIN). In all, over thirty existing standard methods are referred to in this work (which is printed according to DIN B5).In recommending a book of this kind to chemists in British iron and steel works, one has to consider how strong and relatively local are traditions in this field. Nevertheless, although the methods selected would not necessarily be the first choice in a British laboratory, they are un- doubtedly well-tried and reliable, and I do not know of any work in English that covers the same field so thoroughly. The method of presentation is uniformly clear and the production and printing are of a high standard. The list of contents a t the beginning is so full that anyone using the book would have little need of the index, especially as there is a neat system of cross-references by means of hieroglyphs. Like earlier volumes of the “Handbuch,” this one is reasonably priced.G. M. HOLMES DIE FLAMMENSPEKTRALANALYSE : GRUNDLAGEN UND VERFAHREN VON FLAMMENPHOTOMETRIE Pp. xii + 258. Price (paper) DM 65.50; (cloth boards) DM 69. The author of volume 48 in the series “Die chemische Analyse” was one of the earliest users of direct-reading flame methods and is therefore well qualified to discuss flame-emission spectro- scopy, a technique that is still developing and rapidly extending its field. This is a rather ele- mentary book and deals with the application of flame methods to practical analytical problems; it is directed to the student rather than to the research worker. There is a tendency for equipment of non-German origin to be discussed only when no corresponding German apparatus is commer- cially available.Thus, all the filter-type flame photometers listed are of German manufacture, whereas almost all the prism or grating instruments mentioned are of non-German origin. The reader is therefore left with a rather unbalanced picture of the equipment currently available. Professor Schuhknecht restricts the term “flame photometry” to filter instruments and uses “flame spectrophotometry” for prism or grating equipment, a distinction with which not all will agree and which seems out of line with common usage. I t is claimed, too, that equipment of the latter type is less convenient to use, again an opinion that will not be universally accepted. Section I, discussing the flame as a light source, deals briefly with practical aspects of burner design, gas supply , flame excitation, sample introduction, interference effects and flame-background factors.Section 11, on flame photometry and flame spectrophotometry, gives a simple discussion of filters and photocells and details a few commercially available instruments. Atomic-absorption methods are treated in less than two pages, and this section concludes with a consideration of the performance of filter and dispersion instruments from the point of view of interference and other factors of practical importance. Section I11 deals in 75 pages with laboratory instructions for the application of flame photometry to various types of material. These instructions are much more detailed than is justified in a book of this nature, including, for instance, the preparation of straightforward standard solutions for each of the nineteen methods described for determining alkalis and alkaline earths in chemicals, minerals and biological and agricultural materials.Section IV deals very briefly with flame spectrography. A bibliography of 496 items, with titles, is much less useful than it might have been, because it is set out in a manner convenient only to the printer. References are arranged by year, alpha- betically by author’s name within each year. They are printed running on and referred to in UND FLAMMENSPEKTROGRAPHIE. Stuttgart : Ferdinand Enke Verlag. By Prof. Dr. WOLFGANG SCHUHKNECHT. 1960. The book is divided into four sections.August, 19611 BOOK REVIEWS 555 the text by a serial number. The accuracy of the references has not been fully checked, but the page number was omitted from the first one I consulted.This book, within its limited field, will serve a useful purpose, but there have in the past few years been several more comprehensive and more fundamental monographs on flame photo- metry. R. L. MITCHELL CONTRIBUTI TEORICI E SPERIMENTALI DI POLAROGRAFIA. Volume V. Supplemento a “La Padova, Italy: Centro di Studio per la Polarografia del This supplementary volume of La Ricerca Scientifica, which is similar in format to volume IV (see Analyst, 1960, 85, 159), contains the proceedings of two symposia organised jointly by the Italian Polarographic Research Centre and the University of Padua and held a t Bressanone in August, 1959. The 12 papers (all in Italian) contributed to the first symposium on the correlation of chemical constitution with physical properties are largely theoretical in nature, but cover a wide range of physical techniques, including dipole moments and ultra-violet spectroscopy.Those (6 in English, 2 in Italian and 1 each in French and German) submitted to the second symposium, at which relationships between polarographic constants and molecular structure were discussed, are by well known polarographers, such as Elving, Randles, Semerano and Zuman, and will interest analysts concerned with the polarography of organic substances. Ricerca Scientifica.” C.N.R. 1960. Price 2500 Lire. Pp. 315. J. E. PAGE SCIENTIFIC RUSSIAN: A TEXTBOOK FOR CLASSES AND SELF-STUDY. By JAMES W. PERRY. Second Edition. Pp. xxviii + 565. New York and London: Interscience Publishers Inc. 1961.Price $9 50; 72s. Until comparatively recently, few people were aware of the enormous output of research by Russian scientists during the post-war years, and there was a general tendency to underrate their scientific achievements and progress. The increasing number of cover-to-cover translations of Russian scientific journals has done much to alter this situation, but the serious research worker will increasingly experience the need to read papers and books in the original. Indeed, an ability to translate Russian with reasonable facility is becoming as important as the ability to read German. For those, too, who believe that the study of a foreign language is in itself a valuable intellectual discipline, Russian has many features that make it a most attractive alternative to Latin or Greek.But, whatever may be the reason for a student commencing a serious study of Russian, he could find no sounder guide than Perry’s excellent book. Although great experience and care has gone into planning the order of presentation, the treatment is never superficial and the book makes no concessions to the dilettante; although it is broken up into a series of relatively short lessons, subdivided into paragraphs admirably adapted to short spells of spare-time study, industry and application are still essential. Starting with lists of foreign technical terms that have been taken over substantially unchanged into the Russian language, the student is quickly familiarised with the Russian alphabet.At this stage the association of each Russian word with its transliteration (following the system of letter equivalents standardised in Chemical Abstvacts) is most helpful. Throughout the book all words are shown with their accents, which enables the student to associate it with its sound-a powerful method of memorisation. Paradoxically, this becomes of great value when one is confronted with the typical Russian text-book or paper from which all such accents have been omitted by the printer! Every aspect of the Russian language-so far as it concerns the problems of translation from technical literature-is dealt with in the forty-odd lessons that follow, each of which contains reading and translation exercises. The book concludes with a good index and an extensive Russian - English vocabulary of over 5000 words. For those who learnt their Russian from the first (1950) edition of this book, the most striking improvement in the second edition will .appear in the excellent lay-out achieved by the English firm of printers and in the new typography, particularly in the Russian portion of the text. This has made it possible to decrease the number of pages from 8 16 to 565, despite the increased number of notes and cross-references, and yet there is an outstanding increase in legibility. Misprints from the first edition appear all to have been corrected, and the more topical reading exercises on Atomic Energy and Interplanetary Travel have been revised. Whether for class work or for independent study, this text is hard to beat, and the price is in no way excessive. H. IRVING

ISSN:0003-2654    

DOI:10.1039/AN9618600552

出版商:RSC    

年代:1961

数据来源: RSC

摘要:

556 PUBLICATIONS RECEIVED [Vol. 86 Publications Received DETERMINATION OF CHLORIDES. By T. A. STRIVENS, BSc. Pp. viii + 55. Cambridge, England : One of a series of Handbooks of Electroanalytical Methods obtainable f r o m the publisher at Prepared by the 1960 Committee on Analytical Reagents (W. STANLEY CLABAUGH, Chairman). Pp. xvi + 564. Washington, D.C. : Applied Publications, American Chemical Society. 1961. Price $10.00. NAME REACTIONS IN ORGANIC CHEMISTRY. By ALEXANDER. SURREY. Second Edition. Pp. x + 278. New York and London: Academic Press Inc. 1961. Price $8.00; 64s. REFERENCE ELECTRODES: THEORY AND PRACTICE. Edited by DAVID J. G. IVES and GEORGE J. JANZ. Pp. xii + 651. New York and London: Academic Press Inc. 1961. Price $20.00. THE RADIOCHEMISTRY OF SELENIUM. By G.W. LEDDICOTTE. Pp. vi + 38. Washington, D.C. : National Academy of Sciences-National Research Council. 1961. Price 50 cents. W. G. Pye & Co. Ltd. 1961. Price 10s. 6d. York Street, Cambridge, England. REAGENT CHEMICALS : AMERICAN CHEMICAL SOCIETY SPECIFICATIONS, 1960. Nuclear Science Series: NA S-NS-3030. THE RADIOCHEMISTRY OF THE TRANSCURIUM ELEMENTS. By G. H. HIGGINS. Pp. vi + 35. Washington, D.C. : National Academy of Sciences-National Research Council. 1960. Price 50 cents. Nuclear Science Series: NA S-NS-3031. AUSTRALIAN JOURNAL OF EXPERIMENTAL AGRICULTURE AND ANIMAL HUSBANDRY. Vol. I, No. 1, May, 1961. Edited by R. N. SANDIFORD. Pp. 55. East Melbourne, Victoria: Australian Institute of Agricultural Science. Annual Subscription (4 parts per year) +3 (Aust.). A new journal: Postal Address: 226 Clarendon Street, East Melbourne, C.2, Victoria.THE RADIOCHEMISTRY OF ALUMINIUM AND GALLIUM. By JOHN E. LEWIS. Pp. vi + 48. Wash- 1961. Price ington, D.C. : National Academy of Sciences-National Research Council. 50 cents. Nuclear Science Series: NA S-NS-3032. REPRINTS OF “COULOMETRIC METHODS IN ANALYSIS” BY D. T. LEWIS REPRINTS of the Review Paper, “Coulometric Methods in Analysis” by D. T. Lewis, published in this issue of The Analyst (pp. 494-506), will be available shortly from the Assistant Secre- tary, The Society for Analytical Chemistry, 14 Belgrave Square, London, S.W.l, at 5s. per copy, post free. A remittance for the correct amount, made out to The Society for Analytical Chemistry, MUST accompany the order ; these reprints are not obtainable through Trade Agents . Methods for the Analysis of Non-soapy Detergent (NSD) Products By G. F. Longman and J. Hilton Society for Analytical Chemistry Monograph No. 1 THE first of the Society’s Monographs is being published simultaneously with this issue of The Analyst. It is “Methods for the Analysis of Non-soapy Detergent (NSD) Products,” by G. F. Longman and J. Hilton, of Unilever Research Laboratory, Port Sunlight, Cheshire, and is an expanded version of the paper delivered by G. F. Longman to the North of England Section in Manchester on Saturday, March 12th, 1960. The Monograph is bound in stout paper covers, uniform in size with The Analyst, and contains in 30 pages a complete practical scheme for the determination of all components of most non-soapy detergents at present marketed.

ISSN:0003-2654    

DOI:10.1039/AN9618600556

出版商:RSC    

年代:1961

数据来源: RSC