摘要:

264 COLSON : THE RAPID MICRO DETERMINATION [Vol. 75 The Rapid Micro Determination of Nitrogen Organic Substances a Simple Modification in of the Dumas Method BY A. F. COLSON SyNoPsIs-The Pregl - micro-Dumas method for the determination of nitrogen in organic substances has been modified to reduce the analysis time by a t least one half. In the modified procedure, the combustion of the sample is carried out in a slow stream of carbon dioxide, and the nitrogen produced is swept out by a much faster stream of carbon dioxide than is used in the normal Pregl method. Complete reduction of oxides of nitrogen produced during the combustion is effected by a four-fold increase in the normal amount of hot metallic copper employed in the combustion tube. Analytical results are given to show that the rapid method is sufficiently accurate for routine purposes and is applicable t o a variety of organic compounds.ATTENTION is being directed to the development of rapid micro-methods for the determination of carbon, hydrogen, nitrogen, sulphur and halogens. For the first two of these elements a promising procedure has already been developed1 and has been in use for a period of about two years. Procedures for the more rapid micro-determination of nitrogen have been reported in the literature,2#3#4 but the time-saving claimed is not very considerable, and the apparatus required is in general more complicated or less convenient in operation than the Pregl apparatus. In our attempts to devise a simple rapid method it was found that, by increasing the length of the column of heated metallic copper used in the Pregls combustion tube, the speed of the stream of carbon dioxide employed to sweep the nitrogen into the nitrometer could be appreciably increased without affecting the completeness of reduction of any oxides of nitrogen produced during the combustion of the sample.On this basis the rapid micro-method described in this paper was developed. EXPERIMENTAL APPARATUS- The assembled combustion train is depicted in Fig. 1. It comprises the following components. 1. A carbon dioxide generator, containing marble chips and hydrochloric acid. 2. A small bulb-tube packed with glass wool for the retention of spray from the carbon dioxide generator. 3. A White - Wright flowmeter modified from the original form by the provision of (a) an aperture at C, to facilitate the introduction of the manometer fluid, and (b) a bulb, D, to retain the manometer liquid in the event of a suck-back. 4.A thick-walled capillary tube connected to the combustion tube through a good- quality rubber bung. 5. An automatic device for advancing the bunsen burner at about 2.5 cm. per minute. The simple arrangement at present in use is depicted in Fig. 1. It comprises the following components- KZaxon motor (K)-Type EKS UBI-M21; 220/240 volts, A.C./D.C., series wound; 10 r.p.m. Pulley wheels (L, M)-These are wooden pulleys 3 inches in diameter. Driving screw (N)-This is f inch in diameter, and has 12 threads per inch. Bunsen burner carriage (P)-This consists of two sleeves connected by a metal plate The sleeve P is threaded to engage with the screw The second sleeve makes a sliding fit on a guide tube (not shown in Fig.1) of the Gear ratios: lst, 40; Znd, 8. Serial number: C.35816. into which the burner, J, is screwed. N. same diameter as the driving screw N.264 COLSON : THE RAPID MICRO DETERMINATION [Vol. 75 The Rapid Micro Determination of Nitrogen Organic Substances a Simple Modification in of the Dumas Method BY A. F. COLSON SyNoPsIs-The Pregl - micro-Dumas method for the determination of nitrogen in organic substances has been modified to reduce the analysis time by a t least one half. In the modified procedure, the combustion of the sample is carried out in a slow stream of carbon dioxide, and the nitrogen produced is swept out by a much faster stream of carbon dioxide than is used in the normal Pregl method.Complete reduction of oxides of nitrogen produced during the combustion is effected by a four-fold increase in the normal amount of hot metallic copper employed in the combustion tube. Analytical results are given to show that the rapid method is sufficiently accurate for routine purposes and is applicable t o a variety of organic compounds. ATTENTION is being directed to the development of rapid micro-methods for the determination of carbon, hydrogen, nitrogen, sulphur and halogens. For the first two of these elements a promising procedure has already been developed1 and has been in use for a period of about two years. Procedures for the more rapid micro-determination of nitrogen have been reported in the literature,2#3#4 but the time-saving claimed is not very considerable, and the apparatus required is in general more complicated or less convenient in operation than the Pregl apparatus. In our attempts to devise a simple rapid method it was found that, by increasing the length of the column of heated metallic copper used in the Pregls combustion tube, the speed of the stream of carbon dioxide employed to sweep the nitrogen into the nitrometer could be appreciably increased without affecting the completeness of reduction of any oxides of nitrogen produced during the combustion of the sample.On this basis the rapid micro-method described in this paper was developed. EXPERIMENTAL APPARATUS- The assembled combustion train is depicted in Fig. 1. It comprises the following components.1. A carbon dioxide generator, containing marble chips and hydrochloric acid. 2. A small bulb-tube packed with glass wool for the retention of spray from the carbon dioxide generator. 3. A White - Wright flowmeter modified from the original form by the provision of (a) an aperture at C, to facilitate the introduction of the manometer fluid, and (b) a bulb, D, to retain the manometer liquid in the event of a suck-back. 4. A thick-walled capillary tube connected to the combustion tube through a good- quality rubber bung. 5. An automatic device for advancing the bunsen burner at about 2.5 cm. per minute. The simple arrangement at present in use is depicted in Fig. 1. It comprises the following components- KZaxon motor (K)-Type EKS UBI-M21; 220/240 volts, A.C./D.C., series wound; 10 r.p.m. Pulley wheels (L, M)-These are wooden pulleys 3 inches in diameter.Driving screw (N)-This is f inch in diameter, and has 12 threads per inch. Bunsen burner carriage (P)-This consists of two sleeves connected by a metal plate The sleeve P is threaded to engage with the screw The second sleeve makes a sliding fit on a guide tube (not shown in Fig. 1) of the Gear ratios: lst, 40; Znd, 8. Serial number: C.35816. into which the burner, J, is screwed. N. same diameter as the driving screw N.May, 19501 OF NITROGEN IN ORGANIC SUBSTANCES 265 6. A silica combustion tube 70.0 cm. long (including the 3.0 cm. x 3-0 mm. neck) and approximately 9.0 mm. in internal diameter. 7. A cylindrical electric furnace 35.0 cm. long and 7.5 cm.in external diameter. The inner tube (Pythagoras or silica) has an internal diameter of 15.0mm. and is wound with Nichrome wire. 8. A micro-nitrometer charged with concentrated potassium hydroxide solution. The improved form of potash reservoir, E, provided with a narrow side-arm, facilitates accurate reading of the gas volume. The reservoir is held, when necessary, in a specially designed, non-corrodible and easily detachable Perspex support, F. The stop-cock, G, on the nitro- meter inlet tube differs from that employed with the Pregl apparatus in that no grooves (for fine adjustment) are cut in the key. REAGENTS- Marble chi@s-Before use, the marble chips, about $ inch in diameter, should be immersed in water (or, preferably, in spent liquor from the carbon dioxide generator) in a desiccator, which should then be evacuated on the water pump until air bubbles no longer rise from the surface of the marble.About 1 hour is usually sufficient to remove air from the marble. HydrochZoric acid-Pure concentrated hydrochloric acid is diluted with an equal volume of water or, preferably, with spent liquor from the carbon dioxide generator, and boiled for a few minutes. A piece of marble is added and the solution cooled rapidly. Copper oxide, coarse grade-Wire-form copper oxide in pieces 3 to 4 mm. long, strongly ignited before use. Copper oxide, fine grade-Prepared by grinding the coarse-grade oxide and sieving to obtain the fraction that passes B.S. sieve No. 40, and is retained on B.S. sieve No. 80. Metallic copper-Copper gauze, B.S.S.40 mesh, copper turnings, or reduced wire-form copper oxide. Potassium hydroxide-Aqueous solution, sp.gr. 1-42. This solution should be filtered before use. SETTING UP THE APPARATUS- Charge the carbon dioxide reservoir with marble and hydrochloric acid and connect the acid reservoir, H, to a second generator, not shown in Fig. 1. Free the generator from air by flushing it out with carbon dioxide at intervals over a period of several hours. Pack the spray-trap, 2, with clean glass wool and connect it to the carbon dioxide generator and to the flowmeter, 3, using thick-walled rubber tubing to effect glass-to-glass connection. The rubber tubing used for this purpose must be impregnated with wax before use. This may be done by placing the tubing in molten paraffin wax contained in a flask heated on a boiling water bath, evacuating the flask with the water pump and, when bubbles of air cease to rise from the rubber, admitting air to the flask.This process should be repeated until air ceases to rise from the rubber tubing. The tubing should then be drained while warm and wiped inside and outside with cotton wool. Pack the clean, dry combustion tube in the following manner- Insert a 2 to 3-mm. plug of fine silver or platinum wire at the constricted end of the tube, and follow this with a 3 to 4-mm. plug of ignited asbestos. Introduce coarse copper oxide over a length of 10 to 11 cm., metallic copper for 16.0 cm., and then sufficient coarse copper oxide to reach the inlet end of the furnace when the combustion tube is inserted so that its exit end projects about 7.0cm.beyond the other end of the furnace. Complete this “permanent” filling by introducing a 1.O-cm. roll of oxidised copper gauze. Add 8.0 cm. of coarse copper oxide, followed by 4.0cm. of fine copper oxide and about 6.0cm. of the coarse oxide. This constitutes the “temporary” filling of the combustion tube. Connect the flowmeter, 3, to the thick-walled capillary tube, 4, with impregnated rubber tubing, to form a glass-to-glass connection, and join this capillary tube to the combustion tube through a good-quality rubber bung. Fill the nitrometer, 8, with potassium hydroxide solution. To prevent bubbles of gas sticking to the mercury surface, a little mercurous oxide may be added to the mercury. Connect the nitrometer to the exit end of the combustion tube with impregnated rubber tubing.266 COLSON : THE RAPID MICRO DETERMIXATION [Vol.75 METHOD CONDITIONING THE COMBUSTION TUBE- Remove the inlet tap from the nitrometer, and pass carbon dioxide through the combustion tube at about 5-0 ml. per minute, by adjustment of tap A and screw clip B. With the furnace adjusted to 725” to 750” C., heat the “temporary” filling of the com- bustion tube with the bunsen burner, starting at the mouth end of the tube. Repeat this operation in the reverse direction, and then re-heat in the same manner. Replace the inlet tap, reduce the rate of flow of the carbon dioxide to about 2-Oml. per minute, and observe the rising gas bubbles in the nitrometer. If the bubbles rapidly assume the acceptable micro dimensions, the determination of the blank value, and the checking of the apparatus by control analyses may be commenced.DETERMINATION OF BLANK VALUE- Sweep all air out of the combustion tube with carbon dioxide. Introduce a sqall volume (0-3 to 0-4 ml.) of air into the nitrometer and measure its volume. Record the temperature and barometric height. Set the bunsen burner at the maximum distance from the furnace entry. Light the burner and pass carbon dioxide through the combustion tube and into the nitrometer at 2.0 ml. per minute.. Immediately after starting the carbon dioxide stream, set the burner in motion and heat the combustion tube up to the furnace entry. Reverse the movement of the burner, and when the original starting-point is reached, stop the stream of carbon dioxide, switch off the motor driving the bunsen, detach the nitrometer, and after 10 minutes measure the volume of gas collected.Correct this volume to N.T.P. , and deduct the volume of air originally introduced. CONTROL ANALYSES- Control analyses are carried out exactly a!; described below for the analysis of routine samples. Pure dinitrobenzoic acid, m-dinitrobenzene , or benzyl thiuronium chloride6 are suitable compounds for use in these control analyses. The apparatus may be regarded as being in satisfactory condition for routine work if not less than three consecutive control tests give results agreeing with the calculated value to Hithin 0.1 per cent. PROCEDURE FOR DETERMINATION OF NITROGEN- Remove the “temporary” filling from the combustion tube and separate the coarse and fine fractions by sifting.Ignite each fraction in a nickel basin over a Meker burner for a few minutes. Introduce about 8.0 cm. of the coarse copper oxide into the combustion tube and follow this with a little of the fine oxide. To the weighed sample (solid) in a mixing tube, add about one-half of the remaining fine copper oxide, mix thoroughly and transfer to the combustion tube. Rinse out the mixing tube two or three times with the remaining fine oxide, and then add the remaining coarse oxide. Solids may be weighed in a nickel trough, which is then introduced into the combustion tube after addition of about one-half of the fine oxide to the tube. The remaining fine oxide is then added, and mixed with the sample by rotating the combustion tube. Place the combustion tube in the furnace (at 725” to 750” C.) so that the exit end of the tube lies just inside the furnace entry.Coanect the mouth of the tube to the capillary tube (4, Fig. 1) and pass carbon dioxide for a’bout 2 minutes at 500ml. per minute. Slide the furnace along until the combustion tube projects about 7-0 cm. beyond the furnace exit. After about 2 minutes, connect the nitrometer to the neck of the combustion tube, and reduce the carbon dioxide flow-rate to about 2.0ml. per minute. Continue the passage of carbon dioxide until micro-bubbles are obtained. Adjust the speed of the carbon dioxide stream so that gas bubbles enter the nitrometer at a rate of about 2 per second (corresponding to about 1.5 ml.per minute for the nitrometer used in this work). Remove all air from the nitrometer, and set the lighted burner, J, in motion, starting from the extreme right-hand position. Calculate the volume of air at N.T.P. The blank value obtained in this .way should not be greater than 0.01 ml. Liquid samples are manipulated as in the usual Pregl method.May, 19501 OF NITROGEN IN ORGANIC SUBSTANCES 267 When gas bubbles begin to enter the nitrometer at the rate of about 4 per second (this occurs when the burner reaches the sample), stop the movement of the burner, and lower the flame slightly until the bubble rate falls to the initial value. Repeat this adjustment, if necessary, until the burner has reached the furnace entry. Reverse the motion of the burner, and when the diameter of the bubbles entering the nitrometer corresponds to about one-third of a scale division, increase the speed of the carbon dioxide stream to 5.0 ml.per minute. When micro-bubbles are obtained (usually before the burner reaches the end of its traverse) , close tap G, disconnect the nitrometer and, after 10 minutes, measure the volume of nitrogen, and record the temperature and pressure. Calculate the percentage of nitrogen in the usual manner, applying the relevant correction^.^ ANALYTICAL RESULTS The rapid method described in this paper has been applied to the analysis of numerous organic compounds, both solid and liquid, containing nitrogen in the form of nitro-groups, amino-groups, etc. The accuracy attainable is indicated by the values presented in Table I.For the purpose of comparison, a number of results obtained by the usual Pregl method are also given. These latter analyses refer chiefly to compounds of doubtful purity. TABLE I DETERMINATION OF NITROGEN BY THE Compound C H [O] N compounds: Azobenzene . . .. .. .. .. .. .. Mono-anilide of dimethyldihydroresorcin . . .. S-bis-dimethyl-dihydro-resorcyl-m-phenylenediamine Mono-meta base of dimedone . . . . .. Amino azobenzene.. .. .. ,. .. Methyl-urea . . .. .. . . .. Guanazoguanazole .. .. . . .. o-Nitrophenol . . .. .. .. .. m-Nitrotoluene . . .. .. .. .. Nitrobenzene . . .. .. .. .. a-Nitroisobutene . . .. .. .. .. 2-Nitropropane . . .. .. .. .. $-Nitraniline . . .. . . .. .. Di-picrate of 1 : 6-diamino-6-methyl heptane . . Picrate of 4-dimethyl-arnino-butyl-cyanide .. Nitromethane . . .. . . .. .. C H [O] N Br, C. H [O] N S, and C H [O] N S C1 6-Bromoacetanilide . . . . .. .. compounds : C-p-methylthiophenoxy-acetamide . . .. Benzyl thiuronium chloride . . . - .. .. N-p-chlorophen yl-"-dimethyl guar? ylthiourea . . Di(benzy1thiuronium) salt of 6-nitro-6-methyl-undecan dioic acid .. .. . . .. ,. . . . . .. .. .. .. .. .. .. .. .. .. .. .. .. . . . . RAPID MICRO-METHOD Nitrogen found, "/b Nitrogen 5&T-T&z calculated, method method % 15-37 9 16.4 16-5 16.6 6-56 6.7 6-5 8.0 8.1 - 12.20 12.4 12.3 21.31 22.0 22.2 22.2 37-81 37.9 - 67-4 67.6 - 10.07 10.3 10-4 10.21 10.5 10.4 11-37 11.5 - 13.85 14.5 14.7 14.6 15-72 16.6 - 18-60 18.8 - 19.71 19.8 19.9 19.9 20.28 20.4 - 22-95 22-9 - 23.1 - 6.54 6.7 7-72 7.7 7-5 11.52 11.3 11.4 13.83 13.7 - 21.82 21.9 -.DISCUSSION THE BURNING OF THE'SAMPLE AND SWEEPING OUT OF THE NITROGEN- in the analysis of a wide variety of compounds. following simplified procedure- The directions given for carrying out these operations are based on the experience gained It should, however, be noted that satisfactory results can often be obtained by the (1) Charge the combustion tube with copper oxide and sample as already described. (2) Sweep out all air from the apparatus and adjust the carbon dioxide stream to the maximum speed of 5.0ml. per minute.268 COLSON: MICRO DETERMINATION OF NITROGEN [Vol. 75 (3) Light the burner and switch on the Klaxon motor. (4) When the burner reaches the furnace entry, re-heat the combustion tube in the reverse direction.(5) Stop the carbon dioxide stream, detach the nitrometer and finish the determination as already described. For the analysis of a series of compounds of similar constitution, the simplified procedure Some results obtained by this simpler method may sometimes be used with advantage. are given in Table 11. TABLE I1 DETERMINATION OF NITROGEN BY THE RAPID MLCRO-METHOD (SIMPLIFIED PROCEDURE) Compound C H [O] N compounds: Phenacetin . . .. .. .. .. .. .. a-Naphthylamine . . * . .. . a .. .. Acetanilide . . .. .. .. . a .. .. 2 : 4-dinitrobenzoic acid . . .. .. .. .. m-Dinitrobenzene . . .. .. .. .. .. C H [O] N C1, C H [O] N S, and C H [O] N S (3 compounds: Chlorodinitrobenzene . . .. .. .. .. .. Sulphanilic acid .. .. .. . I, .. .. 4-Chloro-3-nitrobenzene sulphonamide ..... .. Thiourea * . .. .. .. . I, .. .. Nitrogen calculated, % 7.81 9.78 10.36 13.20 16.66 13.83 8-08 36-83 11.84 Nitrogen found, Yo 7.7 7.9 8.0 9.8 10.3 10.4 13.2 13.3 16.6 16.5 16.4 13.6 13.8 7.9 37.1 11.9 12.1 THE TIME REQUIRED TO COMPLETE A SINGLE DETERMINATION- The combustion of the sample and sweeping out of the nitrogen can be completed in 15 minutes, or 10 minutes by the simplified procedure, compared with the 40 to 50 minutes required by the usual Pregl method. The complete analysis, including weighing out of the sample, and measuring the gas volume, occupies not more than 35 minutes. The corresponding time by the Pregl method is not less than 1 hour. THE USE OF A FLOWMETER FOR MEASURING THE CARBON DIOXIDE RATE- This flowmeter was introduced in the early stages of the investigation of suitable flow- rates. It could probably be omitted if carbon dioxide of high purity is always employed, since with a little experience the required flow-rate can be estimated, with sufficient accuracy, by observation of the rate of entry of gas bubbles into the nitrometer. REFERENCES 1. 2. 3. 4. 5. 6. 7. Colson, A. F., Analyst, 1948, 73, 541. Rutgers, J. J., Corn#& rend., 1931, 193, 51. Beezley, C. W., Ind. Eng. Chem., Anal. Ed., 1.938, 10, 605. Gysel, H., Helv. Chim. Ada, 1939, 22, 1088. Pregl, F., “Quantitative Organic Micro-analysis,” 4th English Edition, J. & A. Churchill, Ltd., Ogg, C. L., and Willits, C. O., Ind. Eng. Cheni., Anal. Ed., 1946, 18, 334. Pregl, F., “Quantitative Organic Micro-analysis,” 4th English Edition, J. & A. Churchill, Ltd., London, 1945, p. 63. London, 1945, p. 76. IMPERIAL CHEMICAL INDUSTRIES RESEARCH DEPT. ALKALI DIVISION NORTHWICH First submitted, May, 1949 Amended, March, 1950

ISSN:0003-2654    

DOI:10.1039/AN9507500264

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

May, 19501 KENYON, OVENSTON AND PARKER 269 A Modification of the Cobaltinitrite Method for the Absorptiometric Determination of Potassium in the Microgram Range BY C. KENYON, T. C. J. OVENSTON AND C. A. PARKER SYNoPsIs-The novel features of this modification are (a) the inclusion of a “seasoning” process for the sodium cobaltinitrite solution in aqueous alcohol, consisting in overnight cooling a t -5’ C., ( b ) the precipitation of the potassium complex a t a controlled temperature of - 5’ C. and (G) the quantita- tive evaluation of the washed precipitate by the absorptiometric determination of the nitrite by means of Parker’s cc-naphthylamine - alcoholic hydrochloric acid method. A recommended procedure is given in detail, and typical results covering the range 1 to 100 p g .of potassium are tabulated. SINCE the introduction of the cobaltinitrite method for the determination of potassium by Adie and Wood,l there has been much controversy concerning its suitability as a quantitative method owing to the variation in composition of the precipitate produced under different reaction conditions. According to the amount of sodium present in solution, so it has been reported that compositions varying from KNa,[Co(NO,),].nH,O to K3[Co(NO,),]p2H,O have been obtained, the latter only in absence of sodium. The composition of this precipitate was investigated by Vurtheim,, who claimed that a precipitate of constant sodium/potassium ratio could be produced by the use of an excess of sodium reagent, and later by B ~ n n e a u , ~ who showed that a constant composition represented by K,Na[Co(NO,),] .nH,O was obtained when the sodium/potassium ratio of the solution was about 25/1, but that when it fell below 22/1, or when the temperature of the solution was raised, the potassium content of the precipitate was increased.Thus it would appear that a controlled low temperature of precipitation in the presence of a fairly large and constant excess of sodium is desirable, and this has been confirmed by subsequent workers. A review of the large number of variations in methods of precipitation, washing and final estimation has been given by T i n ~ l e y . ~ Absorptiometric finishes have depended upon the determination of the cobalt or of the nitrite content of the precipitate after washing away all excess of reagent.Probably the most sensitive method for determination of cobalt is that using nitroso-R-salt ; this method has been applied to the determination of potassium by Siderk5Y6 lihen the nitrite is determined, the choice has usually been the Griess r n e t h ~ d . ~ However, the sensitivity of the potassium determination is likely to be limited, not by that of the colour reaction employed, but by the appreciable solubility of the cobaltinitrite precipitate ; so that, provided the colour reaction is reasonably sensitive, the principal require- ment is that it should be precise in its quantitative application. I t is considered that the nitroso-K-salt method for cobalt is rather better than the Griess method for nitrite in this respect. In the present investigation the aim was to develop a procedure which provided as high a sensitivity as possible with, at the same time, a fair degree of precision.METHOD OF PRECIPITATION- The precipitation of the double cobaltinitrite was made more complete by Taylor8 by employing a partially alcoholic medium. A further improvement was made by Lohseg who precipitated the salt at a low temperature. These methods were investigated by Robinson and PutnamlO who concluded that, at best, 3 pg. of potassium per ml. was recover- able, and proposed the use of the more insoluble silver cobaltinitrite complex. Sideris6 claimed recoveries one-tenth of this by the normal method when the precipitate was allowed to form in a refrigerator; he did not specify a temperature. For the present work a method of precipitation similar to that employed by Sideris was adopted, but the temperature was controlled at -5” C.This low temperature is made possible by the presence of alcohol in the solution.270 KENYON, OVENSTON AND PARKER : MODIFICATION OF COBALTINITRITE [VOl. 75 I t appears to be usual to add the alcohol to the aqueous sodium cobaltinitrite reagent immediately before use, as the alcoholic solution. is unstable, It has been found, however, that this solution may be stored satisfactorily overnight at -5” C. In fact, in the final procedure here developed, it is essential to do this each time before use. This process serves to “season” the reagent, which was always found to form an appreciable precipitate. The reagent was then separated from this precipitate before use.In carrying out a determination, 10 ml. of this “seasoned” reagent was added to 0.5 ml. of an aqueous solution containing the potassium from the sample. At the same time a blank test was made on 0.5ml. of pure water. This blank always yielded a precipitate, .which gave a figure of about 0-6 pg. in terms of potassium with remarkable regularity. This figure was always deducted from the total potassium found in the sample solution. Before this procedure was used, the “unseasoned” reagent was employed in the normal way, but the blanks were then found to be high and erratic. METHOD FOR FINAL ESTIMATION- The a-naphthylamine method for the absor.ptiometric determination of nitrite recently described by one of usll is not quite so sensitive as the Griess method, but the reaction provides a very stable colour, and it is considered to be more precise for the estimation of the cobalti- nitrite precipitate than either the Griess method or the nitroso-R-salt - cobalt method.Measurements of the specific extinction coefficients of the final solutions at the appropriate wavelengths show that the a-naphthylamine method for nitrite is almost twice as sensitive as the nitroso-R-salt method for cobalt. In addition, the cobaltinitrite ion contains six times as much nitrite as cobalt. For these reasons the a-naphthylamine method was chosen for the present work. RECOMMENDED METHOD REAGENTS- Sodium cobaltinitrite reagent-Dissolve 12.5 g. of pure sodium cobaltinitrite in 100 ml. of water, add 100 ml. of 95 per cent. ethyl alcohol and allow to stand overnight in a refrigerator at -5” C.Settle the precipitate by means of a centrifuge and filter the clear solution through a filter stick in an ammonia-free atmosphere. cc-.NaphthyZamine reagent-Dissolve 1.25 g. of x-naphthylamine (recrystallised from boiling light petroleum, b.p. 40” to 60” C., to give colourless crystals) in 1 litre of ethyl alcohol (95 per cent. alcohol redistilled from caustic soda) and add 10 ml. of 10 N hydrochloric acid. (This reagent is stable in the absence of air and light.) Washing liquid-Make up a 9 : 1 acetone -I water mixture. Standard sodium nitrite solution (for calibration graph)-Dissolve 1.339 g. of pure sodium Dilute 1Oml. of this solution to 1 litre with water. Use immediately. nitrite in water and dilute to 1 litre.Ten ml. of the dilute solution contain 133.9 pg. of NaNO, = 25 pg. of K. PREPARATION OF APPARATUS- Clean all beakers and flasks with chromic acid and follow by repeated washings with distilled water. Stand filter sticks overnight in 5 per cent. hydrochloric acid and finally wash by suction with distilled water. Note-Porcelain micro-filter sticks of low pclrosity were preferred for the present work. However, these appear to be in very short supply, and it was found that satisfactory results could also be obtained by using filters made frorn Celite 535 bedded on glass wool. Gotton wool, paper pulp and asbestos fibre were found to be unsuitable as filtering media. Dry the beakers at 105” C. before use. PROCEDURE- From a known weight of the material under examination, containing not more than about 1OOpg.of potassium, prepare a neutral aqueous solution free from all metals other than the alkali group and from ammonium salts. Evaporate to dryness in a silica beaker of about 20-ml. capacity and dissolve the residue in 0.5 ml. of water. Add 10ml. of filtered sodium cobaltinitrite reagent and enclose in a suitable air-tight vessel, such as an empty desiccator, and place i n a refrigerator at -5” C. for 2 hours. Note-All operations involving the use of sodium cobaltinitrite reagent must be carried out in an atmosphere free from traces of ammonia.May, 19501 METHOD FOR ABSORPTIOMETRIC DETERMINATION OF POTASSIUM 27 1 Attach a porcelain micro- filter stick to a filter flask connected to a water pump and remove the excess of reagent from the beaker through this filter.Disconnect the filter stick from the suction apparatus, leaving it in the beaker, and wash with the washing fluid (contained in a wash bottle), paying particular attention to the beaker wall. Remove the washings by suction and repeat the whole washing operation a further four times, disconnecting the filter stick from the suction apparatus for each wash. Note-A volume of 5ml. of washing fluid is recommended for each wash, a maximum total of 25 ml. being desirable. After the second wash, the filtrate should be quite colourless, and further washes then serve to remove any traces of reagent adhering to the beaker wall or to the filter stick. Once this operation has been completed there is no longer any danger of interference by ammonia.The addition of 10 ml. of sodium cobaltinitrite reagent ensures the presence of a large excess of sodium (about 100,OOOpg.) during precipitation. The presence of comparatively large proportions of sodium in the sample (e.g., 1000 pg,) will not have very much effect on the magnitude of this excess, which can be regarded as approximately constant for practical purposes. Add 4 ml. of hot water to the beaker containing the filter stick and precipitate, and stand the beaker and contents on a boiling water bath for 10 minutes. Transfer the solution through the filter stick into a 20-ml. calibrated flask contained in a wide-necked bottle fitted with (a) a glass capillary tube leading from the filter to the flask, and ( b ) a second tube, with two-way tap, leading to the suction pump.Wash the filter stick and beaker repeatedly with small amounts of hot water, sucking the beaker dry between each washing, until the flask is almost full. Remove the flask, cool to 20" C., make up to the mark with water, and inix. Transfer 5 ml. of this solution by means of a pipette into a 25-ml. calibrated flask con- taining 5 ml. of water. Add 10 ml. of the or-naphthylamine reagent by pipette, mix, and set aside for 10 minutes in the dark. Place in a water bath at GO" C. for 30 minutes, protecting the flask from strong light, and then cool to 20" C. Make up to the mark with water, mix, and measure the extinction of the solution in a 0.5-cm. cell by means of a Spekker absorptio- meter (or similar instrument), using Ilford No.605 (yellow-green) filters and Calorex filters, and either a tungsten filament or mercury arc source. Carry out a blank test on 0.5 ml. of pure water by using the same procedure, and deduct the extinction due to the blank from the extinction due to the sample. Obtain the potassium content of the final solution by reference to a calibration graph prepared by taking amounts of standard sodium nitrite solution covering a range equivalent to 2.5 to 25.0 pg. of potassium, and measuring the extinctions of the solutions obtained after reaction with or-naphthylamine reagent as described above. The potassium content of the sample is then obtained by multiplying this amount by 4. INTERFERING IONS Ammonium is the most likely ion to cause interference and it is therefore necessary to take special precautions to prevent access of traces of ammonia to the vicinity of the bench where the determination is being made.In the present work, the operations which could be performed only outside the protecting vessel, such as washing the precipitate, were carried out in a separate room set aside for the purpose. The seriousness of this interference by the ammonium ion is illustrated by the fact that during the coursc of a series of experiments in which the blanks were regularly showing a value of about 0.6 pg. of potassium per 0-5 ml., some brasswork in the "ammonia-free" room was cleaned with metal polish. The next blank to be measured gave 3.6 pg. of potassium per 0-5 ml., and this was followed the next day by one of 2 4 pg. of potassium per 0.5 ml.The small amount of ammonia present in the polish contaminated the air of the room sufficiently to interfere with the determination. Rubidium and caesium also interfere by producing similar precipitates, but normally these metals would not be encountered in appreciable quantity. Sodium, lithium, magnesium and calcium can be tolerated in large excess. In Table I are given the results obtained on thirty-one tests using known quantities of potassium (in the form of the chloride) ranging from 1 to 1OOpg. Further results were obtained in the presence of milligram additions of sodium, magnesium and calcium, and these are shown in Table 11. Remove the vessel and contents to an ammonia-free room. RESULTS272 KENYON, OVENSTON AND PARKER : MODIFICATION OF COBALTINITRITE [vol.75 TABLE I RESULTS WITH POTASSIUM CHLORIDE ALONE Potassium taken, Pg. 106.6 104.8 100.6 100.2 99.9 85.6 82.3 81-8 72.0 63.1 60.5 43.8 42.0 40.2 40.1 21.4 20.2 19.9 19.8 19.7 10.1 10.0 10.0 5.1 5.0 5-0 5.0 4.0 3.0 2.2 1.3 Potassium found, Pg- 103.6 104.8 102.0 100.0 99.2 854 80.0 80.4 74.4 63-22 63.6 41-21 42.8 40.0 36-14 22.0 214; 204 21.t; 204 11.2 10.0 8.8 5.2 4.8 4.8 3.6 4.4 3.8 2.4 l e t 5 TABLE I1 Error, Fg. - 2.9 nil + 1.4 - 0.2 - 0.7 + 0.1 - 2.3 - 1.4 + 2.4 + 0.1 + 3-1 - 2.6 + 0.8 - 0.2 - 3.3 + 0.6 + 1.4 + 0.9 + 1.8 + 0.7 + 1.1 nil - 1.2 + 0.1 - 0.2 - 0.2 - 1.4 + 0-4 + 0.6 + 0.2 + 0.3 RESULTS WITH POTASSIUM CHLORIDE IN PRESENCE OF OTHER IONS Added ion Potassium taken, Potassium found, Error, PE:. Pg- Pg. 1 mg. of sodium I . .. .. 58-8 58.0 - 0.8 10.0 10-8 + 0.8 2.3 3-2 + 0.9 2.2 2-8 + 0.6 1 mg.of magnesium* (a) . . .. 59.2 68.8 - 0.4 .. 59.2 58.4 - 0.8 - 0.4 (b) * - .. 10.0 9.6 .. 9.9 10.0 + 0.1 (b) * - (4 * * (4 * * .. 2.1 1.9 - 0.2 (4 * - .. 2.0 1.3 - 0.7 1 mg. of calcium . . .. .. 66.6 66.4 - 0.2 10.0 9-0 - 1.0 2.0 2.4 + 0.4 1.9 2-4 + 0.6 * Magnesium (a), (b) and (c) refer to different samples of magnesium salts. All results have been corrected for the blank obtained at the same time, and the results given in Table I1 have also been corrected for the potassium contents of the added salts which were determined by the same method. The potassium contents were as follows- Sodium salt contained 2.4 p g . of potassium Magnesium salt (u) I9 63.6 99 99 )I 1.2 YS 93 99 0.8 Calcium salt n 2.4 39 ” (b) ” (4 n It is remarkable that the high potassium content of the magnesium salt (a) did not affect the recovery of the added potassium, as shown.by the relevant figures in Table 11.May, 19501 METHOD FOR ABSORPTIOMETRIC DETERMINATION OF POTASSIUM 273 Taking all the results from both tables, it is found that for potassium contents above 40 pg., the mean error is 1.2 pg. with a negative bias of 0.4 pg., while for potassium contents below 40 pg., the mean error is 0.7 pg. with a positive bias of 0.2 pg. The change in bias may possibly be an indication of a very slight change in composition of the precipitate over the range. The ratio of sodium to potassium varies from approximately 1000/1 for 100 pg. of potassium to 100,000/1 for 1 pg. of potassium. REFERENCES 1. 2. 3. 4. 5. 6. - , Ibid., 1942, 14, 821. 7. 8. 9. 10. 11. Adie, R. H., and Wood, T. B., J . Chem. Soc., 1900, 77, 1076. Vurtheim, A., Rec. trav. chim. Yays-Bas, 1921, 40, 593. Bonneau, L., Bull, SOC. chim., 1929 (4), 45, 798. Tinsley, J.. Analyst, 1948, 73, 86. Sideris, C. P., I n d . Eng. Chem., Anal. Ed., 1937, 9, 145. Briggs, A. P., J . Biol. Chem., 1923, 57, 351. Taylor, F. H. L., Ibid., 1930, 87, 27. Lohse, H. W., I n d . Eng. Chem., Anal. Ed., 1935, 7, 272. Robinson, R. J . , and Putnam, G. L., Ibid., 1936, 8, 211. Parker, C. A., Analyst, 1949, 74, 112. ROYAL NAVAL SCIENTIFIC SERVICE QUEEN ANNE’S MANSIONS ST. JAMES’ PARK, LONDON, S.W.l Novenz ber, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500269

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

May, 19501 METHOD FOR ABSORPTIOMETRIC DETERMINATION OF POTASSIUM 273 The Influence of Glucose Orp Diphen ylamine Indicators and on Dichromate - Iron Titrations A New Method for the Assay of Official Saccharated Iron Carbonate Preparations BY EDMUND BISHOP AND A. B. CRAWFORD SYNOPSIS-In attempting the accurate analysis, by the dichromate method, of samples of saccharated iron carbonate intended for use by students in lieu of spathic iron ores, it was found that the indicator reactions were strongly disturbed. An accurate investigation of the dichromate - iron reaction in the presence of various amounts of glucose under diverse conditions confirmed the observations of other workers that the results are always high. A study of the indicator reactions led to an explanation of these errors and effects in terms of induced oxidative reactions involving glucose.The replacement of dichromate by an oxidant with a lower normal potential and fewer intermediate oxidation states was expected to eliminate these errors and effects. Vanadate was found to give accurate results, free from interference. The method has been tested and a recommended procedure is described. IN attempting the accurate analysis of some saccharated iron carbonate samples by the dichromate method normally applied to iron ores, the indicator behaved in a peculiar fashion ; it showed colours associated with its destruction and the mercuric chloride effect, and gave wandering and fading end-points. This was at first ascribed to slow consumption of ferric iron or dichromate by the sugars in the solution (glucose and maltose) , but closer examination of the indicator reactions showed typical irreversible oxidation phenomena of very high speed.KO real end-points could be obtained in slow titrations with diphenylamine, diphenyl- benzidine or N-phenylanthranilic acid. The barium salt of diphenylamine sulphonic acid and, to a lesser extent, the sodium salt and the free acid, gave transient end-points of duration up to 115th of a second, and all indicators were rapidly and irreversibly oxidised. These phenomena suggested that some inductive agency was at work, and this idea was confirmed by the fact that neither dichromate alone nor ferric iron alone showed any reaction with glucose under the titration conditions, nor was the indicator destroyed.Further, ferrous - ferric mixtures (samples of ferri carbonas saccharatus) in acid solution in the presence of starch hydrolysates, stored in glass-stoppered volumetric flasks, showed no alteration in the ratio of ferrous to ferric iron over a period of thirty days.274 BISHOP AND CRAWFORD: THE INFLUENCE OF GLUCOSE [Vol. 75 A careful investigation showed further that the results with dichromate were consistently high. As this merely confirms the results of other workers1,2,3 and Heisig’s iodate method4 has been officially substituted6 for the dichromate method,6 this work is not given in detail. The amount of iron has little influence on the error, which is dependent upon, but not strictly proportional to, the amount of glucose present. This makes empirical correction impossible.Moreover, the end-point is dependent on the speed of titration. A method, which need not be described, was developed whereby the error was minimised and the reproducibility improved. Results by this method are included in Table I1 under the heading “dichromate.” These effects must be due to induced reactions, with the probable inclusion of such poly- valent ions as the well-established FeO++’ v 8 and quadri- or quinque-valent chrorni~m.~ $),11~12 It is possible that the rapid, smooth reduction of CrV1 to CrlIE is disturbed’or slowed down by the glucose acting as an inhibitor or negative catalyst, which would favour the transient formation of the CrV or CrlV states in such a fashion that their life-times, normally vanishingly small, are increased to such a value as to enable them to partake in the reaction (but compare SkrabaP on the relation between speed of reaction and stability of reaction products).The simplest explanation would involve two sets of induced reactions- (a) Crvl, as actor, reacting with glucose at; acceptor (a very slow reaction) induced by the fast reaction between Feu as inductor and Crvl; (b) Cry’ as actor reacting with the indicator as acceptor (slow), again with ferrous iron as inductor. Unless double induction, i.e., induction of a reaction (b) by another reaction (a) which is itself induced be admissible, this explanation may be rejected on the grounds that reaction ( b ) does not normally take place, and must be subject to a different inductive or catalytic effect, which in this instance must be connected with glucose.There are many possible interpretations of the data in which the oxidant may be In general terms, taking into consideration the probabilities, and without choosing a (a) chromiumlV,V Or v1 or ironn1 Or IV as actors, glucose as acceptor and iron” as inductor (oxidation of glucose) ; (b) chromiumm* VorV1 or ironZV as actors, the indicator as acceptor and glucose as inductor (destruction of indicator). These are probably preceded by a glucose-stabilised high valency state of chromium or iron, possibly CrV or FeIv, since the simplest oxidation of glucose will involve a two-electron change, whereby C P or Feu are formed. The regeneration of Fe’I would contribute to the fugitive end-points, which might also be favoured by the gradual dissociation, near the end- point, of an ironn - glucose complex.J , or atmospheric oxygen in ~ o l u t i o n , * , ~ ~ , ~ ~ all of which may interchange or CrVI ,V or I V FenI or IV interact in induced reactions. specific actor, we have- THE DETERMINATION OF FERROUS IRON IN THE PRESENCE OF GLUCOSE It appeared that these effects and the error might be overcome by the use of an oxidant of lower normal oxidation potential and fewer reduction stages than dichromate, and vanadate was selected. Unpublished investigations by one of us (E. B.) have shown that the direct titration of ferrous iron by vanadate, using barium diphenylamine sulphonate (the best indicator for this titration) in the presence of phosphoric acid, although accurate, is rendered difficult by the low normal potential of vanadate and the correspondingly slow indicator response.The reverse titration, however, is excellent, rapid and extremely accurate, and there is no indicator error. The best method would therefore be to add a known excess of standard vanadate to an aliquot of the ferrous solution and titrate back with a standard ferrous solution. Vanadate solzction-Approximately 0.1 M sodium vanadate was prepared by weighing ammonium metavanadate, dissolving in water containing a slight excess of caustic soda and diluting to volume. This solution can be standardised against dichromate through a ferrous solution, or by weighed amounts of pure Mohr’s salt or spectroscopically pure iron. AnalaR ammonium vanadate can be weighed directly for routine work.This salt, recrystallised and dried, is suitable for a primary standard, o r a carefully analysed batch can be used by weighing the appropriate amount.May, 19501 OX DIPHENYLAMINE INDICATORS 275 Accuracy-Aliquots of an approximately 0.1 M ferrous solution in 5 N sulphuric acid, standardised against purified potassium dichromate, were acidified with 50 ml. of 4 N sulphuric acid, To this were added 6 ml. of syrupy phosphoric acid, an excess (by pipette) of standardised vanadate and 3 drops of 0.2 per cent. aqueous barium diphenylamine sulphonate solution and the mixture titrated with the ferrous solution, finishing by split drops. The results are the means of five determinations; the greatest deviation from the mean being 0.01 ml. For comparison, the same residual titration technique was applied with dichromate instead of vanadate.Calibrated weights and glassware were used throughout. The results in Table I show that glucose has no effect on the accuracy of the vanadate method and the indicator gave perfect end- points. TABLE I This was repeated with the addition of glucose. With dichromate the usual high results were obtained. Vanadate Dichromate Vol. of Fe++ taken, ml. . . .. . . Vol. of oxidant added, ml. . . . . Fe++ equivalent, ml. . . .. .. Back titration, Fe++, inl. no glucose . . 0.05 g. glucose 0.10 g. glucosa 1-00 g. glucose 5.00 g. glucose .. 20.020 .. 25.076 .. 24-96 .. 4-94 .. 4.94 . . 4.94 . . 4.94 . . 4.94 20.020 25.076 25.32 5.30' 4.50* 4*18* 1.50" 1-98? * Indicator, barium diphenylamine sulphonatc.Indicator, diphenylamine. THE ANALYSIS OF SACCHARATED IRON CARBONATE AND ULAUD'S PILLS Careful analyses oi several samples have been made by the dichromate and vanadate The ferrous iron content o f samples methods and two examples are given in Table 11. kept in ordinary screw capped bottles fell by less than 0.01 per cent. in twelve months. METHOUS Treatment of sample-A quantity sufficient to make an approximately 0.1 solution of total iron, accurately weighed by difference, was transferred to a stoppered flask fitted with a delivery tube dipping into a saturated solution of AnalaK sodium bicarbonate. The sample was treated with water and AnalaR hydrochloric acid, the mixture warmed to complete the reaction and the solution cooled under the bicarbonate trap to exclude air.The solution was then transferred to a volumetric flask and diluted to volume. The experi- ment was replicated. Dichromate method-The modified direct titration method with dichromate,'j referred to above, was used. RECOMMENDED VANADATE METHOD- acid and 6ml. of syrupy phosphoric acid. Total iron was determined by stannous chloride reduction. Ferrous iron-Aliquots of the prepared solution were treated with 50 ml. of 4 A* sulphuric An excess (by pipette) of standard vanadate TABLE I1 COMPARISON OF THE DICHROMATE AND VANADATE METHODS The results from six titrations by both methods on two samples Ferrous iron, highest lowest . . mean . . as FeO as FeCO, Total iron, highest . . lowest . . mean .. as Fe,O, . . Ferric iron (by diff .) Sample A n-e, % o / /O .. . . 22.03 21-09 .. .. 22.01 41 *06 .. . . '32.03 21.07 . . . . 28.35 27.04 .. . . 45.70 43.70 .. . . 38-79 37-76 . . . . 38.78 37.74 .. . . 38-71) 37.76 . . . . 16.76 16.69 . . . . 23.91 m a 1 Sample E Dichromate, Vanadate, - % % 21.58 20.58 21.58 20..58 21.58 20.58 27-76 26.47 44.77 42-68 38.75 37-58 38.69 37.50 38.72 37.5 1 17.14 16-94 -74.44 24.16276 HOLNESS AND TREWICK QUALITATIVE SEPARATION OF [Vol. 75 and 3 drops of indicator solution were added and. the purple colour allowed to develop com- pletely (if a full, rich purple colour does not develop, more vanadate is added). The solution was then titrated back with 0.1 M ferrous solution standardised by running a blank on the vanadate alone, finishing with divided drops to a clear green colour. Total iron-Aliquots, treated with an equal volume of AnalaR hydrochloric acid, were reduced at the boiling-point by dropwise addition of 1 per cent.AnalaR stannous chloride solution, using 2 drops in excess. The solution was cooled rapidly, 10ml. of 5 per cent. AnalaR mercuric chloride and 50 ml. of water added and allowed to stand for 10 minutes. An excess of vanadate, 6ml. of syrupy phosphoric acid and 3 drops of indicator solution were added and the solution titrated back as before. The end-points were excellent. For samples containing about 50 per cent. of ferrous carbonate, aliquots of 50ml. for ferrous and 25ml. for total iron give reasonable titrations, corresponding to about 14g. of sample per litre, or individual weighed amounts of 0.8 and 0.4 g.respectively. Note-Since this work was completed, a reference has been found tda paper by G. G. Rao and J. V. S. Ramanjaneyulu (Current Sci. (India), 1949, 18, 72) in which positive errors in dichromate - iron titrations in the presence of ,alcohol are described. Correct results were obtained by the use of vanadate, but no details of the method are given. ‘ REFERENCES 1. Morton, C., and Harrod, D. C., Quart. J. Pharwz., 1936, 9, 480. 2. Lyons, C. G., and Appleyard, F. N., Ibid., 1936, 9, 462. 3. Hartley, F., and Linnell, W. H., Ibid., 1934, 7, 549. 4. Heisig, J., J . Amer. Chem. SOC., 1928, 50, 168;‘. 5. “The British Pharmacopoeia,” 1948, pp. 204 and 417. 6. Ibid., 1932, pp. 185 and 339. 7. Bray, W. C., and Gorin, M. H., J . Amer. Chein. SOC., 1932, 54, 2124. 8. Manchot, W., and Schmid, H., Ber., 1932, 65, 98. 9. Wagner, C., 2. anorg. Chem., 1928, 168, 279. 10. Stefanovskii, V. F., and Zenko, A. M., Acta Physicochim. U.R.S.S., 1938, 9, 635. 11. Lang, R., Mikrochim. Acta, 1938, 3, 113. 12. Westheimer, F. H., and Novick, A., J . Chem. Physics, 1943, 11, 506. 13. Skrabal, A., 2. Elektrochem., 1905, 11, 653. 14. Kolthoff, I. M., Z. anal. Chem., 1924, 64, 185. 15. Bray, W. C., and Ramsey, J. B., J . Amer. Chsm. Sac., 1933, 55, 2279. UNIVERSITY OF DURHAM KING’S COLLEGE, NEWCASTLE-UPON-TYNE December, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500273

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

276 HOLNESS AND TREWICK QUALITATIVE SEPARATION OF [Vol. 75 Qualitative Separation of the Copper and Arsenic: Groups BY H. HOLNESS AND $1. F. G. TREWICK SYNoPsIs-The behaviour of solutions of the polysulphides of ammonium, lithium, sodium and potassium has been examined from the standpoint of their use as reagents for the separation of sulphides of metals of the copper group from those of the arsenic group, and all have been found to have disadvantages. Aqueous solutions of the hydroxides of sodium, lithium, barium and calcium have also been examined for the same purpose. Lithium hydroxide in 1 per cent. solution with 5 per cent. of potassium nitrate was found to separate cleanly all mixtures of sulphides of the metals of the hydrogen sulphide group when used in 25-ml. portions and heated just to boiling.On the scores of efficiency, pleasantness in use, cost and specificity, this reagent is recommended for general laboratory use in place of the more usual yellow ammonium sulphide. A REVIEW of the methods in common use for the separation of the copper group from the arsenic group in qualitative analysis shows that the most generally used reagent is yellow ammonium sulphide. Alternatives to this reagent are solutions of caustic soda, solutions of sodium sulphide and mixtures of caustic soda with yellow ammonium sulphide, but in no instance can it be said that a clean separation results from the use of any of these reagents. The metallic sulphides that cause trouble with one or other of the above reagents areMay, 19501 THE COPPER AND ARSENIC GROUPS 277 those of copper, mercury and stannous tin, while cadmium sulphide is frequently rendered colloidal by the treatment.With yellow ammonium sulphide, copper sulphide dissolves to some extent, and to a greater extent if the copper was originally present in association with members of the arsenic group, e.g., as copper arsenate. Furthermore, the subsequent treatment of this reagent with acid causes the liberation of considerable quantities of hydrogen sulphide together with a large amount of sulphur. While the former is both poisonous and objectionable, the latter can easily obscure small amounts of the sulphides of arsenic and tin. With solutions of caustic soda, stannous sulphide does not dissolve without boiling for several minutes, and this treatment dissolves mercuric sulphide in amounts dependent on the concentration of the alkali.Compromises between the two types of reagent, while working well in experienced hands with known mixtures of sulphides, are apt to be worse than either reagent when used by students in exercises on unknown substances. It was the purpose of this work to examine qualitatively the several possible alternatives to the above reagents with a view to selecting the one offering the least interference and, if possible, to find a reagent that would always achieve a clean separation of the two groups from each other. EXPERIMENTAL Stock solutions were prepared, each containing 0-1 per cent. of the several usual metals of the hydrogen sulphide group in the form of its chloride.After suitable adjustment of the acidity, hydrogen sulphide was passed into 25-ml. portions of each of these solutions. The precipitated sulphide was filtered, washed once with water, and then treated with the particular reagent under review in accordance with established practice, i.e., warmed with a polysulphide or boiled with an alkali. The whole was then filtered and the residue washed once with water. The filtrate was just acidified with hydrochloric acid, boiled and examined visually for the presence of sulphides of the copper group. The sulphides of mercury=, lead, copper, cadmium, bismuth, stannous and stannic tin, arsenic and antimony were each separately warmed with 10 ml. of solutions of each of the polysulphides of ammonium, lithium, sodium, potassium and calcium in the manner described above and with the result shown below.The polysulphide solutions were used in 8 per cent. strength and were prepared by the digestion of flowers of sulphur with the respective alkali followed by removal of the surplus sulphur, except in the case of ammonium and potassium, when the commercially available material was used. Further samples of the metallic sulphides, prepaxed as above, were each separately heated just to boiling with 25 ml. of solutions of the hydroxides of sodium, lithium, barium and calcium with the result shown below. RE s ULTS POLYSULPHIDE REAGENTS- Each of the polysulphides examined readily dissolved the sulphides of the arsenic group, but in addition, they each dissolved some copper sulphide and a small amount of mercuric sulphide, and rendered some of the cadmium sulphide colloidal.Judged by the amount of copper sulphide dissolved, calcium polysulphide appeared to be the most satisfactory of this group of reagents. ALKALINE HYDROXIDE REAGENTS- Sodium hydroxide-A 5 per cent. aqueous solution was first used in these experiments but was found to dissolve appreciable quantities of mercuric sulphide whilst stannous sulphide dissolved only after considerable boiling. A reduction in strength to 1 per cent. left the mercuric sulphide insoluble under ‘the conditions of the experiment, but moreover, had little action on stannous sulphide. To dissolve stannous sulphide it was found necessary to boil the mixture for several minutes, and this treatment also caused some mercuric sulphide to dissolve.Solutions of caustic soda will not, therefore, effectively separate mercuric sulphide from stannous sulphide. Lithium hydroxide-A 5 per cent. aqueous solution was found to dissolve readily all the sulphides of the arsenic group, including that of stannous tin. At the same time little or no solvent action was observed with mercuric sulphide unless the suspension was boiled for 2 or 3 minutes. With a 1 per cent. solution, stannous sulphide readily dissolved as soon278 HOLNESS AND TREWICK : SULPHIDE SEPARATIONS [Vol. 75 as the solution reached its boiling-point and und.er this condition no mercuric sulphide was found to have dissolved. Barium hydroxide-A saturated solution of harium hydroxide was used in these experi- ments, and although it showed no solvent action towards the sulphides of the copper group, it reacted with both the antimony and tin sulphides to form a white precipitate, probably of the hydrated oxides.Calcium hydroxide-Owing to the low solubility of this reagent, a suspension or milk was used. This dissolved arsenic and antimony sulphides, but converted the tin sulphides into what was probably hydrated stannic oxide. TABLE I SUMMARY OF RESULTS Copper group Arsenic group A A I - f \ Reagent HgI* Pb Bi cu Cd As Sb Snn SnIv (NH,),Sn . . ss NS NS p:j col S S S S KzSn - . . . ss NS NS p:s col S S S S 1%NaOH .. NS NS NS N S NS S S NS S 2 LbSX , . . . ss NS NS p!j col S S S S g NaaSx . . ss NS NS p!s col S S S S 9 CaSx . . . . ss NS NS ps col S S S S S S S S S insol. hydrated oxides S S hydrated oxides NS = not soluble; S = soluble; ps = partly soluble; ss = slightly soluble; col = colloidal solution From the results, which are summarised in Table I, it appeared that a 1 per cent.solution of lithium hydroxide offered the possibility of a clean separation of the sulphides of the copper group from those of the arsenic group. Accordingly, mixtures of compounds of the elements of the hydrogen sulphide group were made up in. order to test exhaustively this possibility. First results were promising, but it was found that, whenever cadmium sulphide was present, the reagent tended to render it colloidal, so that it became very difficult, and at times impossible, to filter. Attempts to overcome this by adding 5 per cent. of sodium chloride to the reagent were successful, but the presence of chloride ions, probably adsorbed, was found to cause mercuric sulphide to dissolve to some extent when subsequently treated with nitric acid.When 5 per cent. of potassium nitraie replaced the sodium chloride, no further trouble was encountered. This reagent, a 1 per cent. aqueous solution of lithium hydroxide containing 5 per cent. of potassium nitrate used in 26-ml. quantities and heated just to boiling, has worked excellently with students at all levels and its use has become standard practice in these laboratories. It has been found to give clean separations of the sulphides of all members of the copper group from those of the arsenic group, and even when the less usual elements are present, such as molybdenum, tungsten, selenium and tellurium, its efficiency is unimpaired.More- over, the condition for its use-heated just to boiling-is a definite instruction, whereas warming with a polysulphide is capable of varied interpretations. Again, when stannous tin is present, it is stannous sulphide that is reprecipitated with the sulphides of the arsenic group and its colour can be an aid to its identification. With the polysulphide reagent it is always stannic sulphide that is reprecipitated and no indication is given as to the original state of the tin. When compounds drawn from among the less familiar elements are suggested as reagents, the subject of their cost inevitably comes to mind, and it is of interest to record, therefore, that the cost of this particular reagent, volume for volume, is very much less than that of the more usual yellow ammonium sulphide. Even when the quantities used are taken into account, 25 ml. of the one and 10 ml. of the other, the cost of the larger volume is less than half that of the smaller. Hence, on the scores of cost, efficiency, pleasantness and specificity in use, it is to be recommended in preference to the several reagents at present in general use. c1 i 2% l%LiOH .. NS NS NS NS NS Ba(OH), .. NS NS NS N S NS *? $ 2 Ca(OH), .. NS NS NS N S NS CHEMISTRY DEPARTMENT SOUTH-WEST ESSEX TECHNICAL COLLEGE WALTHAMSTOW, LONDON, E.17 December, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500276

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

May, 19501 KOTES 27 9 Notes THE SEPARi4TION OF LEAD AND BISMUTH IN the analysis of fusible alloys, a nitric acid attack removes tin as metastannic acid, from which the small amount of occluded lead and bismuth are extracted, and added t o the main filtrate. The usual procedure at this point is to add sulphuric acid and evaporate until the acid is fuming, dilute, and filter off the insoluble lead sulphate. This procedure has been much criticised, on the grounds that bismuth is co-precipitated to such an extent as to render the process inaccurate. For this reason other methods in which bismuth is precipitated as an oxy-salt from a faintly acid solution, leaving lead in the filtrate, have been recommended. The conditions necessary for these methods are difficult to adjust, and usually a double precipitation is required, owing partly to the diEiculty of obtaining the correct conditions of acidity, and partly to the nature of the bismuth precipitates, which are not so easily washed as lead sulphate.Moreover, no account is taken of any other metal which may be present, e.g., cadmium in Wood’s metal. Therefore it seemed worth-while to re-examine the sulphuric acid method, since lead sulphate is easily washed free of ions in solution, and only sulphuric acid is present in the filtrate containing other metals which have to be estimated. Tests were made with mixtures of lead and bismuth varying from 30 to 70 per cent. of each metal, The lead sulphate produced was examined for bismuth by digestion with ammoniacal ammonium acetate solution which was found to separate the bismuth quantitatively, leaving all the lead in the filtrate, which can then be converted to lead molybdate in the usual way.‘The bismuth in the precipitate is dissolved by digesting with weak nitric acid, converted to sulphate and added to the main filtrate. The results obtained by this procedure were quite satisfactory within experimental error. However, in the preliminary trials, the main object was to discover whether any variation in technique would produce lead sulphate free from bismuth, and in these trials the extracted bismuth, if any, was identified by passing hydrogen sulpliide through the sulphate solution to give the black sulphide precipitate. It was found that when fuming was conducted a t a low temperature of about 250” C., the lead sulphate was free from bismuth, whereas there was considerable co-precipitation when fuming was vigorous a t a high temperature in the range 360” to 400” C.Since bismuth alone did not become insoluble a t the higher temperature, it is probable that a complex ion of lead and bismuth is produced by over-fuming. This method avoids the complication of extrdction of lead sulphate for co-precipitated bismuth, and correct results were obtained in trials with mixtures of lead and bismuth, and also with synthetic mixtures of lead, bismuth, tin and cadmiuni simulating Wood’s metal. A. T. ETHERIDGE March, 1949 THE NICKEL AND MOLYBDENUM CONTENT OF “NORMAL” HLTMAhT URINE AND FAECES THE occurrence of lead poisoning amongst shipbreakers is a generally-recognised phenomenon.During the breaking up of old British battleships in the Clyde area, the question arose as to whether certain symptoms appearing in some of the workers could be due to metals other than lead, i.e., certain of the constituents of arniour plate. As a result, the excreta of certain of the workers were examined for chromium, nickel and molybdenum. Measurable amounts of both nickel and molybdenum could be detected. Since no “normal” figures for nickel and molybdenum appeared to be available (Monier-Williams1), they were determined. Measurable amounts of both nickel and molybdenum could be detected in the urine and faeces oi normal human beings, and these amounts differed in no way from those of the shipbreakers. EXPERIMENTAL- Both urine (100 ml.) and faeces (1 g.) were ashed in silica dishes and the residue dissolved in dilute hydrochloric acid, and nickel and nzolybdenum were determined colorirnetrically, the former by means of dimethylglyoxime, and the latter by means of potassium thiocyanate and stannous chloride, followed by extraction with ether.The details of the procedures were essentially- those described by Sandell .2280 NOTES [Vol. 75 RESULTS- The following amounts were found in the urine and faeces of normal humans- Nickel. and molybdenum contents r 1 Minimum Maximum Average URINE (12 cases) Nickel . . .. Opg. per litre 55pg. per litre 29pg. per litre Molybdenum . . 6 p g . per litre 43pg. per litre 28pg. per litre Nickel . . . , 220 pg. per 100 g. 1260 pg. per 100 g. 833 pg. per 100 g. Molybdenum .. 50 p g . per 100 g. 450 pg. per 100 g. 234pg. per 100 g. REFERENCES Monier-Williams, G. W., “Trace Elements in Food,” Chapman & Hall, London, 1949. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Interscience Publishers, New DRIED FAECES (12 cases) 1. 2. York, 1944. BIOCHEMISTRY DEPARTMENT ROYAL INFIRMARY GLASGOW S. L. TOMPSETT J. FITZPATRICK August, 1949 CRYSTALLINE POTASSIUM HYDR0GE:N TARTRATE IN TINNED GRAPES EARLY this year a complaint was received that a tin of grapes, a South African product packed in light syrup, contained particles of glass. On examining the contents we found a number of glass- like fragments, all but one of which were in the syrup, the remaining fragment, about 1 cm. long, being embedded in a grape. On closer examination these fragments wen: seen to consist of hard transparent crystals.Qualitative and quantitative tests proved that these consisted entirely of potassium hydrogen tartrate, the accompanying syrup itself containing approximately 24 per cent. of sucrose. Through the kindness of Professor E. G. Cox, of the Inorganic Chemistry Department, Leeds University, arrangements were made for Dr. N. H. Hartshorne, the Lecturer in Chemical Microscopy, t o carry out a crystallographic examiination of the large fragment already mentioned. Dr. Hartshorne found that its intermediate refractive index, /3, and orthorhombic symmetry were consistent with its being potassium hydrogen tartrate. To both Professor Cox and Dr. Hartshorne we express our best thanks for their ready collaboration in this investigation.The presence of cream of tartar (argol) in both unfermented grape juice and the grape itself was somewhat surprising, inasmuch as its deposition was understood to take place during the later stages of fermentation. Its deposition in this instance may, however, have been encouraged by the cold condition in which the tinned grapes had probably been stored. This occurrence may be compared with the presence of struvite (magnesium ammonium phosphate) in canned salmon .l ,2 REFERENCES 1. 2. Manley, C. H., Analyst, 1931, 56, 808; 1933, 58, 337. James, L. H., Ibid., 1933, 58, 222. CITY ANALYST’S LABORATORY 1, SWINEGATE LEEDS, 1 C. H. MANLEY A. ALCOCK November, 1949 A SIMPLE PROCEDURE FOR THE DE.TERMINATION OF MICROGRAM AMOUNTS 017 CYANIDE ALTHOUGH a considerable number of methods exist for the determination of cyanide, none of these is readily adaptable to microgram quantities without the use of special equipment.The following modification of several existing techniques was devised for the determination of microgram quantities of cyanide in animal tissues, but has also been found to work well in general practice, and in other analyses where a micro-modification of a turbidimetric method is desirable. In principle, the cyanide is measured by a Denigks silver nitrate titration, performed in a photo- electric colorirneter or turbidimeter. The reagents involved are the usual concen1:rated aqueous ammonia, 10 per cent. aqueous potassium iodide and 0.001 n/r silver nitrate solutions. One millilitre of this silver nitrate solutionMay, 19501 NOTES 281 is equivalent to 53 pg.of hydrocyanic acid. For our work, a Klett - Summersoil photo-electric colorimeter containing a 420-m p. filter was used, although other similar instruments should serve equally well. A micro-burette was fashioned from a 0-200-ml. pipette which was calibrated in microlitres. The flow from this was controlled with a 0-25-ml. hypodermic syringe, which was connected to the pipette with fine-gauge copper tubing and rubber sleeves. The burette assembly is firmly mounted upon a ring stand, with the delivery tip of the pipette sufficiently high to titrate directly into the colorimeter tube while the latter is in place. This tube should have a capacity of 5 ml., and it remains in situ throughout the titration.&!!I?THOl-, The sample may be prepared in any manner that will give an essentially clear aqueous solution. For tissue samples and many others, steam distillation of the acidified sample into an alkaline trap is advisable. Transfer 5 ml. of the solution to be analysed into the colorimeter tube, add 0.5 ml. of the potassium iodide solution and 1.0 ml. of the concentrated aqueous ammonia, and agitate the solution with a glass rod to cnsure complete mixing. Then take a reading on the colorimeter; we have found it convenient to adjust the instrument to read 100 at this stage. Then add the silver nitrate solution in 0.01-ml. increments; after each addition, agitate the solution xvith the glass rod and note the colorimeter reading. The process is terminated whcn a sizable increase in the colorimeter readings has occurred.The exact end-point may conveniently be determined by plotting the readings against micro- litres of silver nitrate solution, and extrapolating a smooth curve through these points to the base line. From this must be subtracted a blank value obtained by the same process on a water sample. THE BIOCHEMICAL INSTITUTE THE ~JNIVERSITY OF TEXAS ERNEST BEERSTECHER, JUN. AUSTIN, TEXAS November, 1949 ABSO RPTIOPVIETRIC ESTIMA’TION OF AIUSTARD GAS BY MEANS OF IODOPLATINATE ,4ND STARCH Is the course of investigations on the biological efiects of mustard gas, it became necessary to make rapid routine measurements of the concentrations of the reagent or of its hydrolysate in aqueous dutions.It was found that the well-knowm semi-quantitative colorimetric method in which iodoplatinate is added to release iodine, which can be detected by starch, could be made suitable i f the intensity of the resulting blue colour was read objectively with a Spekker absorptiometer, tiif results being reproducible with an error of only 1 or 2 per cent. e\en when ordinary commercial starch was used. Preliminary investigations showed that (i) highest sensitivity is obtained when niossuremeats are made with red light, (ii) sufficient excess of starch is required, and (iii) the iiitimsity of the blue colour varies with time, but is constant sufficiently long for easy manipulation. _Z technique based on these findings is descJ5bed below. The method involves no rxw principles ; thc experience gained with it may, however, be of value to other workers.METHOU lxb, \GENTS- lodo@afimte--Add 1 ml. of 5 per cent. platinjc chloride solution to 5.3 ml. of fyeshZ.y pvepaved ptir cent. sodium iodide solution (free from ioditie) and dilute the whole to 180 ml. with distilled ater. Starch-Rub 1 g. of B.D.13. soluble starch into a cream with a little watcr and pour into Acetic ncid-Dilute 50 nil. of glacial acetic acid to 1 litre with distilled water. Staxdnvd inustard gas solution-Weigh accurately about 0.06 g. of pure liquid ,8 P’-dichloro- dwthyl sulphide in a stoppered vessel. Add 2.5 anl. of glacial acetic acid and 2.5 ml. of distilled ivater and allow the solution to stand for 24 hours. Dilute this solution with the 5 per cent. acetic acid to give a final coiicentration of 0-05 rng, of mustard gas per ml.XIS:A)RPTTOX SPECTRA- lodoplatinate + stavch-Two rnillilitres of the iodoplatinate solution +- 10 ml. of 5 per cent. acetic acid -+- 2 ml. of starch gave an absorptioi1 ciirve with a niaximum-not very pronounced 100 id. of boiling water. It is desirable to use freshly prepared starch solutions only. ‘l‘hcn dilute t o 50 ml. and shake well. (.) -at 4700 A .282 NOTES [Vol. 75 (b) Starch blue colour-Two millilitres of iodoplatinate + 10ml. of standard mustard gas At this A red filter transmitting solution + 2 ml. of starch gave a maximum three times more intense at about 6500 A. wavelength, the absorption of the mixture (a) was almost negligible. between 6500 and 7000 A. was used for all. subsequent absorptiometer measurements. VARIATION OF BLUE COLOUR WITH QUANTITY OF STARCH- Using 10 ml.of standard mustard gas solution and 2 ml. of iodoplatinate solution, the optical density of the blue colour was found to increase with the quantity of starch solution added up to 1.5 ml., and then became constant. The proportioins adopted were, therefore, 5 volumes of mustard gas solution to 1 volume of iodoplatinate solution and 1 volume of starch solution. STABILITY OF BLUE COLOUR- When these proportions were used, with the standard mustard gas containing 0.05 mg. per ml., maximum density of the starch colour was observed 3 minutes after mixing, and the density remained constant until the tenth minute, after which it fell off rapidly. With more concentrated mustard gas solutions, the constant density was still reached after 3 minutes, but the colour was a little less stable. Readings were therefore made between the third and fifth minutes after the starch had been added.RANGE OF CONCENTRATION OF MUSTARD GAS- Under the conditions described, with l-cm. cells in the absorptiometer, readings were repro- ducible t o 1 per cent. up to an optical density of 1-0, and to 3 per cent. between 1.0 and 1.5. This sets an upper limit, for this degree of reproducibility, of 0.05 mg. of mustard gas per ml. The calibration curve relating optical density to mustard gas concentration was not a straight line, but deviated so little from it as to permit of linear interpolation between calibration points spaced as described below. PROCEDURE ADOPTED FOR UNKNOWN SOLUTIONS-- Dilute the solutions, prepared in 5 per cent.acetic acid, to contain between 0.01 and 0.025 mg. of mustard gas per ml. Prepare a series of standards containing 0-005, 0.01, 0-015, 0.02, 0.025, 0.0375 and 0.05 mg. of mustard gas per ml., and keep these ready for use. Place 10 ml. of test solution and 10 ml. of 5 per cent. acetic acid respectively in two graduated test tubes. Then put 2 ml. of starch solution into each tube and note the time. Mix the contents by inverting the tubes two or three times. Transfer the solutions to two 1-cm. absorption cells and place them in the absorptiometer. Read the optical density of the test solution on the instrument scale between 4 and 5 minutes after adding the starch, using the solution of t.he iodoplatinate control for comparison.(As this operation can be done in a few seconds, a batch of blue solutions could be examined at one time, and compared with the same iodoplatinate.) CALIBRATION- The blue colours produced by the standard mustard gas solutions were measured in the same way and the calibration curve was constructed. The process was repeated two or three times for each solution examined, and concordant readings were usually obtained, provided the time restriction was carefully observed. -This method was developed in the course of experiments made a t the Strangeways Research Laboratory, Cambridge, between 1939 and 1944 for the Chemical Defence Department, Ministry of Supply, and was described in detail in a confidential report to the Ministry.l I am indebted to the Chief Scientist, Ministry of Supply, for permission to publish this summary. Add 2 ml. of iodoplatinate to each and shake the contents. REFERENCE 1. “A Note on the Estimation of Mustard Gas by the Iodoplatinate Method using a Hilger Absorptio- meter.” Dated June, 1941. GUY’S HOSPITAL MEDICAL SCHOOL LONDON, S.E.1 C. B. ALLSOPP December, 1949May, 19501 MINISTRY OF AGRICULTURE 283 Official Appointments PUBLIC ANALYST APPOINTMENTS NOTIFICATION of the following appointments has been received from the Ministry of Food since the last record in The Analyst (1950, 75, 111). Public Analyst Appointments ALLEN, David George (Deputy) .. CHALMERS, Frederick Grant Duncan . . CHILDS, Hugh . . . . . . . . JENKINS, Daniel Ceiriog Evans (Deputy) JONES, Archibald Orton (Deputy) . . WOOD, Eric Charles . . .. . . WOOD. Eric Charles . . .. . . Woon, Eric Charles . . .. .. . . County of Southampton.* . . Borough of Hereford. . . . . Borough of Hereford. . . . . . . Borough of Lowestoft. . . County of &-orfolk. County Borough of Sheffield. County Borough of Sheffield. County Borough of Great Yarmouth. * The Ministry of Food regret that in their notification dated February %th, 1950, this authority was incorrectly described as the Southampton County Borough Council.

ISSN:0003-2654    

DOI:10.1039/AN9507500279

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

May, 19501 MINISTRY OF AGRICULTURE 283 Ministry of Food STATUTORY INSTRUMENTST 1950-No. 362. The Shredded Suet (Revocation) Order, 1950. Price Id. This Order, which came into force o n March 18th, 1950, vevokes the Shredded Suet Order, 1949 (S.I., 1949, No. 1497), but without prejudice to proceedings in contravention thereof. - No. 379. The Citrus Fruit (Amendment and Revocation) Order, 1950. Pricc Id. This Order, which comes into force for grapefruit as from March 21st, 1950, and for ovanges (sweet and bitter) as from May 7th, 1950, (a) amends the Citrus Fruit Order, 1945 (as amended), by defining “citrus fruit” as sweet and bitter oranges and by deleting the definition of “grapefruit” in Article 1, and (b) prohibits the sale of citrus f m i t mixed with any other article or oiseveet oranges mixed with bitter oranges.- No. 409. The Milk (Special Designation) (Pasteurised and Sterilised Milk) (Amendment) This Ovder, which came into .force on March 26th, 1950, provides (a) that the glass-distilled water and the stoppered~t’ask used respectiuely in the preparation and for the storage of the reagent used in the methylene blue test for pastei.~rised milk shall be sterile, and (1)) alters the weight of a ~ n w o ~ i i u i n sulphate to be used in carryirzg out the ttwbidaty test for sterilised milk from 3 r 0.1 g. lo Regulations, 1950. Price 1d. 4 & 0.1 g. Ministry of Agriculture STATUTORY INSTKUiMENTt 1950-No. 410. The Milk (Special Designation) (Raw Milk) (Amendment) Regulations, 1950. Price Id. This Order, which came into force on March 26th, 1950, repeats the requirements as to stcriZit,v c/ the glass distilled water and stoppered $ask contained in Order N o . 409. t Obtainable from H.M. Stationery Office. Italics indicate changed wording.

ISSN:0003-2654    

DOI:10.1039/AN9507500283

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

284 REVIEiWS [Vol. 75 Revie:w THE CHEMICAL ANALYSIS OF FOODS. By H. E. COX, Ph.D., D.Sc., F.R.I.C. Fourth Edition. The chemical analysis of foods now covers such a wide range that it is not possible for a moderate- sized book to deal adequately with all varieties. In addition, it is doubtful if one man would be able to write authoritatively and critically on the whole of the field. Nevertheless, it is desirable that experts should make available to others their specialised knowledge and accumulated experience in those branches which they have made their own. For this reason a new edition of “Cox” is very welcome. At first glance this fourth edition appears to be a reprint of the third (see review, Analyst, 1946, 71, 345) ; but closer examination shows that here and there methods have been replaced by more moderli ones, and now and then new data have been given for the constituents of foods.The bulk of the previous matter, however, remains unchanged. This is a tribute to the care taken and to the selection made in previous editions, factors which are well known to readers of the book and which are to be expected from the author. But improvements are continually being made in many determinations and some of the methods included, although useful, are not really good enough. Reference is made in one place to 1936 for a method for determining citric acid; but in three other places the methods are still those of 1907, 1912 arid 1913. The Denigks reagent cited for methyl alcohol is not the modem form which has been in general use since at least the B.P.1932. In these days, the newer instruments which are necessary for particular determinations or for more exact measurements are very costly and may not be an economic proposition for some analysts. But certain instruments are essentiakimd these must be used to the best advantage in cases of dispute. This applies particularly to the polarimeter, where accuracy of a high order may be required, and this will depend partly upclri the instrument, partly on the technique and partly on the reliability of the calculations based 011 the angular rotations of the specific substances being determined. The International Commission for Uniform Methods of Sugar Analysis has done much to standardise polarimetric determinations and it is important that a book of the standing of “Cox” should follow the Commission’s recommendations and should give sufficient information to enable the internationally agreed temperature of 20” C. to be used. It is hoped that in the next edition the section on Polarimetry will be brought up to date. “COX” has become an essential volume for food laboratories because of its practical outlook. It is known that the author ploughs a much wider field than that which he has reaped for the benefit of others. Pp. 340. London: J. & A. Churchill Ltd. 1950. Price 28s. Perhaps in the near future we may look for an extension of the harvest. J. R. NICHOLLS

ISSN:0003-2654    

DOI:10.1039/AN950750284b

出版商:RSC    

年代:1950

数据来源: RSC