摘要:

88 SAINT: A MICRO PROCEDURE FOR THE ELECTROLYTIC [Vol. 83 A Micro Procedure for the :Electrolytic Determination of Lead in Copper-base Alloys BY H. C. J. SAINT (Research De@artment, Imperial Chemical Industries Ltd., Metals Division, Kynoch Works, Witton, Birmingham) A study has been made of the conditions affecting the electrolytic deposition of micro amounts of lead as peroxide in copper-base alloys, par- ticularly brass, by using a sample weight of 5 mg containing a maximum of 250 pg of lead. The significance of factors such as the time of the electrolysis, the temperature of the electrolyte and the presence of other alloying metals and impurities has been separately investigated. It has been shown that, when the temperature of the electrolyte is 80" C and a potential difference of 2 volts is applied, reliable results can be obtained by using the specially designed micro apparatus with a platinum wire as anode, provided that manganese and arsenic are absent.In the presence of man- ganese the same method is applicable, but the temperature of the electrolyte must be maintained at 20" C. When arsenic is present, the results are slightly high. MICRO procedures for determining lead by electro-deposition as peroxide have been pub- lished,l-2 and the results given in the milligram range are satisfactory. Laboratories occasion- ally receive samples for analysis that involve the determination of lead in the microgramFebruary, 19581 DETERMINATION OF LEAD I N COPPER-BASE ALLOYS 89 range, and, as far as is known, no results have been given for the accuracy of the electro- deposition procedure for these smaller amounts. With the object of developing a method that could be applied to 5 to 10-mg samples of copper-base alloys, particularly brasses con- taining about 3 per cent.of lead, a further study of this procedure was necessary. EXPERIMENTAL APPARATUS- The modified Benedetti-Pichler type of micro electro-deposition3 apparatus, shown in Fig. 1, was used for these experiments and losses on transference of the sample were avoided by weighing it in a small glass dish, which was then transferred to the electrolysis cell and remained there during dissolution of the sample and electrolysis. The electrolysis cell was a Pyrex-glass test-tube, 1.6 cm x 12.5 cm, and the receiver was a Pyrex-glass boiling-tube, 2.5 cm x 15.5 cm.A piece of No. 6 s.w.g. platinum wire was used as an anode, the lower end being wound in the form of a spiral. The cathode was a small platinum gauze cylinder, 2.7 cm x 1-4cm, separated from the anode by a glass sleeve. Agitation was effected by gently passing a stream of air through a narrow syphon tube situated in the centre of the spiral anode. During weighing, the anode was counterpoised against a platinum wire of similar shape and weight. Fig. 1 Fig. 2(a) Fig. 2(b) Fig. 1. Apparatus for the micro electro-deposition of lead: A, platinum hooks on apparatus stand; B, and B,, clamps; C, syphon tube; D, electrolysis cell; E, water bath; F, cathode ; G, anode ; H, glass sleeve; I, sample dish ; J, air inlet and suction tube ; K, receiver for electrolyte Fig.2(a). Fig. 2(b). Glass hook for transferring anode Glass stand for suspending anode in oven or desiccator TESTS ON STANDARD SOLUTIONS- To test the efficiency of the electro-deposition procedure in the microgram range, a standard solution of high-purity lead in dilute nitric acid was prepared. By using amicro weight-burette, aliquots representing the lead content of 5 or 10-mg samples of brass were transferred to test-tubes, and copper nitrate solution was added to prevent cathodic deposition of lead. This combined solution was evaporated to dryness in a bath of glycerol and the residue was dissolved in 5 to 10 ml of 1 per cent. v/v nitric acid. The conditions of electrolysis that were found to be suitable were (a) a temperature of about 80" C, (b) a potential difference of 2 volts, and (c) a current density of 8 mA per sq.cm. These conditions are similar to those applied by other workers.1~2~4 After 10 minutes, the anode and the walls of the tube were washed with a jet of water and the electrolysis was90 SAINT: A MICRO PROCEDURE FOR THE ELECTROLYTIC [Vol. 83 continued for a further 10 minutes. The anode containing the deposit was dried at 180" C, and, in calculating the results, a theoretical factor of 0.8662 was used. This factor was used throughout the experimental work. Recoveries firom solutions containing known amounts of lead were as follows- Lead present, mg . . * . 0.128 0.101 0.148 Lead found, mg . . .. 0.131 0.169 0*163 Difference, mg . . .. . . + 0.003 -- 0.002 + 0.005 The slight differences observed were due largely to weighing errors.For a 10-mg sample of brass containing 3 per cent. of lead, the greatest difference would be equivalent to about 0.05 per cent. of lead. Electrolysis at room temperature was equally satisfactory, but a total time of at least 40 minutes was necessary. No interference was observed from nitrous fumes formed during electrolysis and the possibility of retention of water by the very small deposit was ignored. ANALYSIS OF LEADED BRASS- A leaded brass ingot containing 2-93 per cent. of lead, determined as molybdate on the macro scale, was examined. The surface of the ingot was skimmed and fine drillings were taken from several positions. These were well mixed and a weighed amount was used for preparing a standard solution, portions of which were examined under conditions identical with those mentioned previously.The results are shown in Table I. TABLE :[ DETERMINATION OF LEAD IN SOLUTIONS OF A COPPER - ZINC - LEAD ALLOY CONTAINING 2.93 PER CENT. OF LEAD Equivalent weight of alloy taken, Lead present, Lead found, Difference, Lead found, 4.077 0.138 0.141 + 0.003 3.02 5.057 0.149 0.152 + 0.003 3.02 5.986 0.176 0*1;30 + 0.004 3.01 3.775 0.111 0.1 16 + 0.006 3.07 5.423 0.100 0.1186 + 0.005 3.03 mg *g mt: mg % Separate 5-mg portions of the mixed drillings were dissolved in 0-6 ml of diluted nitric acid (1 + 3) and the solutions were evaporated tio dryness. The residues were dissolved in 1 per cent. v/v nitric acid and the solutions were examined as before, the results being as follows- Weight of sample, mg .. 4.982 6.462 4.891 6.231 4.904 4.731 Lead found, mg . . 0.146 0.108 0.136 0.154 0.161 0.136 Lead found, yo.. . . 2.94 3-04 2.76 2.96 3-07 2.88 The average of these results is 2-94 per cent., which agrees reasonably well with the accepted value obtained by a macro procedure. The variation in results is probably due to differences in the lead content of the small samples. INFLUENCE OF IMPURITIES- High results by the electro-deposition procedure have been reported by some workers and are attributed to the presence of other elements, notably manganese, bismuth, arsenic, antimony and phosphate ; the influence of these elements was separately investigated. Because tin and silicon may be present in alloying amounts in some copper alloys, inter- ference by these metals was also investigated. INFLUENCE OF MANGANESE- By using the electrolytic procedure, Tucke+ o'btained results that indicated that recovery of lead is almost quantitative provided that (a) copper ions are present, (b) the solution is cold at the beginning of the electrolysis, and (c) ra.pid deposition of lead is made from 5 to 10 per cent.v/v nitric acid. Tests were accordingly made to assess the effectiveness of this separation under various conditions of acidity, when only microgram amounts of lead are present, by using solutions prepared from high-purity lead, zinc, copper and manganese. The equivalent of 5-mg samples of brass containing 3 per cent. of lead and 3 per cent. ofFebruary, 19681 DETERMINATION OF LEAD IN COPPER-BASE ALLOYS 91 manganese were electrolysed for 40 minutes at room temperature, i.e., an initial period of 20 minutes and 20 minutes after washing the electrodes.The results in Table I1 show that (i) 3 per cent. v/v nitric acid is suitable, (ii) lead is not deposited when the concentration of acid is increased to 10 per cent. v/v, (iii) manganese is not co-deposited with lead (manganese was not detected by a chemical test on one of the deposits), and (iv) all recoveries are as good as those obtained in the absence of manganese. Further proof of the absence of interference by manganese at this concentration was obtained by analysing weighed amounts of drillings to which the equivalent of 3 per cent. of manganese had been added. TABLE I1 INFLUENCE OF MANGANESE ON THE DETERMINATION OF LEAD BY ELECTRO- DEPOSITION AT 15" TO 20" C FOR 40 MINUTES Concentration of nitric acid, Manganese added, % v/v mg 10 0.160 3 0.160 3 0.160 3 0.160 3 0.150 Weight of deposit Lead taken, calculated as lead, mg mg 0.104 Nil 0.160 0- 162 0.153 0.163 0.162 0.156 0.158 0*166* * Manganese was not detected in this deposit.INFLUENCE OF OTHER ELEMENTS- The presence of 1 per cent. of silicon and 0.1 per cent. of phosphorus (as phosphate) had little effect on the recovery of lead; 3 per cent. of tin and 0.1 per cent. of bismuth caused slightly high results and 0.1 per cent. of antimony caused slightly low results. These amounts of bismuth and antimony, however, are greater than would be encountered in practice and only in the presence of arsenic were results unacceptable.When co-deposition of other elements occurs, eg., arsenic, an alternative method for completing the determination of lead in the deposit must be used. METHOD REAGENTS- Nitric acid, 25 per cent. v/v--Add 25 ml of nitric acid, sp.g 1-42, to 75 ml of water. Nitric acid, 1 per cent. v/v-Add 10ml of nitric acid, sp.gr. 1.42, to 90ml of water, Boil to remove nitrous fumes and make up to volume again with water. For use, dilute 10 ml to 100 ml with water. Hydrochloric acid, M) per cent. v / v . Potassium iodide. Carbon tetrachloride. PREPARATION OF THE APPARATUS- Bend the upper portion of the anode to form a loop and suspend it by a glass hook (see Fig. 2) in a boiling-tube containing 50 per cent. v/v hydrochloric acid and a few crystals of potassium iodide.Then wash the wire with a jet of water and suspend it on a glass hook in a drying oven maintained at 180" C . After 16 minutes, transfer the anode to the hook of a glass stand in a desiccator (without desiccant) and allow it to stand for 5 minutes. Finally, transfer the anode to the pan of a microbalance and weigh it after 5 minutes; use an identical platinum wire as counterpoise. Straighten the upper portion of the wire, pass it through the cleaned cathode and glass sleeve, already attached to the apparatus, and arrange the spiral over the lower portion of the air-supply tube. Attach the anode to the positive terminal of the apparatus, already connected to a 2-volt accumulator, and gently pass a stream of air through the apparatus. Note that, after removing the lead peroxide from the anode with the hydrochloric acid - potassium iodide solution and then washing, the bent portion of the stem should be heated in the flame of a bunsen burner to anneal it, thereby preventing early fracture.PROCEDURE FOR PREPARING AND WEIGHING THE SAMPLE- Use a drop of mineral oil at the position of sampling to prevent the drillings from scattering.92 SAINT [Vol. 83 Lightly skim the surface with a small drill, a dental drill is suitable, and wipe the area clean. Transfer the drillings to a 2-ml centrifuge tube containing carbon tetralchloride, wash them about six times with the solvent, dry them at room temperature for 15 minutes and then remove any particles of iron with a magnet. Prepare a small hemispherical glass dish having a diameter of between 10 and 15 mm and weighing about 0.3 g from the blown-out end of a sealed glass tube.Clean the dish and dry it in an air oven. Allow it to cool in a desiccator for 5 minutes, transfer it to the micro- balance and weigh after a further 5 minutes. Introduce 5 to 10 mg of the prepared drillings and re-weigh. PROCEDURE FOR DETERMINING LEAD- Transfer the glass dish and sample by means of forceps to the mouth of the electrolysis cell, while holding the cell horizontally. Raise the cell slowly to the vertical position so that the dish slides to the bottom and remains upright. Add 0-6ml of 25 per cent. v/v nitric acid by introducing a pipette into the cell until tlhe tip is about + inch from the glass dish, and then allow the acid to fill the dish to prevent it floating. Allow the sample to dissolve and transfer the electrolysis cell to a bath of glycerol heated to 105" C.Evaporate the solution to dryness by gently passing a stream of air filtered through cotton-wool in a glass tube having one end drawn out to a capillary. Add 5 to 10 rnl of 1 per cent. v/v nitric acid, i e . , only sufficient to cover the gauze cathode, rotating the tube during the addition to wash the walls of the tube. For alloys containing manganese, dissolve the dried residue in 3 per cent. v/v nitric acid and proceed with the electrolysis at room temperature for a total time of 40 minutes. Slide the test-tube up over the electrodes and, when they are almost touching the glass dish, clamp the test-tube in position. Place a 100-ml squat-form beaker, containing water at 80" C, around the lower part of the test-tube and maintain at this temperature with a microburner flame.Electrolyse the solution for 10 minutes, and then wash the inner wall of the test-tube, the air-supply tube and the upper portion of the electrodes with a few millilitres of water. Remove the hot-water bath and replace it by a beaker of cold water. Disconnect the air supply and transfer the electro- lyte by gentle suction to the larger tube used as a receiver. Half fill the electrolysis cell with water and continue agitation by passing air for a few seconds. Remove the water bath and electrolysis cell. Immediately disconnect the 2-v olt supply and gently remove the anode. Bend the upper portion of the wire through 180", suspend it on a glass hook and wash the entire length of the wire, including the spiral, by gentle application of a jet of water. Place the anode on the glass stand in an air oven at 180" C for 15 minutes, transfer it to the desiccator and allow it to stand for 5 minute:;, then to the microbalance for 5 minutes and then weigh. Calculate the weight of lead deposited by multiplying the weight of lead peroxide deposited by 0.8662. I thank Mr. W. T. Elwell and other colleagues for their assistance, suggestions and practical help. Add a further drop of oil and continue to drill the cleaned area. Heat in a hot-water bath until all :;ah are dissolved. Continue the electrolysis for a further 10 minutes. REFERENCES 1. 2. 3. 4. 6. Lindsey, A. J., Analyst, 1935.60, 744. Cimerman, C. H., and Ariel, M., Anal. Ckinz. Ada, 1956, 15, 207. Benedetti-Pichler, A. A., "Micro Technique of Inorganic Analysis," John Wiley & Sons Inc., Sand, H. J. S., J . Chew. SOC., 1907,91, 386. Tucker, P. A., Analyst, 1946, 71, 319. New York, 1942, p. 226. Received February 7th, 1967

ISSN:0003-2654    

DOI:10.1039/AN9588300088

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

February, 19581 HUNTER 93 The Micro-determination of Magnesium in Presence of Known Amounts of Calcium BY G. HUNTER (CowZey Road Hospital, Oxford) A method for the determination of magnesium in the range 0 to 1.0p.g in presence of between 0 and 5.0 pg of calcium has been developed with the use of a colour comparator. Calcium is determined with murexide as indi- cator and magnesium plus calcium, expressed as magnesium, with Eriochrome black T as indicator. The interference of calcium is much less than when ethylenediaminetetra-acetic acid is used. In twenty-one known mixtures of calcium and magnesium, the mean error of the results by the method was found to be 2.3 per cent. in the range -10 to f 2 per cent. The method is applicable to the determination of calcium and magnesium in all proportions in organic solutions free from interfering ions and to deproteinised blood and cerebrospinal fluid, and its extreme simplicity makes possible multiple determinations of calcium and magnesium in small amounts in such fluids.HUNTER and Stottl have described a micro method for the determination of magnesium in serum, in which the Eriochrome black T colour change is determined by using a new colour comparator. For the determination of magnesium in serum, the method has the chemical advantage of low affinity of calcium for the indicator compared with ethylene- diaminetetra-acetic acid (EDTA) , which titrates calcium and magnesium equivalently. In normal human blood plasma there are about 2.5 milli-equivalents of calcium and 0.8 milli- equivalents of magnesium per litre, so that rather more than 75 per cent.of the total EDTA titration is due to calcium, as contrasted with a corresponding value of about 15 per cent. in the method to be described. The most uncertain and time-consuming part of the method of Hunter and Stott is the removal of calcium with oxalate. Owing to the marked com- plexing action of oxalate, it is easy to introduce a greater error from this source than that due to the presence of the calcium. It appeared that the removal of calcium might be unnecessary if it were possible to measure the magnesium equivalent of a known amount of calcium in the test portion used -known from a titration with murexide as indicator on a separate test portion. Since the colour comparator is sensitive to less than 0.01 pg of magnesium, it has been found to be possible to prepare a graph from which, after titration of mixtures of calcium and magnesium with a standard magnesium solution in presence of Eriochrome black T, the precise value for magnesium may be read.As the method has general application to mixtures of calcium and magnesium, and as it is both simple and time saving, the conditions for its success have been studied in some detail. METHOD APPARATUS- The new colour comparator described by Stott,2 which is being made available com- mercially by Messrs. Baird and Tatlock Ltd., was used. The comparator has two glass cups held vertically over a pair of barrier-layer photo cells, which are connected in series to a microammeter incorporating a transistor amplifier.Since the light beam is passed down the axis of the cups, the optical density is largely independent of dilution, and titration of the contents of one cup to match the other becomes possible. The titrant is delivered from a 5-ml burette, graduated in 0.01 ml, in approximately 0.01-ml drops, which pass through a slotted glass prism to avoid interference with the light beam. Provision is made for agitation of the cups. REAGENTS- All reagents must be prepared with water free from magnesium and calcium. CaZcium stock solution-Dissolve 250 mg of calcium carbonate in 5 ml of N hydrochloric acid, add 1 ml of 0-1 per cent. w/v solution of thiomersal (sodium ethylmercurithiosalicylate) and dilute with water to 100ml. 1 ml = 1 mg of calcium.94 HUNTER : THE MICRO-DETERMINATION OF MAGNESIUM vol.83 Calcium wmkiqg solution-Dilute a portion of the calcium stock solution with water, so that- 1 ml = 10 pg oE calcium. Murexide indicator solution-Dissolve 5 mg of ammonium purpurate in water and dilute to 250ml. This reagent will keep for several days if stored cold in the dark. Sodium hydroxide solution, N. Magfiesium stock solution-Dissolve 100 mg of magnesium ribbon in 9 ml of N hydro- chloric acid, add 1 ml of 0-1 per cent. w/v solution of thiomersal and dilute with water to 100 ml. 1 ml = 1 mg of magnesium. Magnesifcsn working solutio.n-Dilute a portion of the magnesium stock solution with water, so that- 1 ml = 1 pg of magnesium. Eriochrome bZack T indicator solution A-Dissolve 100 mg of Eriochrome black T in 100 ml of methanol.Allow the solution to stand, with occasional shaking, for several days, then decant it from any inorganic residue and store it in a refrigerator. Eriochrome black T indicator solution 23-Dilute 10 rnl of the indicator solution A to 100ml with methanol. Bufer soldm-Mix 85 parts by volume of 1.5 M ammonium hydroxide with 15 parts of 1.6 M ammonium chloride. This solution has a pH of 10-25. &fleer - indicator solzction-Mix 10 ml of Eriochrome black T indicator solution B with 20 ml of the buffer solution and 20 ml of water. This solution, if cooled below 20" C, is usable for at least 1 hour. PROCEDURE FOR DETERMINING CALCIUM- To each cup of the comparator add 2.5 ml of murexide indicator solution and 0.3 ml of sodium hydroxide solution. Mix the solutions arid adjust the pointer of the ammeter to zero with an Ilford No.110 light filter in position. To the right-hand cup add a suitable volume of test solution, mix and add calcium working solution to the left-hand cup until the ammeter pointer returns to zero. The final volumes in the cups should be about equal; if they differ by more than 0-6 ml, a compensating volume of water should be added to the appropriate cup and the titration repeated. The amount of calcium present in the test portion is given directly by the burette reading. Maghcrlum plus akiurnl (as magncrlum), f l E I - T 9.2 0.4 0.6 0.8 - s : -0.10 - - s s g -0.05- 2 E c L 'SL r -0.15 - L -0.20L I I Fig. 1. Calibration graph for magnesium PROCEDURE FOR DETERMINING MAGNESIUM PLUS CALCIUM- To each cup of the comparator add 2.5 ml of buffer - indicator mixture.Set the am- meter to zero with an Ilford No. 204 light filter in position and add to the right-hand cup a suitable volume of test solution. Add magnesium. working solution to the left-hand cup until the ammeter pointer returns to zero. The volume added from the burette is a measure of magnesium and calcium present in the test portion. The amount of calcium present in this is known from the titration with murexide as indicator and the magnesium present can then be read from a calibration graph such as that shown in Fig. 1.February, 19581 IN PRESENCE OF KNOWN AMOUNTS OF CALCIUM 96 PROCEDURE FOR PREPARING THE CALIBRATION GRAPH- The calibration graph is an arbitrary construction dependent primarily on pH, but also affected by a number of other factors, such as concentration of indicator, ionic concen- tration and temperature.These factors are controllable within tolerable limits and values are reproducible with great precision. In the absence or presence of calcium there is a linear relationship between the titres and amounts of the metals present. Magnesium equiva- lents were determined, as described in the procedure for determining magnesium and calcium, in the absence of magnesium in the right-hand cup, for 1, 2, 3, 4 and 5 pg of calcium. This gave the series of points along line AB. A buffer-indicator mixture was prepared con- taining 0-5 pg of magnesium in 2.5 ml by mixing 10ml of Eriochrome black T indicator solution B, 20ml of buffer solution and 10ml of magnesium working solution in a 50-ml calibrated flask and diluting to the mark with water.With 2.5 ml of this mixture in each cup, the pointer of the ammeter was adjusted to zero. From 1 to 5 pg of calcium were added to the right-hand cup as before and the left-hand cup was titrated with standard magnesium solution. By adding the readings to 0.5 pg, a series of points was plotted along line CD. Similarly, a buffer - indicator mixture was prepared containing 1-0 pg of magnesium in 2.5 ml and from 1 to 5pg of calcium were added to the right hand cup and titrated as before. Again, by adding the titres to 1.0 pg, a series of points was plotted along line EF. Several intermediate magnesium levels have been checked and found to be linear with the values shown. Similar calibration graphs have been prepared at pH values other than 10, with buffer solutions of different ionic concentrations and at different temperatures, and they were found to have similar linear relationships.The necessary observations for the construction of a calibration graph can be made in about 2 hours. Examples of the m e of the calibration gYaph-Let it be assumed that the magnesium plus calcium titre is 0.6 ml, or 0.6 pg as magnesium. Then if the test portion used con- tained 1 pg of calcium, a value of 0.04 pg has to be subtracted from the value for magnesium plus calcium, as magnesium, to give the true value of 0.56 pg of magnesium. If the test portion contained, successively, 2, 3, 4 and 5 tcg of calcium, the values to be subtracted, corresponding to the same magnesium plus calcium titre as before, are 0.07, 0.10, 0.13 and 0-21 pg, which give true values of 0.53, 0-50, 0.47 and 0.39 pg of magnesium. Fractional values of calcium in the test portion can of course be readily interpolated.The calibration graph, shown in Fig. 1, was constructed as follows. RESULTS As a test of the accuracy of the calibration graph, a series of twenty-one solutions containing known amounts of calcium and magnesium with ratios of calcium to magnesium ranging from 100 to 1 was prepared. The calcium in each solution was determined by the proposed method. The amounts found in 0.2 ml of test solution are shown in Table I. Tripli- cate determinations of magnesium plus calcium were carried out by the method on 0.2 ml of each solution (the test portion of blood filtrates we normally use), and the mean titration values were used to determine from the calibration graph the correction to be subtracted to give the true values of magnesium.The results are shown in Table I, and it can be seen from these results that the mean errors of the determinations of calcium and magnesium are 1-6 and 2.3 per cent., respectively. DISCUSSION OF THE CONDITIONS OF THE METHOD Schwarzenbach3 notes that the metal indicators act as chelating agents in a way similar to the complexones, such as EDTA. Most analysts, however, have used EDTA with murexide as indicator for determining calcium or with Eriochrome black T as indicator for determining calcium plus magnesium. Smith* has described a method for determining magnesium in blood serum, after removal of calcium with oxalate, in which the change of colour of Erio- chrome black T with magnesium is measured in an absorptiometer.The method described by Hunter and Stottl made use of the new colour comparator described earlier, p. 93. The proposed method depends primarily on the accurate measurement of calcium in terms of magnesium in presence of various amounts of magnesium. It will be seen from Fig. 1 that the magnesium equivalent of calcium decreases as the amount of magnesium present increases.Calcium present, PLg 5.0 3.0 1.0 Calcium found, PLg 5.00 5.00 5-00 4.95 5.05 5.10 5.00 2.93 3-00 3.02 2-97 2.92 3-05 3.00 0.96 1.02 0.98 0-96 1.00 0.98 0.94 Error, % 0 0 0 -1 + I $2 0 -2 0 tl - 1 -3 +2 0 -4 + 2 -2 -4 0 -2 -6 Magnesium present, Pg 0-05 0.10 0.20 0.40 0.60 0.80 1.00 0.05 0.10 0.20 0.40 0.60 0-80 1.00 0.05 0.10 0-20 0.40 0.60 0.80 1.00 96 [Vol.83 HUNTER : THE MICRO-DETEKM1:NATION OF MAGNESIUM TABLE I[ DETERMINATIONS OF CALCIUM AND MAGNESIUM IN KNOWN MIXTURES The test portions used for the determinations of calcium were of 0.2, 0-4 and 1.0 ml for the solutions containing 5,3 and 1 pg of callcium, respectively; 0.2-ml test portions were used for all the determinations of magnesium plus calcium Magnesium found, Ccg 0.045 0.100 0.195 0.385 0.610 0.805 1.015 0.050 0.090 0.190 0.390 0-585 0.800 1.000 0.050 0.100 0.195 0.400 0.600 0-790 1~000 Error, - 10 0 -3 -4 +2 -t 1 i-1 0 - 10 - 5 - 3 -2 0 0 0 0 - 3 0 0 -1 0 Yo Ratio of calcium to magnesium 100 50 25 12.5 8.3 6.2 5.0 60 30 15 7.5 5-0 3-7 3.0 20 10 5 2.5 1.7 1.2 1.0 Mean error of the determination of citlcium = 1.6 per cent.Mean error of the determination of magnesium = 8.3 per cent. The method is suitable for the determination of calcium and magnesium in all proportions in inorganic solutions free from interfering ions. It is probably applicable to most water supplies. I t has been used for over 2 years for the determination of calcium and magnesium in blood and cerebrospinal fluids that have been deproteinised by the method of H ~ n t e r , ~ and in the course of this work the main factors affecting its accuracy have been determined. I, 0 -5 0-0 I I 0 0.1 0.2 0.3 0.4 0.5 0.6 0 Amount of magnesium equivalent to 5 pg of calcium, pg Fig. 2. Effect of pH on magnesium equivalent t:o 5p.g of calcium r Efect of PH-The effect of pH, at constant ionic concentration, on the magnesium equivalent of 5 pg of calcium is clearly shown in Fig.2, the magnesium equivalent rapidly decreasing with pH. It might appear that at pH 9.0 calcium will not interfere at all, but unfortunately the sensitivity of the determination of magnesium likewise decreases and the end-point becomes uncertain below a pH of about 9.7. To keep a sharp end-point, a reaction mixture of pH 10.0 has been chosen.February, 19581 I N PRESENCE OF KNOWN AMOUNTS OF CALCIUM 97 The change in the value of the magnesium equivalent of 5 pg of calcium with a small shift of pH might make the method appear to be impracticable. Yet this is not so. Relatively large amounts of acid are required to shift the pH appreciably.For example, the addition of 0.2 ml of 0.1 N hydrochloric acid to 2.5 ml of the buffer - indicator mixture decreases the pH by less than 0.05 and the addition of 2ml of blood filtrate (ten times the amount we normally use) has no measurable effect on the pH. It may be noted that, in special circum- stances, the buffering capacity can be doubled by mixing 1 part of indicator solution with 4 parts of buffer solution. Concentration of indicator-The concentration of indicator in the buffer - indicator mixture, 20 pg per ml, is rather more than is necessary for the titration of 2 pg of magnesium. The choice of 5 pg of calcium to illustrate the effect of pH, concentration of indicator and ionic concentration was brought about by the appreciable magnesium equivalent titration obtained with this amount, although it is two to five times greater than the calcium present in the test portions used for the determinations of magnesium in blood and cerebrospinal fluid.The effect of different concentrations of indicator was investigated by determining the magnesium equivalent of 5 pg of calcium with buffer - indicator solutions containing 20, 15 and 10 pg of indicator, the results being 0.21, 0.18 and 0.14 pg, respectively. The concen- tration of indicator in the buffer - indicator mixture was decreased without changing the volume of methanol present or the ionic concentration of the mixture. I t can be seen from the results that decreasing the concentration of indicator decreases, as might be expected,. the magnesium equivalent of calcium.Dilution of the solution in one cup by 50 per cent. of its volume without changing the volume in the other cup introduces an error of about 15 per cent. with 5 p g of calcium. Such differences in volume are extreme and the error would be less with less calcium, but for precise work, especially with large amounts of calcium, the volumes should be made approximately equal by adding water. Ionic concentration-The effect of change in ionic concentration was investigated by determining the magnesium equivalent of 5 pg of calcium with buffer - indicator mixtures of ionic concentrations 1.2,0.6 and 0-3 M , the results being 0-18,0.20 and 0.23 pg, respectively. It can be seen that dilution of the 0.6 M mixture by 50 per cent. increases the magnesium equivalent of 5 pg calcium by about 7 per cent. This is about half the effect of diluting the indicator, but it is in the opposite direction, so that the effects tend to compensate each other.Eflect of concentration of methanol-The buffer - indicator mixture contains 20 per cent, of methanol. This has the effect of decreasing the pH of the buffer solution in the buffer - indicator mixture from 10.25 to 10.00. Observations with different concentrations of methanol of from 2 to 42 per cent. indicate that changes in the magnesium equivalent of calcium over this range are attributable to changes in pH. The presence of 20 per cent. of methanol has, however, the beneficial effect of sharpening the end-point of the titration of calcium - magnesium mixtures.This has been noted by Betz and NoK6 E$ect of temperature-If the pointer of the ammeter is set at zero with 2.5 ml of buffer - indicator mixture in each cup and an equal volume of standard calcium solution is added to each cup, the pointer of the ammeter remains at zero at widely different temperatures, say, 15" and 25" C. The same is true with an equal amount of magnesium added to each cup. But if 5 pg of calcium are added to the right-hand cup and balance is achieved by adding mag- nesium to the left-hand cup at 15" C and the temperature is then changed to, say, 25" C, the ammeter indicates a slight over-titration. With 5 pg of calcium alone, this temperature difference introduces a titration difference of about 10 per cent. In presence of magnesium, the error is less and a scarcely perceptible difference was found between calibration graphs prepared at 17" and at 25" C.A calibration graph prepared for a prevailing laboratory temperature would therefore appear to be valid to within k5" C. CONCLUSIONS The foregoing experimental work deals primarily with the determination of magnesium. in presence of calcium in simple solutions. It is, however, indicated that deproteinised' blood and cerebrospinal fluids may be regarded, for practical purposes, as such simple solutions, since the presence of phosphate, interfering metals and complexing agents such as citrate are negligible at the dilutions used. However, the method is not yet directly applicable98 BANKS AND ROBSON: RA.PID METHODS OF [Vol. 83 t o urine, owing to the presence of excess of phosphate, nor to ashed tissues, owing to inter- ference from other metals such as iron. I am indebted to Mrs. K. M. Nunn for technical assistance. The work described is part *of a programme in the development of methods for the study of the blood - cerebrospinal fluid barrier in association with Dr. H. V. Smith, Reader in Medicine in the University of Oxford, and is made possible by a grant from the Nuffield Foundation. REFERENCES 1. 2. 3. 4. 5. 6. Hunter, G., and Stott, F. D., Biochem. J., 1956, (52, 2 9 ~ . Stott, F. D., Ibid., 1956, 62, 3 6 ~ . Schwarzenbach, G., “Die Komplexometrische Titratiola,” Ferdinand Enke Verlag, Stuttgart, 1955, Smith, H. J., Biochem. J., 1955, 60, 522. Hunter, G., J . Clin. Path., 1957, 10, 161. Betz, J. D., and Noll, C. A,, J . Amer. Wut. WRs Ass., 1950, 42, 49. p. 25. Received August 28th, 1967

ISSN:0003-2654    

DOI:10.1039/AN9588300093

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

98 BANKS AND ROBSON: RA.PID METHODS OF [Vol. 83 Rapid Methods of Boiler-water Analysis BY J. BANKS* AND J. ROBSON (CentraZ Electricity Authority, East Midlands Division, North Wilford Power Station, Colliery Road, Nottingkam) Methods are described for the rapid analysis of boiler-water samples, particularly those from high-pressure boilers. The total acidity of a sample that has been treated with a cation-exchange resin is found, after which the phosphate is determined by a simple titrimetric procedure. The chloride concentration is found by the oxycyanitle method and the sulphate by difference. Possible interference is examined, and a method for the deter- mination of sulphite is suggested. WATER samples from high-pressure boilers (abovle 400 lb per sq. inch) and, in particular, power-plant boilers, usually contain only hydroxide, orthophosphate, sulphate, chloride and silicate, normally as their sodium salts.Sometirrtes, in plant working below 600 lb per sq. inch, sodium sulphite can be used as an oxygen scavenger and, in a few very isolated instances, tannins, often combined with hexametaphosphate, can be added. Tannins are, however, generally little used in high-pressure plant, and siilphite may even be deleterious. Boiler-wat er analysis, therefore, consists in determining ort hophosphat e, sulphat e, chloride and silica, hydroxide often being found by computation. This is a lengthy task when carried out by conventional analytical methods and some of the results may be of doubtful value. It appeared that a cation-exchange process could be usefully applied in the determination of a t least some of the constituents of a normal boiler-water sample.Although apparently not widely applied in this country to boiler water, ion-exchange methods have been reported in another c0untry.l It was also intended to examine and apply, if satisfactory, methods for phosphate and chloride determination not generally used in boiler-water analysis. EXPERIMENTAL CHLORIDE- It has been customary in boiler-water analysis to determine chloride by the classical Mohr titration, after first suitably adjusting the pH value, but this method has been much criticised and its unsuitability is well known. It was considered that a more reliable method was required. Of the titrimetric methods available, that due to Viebock2 appeared most attractive.In this procedure, the neutralised sample is allowed to react with a solution of mercuric oxycyanide. and the liberated sodium hydroxide is titrated yith standard acid. Station, Back Lane, Castle Donington, nr. Derby. The reaction involved i:+- HgO.Hg(CN), + 2NaCl+ H20 = HgC1, + Hg(CN), + 2NaOH * Present address : Central Electricity Authority, East Midlands Division, Castle Donington PowerFebruary, 19581 BOILER-WATER ANALYSIS 99 In the routine analysis of boiler water, it is essential to determine the total alkalinity in order to determine later the sodium hydroxide concentration. In our laboratory, alkalinity is determined by titration with 0.1 N sulphuric acid, B.D.H. “4.5” indicator being used. This appeared to be a convenient point at which to apply the Viebock reaction, the liberated alkali being titrated to the end-point of B.D.H.“4-5” indicator. Experimental work confirmed that chloride could be determined in this way. Belcher, Macdonald and Nutten3 have studied the Viebock reaction and have found that it is not always stoicheiometric. They have suggested applying a comparison titration procedure with a standard solution of sodium chloride. This method was examined, but, despite its excellence, it was decided that the small errors involved in the present work did not justify the extra expenditure of time, and the direct titration was adopted. PHOSPHATE- It is usual in boiler-water analysis to determine phosphate by one of the conventional methods depending on the formation of the molybdenum-blue colour. Results are excellent when a photometric finish can be applied, but when visual comparison is necessary an alternative method would be valuable.Tollei‘has suggested that for phosphate contents greater than 5 p.p.m. as P,O, (approxi- mately 10 p.p.m. as Na,PO,) the absorptiometric method is not sufficiently accurate and has suggested the use of a simple volumetric procedure. In this, the phosphate is brought to the dihydrogen stage and sufficient silver nitrate is added to precipitate all the chloride and phosphate. The phosphate reacts as follows- NaH,PO, + 3AgN0, = Ag,PO, + NaNO, + 2HN0, and the liberated nitric acid is titrated with standard alkali. Previous works had proved the accuracy and convenience of this procedure and originally the method was designed around this reaction.More recently, however, a similar procedure has been described,s in which silver nitrate is replaced by cerous nitrate. The mechanism is similar, but, since cerous chloride is soluble, the bulk of the precipitate formed is considerably less and end-points are much more distinct. In the phosphate determination it is immaterial whether the original solution is acid or alkaline, since it is brought to the dihydrogen phosphate stage, Le., pH 4.5, before the reaction with cerous or silver nitrate. By passing the boiler water through a small column of cation-exchange resin in the hydrogen form, the total acidity owing to the conversion of all the salts present to the corresponding acids could be found and used in the subsequent determination of sulphate.The sample, after being treated with a cation-exchange resin, was boiled to remove carbon dioxide and then titrated with standard alkali to the end-point of methyl red - methy- lene blue indicator. Excess of a neutralised cerous nitrate solution was added and the liberated nitric acid was titrated with 0.02 N alkali to the same end-point. The original total-acidity titration was normally carried out with 0.1 N alkali, but, with samples of low dissolved-solids content, it was found preferable to use 0.02 N alkali. When 0.1 N alkali was used, it was found advisable to add a drop of 0.1 N acid, and to adjust the end-point with 0.02 N alkali to guard against over-titration. In all the titrimetric work described micro-burettes were used.Some slight difficulty was experienced in dealing with very low phosphate contents, particularly when silver nitrate was used, but this was readily overcome by first adding a known amount of a standard phosphate solution previously neutralised to methyl red - methylene blue indicator. The necessity for the addition of standard phosphate may readily be judged from the amount of colour change of the indicator after adding the silver or cerous nitrate solution. With cerous nitrate solution the lower limit of phosphate determinable was considerably less than with silver nitrate and the results were good even with only 1 or 2 p.p.m. (as Na,PO,) present. Because of its many advantages, the use of cerous nitrate has been adopted as routine practice. SULPHATE- The conventional methods of determining sulphate in boiler water are the gravimetric barium sulphate method or the benzidine procedure with a volumetric finish.The distinct It was at this stage that an ion-exchange technique was applied. The results have been excellent with both silver and cerous nitrates.100 BANKS AND ROBSON: RAPID METHODS OF [Vol. 83 solubility of the precipitate renders the benzidine method of doubtful value. In the presence of phosphate, the classical barium sulphate precipitation is open to considerable criticism. Previous work5 had shown that the results are erratic and, in the present investigation, a barium sulphate precipitate was found to contain a considerable amount of phosphate. In the proposed method, when the total acidity after cation exchange and the acidity equivalent to the chloride and phosphate by the Viebock and cerous nitrate titrations, respectively, have been determined, the acidity equivalent to the sulphate, normally the only other anion present after ion exchange, may ‘be found by difference.(In this connection it must be remembered that, although two equivalents of nitric acid are liberated in the phosphate method, only one equivalent of the phosphoric acid formed by cation exchange is titrated.) In order to assess the accuracy of the proposed methods, artificial boiler-water samples were prepared and analysed. A sufficient range of composition was chosen to simulate conditions likely to arise in modern steam-raising practice. Pure sodium chloride, disodium hydrogen phosphate and sodium sulphate were used in the preparation of the samples. The results are shown in Table I.These results indicate that the methods are satisfactory over a wide range of boiler-water composition. The figures for phosphate determination show that cerous nitrate yields somewhat better results than silver nitrate and it is considerably easier to find the correct end-point: when cerous nitrate is used. TABLE r ANALYSIS OF SYNTHETIC BOILER-WATER SAMPLES Chloride, Phosphate, Sulphate, p.p.m. as NaCl- p.p.m. as Na3P04- p.p.m. as Na2S04-- -7 - - r A \ Sample No. 1 150 2 100 3 20 4 20 6 6 6 20 7 80 8 50 150 80 150 100 5 101 20 50 19 20 20 20 7 10 7 19 80 20 80 1 81 50 5 49 . . present found present found by- 7 - 7 silver cerous nitrate nitrate method method 79 79 78 80 4 5 4 5 49 50 50 50 19 19 21 19 10 10 8 9 79 80 78 79 0 2 0 2 4 5 6 5 95 per cent.confidence limits REAGENTS- Sulbhuric acid, 0-1 11-5 METHOD and 0.02 N. present found by- r 130 100 50 20 5 20 80 100 silver nitrate method 130 133 104 99 52 50 22 20 5 4 19 19 80 82 103 104 2 4.0 cerous nitrate method 130 133 100 99 49 51 22 20 5 4 20 19 80 81 102 103 2 3.0 Sodiztrn hydroxide, 0.1 and 0.02 N. Mercuric oxycyanide solution-Prepare a solution containing 20 g per litre. Remove 10ml and add to 50ml of distilled water neutralised to B.D.H. “4.5” indicator. Titrate this aliquot with 0.02 N sulphuric acid to the end-point of B.D.H. “4.5” indicator, using a micro-burette. Now add the calculated volume of 0.02 N sulphuric acid to the remainder of the oxycyanide solution and mix thoroughly.Test for neutrality by adding 10ml to 50ml of distilled water neutralised to B.D.H. ‘ ‘ 4 4 ’ ’ indica-tor. No colour change should occur. (This method of neutralisation is better than straightforward titration of the main solution, owing to the buffering action of mercurilc oxycyanide.)February, 19581 BOILER-WATER ANALYSIS 101 Cerous nitrate solution-Dissolve 25 g of cerous nitrate free from other rare-earth metals in distilled water. Neutralise to methyl red - methylene blue indicator and dilute to 1 litre. Methyl red - methylene blue indicator-Dissolve 0.125 g of methyl red in 50 ml of ethanol; dissolve 0-083 g of methylene blue in 50 ml of ethanol. Mix equal volumes of these solutions. PROCEDURE FOR DETERMINING CHLORIDE- Titrate a 50-ml sample of boiler water with 0.1 N sulphuric acid (or 0.02 N if the alkalinity is low) to the end-point of B.D.H.“4.5” indicator. Add 10ml of neutralised mercuric oxycyanide solution and mix well. Titrate with 0.1 or 0.02 N sulphuric acid to the same end-point as before. The volume of 0.1 N sulphuric acid, in ml, multiplied by 117 (or 23.4 for the 0.02 N acid) gives the concentration of chloride, as p.p.m. of NaCl. PROCEDURE FOR DETERMINING TOTAL ACIDITY AND PHOSPHATE- Pass the boiler water through a column of Zeo-Karb 225 resin (hydrogen form), approxi- mately 20 cm x 1.5 cm. Reject the first 20 to 30 ml of effluent and collect 50 ml over a period of 2 to 3 minutes. Boil the effluent to remove carbon dioxide (and sulphur dioxide if sulphite is present), cool and add methyl red - methylene blue indicator.By using a micro- burette, titrate with 0.1 N sodium hydroxide (or 0.02 N if a small titre is expected). This gives the total acidity. If 0.1 N sodium hydroxide is used, add a drop of 0.1 N sulphuric acid after the total acidity has been determined and adjust the end-point with 0.02 N sodium hydroxide before proceeding with the phosphate titration. Add 5ml of cerous nitrate solution, mix well and titrate to the same end-point with 0.02 N sodium hydroxide. The volume of 0.02 N sodium hydroxide, in ml, multiplied by 32.8 gives the concentration of phosphate, as p.p.m. of Na,PO,. CALCULATION OF THE SULPHATE CONTENT- acid, then- [A - (C + 0.5 P)] x 28-4 = amount of sulphate, as p.p.m. of Na,SO,, where A = volume of 0.02 N sodium hydroxide, in ml, used for total acidity, Convert the total acidity and chloride titrations to the equivalent volumes of 0.02 N C = volume of 0.02 N sulphuric acid, in ml, used for chloride, and P = volume of 0.02 N sodium hydroxide, in ml, used for phosphate.POSSIBLE INTERFERENCES IN THE METHOD Of the constituents found in boiler water apart from those already mentioned, only silicate, nitrate, sulphite and tannin will appear in amounts likely to affect the determination under consideration. Silicate is present in boiler water in amounts ranging from 1 or 2 to between 20 and 30 p.p.m. (as SiO,), and, especially at pressures exceeding 600 lb per sq. inch, its ingress is closely controlled. In the present series of experiments, up to 20 p.p.m.of SiO, had no deleterious effect. The presence of nitrate in a high-pressure boiler arises only from contamination by cooling water and the nitrate content is unlikely to rise above a few parts per million. Such amounts are of no operational consequence, and would be calculated as sulphate. Should nitrate be present in significant amounts, it would have to be determined separately and its equivalent acidity deducted from the acidity equivalent to the sulphate. A method for the titrimetric determination of nitrate has been proposed by Ungar.7 As indicated earlier, the use of sulphite has largely disappeared, but it is employed in some units that work below 600 lb per sq. inch. It is normal in these waters to add 10 p.p.m. of sulphite (as Na,SO,), but amounts up to 50 p.p.m.are not unknown. No sulphite is used in our boilers, but it was deemed advisable to investigate its effect. There is no interference from sulphite in the determination of phosphate, since, after cation exchange and boiling, sulphur dioxide is expelled. In the Viebock chloride method up to 50 p.p.m. of sodium sulphite had no ill effect. During the period in which the methods above were being examined by other Divisions of the C.E.A., it was reported8 that, for a boiler in which a proprietary compound consisting102 BANKS AND IROBSON [Vol. 83 of tannin and sodium hexametaphosphate (Alfloc “28” powder) was used, the results were very erratic when the original Toller method4 was applied. This was, in fact, found to be so. The tannin interference can, however, be removed by treating the sample with charcoal free froim inorganic impurities.The boiler-water sample is then filtered and passed through the col.umn of Zeo-Karb 225, and the total acidity and phosphate are determined by the proposed method. Tannins do not interfere in the Viebock chloride method. The sodium hexametaphosphate present in the Alfloc “28” powder is, under boiler operating conditions, rapidly hydrolysed to orthophosphate and is therefore included in the phosphate determination. DISCUSSION The method as outlined has been used successfully in the authors’ laboratory for many months. It has yielded excellent results, which relate to operational practice more closely than when conventional methods were used. It is found convenient to use a set of four columns.When a sample has been dealt with on a column, it is found that the rejection of the first 20 to 30ml of the subsequent sample provides sufficiently thorough rinsing of the column. The frequency of regeneration of the columns with dilute acid is a matter of experience and depends on the salt concentration of the water samples. One ion-exchange process and four simple titrations enable all except the silica and sodium hydroxide concen- trations to be determined, compared with the motre laborious processes of dealing separately with each ion by titrimetric, colorimetric or gravimetric methods. We have found a Kipp tilt-measure of appropriate capacity excellent for addition of the mercuric oxycyanide and cerous nitrate solutions, the volumes of which need not be exact.Although the use of silver nitrate for phosphate has been abandoned, it still yields good results. The saving in time over the conventional methods is substantial. I t is used in the same concentration as cerous nitrate. SUGGESTED METHOD FOR SULPHITE Neutralise a 50-ml sample of boiler water to thymol blue - phenolphthalein indicator. Add a solution of formaldehyde (2 g per 100 ml) neutralised to the same indicator. Titrate the resulting alkaline solution with 0.02 N acid to the same end-point. The volume of 0.02 N acid, in ml, multiplied by 50.4 gives the concentration of sulphite, as p.p.m. of Na,SO,. As previously stated, sulphite is not used in our boilers, but the results with synthetic solutions are encouraging. We thank Mr. L. F. Jeffrey, Controller, East Midlands Ilivision of the Central Electricity Authority, for permission to publish this paper and colleagues in this and other divisions for examination and criticism of the proposed methods. REFERENCE s 1. 2. 3. 4. 5. Banks, J., unpublished work. 6. 7. 8. Leick, J., Mitt. Ver. Grosskesselbesitzer, 1964, 67. ViebGck, F., Ber., 1932, 65B, 496. Belcher, R., Macdonald, A. M. G., and Nutten, AL. J., Mikrochim. Ada, 1964, 104. Taller, W., Mitt. Ver. Grosskesselbesitzer, 1946, 813. Rancke-Madsen, E., and Kjaergard, T., Acta Chmz. Scand., 1953, 7 , 736. Ungar, J., J. AppZ. Chem., 1966,6, 245. Central Electricity Authority, South Wales Division, unpublished work. Received May 16th, 1967

ISSN:0003-2654    

DOI:10.1039/AN9588300098

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

February, 19581 CLARK 103 o-Dithiols in Analysis Part VI.* Diacetyltoluene - 3 : 4 - dithiol as a Coagulant, Catalyst and Precipitant for Sulphur and Metallic Sulphides BY ROBERT E. D. CLARK (Department of Science and Technology, Cambridgeshire Technical College and School of A r t , Cambridge) When toluene-3 : 4-dithiol, or diacetyltoluene-3 : 4-dithiol, is added to a suspension of sulphides formed in acid solution, coagulation is hastened and the precipitated sulphides become water-repellent. Of the various sulphur compounds tried, none was as effective as these. Diacetyltoluene-3 : 4-dithiol is more stable and less easily lost by volatilisation than toluene-3 : 4-dithiol. Addition of a trace of diacetyltoluene-3 : 4-dithiol to solutions before precipi- tation of group 2 cations as the sulphides improves coagulation, and the reagent acts as a catalyst for the decomposition of arsenates and molybdates.Tungsten is not precipitated as the sulphide, but excess of diacetyltoluene- 3 : 4-dithiol precipitates tungsten as a complex that is soluble in alkali. In presence of silicate the precipitation of tungsten is incomplete. The addition of traces of diacetyltoluene-3 : 4-dithiol in ethanolic solution during precipita- tions of group 2 cations is recommended. This reagent is also a coagulant for colloidal sulphur in acid solution. SINCE toluene-3 : 4-dithiol (dithiol) often displaces hydrogen sulphide from metallic sul- phides,1,2 it was thought that its addition to an aqueous suspension of an insoluble sulphide might coat the particles of the sulphide with a layer of strongly lyophobic mercaptide, and so aid coagulation. Dithiol was found to be a powerful coagulant for sulphide precipitates in acid solutions, and in no instance was interference to the usual procedure of dissolution and separation of the sulphides noted.The presence of a lyophobic surface on particles of sulphides was easily shown by the following experiment. A mixture of all the more common group 2 cations was precipitated by hydrogen sulphide under the usual analytical conditions and a trace of ethanolic dithiol was added. By adding ethyIene dichloride or benzene, with shaking, the precipitate entered immediately and entirely into the organic layer. In absence of dithiol, however, it remained in the aqueous layer or collected at the interface.A solution of arsenious sulphide prepared by saturating 100 ml of a hot 2 per cent. solution of gum arabic containing 0.1 g of arsenious oxide with hydrogen sulphide was used. To 2 ml of the solution in a test-tube, 0.4 ml of 2 N hydrochloric acid and a trace of the sulphur compound dissolved in 1 drop of ethanol were added. The test-tube was heated in a water bath and the solution was compared with a blank solution at intervals during a period of 20 minutes. In many tests comparisons were also made by passing hydrogen sulphide through 0.35 N hydrochloric acid containing all the more common cations (a precipitate of silver chloride also being present) and noting whether complete clarification occurred within 1 to 2 minutes. The sulphur compounds found to be either inactive or only slightly active relative to dithiol were S-benzylisothiouronium chloride, l-o-carbomethoxy-4-phenylthiosemicarbazide, diphenylthiocarbazide, sym.-diphenylthiourea, m-mercaptobenzoic acid disulphide, methylene blue, phenylthiourea, sodium diethyldithiocarbamate, sodium toluene-fi-sulphinate, thio- acetamide, thiosalicylic acid and thiourea.1 : 8-Dimer~aptonaphthalene,~,~ quinoxaline-2 : 3-dithio16 and dibenzoyltoluene-3 : 4-dithiol (dibenzoyldithiol)6 showed considerable activity, although less than that of dithiol. Dithio- oxamide exhibited a remarkable power of precipitating arsenious sulphide from solution, far surpassing that of dithiol, although, in general, it is inferior to dithiol-it does not cause precipitates to be wetted by benzene, sulphide precipitates formed in its presence are bulky, it does not catalyse the decomposition of arsenates and it does not precipitate tungstates (see later, p.105). Experiment at once confirmed this expectation. Experiments were then conducted with other sulphur-containing compounds. * For details of previous parts of this series, see reference list, p. 106.104 CLARK: O-DITHIOLS IN ANALYSIS. PART VI [Vol. 83 Diacetyltoluene-3 : 4-dithiol (diacetyldithiol)g was fully as active as dithiol, especially in acid solutions, and had the great advantage that, once a.dded to a solution, it was much less readily destroyed or lost by volatilisation. In addition, it does not give rise to highly coloured compounds with, for example, molybdenum and tungsten, as does dithiol.DIACETYLDITHIOL AS A COAGULANT FOR SULPHUR AND SULPHIDE PRECIPITATES Diacetyldithiol was found to be a powerful coagulant even in presence of 1 per cent. solutions of starch, Teepol, gum arabic and similar materials. In one experiment, 10 ml of a solution containing about 0.2 per cent. of each of the salts of all the more common cations, 1 per cent. of arsenious oxide and 1 per cent. of gum arabic was made 0.3 to 0.4 N with respect to hydrochloric acid, the permanent precipitate formed being ignored. The mixture was heated and hydrogen sulphide was passed through. Much colloidal sulphide remained after 20 minutes. An exactly similar mixture containing 1 drop of 1 per cent. ethanolic diacetyldithiol became completely clear in 10 minutes.Coagulation could not, however, be effected in presence of 5 per cent. of gum arabic. A trace of diacetyldithiol can be used with advantage in the course of ordinary qualitative analysis, when it causes an almost immediate coagulation of the sulphides of the group 2 metals. The sulphides of the more common group 2A metals usually separate so effectively that the supernatant liquid can be removed with the aid of a dropper without the need for filtration or centrifugation. Group 2~ sulphides adso coagulate rapidly, but are more bulky. PRECIPITATION OF SULPHUR- Diacetyldithiol, like dithiol, was also found to be a useful coagulant for elemental sulphur. A hot 1 per cent. acidified solution of sodium thiosulphate deposited its sulphur completely and in easily filterable form within 6 minutes when a trace of the reagent was added.No noticeable coagulation was observed with selenium in acid solution or with sulphur or sulphides in alkaline solution. DIACETYLDITHIOL AS A CATALYST FOR PRECIPITATION OF ARSENIC AND MOLYBDENUM SULPHIDES REDUCTION OF ARSENATE- When dithiol or diacetyldithiol was added to a hot solution of an arsenatein dilute acid and hydrogen sulphide was passed through, or thioacetamide was added, the yellow colour of arsenious sulphide appeared earlier than usual. and complete precipitation was speeded up (see Table I). TABLE I EFFECT OF ADDING CATALYSTS TO ACIDIFIED ARSENATE SOLUTIONS IN PRESENCE OF HYDROGEN SULPHIDE Experiment Time to first appearance Time for complete precipitate, arsenious sulphide, of bright yellow precipitation of seconds minutes (a) 10 ml of a 0.5 per cent.solution of arsenic oxide in 0.35 N hydrochloric acid were heated to boiling in a water bath and hydrogen sulphide As for ( a ) , but with a trace of zinc dithiol sprinkledl on the surface 2 or 3 times during precipitation . . As for (a), but with one drop of a 1 per cent. solution of diacetyldithiol added . . .. . I 3.5 to 40 was passed through briskly . . .. . .. 90 (b) (c) 40 15 or more 6 6 In these and similar experiments it appeared likely that the reagents were acting as catalysts to promote reduction by hydrogen sulphide. In support of this view it was found that other reducing agents could sometimes be substituted for hydrogen sulphide. Thus, when excess of stannous chloride was added to hot 3 to 4 AV hydrochloric acid containing much arsenicv, no change was observed.When, however, a trace of the zinc complex of toluene-3 : 4-dithiol (zinc dithiol) or diacetyldithiol was added, the solution rapidly became red and metallic arsenic was soon precipitated.February, 19581 CLARK: 0-DITHIOLS IN ANALYSIS. PART VI 106 PRECIPITATION OF MOLYBDENUMVI- MolybdenumvI, 0.003 to 0.03 M , in hot 0.3 to 0.4 N hydrochloric acid rapidly gave a precipitate resembling molybdenum sulphide when thioacetamide and a trace of diacetyl- dithiol were added. The filtrate was colourless and free from molybdenum after 3 to 5 minutes. When hydrogen sulphide was used, coagulation was rather less rapid. The usual blue colloid was not formed and the filtrate was colourless and practically free from molyb- denum (it gave a very pale yellow, but no green colour on treatment with zinc dithiol) in 4 to 6 minutes.The presence of the more common group 2 metals did not interfere with the precipitation. In hot 0.3 to 0-4 N hydrochloric acid containing molybdenumv1, it was found that the development of colour (blue, yellow or pink) due to the reduction of molybdenumvI was markedly catalysed by the addition of a trace of diacetyldithiol. The effect was observed with phosphorous acid, hypophosphorous acid mixed with phosphorous acid (reduction with hypophosphorous acid alone was too rapid for observations to be made), hydrazine,. paraformaldehyde and formic acid. When blue solutions were formed in absence of a catalyst, they were found to be unstable or much less stable than when the catalyst was added, and a dark precipitate always formed rapidly.The effect was observed with phosphorous acid, hypophosphorous acid, hexa- methylenetetramine, paraformaldehyde, hydriodic acid and hydroxylamine, as well as with hydrogen sulphide. DIACETYLDITHIOL AS A PRECIPITANT FOR TUNGSTENVI On addition of diacetyldithiol to hot 0.3 to 0.4 N hydrochloric acid containing tung- stenv1, e.g., 0.1 to 1 per cent. of sodium tungstate, a brick-red precipitate of low colour intensity, previously overlooked,6 forms slowly and coagulates readily. After 5 to 8 minutes the filtrate is free from tungstate. Large excesses of phosphate, borate and tartrate, present together, do not interfere, and the appearance and properties of the precipitate are unchanged if hydrogen sulphide is also passed into the solution, or if thioacetamide is added.The precipitate mainly dissolves in warm 2 N potassium hydroxide, leaving a little oily organic matter free from tungsten. The alkaline extract gives a strong tungstate reaction with zinc dithiol and excess of hydrochloric acid.' When excess of silicate was present, in addition to phosphate, borate and tartrate, the reaction was similar, but tungsten was not completely removed from the solution. With a similar mixture acidified to 5 to 6 N with hydrochloric acid, the addition of diacetyldithiol resulted in the slow formation (15 to 25 minutes) of a bulky green precipitate, probably consisting mainly of silica, but a trace of tungstate still remained in the filtrate.In 0.3 to 0-4 N hydrochloric acid, with tungstate and excess of phosphate present, dibenzoyldithiol gave a brick-red precipitate in 20 to 30 minutes, but in the presence of silicate no coloured product was formed within 3 hours. DISCUSSION It is of interest that the compounds found to function as highly reactive coagulants for sulphides in acid solution are all such as are capable of chelating with thiophilic atoms to. form heterocyclic rings containing two sulphur atoms. The results described would appear to indicate that diacetyldithiol is a useful analytical: reagent (a) for the coagulation of sulphides in acid solution, (b) as a coagulant for sulphur in acid solution, (c) as a catalyst for the reduction of arsenates and molybdates in presence of hydrogen sulphide and their subsequent rapid precipitation as sulphides, and (d) as a precipitant for tungstenv1 in dilute acid solution in the absence of silicate.Its reactions with some rarer cations have previously been described.6 REAGENT- ethanol. PROCEDURE- to 0.3 to 0.4 N with hydrochloric acid and heat to boiling. solution and either pass hydrogen sulphide through or add thioacetamide. METHOD OF USING DIACETYLDITHIOL IN ANALYSIS Diacetyldithiul solution-Dissolve 1 to 5 mg of diacetyldithiol in 0.5 ml of 95 per cent. Precipitation of group 2 cations by hydrogen szclphide-Adjust the acidity of the solution! Add one drop of diacetyldithiok106 CURRY COMPOSITE ANSORPTIOM'ETRY [Vol. 83 Coagulation of the sulphides is usually extremely rapid and it is often possible to deter- mine when precipitation is complete by passing hydrogen sulphide into the clear supernatant liquid.Finally, add one more drop of reagent and boil to remove ethanol (see Note). Filter or spin in a centrifuge and proceed as usual. If the final addition of the reagent gives rise to a coloured precipitate within 1 minute, tungstate is probably present. Add diacetyldithiol in excess and boil to complete the precipitation. Precipitation of szllphzlr-Acidify the solution and add one drop of reagent. Maintain at the boiling-point, adding more reagent if necessary until precipitation is complete, NOTE-AS diacetyldithiol is slightly soluble in hot dilute ethanol, the ethanol is removed by boiling, otherwise an excess may separate on cooling and cause a faint white turbidity. I express my gratitude to Professor H. J. Emelkus, Mr. P. S. Jewell and Dr. F. G. Mann for their interest, and to Hopkin & Williams Limited for a gift of chemicals. REFERENCES 1. 2. 3. 4. 5. 6. 7. NoTE-References 1, 2Kand 6 are to Parts I, 111 and IV, respectively, of this series. Clark, R. E. D., Analyst, 1936, 61, 242. - , Ibid., 1957, 82, 177. Price, W. B., and Smiles, S., J . Chern. SOC., 1928, 2372. Obellianne, P., Bull. SOC. Chim. France, 1952, 14'7. Morrison, D. C., and Furst, A., J . Org. Chem., 1956, 21, 470. Clark, R. E. D., Analyst, 1957, 82, 182. Hamence, J. H., Ibid., 1940, 65, 152. Received Septembev 16th, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300103

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

106 CURRY COMPOSITE ANSORPTI0M:ETRY [Vol. 83 Composite Absorptiometry for the Control Laboratory Examination of Calculating Pr0cedi;lres and Modification of the Spekker Absorptiometer for Use with Interference Filters BY D. R. CURRY* (Bragg Laboratory (N.O.I.D.), ShefieEd, 9) Attempts are being made to adapt composite spectrophotometric pro- cedures for use in control laboratories and two essential preliminaries to this work are described. Graphical methods of solving simultaneous equations have been studied, and a triangular net and a “slide-rule” method have been examined practically, The Spekker absorptiometer has been modified to permit the use of a more powerful tungsten-filament lamp as the light source. The characteristics of some commercially available interference filters have been determined and their use in absorptiometry is discussed. DESPITE the advent of quantometry in metallurgical control, it does not satisfy the demands of all laboratories and, when the application of chemical analytical procedures is still essential, the use of absorptiometry to speed up control methods it; now well established.When pelleted reagents are used in these methods, a great deal of the time and chemical skill required is reduced, and so the economy of the work is enhanced. These absorption methods are generally monocolour and require either a separate sample or an aliquot, removed by means of a pipette from a master solution of the alloy, for each item determined. In recent years in academic and research laboratories, spectrophotometric methods have been evolved for binary1 and ternary2 coloured systems.I n these, a mixed colour that is additive for the various coloured ions produced is developed, the assumption being made that there are no chemical interactions. This mixed colour can be analysed into its n components by monochromatic measurements at N, wavelengths at which the relative trans- mission coefficients of the separate species are already determinant. The purity of the light * Present address : Mineral Resources Division, Overseas Geological Surveys, Imperial Institute Building, London, S.W.7.February, 19581 FOR THE CONTROL LABORATORY 107 used for these measurements must be of a high order, as the theoretical additivity of Beer's law only holds for monochromatic light. Given such purity in a spectrophotometer, these methods have proved satisfactory and have also been used in differential procedures in which standard coloured solutions are used as blanks3 Differential methods will not be dealt with here, as, although of high precision, they are necessarily time-consuming and unsuitable for the control laboratory except in limited circumstances.GENERAL AIMS- The use of multi-component systems in which the chemical pre-treatment of the samples can be minimised could obviously be of advantage in the control laboratory, to give two determinations for the price of one. However, in view of the final goal of simple rapid methods, the following limitations must be made- (i) they should be capable of operation by semi-skilled staff and accordingly must have simple calculations that do not involve simultaneous equations, (ii) the apparatus required should be simple, reliable and not too expensive (although capital outlay may well be offset rapidly by reduced running costs), (iii) the procedures should be more rapid than the individual monocolour methods they replace, and (iv) the ultimate scope of the procedures should be as wide as possible and free from interferences by other components normally found in the base material studied.It is intended that practical applications of these procedures as applied to metallurgical inspection should be the subject of later communications from this laboratory. The present purpose is to describe work on two of the essential preliminaries as indicated above, viz.- (a) an investigation of the graphical methods available for the solution of the simul- taneous equations in two or more variables, and ( b ) an investigation into the possibilities of converting the Spekker photo-electric absorptiometer for use in these procedures by using interference filters.* GRAPHICAL SOLUTION OF SIMULTANEOUS EQUATIONS BINARY SYSTEMS- Simple four-scale parallel nomograms6 can readily be drawn for the solution of this problem; they will have direct-reading scales for concentrations [ A ] and [B] with setting from optical-density scales D, and D, (where D, and D, are measured optical densities a t wavelengths A, and A,, and A and B are coloured components to be determined with absorption maxima at A, and A,, respectively).A nomograph, however, is not precise and physiological problems arise when its size is increased to give scale accuracy.When the two setting scales are so far apart that they cannot be viewed simultaneously, any setting of the straight edge becomes a series of approximations as each point in turn is adjusted. Further, any mechanical setting errors are reflected in the result, possibly magnified. The nomograph is not amenable to reproduction by any procedure in which there is the possibility of paper shrinkage, which may not be uniform. A triangular net, previously describedJB has been in use for some years and, as it has proved sufficiently reliable and amenable to its semi-skilled users, it is put forward again. The original problem was not connected with absorptiometric methods or simultaneous equations, but the same principle is applicable.Although it requires high-quality draughtsmanship in preparation, a net can be repro- duced by any method without impairing its accuracy. From the theoretical standpoint the accuracy should be high, as no physical alignments are required: only a point has to be visualised relative to a small triangle as co-ordinates. For the solution of the simultaneous equations arising in the present problem, two inter-related triangles are required to give the solutions- [ A ] = CD, - C'D, [B] = C"D, - C""A] Space can be conserved when the variables have specified ranges by drawing only part of the triangles. An alternative treatment to that of the triangular net on the same mathematical results is to prepare a "slide rule" working on the principle indicated in Fig.1. Its operation involves subjective alignment of two scales, which is complicated in practice by the size of the rule.108 CURRY : COMPOSITE ABSORPTIOMETRY [Vol. 83 Practical comparison of performance of the two procedures (net and slide rule) was made on a series of results with several operators. The errors were examined statistically and showed no significant difference between the best pt:rformances of the two methods. Some operators, however, had significantly greater errors with the net than others, whereas the slide-rule operation showed a more consistent level of performance. A general preference for the slide rule was expressed, owing to the fatigue resulting from use of the net, which was of closer mesh than in the previous work6 because of the accuracy required.D 43!) I200 IIIIIII I l l I I I I I I I I O(%% I 200 I Fig. 1. Schematic diagram of slide rule Slide rule prepared for solution of the equations- Chromium, yo = 3-490,,g - 0.65705,, Manganese, % = 1.28413,,, - 0-0478D4,, The density scales are multiplied by a factor of 1000 to eliminate the decimal point, i.a., an To use, set 0439 (scale 1) opposite 0522 (scale 2) and read manganese % (scale 3) a t the optical density of 0.10 is read as 100. arrow; without moving the slide, read chromium yo (scale 5) opposite 0439 (scale 4) If reproduction is not therefore a vital factor, it would seem preferable to use the slide-rule method of evaluation. TERNARY SYSTEMS- These have been only superficially examined.from the standpoint of calculation, as the chemical complication of binary systems provided sufficient problems. There is no reason why the net system should not be applied to equations of more than two variables having solutions of the form- [C,] = k1DI - k2D2 - k,D, = k,Dl - ( W 2 + W,) - triangle 1 L I triangle 2 The slide-rule system would demand two slides, and it may well be that at this level the complication of extra parallel scales will confuse the issue to an extent that will render the slide rule less satisfactory than the net. A word should be said here about a binary system involving a single reagent reacting with two ionic species, such as the system molybdenum - vanadium - catechol.7 If both reactions should be reversible and achieve dynamic equilibria of the form- where AC is the coloured complex, A+ the ion -to be determined, and C' the reagent that similarly reacts with B+, the analysis of the mixed colour will not be possible with two measurements, but will require a third optical-density measurement to define the concen- tration of unreacted reagent, and thus the calculation will be a ternary one.A+ + C' f' AC, ADAPTATION OF THE SPEKKER ABSORPTIOMETER FOR USE WITH INTERFERENCE FILTERS General details of the Spekker photo-electric absorptiometer are so familiar that elaboration is not required. In view, however, of the requirements of precise optical-density measurement for composite procedure, because of the possible errors in differences, attention is drawn to the measuring diaphragm of the Spekker model H760.This design of a wear-free backlash-free cam with a long reading scale must contribute, in my opinion, to more satis- factory long-term reproducibility than the potentiometer transmission controls of the spectrophot ometer.February, 19581 FOR THE CONTROL LABORATORY 109 As is well known, use of the mercury-vapour lamp with customary gelatin narrow-band filters gives monochromatic light of adequate purity to ensure linear calibration curves for concentration against optical density, which are essential for composite work. The choice of mercury lines is limited, however, and, when the tungsten-filament lamp is substituted, the polychromatism of the light passed by the filters may give rise to curvature of the calibration curves. A further limitation of the normal tungsten-filament lamp used is the low intensity, particularly at the blue end of the spectrum.The use of any narrow-band interference filters with this source would reduce the signal available from the photocell for accurate balancing. Hence, before such filters could be used, a lamp of higher intensity had to be introduced. The use of a more sensitive galvanometer is excluded by cost and lack of stability under severe working conditions. It was found that a standard projector lamp was available (Ediswan S1/162) with identical filament height to the standard, but rated at 500 watts as opposed to 100 watts. Although this does not require forced-draught cooling to maintain the lamp efficiency, it was found that the lamphouse became overheated with the extra power involved.Accord- ingly, a forced draught was introduced to facilitate cooling. An air blower was fitted with a nozzle leading into the base of the instrument. Deflectors sent the air stream through the holes in the lamp bracket up into the lamphouse. The inner light shield was modified to take in part of the air stream to reduce the tendency of the lamp envelope to bulge during use. The lamphouse could now operate for some hours without overheating, but, when the shutter was opened, the heat filters at the sides of the lamphouse cracked with the intense heat of the beam. A system of flow-through water cells fitting inside the lamphouse was therefore designed, and these adequately protected the glass optical system from heat without unduly reducing the intensity of the visible beam.It was realised that the intro- duction of the water cells approximately 1 cm deep with 1-mm glass faces would disturb the focus of the instrument slightly, but it did not apparently affect the accuracy of the readings. G.A.B. INTERFERENCE FILTERS- Two pairs of interference filters were purchased from Geraetebau-Anstalt Balzers (G.A.B.), Liechtenstein, and before they were used in the Spekker absorptiometer a general examination of their characteristics was made. Two procedures were used- (a) By using a large stigmatic grating spectrograph,s the transmission of the filters to tungsten light was determined for the whole filter area. The normal arc stand was replaced by the 500-watt tungsten-filament lamp and the filter to be examined was placed next to the condensing lens, in order to combine light from different parts of the filter on to the slit.Step spectrograms of equal duration were taken for the unfiltered light and that transmitted by the different filters. Exposures were also made for com- parison with the copper d.c. arc by using a Hartmann diaphragm to establish the exact wavelength of peak transmission. The step spectrograms were examined with a microphotometer to give trans- mission curves of the filters. The diaphragm exposures were measured to determine the position of peak density and the wavelength was assessed by interpolation between known arc lines. (b) A Unicam SP600 spectrophotometer was used to examine the heterogeneity of the filter. With the wavelength set at the mean peak value determined in (a), the transmission at various positions across the filter along two perpendicular axes was measured.When a significant variation was shown in the transmission at different positions, the transmission - wavelength relationship was determined at selected extreme positions. Filters 1 and 2 showed a few pinholes, but these were not a significant proportion of the total area as confirmed by the absence of any detectable shoulder or background to their transmission curves. The filters are shown to be of high quality with notable sharp cut-off at the base of the transmission curves compared with those available from British sources. Some hetero- geneity is shown, the importance of which will depend on the purpose for which they are The results of these examinations are summarised in Table I.110 CURRY: COMPOSITE ABSORPTIOMETRY [vol. 83 required.There is some divergence from the marked wavelength, which could prove serious if the filters were to be used for isolation of line spectra, e.g., in flame photometry. Filters 3 and 4 were generally of poorer quality than 1 and 2, particularly in respect of surface variations, despite the absence of pinholes from these. The somewhat disturbing variations in transmission at the mean wavelength may not for some purposes be serious when this reflects a change in peak wavelength rather than a variation in total optical density across the filter. TABLE I Over-all mean transmission determined by Surface variations determined Nominal method (a) by method (b) A > I A 1 Maximum Bandwidths, mp* Transmission + f Nominal band- Filter peak, width, Amax, transmission, /I---A-, at Amax, Amam No.mp mp mp % 1/2 1/5 1/10 1/20 1/50 % mp 1 438 9 437 28 5 9 12 15 - 7%inone 436-437 2 439 6.5 439 30 5.5 9 12 15.5 - negligible - axis 3 521 12 616 33 10.5 16.5 21 25 33 II 7to10% 516-517 in both axes both axis tions up to 4 522 11 518 31 10.5 16.5 21 25 33 21 15% in Widevaria- 6 mp * Bandwidth is defined as the width of the transmission curve in m p when the transmission is a specified fraction of maximum transmission. PRACTICAL APPLICATION OF INTERFERENCE FILTERS IN THE SPEKKER ABSORPTIOMETER It is intended to deal with the chemical results of a composite procedure in which these filters are used in later papers.The present discussion will cover their physical performance in view of the results outlined above. The sensitivity of the modified Spekker absorptiometer with the 500-watt lamp and the G.A.B. filters in both sides of the instrument was measured. Solutions of various concen- trations were prepared and their optical densities were determined relative to water by the zero setting m e t h ~ d . ~ The right-hand drum was then set 0.01 optical-density unit (1 drum division) from the balance position and the deflection on the galvanometer was noted. The sensitivity can then be expressed as mm per drum division, with the following results- Optical density . . .. . . 0.2 0.45 0.7 1.0 division . . .. .. .. 12 6 4 2 division . . .. .. .. 15 7 6 3 Sensitivity at 522 mp, mm per drum Sensitivity a t 439 mp, mm per drum The galvanometer used was a standard Cambridge spot galvanometer normally supplied and the sensitivity control on the instrument was used at its maximum position.A vital feature in filters for the absorptiometer is homogeneity in view of the method of measuring optical density. The accurately calibrated drum assumes perfectly even illumination over an area approximately 1.2 cm x 1 cm, followed by a filter uniform over a similar area. Lack of uniformity in the filter will cause errors in the optical-density measure- ment (for absorptiometers with photocell response for direct optical-density measurement this error would not arise). In assessing the possible chemical errors that may arise, however, consideration must be given to the absorption curve of the species being determined.If the curve is smooth and relatively level at the point of measurement and any filter variations are of wavelength rather than total optical density, errors in the final chemical result will be small. If, however, the absorbent has a complex absorption curve, such as permanganate , wavelength deviations could be serious. The modified Spekker absorptiometer was used with the right-hand filter being in turn either No. 3 or No. 4 in their four possible orientations. The left-hand filter was the remaining filter of the pair; the orientation of this filter is unimportant, as the left-hand side of the instrument is static for the two stages of reading. Indeed, it may well be that a simple gelatin filter will be adequate in future for reliable compensation of light-source fluctuation. Ten accurate standard solutions of oxidised potassium permanganate were measured withFebruary, 1958 J FOR THE CONTROL LABORATORY 111 each filter arrangement and the calibration graph of concentration against optical density was plotted.The calibration graphs obtained all had deviations from linearity, some as large as 0.02 optical-density unit. I t was found impossible, however, to relate these variations directly to those shown in the filters by the spectrophotometer. This may well be due to the complex nature of the permanganate, as already mentioned. In use, it has been found possible to select an optimum orientation of the filter and to evolve a correction graph or table. Time alone can prove the ageing characteristics of these filters, but it is to be hoped that such corrections will remain constant.In predicting ageing properties, an important difference from gelatin filters is that radiation that is not transmitted is reflected and not absorbed. Consequently, less energy is taken up by the filter and there is less likelihood of the nature of the filter being changed than with a gelatin filter, in which the chemical dye may dissociate after absorbing energy. CONCLUSIONS The work described makes possible practical consideration of composite spectrophoto- metric methods for the control laboratory, provided that the over-riding economics of such procedures are borne in mind. The problems of evaluation can be simplified, without resorting to calculating machines, by graphical means adapted for the particular problem. It is possible to convert the simple Spekker absorptiometer for use with interference filters so as to give complete freedom of choice for the wavelength of measurement. The inter- ference filters examined have been found suitable provided their mode of use is standardised. I thank Miss J. T. King-Cox, who assisted in the practical examination of the inter- ference filters, and P. H, Scholes, Esq., B.I.S.R.A., for helpful interest and comments. The work on graphical calculation was made possible by the tolerant co-operation of the staff of the I.N.O. (Sheffield) Drawing Office, who produced both net and “slide rule.” They also assisted by designing the flow-through water cells for the absorptiometer lamphouse. REFERENCES This paper is published by permission of the Admiralty. 1. Lingane, J. J., and Collat, J. W., Anal. Chem., 1950, 22, 166. 2. Weissler, A., Ind. Eng. Chem., Anal. Ed., 1945, 17, 695. 3. Hiskey, C. F., and Firestone, D., Anal. Chem., 1952, 24, 342. 4. B.I.S.R.A. Study Group, Special Report No. 55, The Iron and Steel Institute, London, 1956, p. 19. 5. Johnson, L. J., “Nomography and Empirical Equations,” J. Wiley & Sons Inc., New York, 1952. 6. Curry, D. R., Electroplating, 1952, 5, 397. 7 . Patrovskl, V., Chem. Listy., 1955, 49, 854; Chem. Abstr., 1955, 49, 13018f. 8. Jarrell, R. F., J. Opt. SOC. Amer., 1942, 32, 666. 9. See ref. 4, p. 4, method lb. Received April 16th, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300106

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

February, 19581 FOR THE CONTROL LABORATORY 111 Notes DERIVATIVES OF l-AMIN0-2-NAPHTHOL-4-SULPHONIC ACID AS REAGENTS FOR THE COLORIMETRIC DETERMINATION OF ZINC COMMERCIALLY available dye-stuffs derived from l-amino-2-naphthol-4-sulphonic acid have been examined as potential reagents for the colorimetric determination of zinc. One of these, chrome fast black CAT, is exceptionally sensitive for visual determinations. Methods of using chrome fast cyanine B (Clayton Dyestuffs Co. Ltd. ; Colour Index No. 202), chrome fast black CAT (Clayton Dyestuffs Co. Ltd. ; Colour Index No. 203) and Solochrome red ERS (Imperial Chemical Industries Ltd. ; Colour Index No. 652) are described and the concentrations at which some common metals interfere are given. The dye-stuff 1-( l-hydroxy-2-naphthylazo)-5-nitro-2-naphthol-4-sulphonic acid (Colour Index No.203) has been recommended by various workers as an indicator for the titration of metals, especially magnesium and zinc, with sequestering agents.1p2 7 3 9 4 This dye-stuff, together with the closely related 1-( 1-hydroxy-2-naphthylazo) -2-naphthol-4-sulphonic acid [chrome fast cyanine G (Clayton Dyestuffs Co. Ltd. ; Colour Index No. ZOl)], 1-(2-hydroxy-l-naphthylazo)-2-naphthol- 4-sulphonic acid [chrome fast cyanine B (Clayton Dyestuffs Co. Ltd.; Colour Index No. 202)] and 1-( 2-hydroxy-1-naphthylazo) -5-nitro-2-naphthol-4-sulphonic acid [Solochrome black A (Im- perial Chemical Industries Ltd.; Colour Index No. 204)], have been found t o be sensitive reagents112 NOTES [Vol. 83 for the colorimetric determination of very small amounts of zinc.Although they are not specific for this element, the fact that they are simpler to use than dithizone6t6 and more sensitive than the best of the styryl dye-stuffs7v8 may make them valuable in some circumstances. The trans- mission curves of these dye-stuffs and their zinc complexes have been measured on a Beckman DU spectrophotometer. With each i t was seen that the greatest difference between a blank and the zinc complex was at a wavelength at which the transmission of the blank had a low reading and that of the zinc complex a high reading. A more suitable reagent for use with the Spekker or similar type absorptiometer is 1-( 5-hydroxy-3-methyl- l-phenylpyrazolylazo)-2-naphthol- 4-sulphonic acid (Solochrome red ERS) the yellow colour of which is changed to a bright bluish red by zinc and so the absorption is higher when more zinc is present.The most sensitive of these dye-stuffs for use with an absorptiometer are C.I.203 and C.I.652; they will each determine zinc at a concentration as low as 0.01 pg per ml when a cell having a 4-cm light path is used. C.I.202 is less sensitive and may conveniently be used for the deter- mination of zinc at concentrations between 0-5 aind 4.0 pg per ml. For visual determination in Nessler glasses, the most suitable dye-stuff is C.I.203 and a solution having a concentration of zinc as low as 1 part in 200,000,000 can be distinguished from a blank solution. METHOD The methods of using the most satisfactory of the dye-stuffs follow. PROCEDURE- (a) Use of C.I.652 to determine 0.1 to 1.0 pg of Zn2+ per ml-Put 1 ml of a 0.1 per cent. solution of C.I.652 in a 25-ml calibrated flask and add 1 ml of a 2-5 per cent. solution of anhydrous sodium carbonate.After 5 minutes take readings for the solution against a blank solution in a Spekker absorptiometer in 1-cm cells, using an Ilford No. 604 green filter. (b) Use of C.1.203 to determine 0.1 to 1.0 pg ofZn2+per ml-Put 0.75ml of a 0.1 per cent. solution of C.I.203 in a 25-ml calibrated flask and add 1 ml of a 2.5 per cent. solution of anhydrous sodium carbonate. After 5 minutes take readings against a water blank in a Spekker absorptiometer in 1-cm cells, using an Ilford No. 608 red filter. (c) Use of C.I.202 to determine 0.5 to 4.0 pg of Zn2+ per ml-Place 2.0 ml of a 0.1 per cent. solution of C.I.202 in a 25-ml calibrated flask and a.dd 2 ml of a 2.5 per cent.solution of anhydrous sodium carbonate. Take readings against a water blank after 5 minutes in 1-cm cells, using an Ilford No. 608 red filter. ( d ) Use of C.I.652 to determine 0.01 to 0.1 pg of Zn2+ per ml-Put 0.5 ml of a 0.02 per cent. solution of C.I.652 in a 25-ml calibrated flask and add 0.5 ml of a 0.5 per cent. solution of anhydrous sodium carbonate. After 5 minutes take readings against water in 4-cm cells, using an Ilford No. 604 green filter. (e) Visual determination of 0.0025 to 0.01 pg of Zn2+ per ml-Add 1.5 ml of a 0-02 per cent. solution of C.I.652 or 1 ml of 0.02 per cent. solution of C.I.203 to 1-5 ml of a 0-5 per cent. solution of anhydrous sodium carbonate in a 100-ml Nessler glass.Add the sample under test and make up to 100 ml. Stir well and set aside for 5 minutes before comparing the colour with those of standards. The relative shades of colour remain unchanged for several hours. In all these tests the use of an excess of sodium carbonate solution over the amount specified does not spoil the results. Add the solution containing the zinc and dilute to the mark. Add the solution containing the zinc and dilute to the mark. Add the solution containing the zinc and dilute to the mark. Add the sample containing zinc and dilute to the mark. INTERFERING METALS The amount of metal (in pg per ml) indicated in the following Table gives a colour approxi- mately equal in shade and intensity to that given b y 1 pg of zinc per ml- C.I.201 C.I.202 C.I.203 C.I.204 C.I. 652 Mg2+ Cae+ Bag+ Sr2+ Cd2+ Zr'+ vo,p+ coa+ Ni2+ Hg2+ 2 > 1000 > 1000 > 1000 1 > 500 60 0.3 1 > 1000 3 200 > 1000 > 1000 4 > 500 80 0.8 1 > 1000 3 30 > 1000 > 1000 4 > 500 80 0.5 1 > 1000 4 200 > 1000 > 1000 2 > 500 80 0.8 1 > 1000 6 > 1000 > 1000 > 1000 3 > 500 80 0.8 0.6 30February, 19581 NOTES 113 The following metals form dark colours or precipitates that mask or prevent the zinc colour reaction- C.I.201 C.I.202 C.I.203 C.I. 204 C.I.652 Bea+ 250 250 250 250 260 Pbe+ 6 80 20 60 20 40 40 40 40 40 ;“3+ 20 20 20 20 20 It will thus be seen that very sensitive methods for the determination of zinc have been described; they may be of use under conditions in which interfering metals are absent. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. Belcher, R., and Spooner, C. E., Metallurgia, 1944, 30, 49. West, T. S., School Sci. Rev., 1952, 33, 25. Debney, E. W., Nature, 1952, 169, 1105. Harvey, A. E., Komarmy, J. M., and Wyatt, G. M., Anal. Chem., 1953, 25, 498. Fischer, H., and Leopoldi, G., Chem.-Ztg., 1940, 64, 231. Fischer, H., Mikrochemie, 1930, 8, 319. Krumholz, P., and Krumholz, E., Ibid., 1935, 19, 47. Wenger, P., Duckert, R., and Rieth, D., Helv. Chinz. Acta, 1942, 25, 406. H. F. LIDDELL SYLVIA M. WILLIAMS CHEMICAL DEFENCE EXPERIMENTAL ESTABLISHMENT PORTON, WILTS. Received June 27th, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300111

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

February, 19581 NOTES 113 Apparatus MODIFICATION TO APPARATUS FOR THE MICRO-DETERMINATION OF OXYGEN, NITROGEN AND HYDROGEN IN TITANIUM AND SOME OTHER METALS IN a recent publication by Booth, Bryant and Parker,l an apparatus is referred to for the micro- determination of oxygen, nitrogen and hydrogen in titanium and some other metals. This apparatus is based on a design described by Gregory, Mapper and Woodward2 and incorporates a “tree” for retaining samples before analysis. Fig. 1. Side view of apparatus The detachable “tree” usually contains twenty-four cone and socket joints, which must be This is time-consuming and, unless sealed with wax each time a batch of samples is examined. the joints are cleaned regularly, leakage may be experienced.114 APPARATUS [Vol. 83 An improved type of “tree,” shown schematically in Figs.1 and 2, has been designed and made in these laboratories. This modified component achieves the same analytical objective with only four cone and socket joints, thus minimising the possibility of leaks, and has the added advantage that it may be conveniently loaded in situ in about 5 minutes. Fig. 2. Top view of apparatus Samples, each weighing about 30 mg, and about 7 g of platinum wire are introduced separately into the twenty-four compartments by way of glass funnel A. Each compartment has a loose fitting glass plunger, B, with a soft-iron core, C, sealed into the stem, which enables it to be raised magnetically by means of a solenoid, D. When the plunger is raised, the sample is released from its compartment and guided into the crucible by a long silica funnel, previously lowered to within about 5 mm of the mouth of the crucible by magnetic actuation of a soft-iron counterweight.REFERENCES . ’ 1. 2. Booth, E., Bryant, F. J., and Parker, A., Analyst, 1957, 82, 50. Gregory, J. N., Mapper, D., and Woodward, J. A., Ibid., 1953, 78, 414. ANALYTICAL SECTION, RESEARCH DEPARTMENT IMPERIAL CHEMICAL INDUSTRIES LIMITED METALS DIVISION, KYNOCH WORKS WITTON, BIRMINGHAM D. A. SWANN D. A. WILLIAMS Received August 20th, 1957 A FLOW-THROUGH CELL FOR USE ’WITH SCINTILLATION COUNTERS THE beta particles emitted by sulphur-35 have such a low energy (0-167 MeV) that the use of the usual liquid-counting techniques is precluded, since the particles are all absorbed in the glass walls of the counter. Recourse is usually made to such methods as precipitation of the sulphur-35 as barium sulphate, collection on a planchette and (counting by means of an end-window Geiger - Miller tube, preferably with a mica window.During studies that involved the use of sulphiur-35 as sulphuric acid, it was found desirable to avoid the use of solid counting. It was thought that scintillation counting techniques might offer a means of achieving this end, since a cell t:o hold liquid could be constructed from phos- phorescent material, thus affording a means whereby the energy of the weak beta particles could be converted to light energy, which could be measured by means of a photomultipliertube. AFebruary, 19581 APPARATUS 115 cell was constructed of Pamelon (obtained from Isotope Development Ltd., Aldermaston, Berks.) and experiments were undertaken to determine the sensitivity of the method.It soon became apparent that, although the method had sufficient sensitivity for the purpose, it was not without difficulties. It was necessary to orientate the cell on top of the photomultiplier tube, within the lead castle, in exactly the same position each time a sample was counted. Optical contact had to be made between the cell and the tube; this involved dispensing silicone oil in semi-darkness. The major difficulty was due to the Pamelon cell being light sensitive; that is, when exposed to light and replaced within the counter, the cell gave an abnormal background count, which slowly decayed after several hours, only to increase again on further exposure to light.The cell can be placed in position within the lead castle and can be filled, emptied and washed without being removed from the counter. If necessary, the whole assembly can be dismantled quickly. These difficulties were overcome by using the flow-through cell shown in Fig. 1. Glass tubes Dainted I r matt b&ck I n Slit sealed with 3 inches of ad hesive tape Lead shot -3 Slit for syphon tube Film of silicone Aw'" oil ,Tin plate IIUi=! \Rubber bunn lnterlockinn lead counter Fig. 1. Diagram of apparatus In use, 40ml of the radioactive solution to be counted are poured into the cell by way of funnel C. After counting, 5ml of water are added, which cause a syphon to operate by way of tube A. Wash solution is then admitted to the cell by way of funnel C and the sequence of operations is repeated until the background count is reduced to normal.Tube B is merely a pressure-balance tube. The apparatus described has been found to be very convenient in use and with its aid linear relationships between counting rates and concentrations of sulphur-35 have been obtained. The results of a typical experiment are shown in Table I. The cell is emptied to waste, except for 1 or 2 ml.116 BOOK REVIEWS [Vol. 83 TABLE I RELATIONSHIP BETWEEN COUNTING RATE AND CONCENTRATION OF SULPHUR-% Sulphur-35 present, microcuries per 40 ml 0.000 0.375 0.750 1.125 1-500 1-875 2.250 2.625 3.000 Activity, counts per minute 165 843 1461 2175 2810 3460 4133 4750 5432 Activity less background, counts per minute 678 1296 2010 2645 3295 3968 4585 5267 - I thank the authorities of Wolverhampton and Staffordshire College of Technology, who pro- vided the facilities for this work, and also Mr. F. Scott for his helpful advice. WOLVERHAMPTON AND STAFFORDSHIRE WULFRUNA STREET ALBRIGHT & WILSON (MFG.) Co. LTD. COLLEGE OF TECHNOLOGY WOLVERHAMPTON and OLDBURY BIRMINGHAM S. GREENFIELD Received August 30th, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300113

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

116 BOOK REVIEWS [Vol. 83 Book Reviews ANALYTICAL MICROSCOPY: ITS AIMS AND METHODS IN RELATION TO FOODS, WATER, SPICES AND DRUGS. Second Edition. Pp. viii + 215. London: J. & A. Churchill Ltd. 1957. Price 25s. The first edition of “Analytical Microscopy” was published in 1923, and in the intervening years the accumulation of information and new methods (many developed by the author) has necessitated such extensive revision that this edition is almost a new book, although it incorporates the subject matter and illustrations of the original work. In a long and distinguished scientific career Dr. Wallis has made the subject of Microscopy as applied to the examination of Vegetable Drugs, Foods and Water Deposits peculiarly his own, and in this book he gives a practical exposition of a valuable adjuvant to Food and Drug Analysis.The treatment is practical throughout, most of the chapter headings referring to particular procedures, which are first described, and their application is then exemplified by reference to the examination of particular objects or commodities. Thus after an Introduction, the second chapter is headed Simple Methods of Preliminary Treatment and includes instructions for the examination of seeds and fruits and gives 77 diagrams of varieties, many being those of weeds liable to occur in agricultural produce. The same chapter also embraces descriptions of infesting insects as well as covering an account of powdered lucerne. Particularly valuable to the food analyst is the third chapter entitled Surface Preparations and Sections, because it is mainly devoted to copiously By T.E. WALLIS, D.Sc., F.R.I.C., F.P.S., M.I.Biol., Hon. F.R.M.S.February, 19581 BOOK REVIEWS 117 illustrated descriptions of the culinary herbs and their common adulterants. Here we have fine drawings of the macroscopical and microscopical characters of important commodities, notably mint and sage, and also of adulterants such as ailanthus, mulberry, phlomis and cistus. The following chapter, designated Sedimentation and Centrifugation, is concerned with deposits found in waters, including diatoms, algae, fungi, protozoa, water fleas, rotifers, moth scales, worms, mites and insects, all finely illustrated. So the book goes on, giving accounts of the microscopy of commercial starches, cereals, honey, infesting mites of flour and fodder, all complete with diagrams.In a chapter entitled Micromorphology allusion is made to the author’s discovery that traces of dissolved silica influence the crystalline structure of the scale that is formed when natural waters are boiled. A more complete account of this phenomenon was recently given by Dr. Wallis in his Presidential Address to The Royal Microscopical Society and has now been published (March, 1957) in that Society’s Journal. Although the book covers the examination of a large number of commodities, and especially foods, it should be emphasised that no attempt has been made in it to provide a comprehensive atlas of drawings to illustrate all substances requiring microscopical examination for their asess- ment.The aim has been to tell the reader how to prepare material for microscopical observation and to offer guidance in the selection of clearing and mounting agents and in methods of manipula- tion. The three concluding chapters treat of more specialised Microscopy and deal with Measure- ment and Drawing, Numerical Values, including palisade ratios, the stomata1 index, vein islet numbers and, lastly, a chapter on Quantitative Microscopical Analysis. There are two appendixes, the one giving numerical data, the other formulae for reagents mentioned in the text, and the book closes with a useful classified bibliography and a comprehensive index. One ventures to remark that the value of the last two chapters would have been enhanced by including a wider selection of references to the original scientific literature.Modesty on the part of the author doubtless has something to do with this omission, since much of the work is his own. Both author and publisher are to be congratulated on the fine production of this handsome volume, which is offered at a very modest price. I understand that the preparation of the manu- script and drawings has occupied most of the author’s leisure during the last 6 years; many fellow scientists will have cause to be grateful to Dr. Wallis for spending his time so generously to their profit. N. L. ALLPORT SPOT TESTS IN ORGANIC ANALYSIS. By FRITZ FEIGL, Eng., D.Sc. Translated by RALPH E. OESPER, Ph.D. Fifth English Edition. Pp. xx + 616, Amsterdam, London and New York : Elsevier Publishing Company; London : Cleaver-Hume Press Ltd. 1956.Price $10.00; 55s. In the first three English editions of Feigl’s “Spot Tests” the application of spot tests to organic analysis was included as a comparatively small, but growing, section. Three years ago the increase of material warranted the division of “Spot Tests” into two volumes, the second of which dealt solely with organic analysis (for review see Analyst, 1954, 79, 722). This volume has now been expanded into a separate work. The arrangement of the contents-survey of the subject, technique, preliminary tests, tests for functional groups and for individual compounds, technical applications-remains the same as in the previous edition. Much of the text has, however, been rewritten, and the number of tests reported has been increased by nearly half.As before, full practical details are given of each test, with limits of identification, known interferences and, when possible, the reactions underlying the test. Each test is well documented, and a new chapter gives nearly a hundred references to other general studies of organic spot tests. The translation is a model of clarity, and there is an admirably detailed subject index. The expansion of the book in such a short time is a measure of the increase of information on spot tests applied to organic compounds. Much of this has come from Professor Feigl and his co- workers, but, if the subject arouses the same interest in other workers as inorganic applications of spot tests did, we can expect a much greater increase in the future.It is pointed out in the book how much remains to be done in the way of improving tests, of adapting known reactions as spot tests and of finding new tests. Workers who are attracted to this field will have cause to be grateful to Professor Feigl for the groundwork he has done. Analysts will, in the meantime, find in the book much useful practical information that will affect classical organic analysis as well as the particular field of spot tests. This is highly likely. DAVID W. WILSONBOOK REVIEWS [Vol. 83 By L. SAUNDERS and R. FLEMING. Pp. x + 257. London: The Pharmaceutical Press. 1957. Price 27s. 6d. The authors have undertaken the task of attempting to remedy the deficiencies in mathematical and statistical knowledge and understanding often encountered in students and graduates of pharmacy, biology and allied branches of science.In general, the mathematical chapters of the book follow the conventional pattern and justi- fiably, in view of the readers for whom they are intended, dispense with the niceties of mathe- matical rigour, appealing more to the reader’s intuition and intelligence. Unfortunately, however, the conventional appi-oach adopted perpetuates the lack of balance in mathematical material that is too often found in books intended for such readers, particularly at the expense of practical computation methods. Thus a whole chapter is devoted to the little used topic of trigonometry, whereas the very important subject of simultaneous linear equations is relegated to a two-page appendix on determinants.Similarly the fitting of equations to experi- mental results is dealt with in a chapter of only seven pages, without any indication even of the existence of simple numerical methods for curve-fitting. Deficiencies in the treatment of com- putational methods are also evidenced by the lack of any discussion of interpolation, of the solution of simple non-linear equations by successive approxirnation and of elementary numerical integration and differentiation. For example, on p. 40, the ridiculous statement is made that the partial sums of a convergent series never exceed their limit. This is unfortunately almost immediately followed on p. 42 by the series 7r = 4(1 - + - . . . .), which is anyhow only conditionally convergent and, if summed in the order given, would yield partial sums alternately greater and smaller than their limit.Insufficient warning is given that power series have their limitations, and indeed on p. 84 the dangerous statement is made that any function can be so represented. The statistical chapters are moderately successfiil in the treatment of probability and are well supplied with useful practical examples. The object of randomisation and experimental design, so important in biology and biological assay, is, however, never clearly explained or even stated. Such discussion as does appear is more likely to confuse than help. Thus, for example, on p. 169 it is contended that randomisation can be ensured by using a Latin square. This statement is immediately followed by a Latin square, obviously systematic in layout, and later on p.173 by a 4 x 6 table of results, claimed to be a Latin square and analysed as if no design had been employed. Sometimes generality of method is claimed or supplied when no such generality exists. In addition to the alleged generality of the power series already discussed there is, for example, the claim on p. 175 that the underlying normal tolerance distribution is assumed in the interpretation of all quanta1 assays. Similarly on p. 13 no warning is given that dimensional analysis generally leads to undetermined functions rather than constants ; incidently the example given would be such an instance, but for the non-dimensional assumption of dependence on inverse tube length. As the symbolism of mathematics is one of the main hurdles the non-mathematical practical reader has to overcome, a list of symbols would have been helpful particularly, as near-synonyms (e.g., C and S for summation and a Gothic p, p and P for probability) are often employed.On the whole the book is rather patchy in quality and would probably have been more success- ful if a less ambitious coverage had been attempted. The lack of practical numerical methods is likely to mean that, even though readers may learn to formulate their problems in mathematical terms, they will be ill equipped to solve them practically. MODERN CEREAL CHEMISTRY. 118 MATHEMATICS AND STATISTICS FOR USE IN PHARMACY, BIOLOGY AND CHEMISTRY. The treatment of series is marred by several inaccuracies. J. P. R. TOOTILL By D. W. KENT-JONES, Ph.D., B.Sc., F.R.I.C., and A.J. AMOS, Ph.D., B.Sc., F.R.I.C. Fifth Edition. Pp. x + 817. Liverpool: The Northern Publishing Co. Ltd. 1957. Price 105s.; $15.25. Cereals furnish staple foods for the inhabitants of all parts of the world and contribute sub- stantial items to the diet of domestic animals and poultry. In consequence they and their products have come under the critical eye of all concerned with nutrition, and the resulting picture-some- times white and sometimes brown-has not always been free from distortion. It is therefore important that the latest information should be readily available so that what is seen may accord with the facts. “Modern Cereal Chemistry” sets out to supply this information, and previous editions have appeared a t approximately ten-year intervals, each of wider scope than its predecessor and adequately justifying the term “modern.” The general arrangement of this new edition is substantially the same as that of the last, which became so familiar to all food chemists and which was recognised as the standard work on theFebruary, 19581 BOOK REVIEWS 119 subject. The old chapter on Some Physico-Chemical Aspects of Flour has been deleted-parts being incorporated elsewhere-and a new one included on Infestation by Insects and Mites.Additional material appears in every chapter, and as a whole the increase is approximately 25 per cent. During the decade since the last edition much important work has been published, and this has now been incorporated with critical appraisals. One is conscious that the long practical experience of the authors has been brought to bear on their comments.No better summing-up of this book can be given than by saying that the dust-cover is fully justified in claiming that “this new edition is probably the most comprehensive book on the subject and one which no modern cereal chemist should be without. Covering the wide subject so com- pletely it is invaluable also to the miller, the baker, and the many others whose work deals with cereals, and indeed to all those concerned in food technology.” J. R. NICHOLLS QUANTITATIVE INORGANIC ANALYSIS : A LABORATORY MANUAL. F.R.I.C., F.Inst.F., and A. J. NUTTEN, B.Sc., Ph.D., F.R.I.C. Butterworths Scientific Publications. 1955. Price 25s. By R. BELCHER, D.Sc., Ph.D., Pp. viii + 337. London: The authors have written a book that is designed primarily as a practical course for teaching quantitative inorganic analysis in Universities and Technical Colleges.The main sections deal with laboratory apparatus and technique, gravimetric, titrimetric and colorimetric analysis, and a few applied analyses. As might be expected, the range of determina- tions in all sections is limited, there being no inclusion in the gravimetric section, for example, of determinations of the less-common elements. Here the authors are wise, as there is never time for such analyses in the ordinary college course and the better-known elements and radicles can provide ample opportunity for the beginner to gain experience. Minor consideration is given to the so-called physical methods of analysis in spite of their importance in industry today, but the authors are right in maintaining, as they do in their preface, that in a University course classical analysis must always predominate.The practical instructions provided for the various determinations are supported by an adequate amount of general theory, by notes on points of special practical importance, and by a selected number of references, but a detailed physico-chemical treatment is left to the lecture course or to the student’s own study of T. B. Smith’s well-known text-book. In the gravimetric section of this present book, determinations other than those dealt with are often mentioned. This information, although of interest, is more suitable for a text-book than for a laboratory manual such as this purports to be.In its place, I would have preferred to see an adequate treatment of the effect of variations in temperature on the measurement of volumetric solutions, a subject that receives only incidental mention. Much useful instruction is supplied for the student, but there are many points on which I disagree with the authors. When aperiodic balances are not available, the method of swings should, in my opinion, always be insisted upon and not the method of moving the rider along the 0. 1-mg notches until deflections coincide ; before being weighed, weighing bottles should never be handled by bare fingers, and the weighing-out of an amount of substance required for an exactly decinormal solution should be discouraged. Neither have I found i t necessary to invert a stoppered calibrated flask, filled to its mark, thirty to forty times to ensure thorough mixing of the contents (p.141). The preparation of constant-boiling hydrochloric acid is not particularly troublesome (p. 157); indeed, it is an excellent method for preparing standard acid and one well withinthe capacity of a second-year student. A curious omission in the text is that no mention is made of the need for rinsing burettes and transfer pipettes with the solution to be used before a measurement is carried out. The short chapter on desiccators emphasises the difficulties associated with the proper use of desiccants, but, as specified cooling times are seldom laid down, the student may find i t difficult to know when to remove his crucible “as soon as it is cool,” or, p.53, to “weigh to constant weight, when it has cooled to the temperature of the balance.” The inclusion of a description of the Hartley funnel, an improved type of Buchner funnel that deserves to be better known and more widely used than it is a t present, is a good feature. So is the direction of the reader’s attention to the appreciable solubility of the nickel - dimethylglyoxime complex in the hot mother liquid and the need for cooling before the filtration is carried out. The description of the silver reductor for determining iron is a welcome inclusion in a text of this kind, but the instruction to use a glass tube of the same dimensions as those of the Jones reductor nullifies one of its advantages over the latter. I agree with the authors’ Many points in the text will commend themselves to the reader.120 BOOK REVIEWS considered opinion “that adsorption indicators ha,ve been somewhat overrated” and “only very few give good end-points.” Some of the statements in the text are open to question : thus on p.94 we are told to “Dissolve the residue remaining in the beaker in the minimum volume of hot distilled water and transfer it quantitatively to the filter collecting the filtrate in a beaker” ; that solid ferrous ammonium sulphate is stable (p. 206) ; that the precipitate obtained with ferric ions and N-benzoylphenylhydroxylamine must be ignited to ferric oxide (cf. Shome, Analyst, 1950, 75, 27); that MgHPO, does not go to the pyrophosphate on ignition (p. 77); that the ferric iron - cupferron complex may be weighed after drying at a suitable temperature (p.70) ; and finally, that felspar should be heated to 250” C for 1 hour before a determination of the alkalies in it is made (p. 316). Typographical mistakes are few in number, but; the formulae given for uranyl and zinc acetates on p. 88 are wrong; the one-hole rubber bung of the text on p. 167 has two holes shown in the diagram on p. 168, and the 24(NH4),Mo04 of the equation on p. 174 should be 12(NH4),Mo0,. The figures relating to the equivalent weights of certain substances might well be checked by re-calculation, for they do not always agree with those calculated from the 1952 atomic weights given in the appendix. In the description of the determination of the alkalies in felspar a paragraph seems to be out of its order in the text, and the Kalignost solution to be added is not described; presumably it is the reagent described on p. 96. In order to complete the determination, reference must be made to p. 96 and not to the p. 90 indicated. Fifty-four review questions are provided after the text, whereby the student can test his knowledge of the subject. One wonders what hi.s answer will be to the question “How many methods, based on the method of completion, do you know for the determination of iron?” The book will be of interest more to the teacher of analytical chemistry than to the practising analyst, to whom, I imagine, it will not have a wide appeal. It is a reflection on the times and on present-day teaching that these authors, in urging the student-reader to present his experimental results in such a way that they can be assimilated at a glance, can tell him candidly the melancholy truth. when they write “It should be borne in mind that the supervisor rarely has time to read carefiilly through the notebook. . . .” Students have little success in using many of these indicators. L. S. THEOBALD

ISSN:0003-2654    

DOI:10.1039/AN958830116b

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

120 BOOK REVIEWS Erratum DECEMBER (1957) ISSUE, p. 813, 3rd h e from foot of page. For “17 ml” read “170 ml”.

ISSN:0003-2654    

DOI:10.1039/AN958830120c

出版商:RSC    

年代:1958

数据来源: RSC