摘要:

296 PANICKER AND BANERJEE : THE RAPID DETERMINATION OF CARBON AND [VOl. 83 The Rapid Determination of Carbon and Hydrogen in Highly Volatile Corn-bustible Organic Liquids BY A. R. PANICKER AND N. G. BANERJEE (Central Fuel Research Institute, Jealgora, Bihar, I n d i a ) A method is described for the rapid determination of carbon and hydrogen in organic liquids of low boiling-point, such as benzene, toluene, gasoline and other petroleum distillates. The method is also applicable to solutions in solvents of low boiling-point and to mixtures of organic liquids or solids of low and high boiling-points. As such samples are too volatile to be weighed in open containers, a special weighing tube has been designed that avoids the difficulty of using a sealed weighing capillary. Moreover, the evapora- tion of the sample is carried out in a current of nitrogen; by accurate control of the rate of flow the evaporation can also be controlled. After the volatile portion of the sample has been carried over and burnt, the non-volatile portion, if any, is burnt a t an elevated temperature in a current of oxygen passed through the sample tube.The test takes about 45 minutes and gives accurate results. Replicate determinations generally lie within k0.14 per cent. for carbon and -10.07 per cent. for hydrogen. THE determination of carbon and hydrogen in highly volatile organic liquids is important in the analysis of petroleum, coal, tar hydrogenation products and so on. One of the earlier methods is that of Liebig. In this method the sample is taken in a bulb, the open end of which is drawn out into a capillary that is sealed before it is weighed.The sealed tube is placed in a boat and is introduced into a combustion tube after the sealed end has been broken. The sample is burnt in a current of oxygen as usual. There are several drawbacks to this method. If the boiling-point of the substance is very low, there is the danger of an explosion caused by rapid evaporation and the resulting back pressure. Also, specks of carbon may remain, which cannot be burnt off without smashing the bulb. Clarke1 has suggested a method in which one end of the combustion tube is drawn out and a ground-glass socket fused to it. The liquid to be analysed is weighed in a U-tube, to one side-tube of which is fused the corresponding ground-glass cone.The stopcock in each arm is greased with soft paraffin wax. The sample is weighed in the U-tube and carried into the combustion tube by the oxygen stream. solution, a part of the sample remains in the U-tube and causes low results. Another point worthy of mention is that, if the sample is highly volatile, some of the air inside the U-tube may be expelled by the vapour from the liquid before the stopcocks are closed. This would give high results. When highly volatile liquids are used the risk of explosion is also present. Many other workers, e.g., Reid,2 Brunn and Faulconer3 and Avery and Hayman! have suggested similar methods, but none is entirely free from the defects previously mentioned. Levin and Uhrig,6 Baily,g Aluise' and Sevag8 have suggested improvements, but in all these methods the difficulties of weighing the sample and the danger of explosion are not completely overcome.Subsequent workers have endeavoured to eliminate these disadvantages. Combustion is reported to take place smoothly. The main drawback to this method is that, if the sample contains any solid The essential features of the proposed method are- (i) easy and accurate weighing of the sample; (ii) the danger of explosion is avoided; (iii) smooth combustion; (iv) accurate analysis of highly volatile pure organic liquids or their mixtures, or solutions of solids in such liquids; (v) simplicity of apparatus and ease of manipulation.&fay, 19581 HYDROGEN I N HIGHLY VOLATILE COMBUSTIBLE ORGANIC LIQUIDS METHOD APPARATUS- 297 The weighing tube consists of a I-mm Pyrex-glass capillary of external diameter 6 mm.Three bulbs are blown in the capillary, as shown in Fig. 1. The largest bulb, C, is 35mm long and has an external diameter of 13 mm, and the two smaller bulbs, A and B, each have an external diameter of 9 mm. The bulb A, in which the sample is weighed, holds 0.15 to 0-2 g. 13 mm Dia. 9 mm I .mm capillary Fig. 1. Weighing tube The purification trains consist of a flow meter to measure the rate of flow of gas, a sulphuric acid bubbler, an empty bubbler to trap the acid spray, a U-tube packed with Carbo- sorb soda asbestos and a U-tube packed with anhydrone to remove carbon dioxide and moisture, respectively, from the gas before it enters the combustion tube. Two purifying trains are used, one to purify oxygen and the other to purify nitrogen.Nitrogen " - in arbosorb AnhYdrone \ , FIOW I meter Sulphuric Empty bubbler Anhydrohe acid bubbler Arrangement of apparatus with purifying and absorption trains acid bubbler Fig. 2. 3; inches -4 b0 inches+9f inchesAk-18f inches-jF81 inches Fig. 3. Packing of the combustion tube: zones A, 1 to 2 inches of oxidised copper gauze; zone B, 16 inches of copper oxide; zone C, 5 inches of red lead; zone D, 3 inches of silver gauze The absorption train consists of a U-tube packed with anhydrone, a Midvale tube packed with Carbosorb soda asbestos, a guard U-tube packed with Carbosorb soda asbestos in one limb and anhydrone in the other and a sulphuric acid bubbler to allow observation of the rate of flow of the exit gas.The temperature of the first furnace can be raised from 30" to 600" C in 30 minutes, the second and longest furnace is maintained at 800" C and the third at 350" C. A silica combustion tube 50 inches long, the first 20 inches of which are transparent, is used. A three-unit Liebig furnace is used for the combustion.298 PANICKER AND BANERJEE THE RAPID DETERMINATION OF CARBON AND [VOl. 83 The combustion tube is closed at the exit end by a rubber bung with a glass tube passing through it for connection to the absorption train and at the inlet end by a rubber bung with two holes. Through one hole a glass tube is fitted for connection to the oxygen purifying train and the weighing tube is fitted through the other, as shown in Fig. 2. This also serves as the nitrogen inlet to the combustion tube. The complete packing of the combustion tube is shown in Fig.3. The red lead is prepared by making a paste of lead peroxide with water, drying it, cutting it into cubes and then heating it at 400" C for 4 to 5 hours in a muffle furnace. PROCEDURE- To start a determination the second and third furnaces are switched on and oxygen is passed through the combustion tube. The empty weighing tube is fitted in position as shown in Fig. 2 and nitrogen is passed through it. When the temperatures of the furnaces have risen to the required values, the rubber bung at the inlet of the combustion tube is taken out and the weighing tube is detached. The combustion tube is now closed with a rubber bung carrying a glass tube connected to the oxygen supply.The flushing of the apparatus is continued. The combustion tube is ready for use when the weighed absorption train connected to the exit end for 45 minutes does not show an over-all gain in weight of more than 0.0005 g. The absorption train is then re-weighed and connected to the combustion tube. The weighing tube is placed in a cradle hung from the pan of a balance and weighed empty. The capillary of bulb A is then dipped into the sample. By mild suction applied at the other end an amount of the sample just sufficient to fill bulb A is drawn in. The tube is then brought to the horizontal position and the ends are wiped clean and dry. It is again weighed and the difference gives the weight of the sample. After weighing, the tube is tilted and tapped lightly to transfer the liquid to bulb C.The two-holed rubber bung is now slipped on to the longer arm of the sample tube and the whole is fitted into the com- bustion tube as before. The nitrogen supply is connected to the sample tube and oxygen to the other inlet tube. The flow of nitrogen is started first, followed almost immediately by oxygen. The flow of nitrogen is adjusted so that the sample evaporates smoothly and the oxygen flow is set at 150ml per minute. For liquids with a low vapour pressure at ordinary temperatures, the flow of nitrogen is increased and, if necessary, the first furnace is drawn over the sample tube and switched on. The temperature is adjusted to about 30" to 50" C below the boiling-point of the liquid. With a little practice, evaporation and combustion can be carried out smoothly.In about 30 to 35 minutes the whole of the volatile portion of the sample evaporates. If any residue remains in the sample tube, the temperature of the first furnace is raised rapidly to 500" C, the flow of nitrogen stopped and the oxygen supply connected to the sample tube. In $about 5 minutes the whole of the carbonaceous matter vanishes. The tube is flushed for a further 5 minutes with oxygen at the increased rate of 250ml per minute. The absorption tubes are then detached and left to cool. One precaution to be observed is that, when the evaporation of the sample is carried out at an elevated temperature, the exposed portion of the combustion tube between the first and second furnaces should be kept heated to a temperature above the boiling-point of the liquid, so that no condensation takes place in this region.This can be done either by wrapping the exposed portion of combustion tube with a heating tape or by playing a small flame on it. During cooling, the pressure in the absorption tubes falls below atmospheric. Therefore, after being wiped clean, the tubes are opened momentarily to the atmosphere to equalise the pressures. The gain in weight of the anhydrone tube gives the water, and the combined gain of the Midvale and guard tubes gives the carbon dioxide formed. From these weights carbon and hydrogen are calculated. A correction to the hydrogen figure is required if moisture is present in the sample. They are then weighed. RESULTS Some of the results obtained by using the proposed method are shown in Table I.It can be seen that for pure compounds the determined results compare very well with the theoretical. For hydrocarbons, such as kerosine and decahydronaphthalene , it is not possible to deduce theoretical values, as impurities are present in the sample, but in these instances the sums of the carbon and hydrogen contsents are very nearly 100 per cent.May, 19581 HYDROGEN I N HIGHLY VOLATILE COMBUSTIBLE ORGANIC LIQUIDS TABLE I ANALYSIS OF HYDROCARBONS BY THE PROPOSED METHOD Carbon ------7 Sample found, theoretical, % % Benzene . . 92-23 92.26 92-25 92-25 92-39 91.18 91.22 90.38 84.59 85.55 85.88 85.6 1 Toluene . . 91.26 91.26 Xylene . . 90.31. 90.50 n-Heptane . . 84.65 84.50 cycZoHexane . . 85.70 85.63 Hydrogen found , theoretical, Carbon Sample found, 7.77 7-74 Kerosine 85.70 7.74 85-59 7.41 85-57 7.79 Decahydronaphthalene 87.78 8-81 8-74 87.88 8-81 rH134 87.54 8-77 87.47 9.55 9.50 H132 87.00 86-88 15-50 Neutral H133(B) 87-12 87-28 H134(B) 88-08 88.06 H133 88.06 87.92 H132(A) 87.08 Oil 1 86.96 9.69 15-54 15.61 14.33 14.37 14.31 14.47 14.36 & % % % 299 Hydrogen found, 14.35 14-38 14-34 12.31 12.30 11.61 11.62 12-96 13.03 12.30 12.26 11-71 11.79 12.10 12.12 12.91 13.05 % We thank the Director, Central Fuel Research Institute, for his encouragement and kind permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. 6. 5 . 8. Clarke, L., J . Amer. Chem. SOC., 1912, 34, 746. Reid, E. E., Ibid., 1912, 34, 1037. Bruun, J. H., and Faulconer, W. B. M., Ind. Eng. Cham., Anal. Ed., 1936, 8, 315. Avery, S., and Hayman, D., Ibid., 1930, 2, 336. Levin, H., and Uhrig, K., Ibid., 1937, 9, 326. Baily, J. R., Ibid., 1933, 5, 171. Aluise, V. A., Ibid., 1938, 10, 57. Sevag, M. G., Ibid., 1929, 1, 16. Received August 22nd. 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300296

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

May, 19581 HYDROGEN IN HIGHLY VOLATILE COMBUSTIBLE ORGANIC LIQUIDS The Determination of Copper in Sea Water, Silicate Rocks and Biological Materials BY J. P. RILEY AND P. SINHASENI (Department of Oceanography, The University of Liverpool) The use of 2:2’-diquinolyl in n-hexanol as a specific reagent for the extraction and spectrophotometric determination of copper in sea water is recommended. The colour of the cuprous - diquinolyl complex can be stabilised by means of hydroquinone. The method gave a coefficient of variation of 2-5 per cent. with sea water containing 27.0pg of copper per litre. The method has also been used for the determination of small con- centrations of copper in silicate and carbonate rocks and in biological materials. The U.S. Geological Survey granite G1 and diabase W1 have been found to contain 15.9 f 0.4 and 121 -J= 3 p.p.m. of copper, respectively.299 COPPER is an important trace constituent of sea water, in which it occurs in amounts variously reported as in the range 0.01 to 0.47 pg-atom of copper per litre. Chow and Robinson1 have reviewed the literature published up to 1952 on the occurrence and determination of copper in sea water. It is probable that all the earlier and much of the later published work is300 RILEY AND SINHASENI : THE DETERMINATION OF COPPER IN [Vol. 83 unreliable, owing to either faulty analytical techniques or contamination of the samples by the sampling apparatus or storage vessels. Suchi effects must be the explanation of the very high copper contents reported by certain workers. Most of the published data are for surface waters, and, although they are of value in biological studies, they give little idea of the copper content of deep ocean water.2 Copper is strongly concentrated by certain marine organisms and organic detritus, and this adsorption can cause considerable local variations in its concentration in sea water.Atkins3 has found a regular annual cycle for copper in the surface waters of the English Channel. Copper occurs in haemocyanin, the respiratory pigment of the blood of many marine invertebrates, which is analogous to the haemoglobin of the blood of mammals. It plays an essential part in the setting of oysters and acts as an oxidation catalyst in certain enzyme systems. In the course of geochemical studies in these laboratories of the inter-relationships of copper in sea water, marine sediments and marine organisms, a sensitive and specific method for its determination was required. The concentration of copper in sea water is too small to be determined directly, and in most of the more recent investigations it has been separated by solvent extraction with either dithizone4s5 or sodium diethyldithiocarbamatels6~7 and then determined photometrically.Both these reagents are extremely sensitive, but neither is specific for copper; by the use of suitable masking agentsssg they can be made more selective. Within the last decade, a number of reagents specific for copper have been introduced, mainly based on 1 : 10-phenanthroline1° y 1 l #l2 9 1 3 and 2 : 2’-diquinolyl.14 p 1 5 9 1 6 9 1 7 J 8 P These compounds react with monovalent copper to form strongly coloured cationic complexes that are soluble in certain organic solvents.EXPERIMENTAL In order to obtain a reasonably sensitive and completely specific method for the deter- mination of copper in sea water, the use of 2 : 2’-diquinolyl was investigated. This reagent has been used for the determination of copper in natural waters,20921 which are generally at least one order of magnitude richer in copper than is sea water. Its sensitivity for copper is about two-thirds of that of most of the substituted phenanthrolines, but it is cheaper and more easily obtainable. The majority of investigators have used isoamyl alcohol as the solvent for diquinolyl, although trichloroethatnol has also been used.21 In order to obtain sufficiently high optical densities, large volumes of sea water (up to 1 litre) must be taken for analysis, and it was apparent that isoamyl alcohol (solubility 2.67 g per 100 ml at 22” C) was too soluble to be used for the extraction in this instance.Several other solvents were tested and it was found that n-hexanol (solubility 0.5 g per 100 ml) was a suitable solvent for the extraction of the cuprous - diquinolyl compliex from sea water. The optimum conditions of pH for the extraction of the cuprous - diquinolyl complex with ut-hexanol were next investigated. Aliquots of distilled water (300 ml) were enriched with 10 pg of copper and 5 ml of 25 per cent. hydroxylamine hydrochloride solution were added, and then the solution was adjusted to the desired pH with sodium acetate and acetic acid.A double extraction with two 5-ml portions of 0.03 per cent. diquinolyl solution was then carried out. The optical density of the combined organic extracts was measured at 640 rnp in a 4-crn cell. It was found that below pH 2-7 practically no copper was extracted, but, above pH 4-0, copper was completely extracted, and in all subsequent work extractions were made at a pH between 4-3 and 5.8. The efficiency of diquinolyl for the extraction of copper, in common with that of many other organic complexing agents, decreases markedly with the copper concentration. Hence, although 1 p.p.rn. of copper can be extracted with 99 per cent. completeness by a single extraction, at a level of 0.01 p.p.m., two extractions are necessary to separate 97 per cent.of the element. For the determination of copper in sea water, three extractions with 8, 3 and 3 ml of diquinolyl reagent were used and gave over 99 per cent. recovery of added copper. In order to ensure that all the remaining copper was in the cuprous state, more hydroxylamine hydrochloride solution was added before the second extraction. In the early stages of this work, with small amounts of copper, fading of the colour of the cuprous - diquinolyl complex was troublesome. The optical densities of the solutions sometimes decreased by as much as 3 per cent. per hour; fading was even more rapid in sunlight. Analogous fading has been encountered with cuprous - phenanthroline deriva- tivesll and is attributed to oxidation of the complexes by atmospheric oxygen.This&hy, 19581 SEA WATER, SILICATE ROCKS AND BIOLOGICAL MATERIALS 301 oxidation can be completely prevented by the addition of 0-5 ml of 1 per cent. hydroquinone solution to the n-hexanol extracts. Guest22 has used diquinolyl for the determination of copper in rocks. Almond23 claims coefficients of variation of about 16 and 6 per cent., respectively, for field and laboratory methods for the determination of copper in rocks in the range 38 to 300 p.p.m. In Almond's procedure the copper in the sample is brought into a soluble form by fusion with potassium pyrosulphate and then determined with diquinolyl. We have found that appreciably higher results were obtained if silicate rocks were opened up by digestion with hydrofluoric and nitric acids instead of by pyrosulphate fusion.This is probably because the fusion process attacks only the sulphide and oxide minerals, if any, in the samples, and will not dissolve copper contained in the silicate lattice. In the work described, complete dissolution of the copper, whatever its mode of combination, has been ensured by evaporating the rock sample to dryness with a mixture of hydrofluoric and nitric acids. The residue is then fused with potassium bisulphate, which also helps to free the residue from nitrate and fluoride. METHOD All solutions and aqueous reagents should be prepared with water that has been distilled from an all-glass or silica still. REAGENTS- DiquinoZyZ reagant-Dissolve 0.03 g of 2 : 2'-diquinolyl in 100 ml of n-hexanol that has been redistilled from sodium hydroxide.Hydroxylamine hydrochloride solution-Dissolve 25 g of analytical-reagent grade hydroxyl- amine hydrochloride in about 80ml of water, filter and dilute to 100ml. If appreciable amounts of copper are present in the reagent, extract it with 10-ml portions of a 0.01 per cent. solution of dithizone in carbon tetrachloride until there is no change in the colour of the dithizone. Extract the solution with carbon tetrachloride until all colour has been removed. Sodium acetate bufeer solution, N-Prepare a solution containing 136 g of sodium acetate trihydrate per litre. If the reagent contains more than a trace of copper, extract with dithizone as described above. Ethanolic hydroquinone solution-Prepare a 1 per cent.w/v solution of hydroquinone in redistilled ethanol. Standard copper sol.ution--Weigh 0.1 g of Specpure or electrolytic copper into a silica basin. Dissolve it in 3 ml of concentrated nitric acid, add 1 ml of concentrated sulphuric acid and evaporate under an infra-red heater until dense white fumes are evolved. Allow to cool and then dissolve the residue in distilled water and dilute to 100 ml. This solution, which contains 1 mg of copper per ml, is used for the preparation of the working solutions, which contain either 2 or 10 pg of copper per ml and should be prepared freshly as required. TREATMENT OF APPARATUS- Clean the digestion flasks and the separating funnels that are to be used for the extraction of the cuprous - diquinolyl complex by setting them aside overnight filled with a (1 + 1) mixture of concentrated nitric and sulphuric acids.Empty them and rinse several times with distilled water. PROCEDURE FOR DETERMINING COPPER IN SEA WATER- Filter the sample through a No. 5 sintered-glass funnel. Transfer 900 ml of the filtrate to a 1-litre separating funnel and add 5 ml of 25 per cent. hydroxylamine hydrochloride solution and 10 ml of N sodium acetate buffer solution. Shake the solution with 8 ml of diquinolyl reagent for 5 minutes and then allow the phases to separate. Run the lower aqueous layer into another separating funnel, add 2 ml of 25 per cent. hydroxylamine hydrochloride solution and re-extract for 3 minutes with a further 3 ml of diquinolyl reagent. Separate the aqueous phase and again extract with 3ml of diquinolyl reagent.Combine the n-hexanol extracts in a 10-ml calibrated flask containing 0.5 ml of 1 per cent. ethanolic hydroquinone solution and dilute to the mark with n-hexanol. Measure the optical density of the solution at 540 mp in a 4-cm cell. Determine the reagent blank in the same manner with water distilled from a silica still. Prepare a calibration curve with 10 and 20p.g of copper added to 900ml of met al-free water.302 [Vol. 83 PROCEDURE FOR DETERMINING COPPER IN SILICATE ROCKS- Add 2 ml of concentrated nitric acid, 15ml of 40 per cent. hydrofluoric acid and set the covered crucible aside overnight on a water bath. Evaporate to dryness on a water bath. Fuse the residue with 1.5 to 2 g of fused potassium bisulphate at dull red heat for 5 minutes. Dissolve the fused cake by warming on a water bath with 100 ml of water containing 1.5 ml of concentrated hydrochloric acid.When cold, transfer the solution to a 250-ml calibrated flask and dilute to the mark with water. Transfer 100-ml aliquots of the solution (not more than 80 pg of copper) to 250-ml separating funncls, and add 2.5 ml of 25 per cent. hydroxyl- amine hydrochloride solution and 25 ml of N sodium acetate buffer solution, Carry out the extraction of copper as described previously for sea water, but use 6, 2-5 and 2 ml of diquinolyl reagent for the three extractions. Combine the three extracts in a 10-ml calibrated flask containing 0.5 ml of 1 per cent. ethanolic hydroquinone and dilute to the mark with n-hexanol. Measure the optical density of the extract at 540 mp in a cell of suitable length (I or 4 cm).Determine the reagent blank in the same manner, but omit the sample. Prepare a calibration curve by using 0, 5, 10, 25 and 50 pg of copper. PROCEDURE FOR DETERMINING COPPER IN CARBONATE ROCKS- Weigh 5 g of the carbonate sample into a silica flask. Dissolve it by gradual addition of 30 ml of 4 N nitric acid. If foam tends to rise to the top of the flask it can be broken by the addition of a drop of octyl alcohol. Cautiously evaporate the solution to dryness on a hot-plate; if organic matter is present, add 10 to 16 ml of concentrated nitric acid and repeat the evaporation. Evaporate the residue twice to dryness with 15 ml of concentrated hydrochloric acid to remove nitric acid. Disso'lve the residue in 100 ml of distilled water containing 1 6 m l of concentrated hydrochloric acid and dilute to 250ml.Carry out the determination of copper in a 100-ml aliquot as described €or silicate rocks. RILEY AND SlNHASENI: THE DETERMINATION OF COPPER I N Weigh accurately 0.6 to 1 g of the finely powdered rock into a platinum crucible, PROCEDURE FOR DETERMINING COPPER I N BIOLOGICAL MATERIALS- Weigh 0.5 to 3 g of the sample (1 to 40 pg of copper) into a 150-ml silica conical flask, add 10 to 15 ml of concentrated nitric acid and heat the flask carefully on a hot-plate. If the reaction becomes violent, the flask should be removed from the heat source and heating resumed only when the reaction has moderated. Repeat the evaporation with 10 to 15-ml portions of nitric acid until the residue is free from organic matter.Add 2 ml of 60 per cent. perchloric acid and evaporate until dense white fumes are evolved. Dissolve the residue in distilled water, filter, and determine copper iii the filtrate as described for silicate rocks. Carry out a blank determination in the same manner, but omit the sample. TABLE I DETERMINATION OF COPPER IN WATER Weight of copper Optical-density increment present, pg Mean optical density" per pg of copper 2 4 6 8 10 12 16 20 25 30 50 70 100 0.080 0.160 0.235 0.310 0.385 0.480 0-640 0.797 0.985t 1.1851. 1.9201. 2.6807 3.840t 0*0400 0*0400 0.0392 0.0388 0.0385 0.0400 0*0400 0.0398 0.0394 0.0395 0.0385 0.0383 0-0384 Mean 0.0393 * Measured in 4-cm cell, less reagent blank of 0.060. t Measured in l-cm cell, calculated for 4-cm cell.BEER'S LAW AND REPRODUCIBILITY OF RESULTS Various amounts of copper from 2 to 100 pg in 500-ml aliquots of redistilled water were extracted in duplicate, as described for sea water (p. 301). The optical densities of theMay, 19581 SEA WATER, SILICATE ROCKS AND BIOLOGICAL MATERIALS 303 extracts were measured at 540mp in 1 or 4-cm cells as appropriate. The results, which are given in Table I, indicate that Beer’s law is obeyed up to at least 10 p.p.m. of copper in the n-hexanol phase. The molecular extinction coefficient calculated from the average slope is 6160 & 109. DISCUSSION INTERFEREYCE- The possible interference of several ions has been investigated by carrying out copper determinations on 250-ml aliquots of redistilled water containing these ions.No interference was experienced with 50-mg amounts of sodium, potassium, magnesium, calcium, strontium , barium, aluminium, iron, manganese or titanium, or with 50-pg amounts of bismuth, cadmium, cobalt, chromium, gallium, lead, mercury, nickel, thallium, silver or zinc. Complete recoveries of 10 pg of added copper were obtained in the presence of 1 mg of phosphorus, as phosphate, 1 mg of silicon, as silicate, and 1 mg of fluorine, as fluoride. No interference can therefore be expected from any of these ions at the concentrations at which they occur in sea water or silicate rocks. APPLICATION OF THE METHOD TO SEA WATER- It was found that the proposed method was not subject to salt error, since the optical- density increment per pg of added copper was the same whether the extraction was from sea water or distilled water.The reproducibility of the method was tested by carrying out six replicate analyses on a sample of surface water collected from the Irish Sea in January, 1956. Before analysis, the water (chlorinity = 18.83 Oleo) was filtered through a No. 4 sintered-glass filter. The results (26.0, 26.6, 27.0, 27.0, 27.4 and 28.0; mean 27.0 pg of copper per litre) showed a coefficient of variation of 2.5 per cent. Similar reproducibility was obtained with a sample of water from the English Channel (chlorinity = 19.36 O/oo) containing 19.0 pg of copper per litre. ACCURACY OF THE METHOD FOR SILICATE ROCKS- In order to test the accuracy of the proposed method, six replicate analyses were carried out on a granite G1 from Westerly, Rhode Island, and on a diabase W1 from Centerville, Virginia.These rocks, which were obtained from the US. Geological Survey, have been used as the basis of a collaborative s t ~ d y ~ * y ~ ~ of the precision of methods for the determination of major and minor components in silicate rocks. AhrenP has summarised the results, both spectrographic and chemical, obtained for these rocks. The figures for copper show very poor reproducibility, the results ranging from 5 to 20 p.p.m. for G1 and from 44 to 130 p.p.m. for W1. Ahrens recommends values of 11 and 110 p.p.rn., respectively, as the most probable concentrations, but states that the former figure may need considerable revision because of poor agreement. The proposed method showed the presence of 15.9 p.p.m.of copper in G1 and 121 p.p.m. in W1 (coefficients of variation 2.5 and 3.5 per cent., respectively). These figures are of the same order as those recommended by Ahrens, but are somewhat higher in both cases. As a further check on the reproducibility of the method, a gabbro from Colmonell, Ballantrae, was analysed; the average of sixteen determinations showed a copper content of 29.8 & 1.1 p.p.m. (coefficient of variation 3.5 per cent.). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. REFERENCES Chow, T. J., and Robinson, R. G., J . Mar. Res., 1952, 11, 124. Richards, F. A., Geochim. Cosmochim. A d a , 1956, 10, 241. Atkins, W. R. G., J . Mar. Biol. Ass., 1953, 31, 493. Buch, K., Finska Kemistsamfundets Medd., 1944, 53 ,25. Morita, Y., J.Chem. SOC. Japan, 1948, 69, 174 and 246. Galtsoff, P. S., Ecology, 1943, 24, 263. Barnes, H., and Lord Rothschild, J . Exp. Biol., 1950, 27, 123. Cluley, H. J., Analyst, 1954, 79, 561. Claasen, A, and Bastings, L., Z. anal. Claem., 1956, 153, 30. Smith, G. F., and McCurdy, W. EL, Anal. Chem., 1952, 24, 371. McCurdy, TIV. H., and Smith, G. F., Analyst, 1952, 77, 846. Smith, G. F., and Wilkins, D. H., Anal. Chem., 1953, 25, 510. Gahler, A. R., Ibid., 1954, 26, 577. Hoste, J., Anal. Chim. A d a , 1950, 4, 23.304 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. BRIGGS, DYKE AND KNOWLES : WIDE-BORE: DROPPING-MERCURY ELECTRODE p o l . 83 Hoste, J., Heinemans, A., and Gillis, J., Microchenzie, 1951, 36/37, 349. Gillis, J., Hoste, J., and van Moffaert, Y., Meded. Vlaamsche Acad. KZ. Wet., 1953, 7, 3. Smith, G. F., and Wilkins, D. H., Anal. Chim. .A&, 1954, 10, 139. Cheng, K . L., and Bray, R. H., Anal. Cham., 1953,25, 657. Hoste, J . , Eeckhoute, J., and Gillis, J., Anal. Chim. Acta, 1953, 9, 263. Gillis, J., Bull. Cent. Belge e’tude et document. Eizux, 1953, 22, 233. Ghyssaert, L., Ibid., 1954, 23, 56. Guest, R. J., Ca%. Dept. Mines and Tech. Surveys, Top. Rep. No. TR-105i62, 1952. Almond, H., U.S. Geol. Survey Bull. No. 1 0 3 6 ~ , 1955. Fairbairn, H . W., and Schairer, J. F., Amer. &Jim, 1952, 37, 744. Fairbairn, H. W., Schlecht, W. G., Stevens, R. E., Dennen, W. H., Ahrens, L. H., and Chayes, F., Ahrens, L. H., _“Quantitative Spectrochemical Analysis of Silicates,” Pergamon Press, London, Received November 4th, 1957 U.S. Geol. Survey Bull. No. 980, 1951. 1954, pp. 24 to 30.

ISSN:0003-2654    

DOI:10.1039/AN9588300299

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

304 BRIGGS, DYKE AND KNOWLES : WIDE-BORE: DROPPING-MERCURY ELECTRODE [Vd. 83 Use of the Wide-bore Dropping-mercury Electrode for Long-period Recording of Concentration of Dissolved Oxygen BY K. BRIGGS, G. V. DYKE AND G. KNOWLES (Water Pollution Reseavch Laboratory, Stevenage, Hevts.) A new type of dropping-mercury electrode, consisting of a capillary of 0-8 mm internal diameter, sloping upwards, which is stable in performance over long periods, has been used for recording the concentration of dissolved oxygen in natural waters for periods of weeks. One form of the equipment records dissolved oxygen directly in parts per million, thermistors in a non- electronic circuit compensating for the effect of any change in water tem- perature. The range is 0.0 to 15.0 p.p.rn. by weight, and the mean value of the errors in two field trials was 0.00 p.p.m., taking into account the signs of the errors; the standard deviation of their distribution about this mean was 0-06 p.p.m.The method can be used either with the water flowing past the electrode or with stationary saimples down to 1 ml. IN studies of the respiration of tissues or invertebrate animals, and the de-aeration and re-aeration of polluted rivers, it is often useful to have a continuous record of the concentration of dissolved oxygen. The proposed method is being used for rivers and effluents and its use is now being extended to laboratory applications, including direct recording of the concen- tration of dissolved oxygen in samples as small as 1 ml. No review appears to exist of instrumental methods of indicating and recording concentrations of dissolved oxygen, so perhaps a very brief selective survey of the field will be useful.SURVEY OF METHODS METHODS IN WHICH ELECTRODES ARE NOT USEID- One methodl depends on removal of the oxygen from the water by means of an inert gas, and determination of the oxygen content of the resulting gas mixture by the para- magnetic method; the equipment is complex and requires mains power. The same dis- advantages are present in a mechanised version2 of the Winkler chemical determination. METHODS IN WHICH ELECTRODES ARE USED- It appears that all the electrode methods measure the current permitted to flow by the reduction of oxygen to hydrogen peroxide or to water. These can be sub-divided into methods in which a solid metal electrode is used and those in which a mercury electrode is used.Solid electrodes-Solid electrodes have been used for the determination of dissolved oxygen by workers in bacteriology,3~~~~~~ biochemistry7 ,* and physi~logy,~ ~ l o ~ l l ,12 9 1 3 and have been applied also to the determination of dissolved oxygen in natural waters1*J5J6 and boiler-feed water.17 In general, solid electrodes require frequent re-calibration, and, when used in natural waters, they tend to become coated with calcium carbonate.May, 19581 FOR LONG-PERIOD RECORDING OF CONCENTRATION OF DISSOLVED OXYGEN 305 Mercury electrodes-A mercury electrode has a great advantage over a solid electrode in that it is easy to arrange for its surface to be renewed automatically, as, for example, the use in ordinary polarography of the dropping-mercury electrode.Dissolved oxygen in natural waters and other aqueous liquids has been measured in this way by many workers, including Seaman and Allen,ls Rand and Heukelekian,lg Moore, Morris and Okun20 and Ingols.21 The impression given in the literature is that dissolved oxygen can be measured satisfactorily by the usual type of dropping-mercury electrode if it is re-calibrated frequently; for example, Seaman and Allen1* recommend that the calibration should be checked several times a day. All the authors mentioned used a standard reference electrode as their second electrode, but Foyn22 used a dropping-mercury electrode with a second electrode consisting of a zinc plate immersed in the water; again, the calibration was checked before each test.AmbuhP renewed the mercury surface by means of a motor-driven wiper, new mercury flowing from a reservoir to replace that wiped away. EXPERIMENTAL UNSUCCESSFUL USE OF SOLID ELECTRODES- When a stationary gold electrode was used in conjunction with a saturated-calomel electrode, currents produced by dissolved oxygen in 0.1 M potassium chloride solution decreased with time. During this experiment the gold electrode became covered with a film, probably of mercury from the calomel electrode. Use of a guard electrode, as described by Giguere and Lauzier,l4 to avoid this deposition of mercury did not prevent the fall in current, which also occurred when a stationary platinum electrode was used with cadmium as the second electrode.SUCCESSFUL USE OF MERCURY ELECTRODE- First experiments on the use of the dropping-mercury electrode for oxygen measurement were with a conventional electrode of internal diameter about 0-04 mm and pointing down- wards, which was placed in Stevenage tap-water and connected by way of a microammeter to a zinc plate in the water. This was the arrangement used by FoynZ2; we found it gave an almost linear relationship between current and concentration of dissolved oxygen in parts per million if the temperature was kept constant. This electrode, however, would not behave consistently for more than a few hours, as the drops became smaller and more frequent, and finally ceased altogether. This is not uncommon with the conventional dropping-mercury electrode and may be more liable to occur in the proposed application, owing to the tendency for calcium carbonate to deposit in or near the very fine bore of the electrode.After many such failures, it seemed to be clear that this form of electrode would not be capable of giving a stable performance if left in continuous operation over a period of days or weeks. This mechanical failure of the conventional dropping-mercury electrode may well have been due to its very narrow internal diameter; trials were therefore made with capillaries of bore between 0.4 and 1.05 mm. It was found to be best to arrange these capillaries so that they delivered mercury upwards to the orifice through which mercury enters the solution. The first successful arrangement was to point the wide-bore capillary vertically upwards with a bevel ground at 45" to the horizontal, but it proved to be equally successful to point the wide-bore capillary upwards at about 45" to the horizontal (see Fig.1). Although all the capillaries in the stated range of internal diameters gave completely stable behaviour, we standardised early on the range 0.8 to 1.0 mm; a number of such electrodes has been in intermittent use, for periods of weeks' continuous operation at a time, for over a year, and none has ever altered in behaviour or calibration, in spite of our practice of the re-use of mercury after simple chemical cleaning and drying, without distillation. At present, a wide-bore dropping-mercury electrode made of Perspex is on trial, and it appears to be as satisfactory as the glass ones.A problem to be solved was to find means of regulating the flow of mercury to these wide-bore capillaries, which, unlike the conventional dropping-mercury electrode, do not provide any significant restriction to flow. One method is to include a capillary restriction between the reservoir of mercury and the electrode. At first it was thought that this additional capillary would itself have to be of rather wide internal diameter and the results were good with a head of mercury of about 15 cm and a long capillary of internal diameter306 BRIGGS, DYKE AND KNOWLES : WIDE-BORIS DROPPING-MERCURY ELECTRODE [VOl. 83 about 1 mm, but later the arrangement shown in Fig. 1 was found to be completely satis- factory, the system being to supply mercury (at a head of 15 cm) to the wide-bore electrode through a flow-restricting capillary about 25 CITL long and of bore about 0.2 mm.Another system found to be satisfactory was to feed the rnercury to the wide-bore electrode by means of a motor-driven Perspex syringe with piston rings of silicone rubber, which holds a %-day supply of mercury. Mercury Overflow water out Water trom Pump A,, A, = Electromagnetic valves E = Zinc electrode B = Platinum contacts F = No. 4 sintered-glass disc C == Glass o r Perspex boat con- G = Dropping-mercury electrode taining mercury H = Thermistors D = Silicone rubber tubing I = Disc with 4 holes for supplying water and mercury. are omitted when temperature compensation is not required Fig. 1. Layout of apparatus showing cell containing electrodes and systems Thermistors All vessels are circular in section. Besides its reliability, another advantage possessed by the wide-bore electrode is that currents are about ten times greater than with the conventional electrode, which means that a polarographic current of about 40 pA is produced by the reduction of dissolved oxygen in air-saturated water at about 20" C; the second electrode used in conjunction with this electrode must be capable of carrying such a cunrent without significant polarisation.From the consideration of a number of polarograms it was decided that the best fixed voltage to apply in order to record dissolved osygen would be -0.5 volt with respect to the zinc electrode (equivalent to -1.5 volts with respect to the saturated-calomel electrode).Use of this voltage, which means working on the second plateau of oxygen reduction, has the advantage that cyanide and sulphide will not interfere, whereas at the more positive potential of the first plateau of oxygen reduction their oxidation waves would be present. The only other source of interference likely t o be encountered is the reduction of metal ions, and in most natural waters it is unlikely that the concentrations of these, which include iron, lead, copper, zinc and nickel, will be so high as to have a significant effect. If work has to be done on solutions containing substances that interfere with the proposed method, the equipment could probably be rearranged so as to avoid error from such a cause. The suggested method of rearrangement is to use two wide-bore dropping-mercury electrodes each with its own second electrode and in its own cell; the solution under examination would be in both cells, but in one it would be de-oxygenated with a stream of electrolytically generated hydrogen, or with sodium sulphite and a catalyst.The difference between the currents fromMay, 19581 FOR LONG-PERIOD RECORDING OF CONCENTRATION OF DISSOLVED OXYGEN 307 the two cells, displayed on a single meter, would then be a measure of the concentration of dissolved oxygen. For placing the electrodes in the sample, a wide variety of sizes and shapes of cells can be used; a type used in our work is shown in Fig. 1, For a given cell and position of electrodes the current corresponding to a given concentration of dissolved oxygen is the same with the water stationary as with it moving through the cell at velocities up to a critical value, above which the current increases somewhat with increase in velocity of the water.For the cell shown in Fig. 1 this critical value is about 40 ml per minute, and for more rapid flow the current increases only by about 0.5 per cent. of its magnitude at zero flow for each 10 ml per minute increase in flow beyond 40 ml per minute, up to a rate of flow of 220 ml per minute. As arranged in Fig. 1, the rate of flow through the cell, governed by the head of water in the constant-head unit on the right, is about 100 ml pcr minute, i.e., the rate at which the equip- ment is calibrated. The constant-head unit is of the type described by Spoor24; this arrange- ment does not expose the water to air.In a 1-ml cell the current was unaffected by an increase in rate of flow of water from zero to 5 ml per minute, and then increased by about 0.3 per cent. of its value at zero flow for every 1-ml increase in rate of flow above 5 ml per minute. The conductivity of the water has no effect provided its value is equal to or greater than 50 micromhos per cm cube. The pH has likewise no influence, at least within the range of pH values so far tested-5 to 8.5-and is probably without effect for pH values higher than 8, although for those much less than 5 it might be necessary to use a less negative value for the potential of the dropping-mercury electrode. APPARATUS- The layout of the apparatus is shown in Fig. 1. Su@Zy of mercury-A constant rate of flow of mercury (6 ml per hour) and a constant drop rate (1 drop every 2.5 seconds) are maintained by supplying mercury to the electrode from a reservoir in which the surface of the mercury is 15 cm above the electrode tip; the level in the reservoir is kept constant by means of platinum contacts, which cause mercury to be admitted as required through an electromagnetic valve, A,, (Londex Ltd., type LF/VA, with 24-volt a.c.coil, but operated on 9 volts d.c.), which, until energised, clamps firmly on a silicone rubber tube (QS166 3-mm Symel sleeving, H. D. Symons Ltd.). The platinum contacts do not carry the current for the electromagnetic valve; they switch on this current by means of a type P.0.3000 relay operating on 9 volts d.c. From the reservoir, mercury flows by gravity through silicone rubber tubing that has a screw-clip or an electromagnetic valve, A,, to stop the flow.At a level below the delivery tip of the dropping-mercury electrode, this silicone rubber tubing joins a 25-cm length of glass capillary tubing of bore about 0.2mm bent to join, by means of poly(viny1 chloride) or polythene tubing, to the lower end of the glass capillary tubing that is the electrode. Both the 0.2-mm and the 0.8-mm capillary tubing are precision-bore Veridia capillary tubing (Chance Brothers Ltd.) . As an alternative to the supply of mercury to the electrode by gravity, supply from a motor- driven syringe, as mentioned previously, has proved to be satisfactory in short-period tests. Whichever method is used t o supply the electrode, the mercury can be re-used several times without distillation, simple chemical cleaning followed by drying by passage through a pin-hole in a filter-paper being sufficient. Reference electrode-The zinc electrode shown in Fig.1 is a convenient reference electrode, capable of carrying, without significant polarisation, the highest current (about 40 PA) that will be encountered; it comprises a pure zinc rod (Imperial Smelting Corporation) in a buffer solution (iV hydrochloric acid plus 0.2 M sodium acetate solution) of pH 5.5, which com- municates with the water through a No. 4 sintered-glass disc. This electrode can be quickly made from a cut-down “filter-tube” ; it has a potential of -1.0 volt with respect to a saturated- calomel electrode and works for at least 1 month without replacement of the buffer solution.I t is possible, however, that a mercury-pool electrode, as used in commercial polarographs, would be equally satisfactory as the second electrode. Equipment not Jitted with temperature compensation-When temperature compensation is not required the thermistors shown in Fig. 1 are omitted, and the circuit to be used is given in Fig. 2. R, is adjusted until V indicates 0-5 volt with the terminal connected to the mercury electrode the more negative one. If all readings are taken with the water a t a standard temperature, a current - dissolved oxygen calibration graph, prepared as described later, gives the concentration of dissolved308 BRIGGS, DYKE AND KNOWLES WIDE-BORE DROPPING-MERCURY ELECTRODE [VOl.83 oxygen from the current shown by the indicator or recorder. Alternatively, the instrument may be calibrated directly in parts per million of dissolved oxygen with the aid of such a graph. B = 2-volt accumulator C = 7500-pF electrolytic condenser M = Microammeter, 25-pA full-scale R, = 10-ohm wire-wound potentio- R, = 8000-ohm wire-wound resistance R, = Wire-wound resistance of such value that, when added t o resistance of M, the total is 5000 f20 ohms V = Voltmeter, I-volt full-scale deflec- tion d ef I ect i o n meter &-------I Mercury elect rode electrode I A,, A, = Connections t o input of amplifier of B = 2-volt accumulator C = 7500-pF electrolytic condenser M input impedance 2 50 kilo-ohms = Microammeter calibrated in parts per million of dissolved oxygen, 5-pA full- scale deflection Rl = 10-ohm wire-wound potentiometer R, =1 Wire-wound resistance] exact value selected f,or i ment, see p.309 about 750 ohms = Wire-wound resistance each equip- about 130 ohms = Wire-wound resistance of value such that, when added t o resistance of M, the total is 5000 & 20 ohms TI, T,, T,, T, = Thermistors (Standard Telephones & Cables Ltd., type F 23 I 1/300), each of resistance about 2000 ohms a t 20" C R, R* V = voltmeter, I-volt full-scale deflection recorder with temperature compensation Fig.2. Circuit for dissolved-oxy- gen indicator and recorder without Fig. 3. Circuit for dissolved-oxygen indicator and temperature compensation If, on the other hand, it is required to take rleadings with the water at other temperatures, then each measured current must be converted to its equivalent value at the standard tem- perature for which the calibration graph is drawn.The factor for this conversion can be obtained from a graph drawn from the following figures, which are the relative currents produced by constant concentration of dissolved oxygen at various water temperatures : 0.730 at 0" C, 0-803 at 5" C, 0.880 at 10" C, 04352 at 15" C, 1.000 at 18" C, 1428 at 20" C, 1.100 at 25" C, 1.171 at 30" C and 1.236 at 35" C. In our experience, these figures apply to all equipments in which the wide-bore dropping-mlercury electrode is used for measurements of dissolved oxygen. The calibration graph itself, for any desired standard temperature, can be prepared by using tap-water at this standard temperature containing known concentrations of dissolved oxygen, as determined by the Winkler chemical method.25 Alternatively, tap-water at any temperature can be used, but each current has to be converted to its equivalent at the standard temperature by means of a conversion factor obtained as described above.May, 19581 FOR LONG-PERIOD RECORDING OF CONCENTRATION OF DISSOLVED OXYGEN 309 Equipment Jitted with temperature co;mpemation-The indicator or recorder can be marked directly in parts per million of dissolved oxygen if arrangements are made for automatic compensation for the increase in polarographic current with temperature.This compensation may be effected by the decrease in resistance, as temperature rises, of thermistors immersed in the water; the circuit is shown in Fig.3. A variable resistance is con- nected temporarily between X and Y instead of R,, R, and the thermistors. Tap-water is supplied to the electrodes at the rate of flow to be used in practice, its temperature being about midway in the range of temperatures to be covered and its concentration of dissolved oxygen being the maximum to be recorded; the variable resistance is now adjusted to give an almost full-scale reading on meter M. The variable resistance is then removed and its resistance measured. If the resistance so measured is R, then the required approximate value of R, will be 2R,, whereas the thermistor, or set of thermistors in parallel, is chosen to have a resistance of somewhat less than 2R, at a temperature midway in the range; thermistors must be of a type aged by the manufacturer. The required approximate value of R, can now be calculated from the equation- The circuit values are easily arrived at by experiment.x~,[R,(R?n + R4 + Ks) - (1 + WJWL + 114) + Rt(Rm + R, + ~ i , ) I l >--& R3= { k[2R,(R, -t R4) -I- R,(R, + R, + R,)1 where R, = the resistance of meter M in ohms, R, = the resistance of the thermistors in ohms at a temperature midway in the range, R4 = the resistance of R, in ohms, x = the fractional rate of increase per "C rise in temperature of the resistance of the thermistors at a temperature midway in the range, and k = the fractional rate of increase per "C, rise in temperature of the current due to a given concentration of dissolved oxygen at a temperature midway in the range.For k it is sufficiently accurate to select the value given against the nearest temperature from the following: 0.0174 at 5" C, 0.0171 at 10"C, 0.0163 at 15" C, 0.0141 at 20°C and 0.0129 at 25" C. O L L i I I I d 0 2 4 6 8 10 12 14 16 Disso!ved oxygen, p.p.m. Relationship between current and concentration of dissolved oxygen for a tem- perature-compensated recorder. The temper- atures corresponding to some points on the curve are: A, 28°C; €3, 1-5"G; C, 4°C; D, 2.6"C; E, 18-8'C; F, 29.9"C; G, 3.5"C Fig. 4. Exact values for R, and R, can now be found by using the circuit of Fig. 3 modified as follows : remove C, replace the electrodes by a meter having a full-scale deflection of about 50 pA to measure the total current and replace R, and R, by variable resistances initially set at the respective approximate values determined as described.Immerse the thermistors in water of temperature about midway in the range to be covered, adjust the total current to, say, 35 PA, and note the reading on meter M. Transfer the thermistors to water of temperature about the maximum in the range and adjust the total current by means of R, to the new value it would have if the current of 35 pA had been due to dissolved oxygen; the factor required to find the new value can be obtained from the graph of relative currents310 BRIGGS, DYKE AND KNOWLES : WIDE-BORE DROPPISG-MERCURY ELECTRODE [VOl. 83 previously described (p. 308). Adjust R, until the original reading of meter M has been restored.Return the thermistors to the water of temperature about midway in the range, adjust the total current to 35 PA, and adjust R, until the original reading of meter M has been restored. Continue in this way, using also water of temperature about the minimum in the range, until the reading of M is unaltered by placing the thermistors in water at any of the three temperatures, provided that the total current is always adjusted to correspond with the temperature. Finally, a calibration graph, of which an example is given in Fig. 4, is prepared by supplying the cell with water, at any temperature, containing known concentrations of dis- solved oxygen. If desired, the scale of meter M, or the recorder charts, can be inscribed directly in parts per million of dissolved oxygen.METHODS OF RECORDlKG When mains power is available, the dissolved-oxygen concentration is continuously recorded on a strip-chart by using a stable d.c. amplifier of the chopper type and a recorder having a full-scale deflection of about 1 to 5 mA. Commercially available equipment or the chopper amplifier described by Davies26 can be used, the circuit being modified from that published to allow rectification of the amplified current by the second pair of contacts of the high-speed relay, and to include one more stage of amplification. On the other hand, when only a 12-volt accumulator is available for power supply, a photograph of the meter indicating dissolved-.oxygen concentration is taken automatically at chosen intervals, usually every 15 minutes or every hour, by a camera that gives 200 pic- tures, each 1 inch square, with one loading of 36-mm film.The sequence of events begins when a clockwork time-switch energises electromagnetic valve -4, in Fig. 1, thereby starting the flow of mercury, and simultaneously starting the 12-volt immersion pump supplying the river water to the equipment and switching on the 12-volt charging unit of an electronic flash unit, as used by photographers. Three minutes later, the timing unit operates the camera shutter, which, by means of its internal synchronised contacts, fires the electronic flash tube, the light from which illuminates the oxygen-indicating meter indirectly by way of the top and sides of the box, which are painted white internally. The whole equipment then ceases to operate until the next sequence begins.Usually, the camera is arranged so that the photographs include not only the dissolved-oxygen indicator, but also the dial of a mercury-in-steel thermometer, a river-level indicator and an exposure-number counter or a battery-operated clock. RESULTS After laboratory tests had shown that the equipnients held their calibrations indefinitely for tap-water, they were used in two field investigations, The first of these was the indication, every 15 minutes, of the concentration of dissolved oxygen in a sewage effluent containing suspended and colloidal material (a switch was pro- vided to change the method of indication to a continuous one when desired). At this time, the method of automatic compensation for water temperature had not been developed, so the temperature of the effluent was also noted when the meter was read, and the concen- tration of dissolved oxygen was found from these data by the procedure described.In a consecutive series of 115 check tests against the izide m~dification~~ of the Winkler method, over a period of 3 weeks, the mean value of the errors of the results given by the equipment was 0.002 p.p.m. by weight, taking account of the signs of the errors; the standard deviation of their distribution about this mean was 0.06 p.p.m. and the maximum error was 0.15 p.p.m. The range of concentrations of dissolved oxygen encountered in this investigation was 3.20 to 7.05 p.p.m., and the range of temperatures 11.6" to 19.0" C. The second field investigation was on a small river at a point downstream of the entry of the effluent from the sewage works of a large industrial town.On this occasion there was less opportunity to make check tests, which numbered 19 over a period of 3 weeks, including 3 checks with tap-water at Stevenage where the equipment mas sent by lorry -a journey of 100 miles. In this investigation, photographic recording was used and compensation for water temperature, over the range 0" t o 30" C, was incorporated ; readings were directly in parts per million of dissolved oxygen. The errors, calculated as above, had a mean of 0.01 p.p.m. by weight, and the standard deviation of their distribution about this mean was 0.07 p.p.m.; the maximum error was 0.20 p.p.m. In these 19 check tests,May, 19581 FOR LONG-PERIOD RECORDIKG OF COXCENTRATION OF DISSOLVED OXYGEN 311 the range of concentrations of dissolved oxygen was 2.20 to 11.00 p.p.m., and the range of temperature was 5" to 20" C.This paper is published by perinission of the Department of Scientific and Industrial Research. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Beckman Instruments Inc., Fullerton, California, U.S.A., Bulletin N o . 113. Briggs, R., Knowles, G., and Scragg, L. J.. Analyst, 1954. 79, 744. Zeidler, H., and Taubeneck, U., Zbl. Bakt. T I , 1956, 109, 516. Oehme, F., and hToack, D., Chem. Tech. Berlin, 1955, 7, 270. Warshowsky, B., and Schantz, E. J., Anal. Chem., 1954, 26, 1811. Todt, F., Teske, G., Windisch, F., Heumann, W., and Goslich, C., Bid. Zbl., 1953, 72, 146. Longmuir, I. S., Biochem. J., 1954, 57, 81. - , Ibid., 1957, 65, 378. Montgomery, H., and Horwitz, O., J . Clin. Iizvest., 1950, 29, 1120. Clark, L. C., Wolf, R., Granger, D., and Taylor, Z., J . A$$. Physiol., 1953, 6, 189. Mochizuki, M., and Bartels, H., Pjlug. Arch., 1955, 261, 152. Hill, D. K., J . Physiol., 1948, 107, 479. Carlson, F. D., Brink, F., and Bronk, D. W., Rev. Sci. Instrum., 1950, 21, 923. Giguere, P. A., and Lauzier, L., Canad. J . Res. B., 1945, 23, 223. Ohle, W., Vom Wasser, 1953, 19, 99. Todt, F., Schwarz, W., and Todt, H.-G., Gesundheitsing., 1954, 75, 224. Freier, R., Tddt, F., and Wickert, K., Chew.-Ing.-Tech., 1951, 23, 325. Seaman, W., and Allen, W., Sewage I n d . Wastes, 1950, 22, 912. Rand, M. C., and Heukelekian, H., Ibid., 1951, 23, 1141. Moore, E. W., Morris, J. C., and Okun, D. A,, Sewage W k s J., 1948, 20, 1041. Ingols, R. S., Ibid., 1941, 13, 1097. Foyn, E., Fiskeridir. Skr., 1955, 11, No. 3. Ambuhl, H., Schweiz. Z . Hydrol., 1955, 17, 123. Spoor, W. A., Science, 1948, 108, 421. American Public Health,,Association, "Standard Methods for the Examination of Water, Sewage Davies, F. S., Lab. Practice, 1957, 6, 97. and Industrial Wastes, Tenth Edition, New York, 1955. Received August 6th, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300304

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

May, 19581 FOR LONG-PERIOD RECORDIKG OF COXCENTRATION OF DISSOLVED OXYGEN 311 Notes A MODIFICATION OF THE VON STIEGLITZ ELECTROBIETRIC METHOD FOR SUGAR TITRATIONS THE well known Lane and Eynonl method for the titration of reducing sugars a t the boiling-point with Fehling’s solution has been modified by von Stieglitz,a who determined the end-point electro- metrically instead of colorimetrically. This has some advantages when the colour of the solution being titrated masks the methylene blue end-point, or when the titration is carried out in unsuitable lighting conditions. In ideal conditions, some operators experience difficulty in detecting the methylene blue end-point. The von Stieglitz method is not well known in this country, as it was described in a journal that is not readily accessible, although i t has also been mentioned by Browne and Zerban.3 This method requires the preparation of a special electrode that has a porous plug of plaster of Paris in the end.We have found that the von Stieglitz electrode can be replaced with advantage by the outer case of a Cambridge pH reference electrode, or one of a similar type. The Cambridge electrode has a ground-glass sleeve that permits electrical contact to be maintained between the inside and the outside of the electrode. The copper wires should be cut from the same roll of 1-mm diameter pure copper wire, and they should be kept clean with fine emery cloth. The sensitive galvanometer has a central zero-point and a sensitivity of about 2 pA per division. The von Stieglitz solution used for filling the Cambridge electrode is prepared by mixing 5 ml of Fehling’s solution No.2 (alkaline tartrate), 5 ml of sodium sulphate solution (39.4 g of the anhydrous salt per litre) and 20 ml of distilled water. This solution is suitable for titres of between 15 and 30 ml, but it is preferable to use 40 ml of distilled water if the titres lie between 30 and 50 ml. The sugar solution is titrated against mixed Fehling’s solution in the usual manner, the end-point being reached when the null-point is recorded on the galvanometer. As the sugar solution is added, the deflections of the needle become less and less; they are not detectable a t The useful life of these plugs is rather short. The apparatus is shown in Fig. 1.312 NOTES [VOl. 83 the end-point, and, finally, a deflection in the opposite direction is given when the end-point is passed. Differences in titre between this method and the methylene blue indicator method have never exceeded 0-1 ml in 25 ml with any of the sugars mentioned in the Lane and Eynon Tables.Copper Tapping key Galvanometer von Stieglitz solution Fehling’s solution Fig. 1. Apparatus for the modified von Stieglitz I t is advisable for each analyst to check the Lane and Eynon Tables for his own conditions. The simplest way of doing this is to follow the method of Zerban, Hughes and Nygren.4 A constant-volume method discussed in the Proceedings of the International Commission for Uniform Methods of Sugar Analysis6 makes it possible to dispense with the use of tables. The electrometric method of end-point detection could be applied to this procedure with suitable checks against a standard invert sugar solution.Correction for variable sucrose is still necessary.6 r7 electrometric method for sugar titrations We thank the Directors of Spillers Limited for permission to publish this Note, REFEREYCES 1. 2. 3. 4. 5 . 6. 7. Lane, J. H., and Eynon, L., J . Soc. Clzenz. Ind., 1923, 43, 3 2 ~ . von Stieglitz, C. K., PYOG. Queensland SOG. Sugar Cane Tech., 1936, p. 101; 1938, p. 29; 1939, p. 43. Browne, C. A., and Zerban, 17. W., “Physical and Chemical Methods of Sugar Analysis,” Third Zerban, F. W., Hughes, W. J., and Nygren, C A., Ind. Eng. C h e w , Anal. Ed., 1946, 18, 64. Proc. I n t . Comm Uniform Meth. Sugar Anal., Eleventh Session, Paris, 1954, p.18. Tbid., Eleventh Session, Paris, 1954, pp. 37 arid 38. Bell, E. V., and Graham, J . C., Int. Sug. J . , 1950, 52, 90. Edition, John Wiley & Sons Inc., New Yorb., 1941, pp. 755 to 757. L. Mi. CHUBB SPILLERS CENTRAL LABORATORY A. W. HARTLEY STATIOX KOAD, CAMBRIDGE Iieceived December 16th, 1957 THE DETERMINATION OF WATER IN GLYCEROL, MARGARINE, OILS AND FATS A METHOD is described for the determination of water in glycerol, margarine, oils and fats. It involves the use of an apparatus that permits a thin film of material to be heated in a vacuum and the loss in weight is calculated as water. In July, 1956, the Karl Fischer method for the determination of water was adopted by the British Standards Institution in place of the International Standards Method of 1911.The latter was generally referred to as the vacuum-desiccator method. The use of concentrated sulphuric acid and the time required to attain constant weight were two disadvantages of the vacuu ni-desiccator method.May, 19581 SOTES 313 The object of this investigation was to find a method that would be suitable as an alternative to the Karl Fischer and be less expensive and more simple in operation. Heidbrink1 published an article entitled “Plate Glass Apparatus as a Simple Tool for the Rapid Determination of Water and Other Volatile Material in Emulsions and Viscous Mixtures.” A similar method adopted for cosmetics by Iwasenko and l<raus2 includes the use of low temperatures, small samples and relatively large surface areas, which permit water to evaporate easily.Heid- brink used two flat glass plates resting in a cradle to which was attached a hook. The sample was placed on the lower plate and covered immediately with the top plate. The two plates were rubbed together to make the sample spread into a thin film, and moisture was determined by evaporation in an air-oven. By using these investigators’ findings as a background, it was found to be necessary in the development of the method to take certain precautions in order to avoid loss of glycerol during heating and to eliminate errors caused by the hygroscopic nature of glycerol. Loss of glycerol was completely avoided by adopting the use of a vacuum-oven. Errors caused by absorption of moisture from the atmosphere were eliminated by completing the determination without delay.The effects of various temperatures, times of heating and also weights of sample were investigated to ascertain the correct conditions for the determination. METHOD FOR GLYCEROL APPARATUS- 11-eighs about 20 g. a rim to prevent the glycerol from creeping over the side (see Fig. 1 ) . The apparatus is similar to that used by previous workers,lJ but is made of aluminium and The tn-o plates are approximately 3 inches in diameter and the lower one has PROCEDURE- Drop on to the lower plate of the apparatus 0.35 to 0.45 g of the sample, place the top plate over the lower plate immediately and rub them together. Fix the two plates in the cradle pro- vided, and weigh quickly and accurately to 0.0001 g. Transfer to a vacuum-oven kept at a tem- perature of 58” C and lift the top plate on to the hook of the cradle.Evacuate the oven. Leave the sample for at least 5 hours, maintaining a vacuum of -27 inches of mercury throughout. Release the vacuum, replace the top plate of the apparatus on the lower and transfer from the oven to a desiccator. Calculate the percentage of water in the sample. The conditions stated must be strictly observed in order to produce satisfactory results. When cool, weigh as soon as possible and note the loss in weight. Fig. 1 . Apparatus for determining water RESULTS The results were compared with those by the Karl Fischer method and also those by sub tracting from 100 per cent. the sum of the glycerol and total residue at 160’ C found by analysis A comparison of the these results is shown in Table I.3 14 NOTE!; TABLE I [Vol.83 COMPARISON OF THE RESULTS OF THE DETERMINATION O F WATER BY DIFFERENT METHODS Sample Crude glycerol , Pure glycerol . Water found by Karl Fischer method, % 4.82, 4.94, 5.09, 4.93 5-11, 5.24 5.04, 4.92, 5.03, 4.82 4.7 1 4.84 5.13 9.90 - Water found by subtracting from 100 per cent. the sum of the glycerol and total residue i3t 160" C, % 4-87 5.02 4-66 4.86 5-02 4.77 503 Water found by proposed method, % 4-70, 4.68, 4.61, 4-93, 4.93 4-72, 5-17 5.05, 4.83 4.81, 4.75 4.58, 4.85, 4-65, 4.52 10.05, 10.09 4.81, 5.07, 5-14, 4.74, 4.74 5-11, 5.11, 5-06, 5-17, 5-15 APPLICATION OF THE METHOD TO MARGARINE, OILS AND FATS The foregoing investigation and procedure for the determination of water in glycerol were With the follovuing modifications, the method can be adopted applied to margarine, oils and fats.for these materials also- (u) weigh accurately 1.76 to 2-Og of sample, (b) place for 1 hour in a vacuum-oven maintained at 70" C. Some results obtained on margarine, oils and fats are shown in Table 11. CONCLUSIONS The method described has the advantages that (a) no great skill is required, j u s t accuracy and care by the operator, (b) the apparatus is simple and easy to clean, (G) only a small amount of manipulative work is required, and (d) the determination is complete in a short time, i.e., 6 hours for glycerol and 1 hour for margarine, oils and fats. COMPARISON OF THE Sample Margarine . . Lard acid oil Tallow . . TABLE :[I RESULTS OF THE DETERMINATION OF WATER BY DIFFERENT METHODS Water found by Dean and Stark Water found by Water found by m e t h ~ d , ~ heating in an ovenJ3 proposed method, % % % 15.3 15.42, 15-46 15.0 15-55, 15.31 ..3.40 3.62 3.44, 3.50 - .. - 10.25 - 10.11, 9.95 .. 0.70 0.90 0.98 0-79, 0.85 0.96 1.11, 0.99 I thank the Directors of J. Bibby & Sons Ltd. for permission to publish this Note, and also Mr. Weatherall for his advice. REFERENCES 1. 2. 3. British Standard 684: 1950. Heidbrink, W., Fette a. Seifen, 1951, 5, 291. Iwasenko, H., and Kraus, S. R., Proceedings of Scientific Section, Toilet Goods Association, New York, 1952. ANALYTICAL DEPARTMENT J. BIEEY & SONS LTD. GREAT HOWARD STREET LIVERPOOL, 3 C. B. STUFFINS Received November 22n4 1957 THE ANALYTICAL APPLICATION OF ORGANIC MERCURY COMPOUNDS ORGANIC mercury compounds of the type R-HgOH can form complexes with thiourea, xanthates, mercaptans and compounds that form ions such as S2-, SZ2-, CS32- and CS,2-. This property can be used analytically for the determination of these types of sulphur-containing compounds.May, 1958; XOTES 316 For titration in aqueous solutions, only water-soluble organic mercury compounds can be used, i.e., those that have phenolic or carboxylic functions and are soluble in alkaline solution.The complexes formed are also soluble in water, and, in the presence of a suitable indicator, a titration is possible. To date, the following compounds have proved to be suitable indicators: sodium nitroprusside, diphenylcarbazone, dithizone, monomercurphenolphthalein, thiofluorescein and certain products that result from the heating of organic compounds with sulphur.If the appear- ance of the complex is sufficiently distinct, as occurs when dimercurfluorescein is used, an indicator is unnecessary. By choosing a suitable indicator it is possible to titrate one sulphur compound in the presence of another, e.g., sulphide can be determined in the presence of xanthates when dithizone is used as indicator, but, when diphenylcarbazone is used, the titre agrees with the total amount of sulphide plus xanthate present. In some instances the titre depends also on the indicator used, e.g., in the titration of sdlphide with a clear 0.002 to 0.05 M solution of o-hydroxy- mercurbenzoic acid in 0.2 N potassium hydroxide, a freshly prepared 0.1 per cent. solution of dithizone in ethanol being used as the indicator.One millilitre of indicator solution and 5 ml of N potassium hydroxide are added to 100 ml of water containing from 50 to 2 x lo4 p g of hydrogen sulphide and the mixture is titrated with the reagent until the colour of the solution changes from yellow to a permanent purple. Under the conditions described, the end-point is sensitive to 0.15 ml of 0,002 M o-hydroxymercurbenzoic acid. The titre agrees with the following reaction- The presence of xanthates, thiourea or the ions SO?-, S,0,2-, S,0,2- and CNS- does not interfere with the titration. When sodium nitroprusside is used as indicator and the reagent is added until the purple colour of the solution disappears, the titre agrees with the reaction- OH-. The use of monomercurphenolphthalein as indicator depends on the formation of blue- An alkaline solution of this compound is purple, and the addition However, in the presence of compounds such as This can be represented coloured mixed complexes.of sulphide changes its colour only slightly. 0- and P-hydroxymercurbenzoic acid, the colour distinctly changes to blue. as follows- -+ "'...- - ".g-s-Hg-c6H6-coo-. Purple Purple Blue The fusion of sulphur with resorcinol a t 250" C, and with phenolphthalein a t 280' C, produces two compounds, which, in alkaline solution, change colour when organic mercury compounds are added; the first from yellow to red, and the second, reduced by boiling with sodium sulphite, from red to blue. An alkaline solution of thiofluorescein is blue, but it changes to very pale yellow when organic mercury compounds or mercury salts are added.Thiofluorescein can be used in the titration of mercury in alkaline potassium iodide solution to a permanent blue colour. I n this way 10 p g of mercury can be determined without interference from other cations, such as divalent copper and lead. An interesting application of dimercurfluorescein is made possible by the fact that, in alkaline solution, i t has a green fluorescence, which changes to a pinkish orange colour when any of several sulphur compounds is added. It is possible to titrate with a diluted solution of dimercur- fluorescein to the first green fluorescence, and in this way from 5 to 50 p g of hydrogen sulphide and very small amounts of many other sulphur compounds can be determined.DEPARTMENT OF CHEMICAL TECHNOLOGY UNIVERSITY OF LoD~, POLAND MIECZYSLAW WRONSKI Received JuZy 5th, 1957316 SOTElj [Vol. 83 THE DETERMINATION OF SMALL AMOUNTS OF CARBON TETRACHLORIDE BY THE FU JIWAliA REACTION THE Fujiwara colorimetric test for chlorinated hydrocarbons, based on their reaction with pyridine and sodium hydroxide, has a large number of modifications. These have involved the concen- tration and relative amount of sodium hydroxide solution used, the time and temperature of heating and the time of standing before comparison of the colour produced. The test is not specific for any one hydrocarbon, but, by modifying the procedure somewhat, conditions can be found that change the relative sensitivity toward:; the individual hydrocarbons.The authors of most of the papers that have been published have elected to adopt a procedure in which the relative amounts of pyridine and aqueous sodium hydroxide are such that the final mixture remains in two phases, the colour of the pyridine layer being evaluated after separation of the two layers. By using a comparatively small volume of aqueous sodium hydroxide, the reaction mixture remains in one phase, thereby making the procedure simpler and more repro- ducible. Variation in the relative amount of water to pyridine results in a change of the relative sensitivity of the test towards some of the chlorinated hydrocarbons. Carbon tetrachloride is particularly dependent on the conditions used, the optimum for which we have found to be, as applied to the determination of its vapour in the atmosphere, as follows.To 10ml of analytical-reagent grade pyridine (free from water), containing 0.1 to 1.0mg of carbon tetrachloride, in a 6-inch x &inch thin-walled test-tube add from a burette exactly 0.4 ml of 0.1 N sodium hydroxide. (A slight turbidity will persist at this point.) Heat the tube in a boiling-water bath for 15 minutes and then add 5ml of water and cool in running water to room temperature. Measure the optical density of the solution in an absorptiometer with use of Ilford No. 604 (green) gelatin filters, and determine the carbon tetrachloride content from a calibration curve prepared from known amounts of carbon tetrachloride in pyridine solution treated in a similar manner. Under these conditions, the test is of about equal sensitivity for trichloroethylene, tetra- chloroethane and carbon tetrachloride, although it is still about three times as sensitive for chloroform.The initial deep colours produced by trichloroethylene and tetrachloroethane gradually diminish during the 15-minute heating period. If the same amount of sodium hydroxide, but in a more dilute form, is used for this test and a shorter heating period is used for the development of the colour, the sensitivity towards chloroform, trichloroethylene and tetra- chloroethane is greatly increased compared with tk.at for carbon tetrachloride, which under these conditions is almost negligible. These facts do not appear to be generally known and it is presumably for this reason that the author of a recently published paper' claims that carbon tetrachloride does not give the Fujiwara reaction.Jensovsky maintains that any colour produced is due to an impurity present, which he suggests may be chloroform. We have, however, shown that, under the conditions of test that me use, as described above, the presencl: of 25 per cent. of chloroform in the carbon tetrachloride would be required to give a colour double that produced by carbon tetrachloride alone. I t is admitted that the test, even under o:?timum conditions, is less sensitive for carbon tetrachloride than for chloroform, but we wish to record that under these conditions it does respond to the Fujiwara reaction and that we have been unable to reduce this response by careful fractionation of the analytical-reagent grade of this material. Infra-red examination of the middle fraction showed the absence of chloroform or related materials. Another recent paper, by Hildebrecht,2 describes a method for the determination of chloroform in carbon tetrachloride by the Fujitvara reaction. The author does not claim that carbon tetra- chloride will not give the reaction, but only that the conditions he has chosen are such that inter- ference by carbon tetrachloride is eliminated REFERENCES Mix thoroughly and lightly stopper with a cork. 1. 2. Jensovsky, L., Coll. Czech. Chem. Comm., 19%, 20, Hildebrecht, C. D., Aizal. Chem., 1957, 29, 1037. IMPERIAL CHEMICAL IKDUSTRIES LIMITED RESEARCH DEPARTMEKT GENER.4L CHEMICALS DIVISION WIDNES LABORATORY \l'IDPiES, L.4NCS. 604. T. E. BURKE H. K. SOUTHERN Received November Sth, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300311

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

May, 19581 BOOK REVIEWS 317 Book Reviews THE ENCYCLOPEDIA OF CHEMISTRY. Editor-in-Chief: GEORGE L. CLARK. Pp. xvi + 1037. 1957. This single-volume encyclopedia-a collection of about 800 articles on chemical and allied topics-is written by some 540 contributors, many of them very well known, including repre- sentatives of Government departments, learned societies, industry, universities and institutions, and some consultants. Some of the topics have several authors, each treating a different aspect. Although the subjects are essentially chemical, there are articles about “over 20 sciences that border on chemistry,” but these are treated from a chemical viewpoint. An essential feature of this book is that clarity and lack of overlap have been the editors’ aims, although these have not always been achieved.The amount of information in it is amazingly large, and this the editors attribute to the “effort and time with no monetary compensation” given by the contributors, most articles being “near miracles of condensation.” The comprehensive nature of the book can be seen from a few random titles: Abrasion Re- sistance, Bacteriology, Bile Acids, Chemical Economics, Cryogenics, Debye - Hiickel Theory, Documentation, Elastomers, Furans, Genes, Histochemistry, Industrial Chemistry, Keratins, Luminescence, Magnetochemistry, Microwave Spectroscopy, Nomography, Ores and Ore Dressing, Photometric Analysis, Positrons, Quanticule Theory, Racemization, Steric Hindrance, Textile Printing, Urea Formaldehyde Resins, U S . Food and Drug Administration and Zsigmondy, Richard (1865-1929).The last two titles are examples of important types of article: first, there are over sixty articles about research institutions and foundations and official organisations in N. America, each containing much useful information about their constitution and services; secondly, there are about seventy interesting and comprehensive biographies, but i t is surprising to find that some famous chemists have not been mentioned. Nearly all the chemical elements and their compounds are treated individually, each fairly fully, although the Rare Earths and the Transuranium Elements are treated as groups; however, there is a separate article on Plutonium, in addition to a substantial entry under its group. The editors and their helpers are to be congratulated on carrying out such an ambitious task as collecting together and dealing with all these articles.I can well imagine their difficulties, especially in view of their desire to make the book “a true and up-to-date representation of the chemistry of 1956.” Inevitably in a book of this type and size, with numerous authors, there are inconsistencies of style and minor errors, especially in the names of organic compounds; for example, the formulae on pp. 748 and 749, the names of the compounds used in weed control on p. 991 and the structural formula for caffeine on p. 797, which contains too many bonds. However, this is a minor criticism and these slight errors in no way detract from the value of the book. Unfortunately, the necessarily high price of the book will restrict its circulation, which is a pity, as most chemists should find a t least some parts of i t interesting, and it will be a valuable addition to any set of reference books.New York: Reinhold Publishing Corporation; London: Chapman & Hall Ltd. Price $19.50; 156s. N. C. FRANCIS INSTRUMENTAL ANALYSIS. By PAUL DELAHAY. Pp. xiv + 384. New York and London: This book is an introduction to instrumental methods of analysis and is evidently the result of the author’s experience in this field. He claims that it is intended for undergraduate and graduate students and that the subject matter can be covered in from forty to sixty lectures. The latter is an extravagant claim, but does not detract from the merit of the book; it is an excellent introduction to the subject. The first two chapters explain the scope of the work and give some theoretical principles.Then there follows a series of short monographs on potentiometry, polarography, voltammetry, with titration methods, electrolytic methods with separations by graded potential, coulometry, conductometry and high-frequency methods. The remaining chapters deal with radiation and similar methods and include fluorimetry, turbidimetry, nephelometry, Raman spectroscopy, emission spectroscopy, absorption spectrometry, X-ray methods, mass spectrometry and nuclear- radiation methods. The electrochemical subjects are much better treated than the others, but all the chapters are good introductions to their subjects and they are characterised by lucid and concise theoretical summaries with useful practical details.Throughout the book the author has adopted the The Macmillan Company. 1957. Price $7.90; 55s. 6d.318 BOOK REVIEWS [Vol. 83 European sign convention for potential and he clearly explains the advantages this has for the practical electrochemist. In a work of this kind there are inevitably omissions that strike the reader forcibly; omissions that are possibly deliberate when the author in so small a space summarises so much. For example, no mention is made of Peters as the originator of the equation dealing with redox equili- brium, and, although internal electrolysis has ample mention, no reference is given to the pioneer work of Sand and his colleagues in this field. Similarly, there is no reference to microchemical electrolytic separations by potential control.Attractive features that will appeal to the teacher of analytical chemistry as well as to the practising analyst are the problems posed a t the end of each chapter and the ample references to the original literature and specialist textbooks. A. J. LINDSEY ORGANIC SYNTHESES. An Annual Publication of Satisfactory Methods for the Preparation of Organic Chemicals. Volume 37. Editor-in-Chief: JAMES CASON. Pp. viii + 109. New York: John Wiley & Sons Inc.; London: Chapman & Hall Ltd. 1957. Price $4.00; 32s. The collection for this year includes norbornylene; 3: 4-dinitrohex-3-ene; 5 : 6 : 7: 8-tetrahydro- 1-naphthol; 2-chloro-2-methylcyclohexanone, 2-methylcyclohex-2-enone and isophorone oxide; stearolic, transdodec-2-enoic and 4-ethyl-2-methy:loct-2-enoic acids and ethyl cc-nitrobutyrate; bert.-butyl ethyl malonate and glutaric acid; ethyl benzoylacetate and diethyl benzoylmalonate ; 4-diethylaminobutan-%one; n-heptamide and paxabanic acid; glutarimide; trichloromethyl- phosphoric dichloride; benzofurazan oxide, nicotinamide-I-oxide, 2-benzoylpyridine, 2-chloro- nicotinonitrile, diaminouracil hydrochloride, S-cyano-3-n-heptylcytosine, 5-amino- 1 : 4-diphenyl- and 5-anilino-4-phenyl-1: 2 : 3-triazole; 4-hydroxylbutanesulphonic acid sultone and 3-methyl- oxindole; psezcdopelletierine. A continuous reactor is recommended for the preparation of benzoylacetanilide and oleoyl chloride.It is particularly advantageous with rreactions that proceed relatively rapidly; the short time of exposure to heat results in fewer side reactions and a better quality product than the usual batch processes.A warning is issued that toluene-p-sulphonmethyl-nitrosamide always should be recrystallised and kept in a dark bottle if it is to be stored for a long period. B. A. ELLIS ORGANIC SYNTHESIS. Volume I : OPEN-CHAIN (SATURATED COMPOUNDS; Volume 11 : OPBN- By New York: Reinhold Price $35.00; k14 the set. When ordering in future, it will be necessary to distinguish very clearly between “Organic Synthesis” and “Organic Syntheses.” This small variation of one vowel may involve not merely a considerable difference in the price to be paid, but also a direct antithesis in the type of book supplied. The second, as is well known, gives detailed and checked directions for the preparation of a very limited number of compounds, whereas in the first, the author has set completeness as his goal; occasionally a procedure is given, but it is usually in quite general terms.This work can best be described as the notes of an enthusiastic research chemist, amplified and set out in narrative form. Three chapters deall with reactions (Grignard, Friedel - Crafts and diene synthesis) ; the remaining 27 chapters are on is chemical basis, covering methods of prepara- tion of the various classes of compounds and then their reactions. Clearly this could lead to duplication, but the author has endeavoured to minimise this by suitable, if arbitrary, arrangement of the material while the division is covered by metans of a very full index.A number of topics, e g . , carbohydrates, polypeptides, ozonides, azulenes and modifications of steroid molecules without change in ring structure, are dealt with at some length in the appropriate places. The subject matter is necessarily compressed, but is well documented; the number of references per chapter may run up to over 700. These volumes serve as a useful starting point for the general worker’s searching ; obviously one specialising in organosilicon compounds would find little to assist him in just over three pages. There are numerous references in passing to hetero- cyclic compounds, but they are not considered systematically; the author contemplates a companion volume for this increasingly important group. The books are marred by a considerable number of minor typing errors, over and above those already listed as errata.The statement that a compound with a cis configuration always CHAIN UNSATURATED COMPOUNDS, ALICYCLICOMPOUNDS, AROMATIC COMPOUNDS. VARTKES MIGRDICHIAN, Ph.D. Publishing Corporation; London: Chapman t 5 t Hall Ltd. Pp. xxx + 833; xvi + 835-1822. 1957.May, 19581 BOOK REVIEWS 319 has a higher density and refractive index than its trans isomeride was evidently written before other consideration indicated that the configurations of the 1 : 3-dimethylcyclohexanes based on this rule had to be reversed. B. A. ELLIS CHROMATOGRAPHY: A REVIEW OF PRINCIPLES AND APPLICATIONS. By EDGAR LEDERER and MICHAEL LEDERER. Second Edition. Pp. xx + 71 1. Amsterdam: Elsevier Publishing Co.; London: Cleaver-Hume Press Ltd.; New York: D.Van Nostrand Co. Inc. 1957. Price 72s.; $11.50. The rapid growth in the development and applications of chromatography calls for frequent editions of any book that aims to give an up to date review of the subject. This new edition is one and a half times the length of the first, and has 4000 references compared with the former 1879. It is a veritable mine of information of the work done up to September, 1956, but even so the field continues to widen and grow so rapidly that one has to consult the literature thoroughly after that date. A brief glance at the number of journals mentioned in the list of references makes one aware how widely spread are the original papers, and how thoroughly the authors have tried to include everything.The book follows the original pattern with five divisions to deal with Adsorption, Ion-Exchange, Partition, Applications to Organic and Inorganic Substances. Among the new features of the first division is its extension to describe the adsorption chromatography of gases with some useful tables of data for the chromatography of gases and low molecular weight organic compounds on active carbon and silica gel. There is also a section that discusses the secondary reactions caused by the adsorbent on substances as they pass through a column. It refers mainly t o oxidation of organic molecules, ammonylysis and polarisation, but it serves as a sharp reminder that such effects can occur, and the need to establish that the zones as they emerge from a column are in the state in which they were in the initial mixture before it was put on the column.A similar section in the ion-exchange part again deals with this topic and emphasises that such resins should be regarded as insoluble acids or bases. The ion-exchange section has been rewritten to give a fuller explanation of some of the principles, and extended to give the new developments of ion-exchange papers and phosphorylated papers for separating inorganic cations. There is also a useful account of the adsorption chromato- graphy on ion-exchange resins of organic compounds, and one on the use of modified resins for specific purposes. The scanty mention of the possibilities of molecular sieves and occlusion probably does not do justice to their potentialities. The comparatively short section on Partition Chromatography, which outlines the main principles and techniques, is followed by the division on the application to organic substances.It is, as is well known, in this branch of chemistry that chromatography received its first impetus; consequently this section is about one-third of the whole book, and includes gas - liquid partition chromatography. The material is collected together in chapters, which are devoted to the normal classification of organic compounds: hydrocarbons, acids, alkaloids, carbohydrates, amino acids, peptides, proteins and antibiotics, to name but a few a t random. Here the organic chemist and the biochemist will find the answer to many problems of separation. The authors are to be congratulated for condensing into a small space such a wealth of information.The last division on the applications to inorganic substances finds space to mention a wide variety of separations that have been achieved, and which will suggest many applications in the fields of radiochemistry, geology and chemical industry. From the point of view of the person who wants to try any particular application of chromato- graphy, the book fails to advise him on which particular technique to use. It is this lack of criticism of published work that the reviewer finds most irksome, and, in fact, may tend to bring chromato- graphy into disrepute. After all, chromatography is a means of separation, and it is the quality of the separation in its broadest aspects and the ideal conditions to bring it about that the reader wants to know. If n methods of separation of a group of materials are given, which of these is the best or are they all equally good ? The reviewer could quote examples of work mentioned that could not be repeated in his laboratory, and yet it is given a prominent position in the book, and of reliable work just barely mentioned. It would be a good thing if the authors could be more critical in their selection of material, though the reviewer realises only too well what a gigantic task this would be. This is a book that many chemists and others will want to buy, for it is undoubtedly the most comprehensive book on the subject. F. H. POLLARD

ISSN:0003-2654    

DOI:10.1039/AN9588300317

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

320 PUBLICATIONS RECEIVED Papers for Publication in The Analyst THE Editor welcomes Papers and Notes for insertion in The Analyst, whether from members of the Society or non-members. They are submitted to the Publication Committee, who decide on their suitability for insertion or otherwise. A copy of the current Notice to Authors, last published in full in The Analyst, 1957, 82, 719, can be obtained on application to the Editor, The Analyst, 14 Belgrave Square, London, S.W.l. All Papers submitted will be expected. to conform to the recommendations there laid down and any that do not may be returned for amendment. Fair Copying 'Declaration THE Society is a signatory to the Fair Copying Declaration, details of which, together with a list of signatories, can be obtained from the offices of the Royal Society, Burlington House, London, W.l, on application to the Assistant Secretary.

ISSN:0003-2654    

DOI:10.1039/AN958830320b

出版商:RSC    

年代:1958

数据来源: RSC