摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300545

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

546 INGLESON DETERMINATION OF LEAD Ih; DRINKING XVATER Determination of Lead in Drinking Water BY H. INGLESON M.A. D.PHIL. INTRoDuCToRY.-It has long been recognised that some waters used for domestic supply attack the lead piping connecting the iron mains to consumers’ taps and become contaminated with small quantities of lead. From time to time, cases of lead poisoning have resulted in many parts of the world and in the localities affected much work has been carried out in attempts to reduce the contamination and to form some estimate of the amount of lead which may be safely tolerated in drinking water. These aspects of the problem and the theories advanced to explain the action of waters on lead have been discussed in an earlier publication.1 In the present state of knowledge the extent to which a given water will be contaminated by lead piping under conditions of supply can be found only by direct experiment; it cannot be predicted with certainty from results of analysis of the water entering the lead pipe.This arises from the large number of variable factors such as changes in the nature and concentration of the substances dis-solved or suspended in the water in temperature and in the conditions of drawing the water from the piping. Though it is recognised that there is some relation between the composition of a water and its action upon lead the primary purpose of water analyses as ordinarily carried out is to ascertain the suitability of the water for drinking purposes and not to determine the probable action of the water upon lead or other metals.It is not surprising therefore to find that attempts to correlate the results of ordinary water analyses with the action on lead service pipes have been only partially successful. This point is well illustrated in a recent paper by Kruse2 in which he gives the history of the serious outbreak of plumbism in Leipzig in 1930. After a detailed consideration of the large amount of data this author is unable to account satisfactorily for the outbreak of poisoning. The analytical results did not appear to show a change in the composition of the water sufficient to explain the change in its action upon lead and Kruse concluded that some substance not identified in the analyses had made its appearance and had acted possibly cata-lytically in speeding up the rate of attack on the lead.The Leipzig outbreak gave rise to legal proceedings by some 70 persons, and Fuchss their legal representative during the five years of litigation has collaborated with Bruns and Haupt in a reviewza of the legal medical and chemical aspects. These authors disagree with Kruse’s catalytic theory of the cause of the epidemic. Since the cases arose in newly-erected blocks of flats it is suggested that the new lead piping was in large measure responsible. They think it probable that changes in water composition had occurred but that the analyses made during the relevant period were either not complete or not frequent enough to permit of a decision on the true cause of the excessive attack. It is usually considered that hard waters have little or no action on lead, but some of these waters have in fact a marked action on the metal.For example, Beale and Suckling3 have encountered several instances of very hard well waters INGLESOK’ DETERMINATION OF LEAD IN DRINKING \VATER 547 which were markedly plumbo-solvent and had given rise to lead poisoning and they issue a warning against making the assumption that because a water is hard it is safe to convey it in lead pipes. It seems certain that all waters under service conditions act upon lead piping, even after the lead pipe has been in use for some time but that in many instances the amount of attack is so minute that it may probably be disregarded. The wide distribution of lead in small amount in drinking waters from public supplies is discussed by Weyrauch and Muller,4 who have determined the lead-content of waters in 21 German towns where no suspicion of plumbism had arisen.They found that traces of lead are by no means rare in water particularly after it has been stagnant over-night in the lead pipes. They consider this to be the main source of the lead usually present in human bones. Many of the waters examined had a high temporary hardness and though the amounts of lead were in general, small they were large enough to be detected readily by the usual analytical methods. Many attempts have been made in the past to find out roughly at what point in the gradual transition from non-aggressive to aggressive waters the possibility of danger to health arises. From the chemical side it may be said that the technique of analysing water for lead is fairly well understood and that the results obtained are reasonably accurate.I t follows that so far as the final chemical result is concerned major importance attaches t o the methods used in taking the samples for analysis. In some areas of supply regular tests are made by the following method:-Specimens of bright lead are suspended in the water with or without free access of air for 12 or 24 hours and at the end the lead in the water is determined. In other towns it is customary to determine the lead in the water after it has remained stagnant over-night in the service pipes; the values obtained by this methodare generally assumed to represent maximum concentrations of lead likely to be encountered.* Whether the results obtained in this way do in fact represent maximum values is open to considerable doubt in the absence of direct experi-mental evidence.The water (volume V) left stagnant in a pipe contains a limited amount of corrosive agent The velocity of attack will be greatest in the early stages of contact when the concentration of the active agent is at a maximum and it will fall with time following a curve the shape of which depends on the rate at which the active substance can reach the lead or otherwise assist the corrosion. On the other hand the amount of corrosion that is the weight of lead changed from the metallic state must increase with time to a limit dependent on the amount of active agent present. I t is important to realise that the corrosion product initially formed may undergo change and may be precipitated on and adhere more or less firmly to, the walls of the pipe.When therefore the volume of water V is withdrawn from the pipe after contact for t hours the amount of corrosion product passing out of the This involves medical and chemical questi0ns.l * In taking such samples it is desirable to have as large a volume of water as possible for analysis. A t the same time it is important that the volume of water drawn off should not exceed the volume of the pipe between the tap and the iron street main otherwise dilution with fresh water will occur. These apparently simple conditions are not always easily satisfied in practice because of the complications introduced by connections and by the uncertainty of the exact length of piping buried in the ground or under floors 548 INGLESON DETERMINATION OF L E .~ I N mtmrmG WATER pipe with the sample is not necessarily equal to the total amount of corrosion product formed in t hours. The fraction of the total removed with the sample depends (a) upon the scouring action of the water during sampling (b) upon the physical nature of the corrosion product and (c) to some extent upon the relative positions of the tap and piping. In a vertical pipe the non-adherent corrosion product will tend to settle during quiescence to the lowest part of the pipe and a sample of water taken a t the base of the pipe may contain more lead than an equal volume of water removed from the top. Lead piping is bent in a very irregular fashion in fixing to house walls and the bends may cause trapping of corrosion product.From what has been said it follows that the relation between the amounts of lead removed from a given tap in samples of water of volume V after times t, t, etc. cannot be stated in general terms but must be determined by experiment. Very little information has been published to show how the length of the period of stagnation affects the lead-content of the samples of water. The concentrations (p.p.m.) of lead in I ‘ stagnation ” samples are not average values and cannot therefore when multiplied by the total volume of water ingested, give the total weight of lead taken into the body per day. Furthermore there is very little published information to show how much water is used in the average household per person per day for drinking and cooking purposes only.Much information is available showing the overall requirements of each consumer and in a recent paper before the American Waterworks Association H. V. Pedersens urged the desirability of finding out the fractions of the total consumption which were used in different ways. He gives an instance of an American family of seven persons using 16.1 gallons per head per day for all purposes. Of this he sets aside 6 per cent. for drinking and another 6 per cent. for cooking and kitchen work. For drinking and cooking, about 1.6 gallons per head per day were used. Observations have been made by Magee6 in two Public Assistance Institutions in England of the daily water ingestion of adults. The figures obtained are much smaller than Pedersen’s results.The volumes used for drinking in beverages, soups stews puddings etc. averaged 0.616 gallon per day for men and 0,635 gallon per day for women. These values are in general accord with my observations on one household. Since water plumbism is a chronic condition arising from the ingestion of small amounts of lead over an extended period it is important in attempting to correlate incidence of the disease with the lead-content of drinking water to have a means of measuring the amount of lead removed from a given water-pipe over a period of time long enough to give a true average result. In the following para-graphs a method is described for determining the average concentration of lead in water withdrawn from household services over long periods.The method has been developed as a result of experiments carried out for the Water Pollution Research Board. EXPERIMENTAL.-h the average household water is withdrawn a t irregular intervals of time and in varying amounts. From the outset the apparatus had to be designed for use by the housewife in the kitchen. The principle of the method is simple. Each small charge of lead present in the water drawn for consumption This would help in checking waste of water INGLESON DETERMINATION OF L1SAD I N DRIN KING WATER 549 (i.e. for drinking or cooking purposes) is taken up by a bed of filtering material,* and at the end of the test the total accumulated lead can be found. The sum of the corresponding small volumes of water is obtained from readings of a water meter.The average lead-content of the water consumed can thus be readily determined. For many years past filters have been used to protect consumers at times when a suspected contamination of a water supply was causing apprehension but they have not hitherto been used as an aid to the analysis of water. The active media in these protective filters have been very varied in character; sand,' c h a i - ~ o a l ~ ~ ~ kieselguhr,1° several types of coke8 and calcium phosphatex1 have all been used. Small amounts of lead can be removed from solution by a large variety of sub-stances e.g. filter-paper,12 peat and peat charcoaP3 arid ferric hydr0~ide.l~ In translating the simple principle of the method into practice attention had to be given to the following points:-The filtering medium selected must ( a ) remove the whole of the lead (6) have a high total capacity for lead (c) not introduce any harmful or objectionable matter into the water being filtered and (d) give the mini-mum of trouble in the final analysis.The filter itself must admit of ready replace-ment at the end of the test and since the co-operation of the housewife is needed, it is important to ensure the greatest possible simplicity in operation a good rate of flow and that the apparatus is as compact as possible. In the preliminary experiments the suitability of different materials as filtering media was judged by placing them in glass tubes and passing dilute aqueous solutions of lead salts through them until lead began to appear in the filtrate.When tested in this way cotton-wool and filter-paper were found to be unsuitable for use under the conditions of the average household. Earlier work carried out in this laboratoryx5 had shown that lead can be readily adsorbed from solution by base-exchange materials ordinarily used for water softening. It therefore appeared that they might be used in a finely powdered condition for the removal of both dissolved and particulate lead from water. If only the lead could be recovered subsequently from the filter by appropriate regeneration a convenient method of sampling could be devised. Experiment showed that though lead could be readily removed from solution it was impossible to recover it quantitatively from the zeolite by the methods commonly used in regeneration of base-exchange materials.In one instance which will serve as an illustration of the general hehaviour a total volume of a 5 per cent. sodium nitrate solution some GO times the bulk volume of the base-exchange material was used in regeneration. Of the 100 mg. of lead known to be present in the zeolite successive volumes of 500 nil. of the nitrate solution removed lst 40-3; Znd 7.9; 3rd 3.4; 4th 3.2; 5th (stationary contact), 5.4 mg. making a total of only 60.2 mg. A further 13.0 mg. of lead could be removed by treating the zeolite with 500 ml. of 5 per cent. acetic acid giving a total of 73.2 per cent. recovery. I t may be noted that this incomplete removal of metals from zeolites on regeneration is not peculiar to lead. I t is difficult to remove calcium barium etc.completely from zeolites even by prolonged and repeated regeneration. Probably this behaviour is more marked with lead than with * I am informed by Prof. T. Campbell James of University College of Wales Aberystwyth, that be bas analysed water samples for lead by using cotton-wool to adsorb the metal from aqueous solution. In this way tedious evaporations are avoided 55 0 INGLESON DETERMINATION OF LEAD IK DRINKING WATER calcium because of the higher atomic weight of lead. It was obvious from the experiments that nothing short of complete disintegration of the zeolite would allow the recovery of the whole of the lead which had originally been removed from solution. The zeolites are materials of high silica-content and since 70 to 100 g. of the substance would be required for each test the removal of this silica c1 iz 'x FILTER M E W - -EAOZli GLICC / $ J TEE-BECE Fig.1. - GLASS CYLINDER -DISCS 'PACK ROD -7 would have involved long and tedious treat-ment during analysis. As sampling agents the zeolites were not therefore further considered. Several kinds of refractory and other bricks and filter-candles were next tested but the results were unsatisfactory either because of the incomplete retention of lead or of the imperfect recovery of the metal from the adsorb-ent. One type of filter-candle used was found to be very fragile whilst the method of jointing the body of another type to the collecting nozzle made it impracticable to treat the candle with acid solutions for the extraction of adsorbed lead.In addition the rate of delivery of water under mains pressure was low. Experiments were then made with a " Netafilter " domestic water filter The principle of the apparatus has been explained by Pickard.16 Though the type used in the experiments (Fig. 1) differs in detail from that described by Pickard it admits of the formation under pressure of a compact layer of filtering medium, supported on a column of closely spaced metal discs. The method of operation is as follows :-The pack rod with the head is removed from the thick glass cylinder and the base of the filter is connected with the water supply. A suitable amount (varied according to the material) of finely powdered filter medium is then made into a thin cream with water (400 ml.) and poured into the cylinder.The pack rod is replaced and adjusted centrally by appropriate tightening of the winged nuts. When the water is turned on it passes at high speed through two small jets on opposite sides of the T-piece. An upward swirling motion is imparted to the suspended filter medium which is deposited in an even layer between the plates and outside the removable gauze cage. If the material should deposit irregularly the water is shut off and the pet-cock is opened. This causes air to leak behind the plates thus loosening the bed which falls away from the cage. It can then be re-made and the filter is ready for use. For the experiments described in this paper the great advantage of this type of apparatus over the compressed block or candle is that the nature of the filter medium can be changed or modified to suit the character of the water supply.When an experiment has been completed the filter material can be removed completely from the supporting pack rod and analysed for lead Two types of " Sletafilter " The tap T in Fig. 2 is at the top on the right-hand in Fig. 3 INGLESON DETERMINATION OF LEAD IN DRINKING WATER 551 The photograph in Fig. 2 gives a general view of the complete apparatus. When water is required for consumption tap T is used. This allows water to pass through the meter and filter to be delivered from the nozzle N. Water for other purposes is taken from tap T,. In some towns it is the practice to solder the bib taps direct into the lead piping instead of screwing them into a brass tail-piece which as in Fig.2 is soldered into the lead pipe. An alternative arrangement shown in Fig. 3 can be used in such cases. Details of construction are shown in the scale drawing in Fig. 4. The makers of the filter supply a special grade of kieselguhr for general purposes of filtration. When used to collect lead kieselguhr gave rise to the same difficulty as had arisen with zeolites; the quantitative recovery of the lead was possible only when the material had been completely disintegrated. This involved the removal of 40 to 50 g. of silica during each analysis. Activated carbon appeared to be a suitable alternative since the finely-powdered material could be made to form a coherent bed after addition of a suitable binding agent such as aluminium hydroxide.Carbon however gave trouble in the analysis because the lead was not readily extracted from it The removal of the carbon involves burning it in air or oxygen and during this process the temperature rises locally to over 1000" C. In these circumstances losses of lead oxide which is appreciably volatile cannot readily be prevented. In addition, trouble arises from the sintering of the ash left by the carbon and by the attack of lead compounds on the silica vessel in which the ignition is carried out. Of the other finely-powdered harmless sparingly soluble substances readily available magnesium oxide* seemed likely to give little trouble in analysis. Its absorptive power for lead was tested in the following way. A compact layer of the wet oxide was prepared on a Buchner funnel and a solution of lead nitrate was filtered through it.The lead in the filtrate was not sufficient to give a colour when tested with hydrogen sulphide. A filter of the kind shown in Figs. 2 3 and 4 was charged with 80 g. of mag-nesium oxide (heavy) and its mechanical behaviour was studied under working conditions in a local household. The main drawback to the use of the oxide was that after a few days the rate of delivery of the filtered water began to fall to such an extent that the apparatus was tedious to use. The graph (a) in Fig. 5 represents the general effect observed. Attempts to increase the rate of flow by the use of material of larger grain size were unsuccessful. The decrease had been caused by two factors (1) Teddington tap water has a high temporary hardness (about 16 p.p.100,000) and this brings about a deposition of calcium carbonatet between the grains of magnesia. The whole filter-bed becomes gradually transformed into a hard mass which at the end of a month's use is not easily removed from the pack rod; (2) deposition of ferric oxide silica and small amounts of suspended * During this investigation HoIl1' published an interesting account of the conditions of He developed Our observations 7 In this connection it is interesting to note that the solubilities (g.-mols. per litre) giver :.n MgCO approx. 9.5 x lo-,; Mg water supply in Heligoland and of the outbreak of water plumbism on the island. the use of burnt magnesite as a protective agent for use in percolating filters.confirm his statement that magnesium oxide is an efficient absorbent for dissolved lead. Abegg's Hnndbuch d. anorg. Chem. I1 (2) are CaCO 0.5 x (OH) 1.5 x 10-4 552 INGLESON DETERMINATION OF LEAD IN DRISKING WATER WATER SAMPLING APPARATUS +- I OVUZALL WIDTH Ion' TO MAINS SUPPLY t-I PLAN Fig. INGLESON DETERMINATION OF LEAD I N DRINKING WATER 553 matter picked up by the water during distribution. With the Teddington supply the second of these factors is of minor importance. Since calcium carbonate was being deposited from the water it appeared that, by using this substance in the form of chalk the major cause of the decreased rate of filtration could be removed. When chalk was used trouble was experienced with turbid water from the persistent presence of particles of chalk in the filtered water.After the water had been turned off the bed began to fall away from the cage and when the water was turned on again more turbid water was obtained. It was decided to make use of the ability of magnesium oxide to remove calcium carbonate from the tap water and thus to bind the chalk grains into a reasonably compact filter-bed capable of remaining in position after the flow of water had ceased. It was found that a mixture of 10 per cent. of MgO and 90 per cent. of CaCO not only formed a coherent bed but also that the filter would allow water to pass after three weeks' use in a household at practically the same rate as at the beginning. This is shown in Fig. 5 graph (b). FIGURE 5 I 380 3eo USING MAGNESIUM OXIDE 340 - (b) DITTO USlNG 109'0 MAG.OXIDE 90% CHALK -(a)b-o- TIMES OF DELfVERV OF I LITRE a 320-0 300-$ 280-4 260-240-WATER PRESSURE 40-50 i b G / y In c 2 2 0 -J - 200-& I80 ->- 160 -$ 140- 3 120- 0" toor 40 ~ - ~ - ~ - ~ ~ 0 0 2 4 6 8 10 I 2 14 I6 I8 2022 24 26 2 8 50 32 34 36 38 40 DAY S Fig. 5 Chalk in admixture with magnesium oxide removed suspended and dissolved lead compounds from water so completely that the filtrate gave no colour with hydrogen sulphide. The mechanism of lead removal by calcium carbonate has not been studied fully but it seems not unlikely that the very low solubility of lead carbonate, PbCO, and of the basic carbonate Pb,(CO,) (OH), which Pleissnerls gives as corresponding to 0-04 mg. of lead per litre may cause a small amount of calcium of the chalk to replace the lead in solution.The solubility product* of lead carbonate (PbCO,) would be 3.3 x 1O-l4 whilst that of calcium carbonate is 2.74 x Since the concentration of carbonate ion will remain constant in the presence of the large excess of chalk at 5.23 x it follows that the con-* Abegg's Hadbuch d . mzovg. Chew. 1909 Vol. 111 (2) 726 554 INGLESON DETERMINATION OF LEAD IN DRISKING WATER centration of the lead cannot exceed 6.3 x 10-1O mols. per litre or 0.00013 mg. per litre. If this view of the mechanism is correct it supplies a ready explanation of the absence of lead as shown by the test with hydrogen sulphide in the filtrate from the chalk. Cooper19 has estimated that the sensitivity of the sulphide test for lead is 1 part of metal in 100 million to 196 million parts of water according to the conditions of the test.The first of these values corresponds with 0.01 mg. of lead per litre or about 80 times the amount calculated as in equilibrium with the calcium carbonate-lead carbonate mixture. ANALYTICAL.-some of the problems of analysis of samples of water and excreta for lead have already been discussed.1 The difficulties arise in general from the smallness of the amounts of lead (rarely exceeding 1 mg.) to be determined. Such small amounts of lead are usually associated with 0.5 to 1 g. of other materials in a large volume of water. To ensure in these circumstances the quantitative separation of lead from other metals requires special technique which must be varied according to the conditions.With the filter already described it is possible to obtain an appreciable quantity of lead from a large volume of water (hundreds of gallons if necessary) and to analyse the filtering medium by well-established methods. In the filtering material in some tests the amount of lead has been as high as 200 mg. The result is that the small gains or losses of lead incidental to analytical procedure are rela-tively unimportant. A disadvantage of the method is that the lead is mixed with 50 to 100 g. of the filter material. It is only necessary to deal here in detail with the methods of analysis adopted when magnesium oxide or mixtures of it with chalk were used as filter media, because the use of the other materials introduced such difficulties in the analysis that determinations were both too lengthy and too tedious to form the basis of a satisfactory method.Magnesium Oxide.-After use in the filtration of 50 to 200 gallons of a tap water the magnesium oxide contains besides lead calcium carbonate silica, oxides of iron small amounts of copper and organic matter. The analysis of the material has therefore to be undertaken in two sections (a) on the portion soluble and (b) on that insoluble in dilute hydrochloric acid. The soluble portion con-taining about 80 g. of the oxide in solution together with the bulk of the lead is treated so as to separate the metal from most of the extraneous matter; the organic matter and silica in the insoluble fraction are removed by wet oxidation by nitric and sulphuric acids and treatment with hydrofluoric acid.The residue freed from sulphuric acid can then be united with the main solution a t a suitable stage in the analysis. The tests to find the most suitable analytical procedure for concentrating the lead present in portion (a) were made by adding known weights of lead (as nitrate) to 80 g. of magnesium oxide before dissolving the oxide in hydrochloric acid. On making the solution diluted to 1 litre slightly acid with hydrochloric acid and treating with hydrogen sulphide only about 60 per cent. of the lead was recovered. This is in accord with the observations of Dede and Bonin20 that high concentrations of neutral chlorides hinder and in some instances, prevent completely the precipitation of lead sulphide from hydrochloric acid solution INGLESON DETERMINATION OF LEAD I N DRINKING WATER 555 The second procedure tried was that used by Allport and Skrimshire21 and is based on the observations of H.Fischer,22 that lead produces a brilliant red com-pound with solutions of diphenylthiocarbazone in organic solvents such as chloro-form. By mechanical separation of this solution the lead can be concentrated. When applied to magnesium oxide the process gave an excellent recovery of the added lead; for example of 109 mg. of lead added as lead nitrate 108 mg. were recovered. The method therefore appeared likely to solve the problem of separating the lead, but it should be pointed out that calcium carbonate is present in the oxide after use and so to be of service the extraction must be quantitative in the presence of calcium salts.Magnesiztm Oxide with Chalk.-The application of the method of Allport and Skrimshire to solutions made by dissolving 110 g. of a mixture of chalk and magnesium oxide (MgO 10 per cent. CaCO 90 per cent.) in hydrochloric acid gave unsatisfactory results.* In most instances the lead recovered did not exceed about 15 per cent. of that known to be present. The reason for this marked differ-ence in the presence of calcium salts is not clear and was not investigated since other methods were tried with success. Two further methods of initial precipi-tation were tried namely the separation of lead as the carbonate by addition of sodium carbonate to the nitric acid solution of the mixture and the removal of the lead as the chromate from the nitric acid solution rendered neutral by addition of ammonia.Both methods gave a satisfactory recovery but whereas the addition of sodium carbonate brought down a considerable amount of calcium carbonate the potassium chromate merely precipitated the lead in admixture with the small quantity of magnesium hydroxide produced during neutralisation of the nitric acid solution. The chromate method is therefore preferred because less unwanted material is precipitated and because it possesses the advantage of giving a preliminary rough idea of the quantity of lead present. The next step in the analysis consists in precipitating the lead (and any copper) as the sulphide from hydrochloric acid solution. This separation besides removing traces of alkaline earths prevents the precipitation with the lead in the later stages of analysis of any manganese that may be present.The next stage consists in dissolving the sulphide in hot dilute nitric acid and then depositing the lead as peroxide on the anode by electrolysis in nitric acid solution. The deposition of manganese in this way was shown by Luckow23 in 1880. The lead can then be determined gravimetrically as the sulphate or if very small in amount colorirnetrically as the sulphide. Details of analysis of Jilter material after we.-After being used in a test the whole of the filter material is transferred to a 4-litre beaker with the minimum volume of water. Owing to its length and weight it is convenient to place the filter pack in a glass pie-dish (10.5 x 6.5 x 2.0 inches) during the removal of the medium from the metal framework.Conc. nitric acid is gradually added until just enough acid has been used to ensure the solution of all save the brown or black flocculent matter which floats about even after several hours' gentle boiling. *It should be pointed out that Allport used this reagent for determining lead in samples of dyestuffs in which the total amount of mineral matter was of a different order of magnitude from that employed in the experiments described in this paper 556 INGLESOX DETERMINATIOK OF LEAD IN DRINKING IVATER The filtrate and washings from this solution (A) are allowed to cool and the solid matter and filter-paper are oxidised with conc. sulphuric and nitric acids to yield a residue (B) which is set aside for a later stage.To solution A (approximately 2 litres) is added ammonia until a faint permanent opalescence is seen; 200 ml. of 1 per cent. potassium chromate solution are added t o produce a yellow cloud of lead (and copper) chromate. The boiled cooled solution is filtered by suction through a fritted Jena glass filter (17G.4) from which, after washing it is dissolved by warm dilute hydrochloric acid and boiling water. The resultant solution is repeatedly taken to dryness with conc. hydrochloric acid until no further evolution of chlorine occurs and the solid is then boiled with water to yield solution C. At this stage a second residue of silica often remains; this is added to residue B for treatment with hydrofluoric acid. The residue from this treatment with hydrofluoric acid consists principally of calcium and magnesium sulphates with small amounts of the lead salt.After the removal by evaporation, of the excess of sulphuric acid the residue is united with solution C in a 2-litre conical flask; 1 ml. of conc. hydrochloric acid is added and the solution is diluted to 1 litre. If there are indications that very little copper is present a small amount of copper (50 to 100 mg.) is added as sulphate to ensure the proper coagulation of lead sulphide. In order to hinder the production of colloidal sulphur during the precipitation of the lead and copper as sulphides it is convenient to use the following procedure:-The acidified solution is boiled for 5 to 10 minutes and while the liquid is still boiling the flask is transferred to an asbestos mat.A rubber bung carrying a glass tube which reaches about 2 or 3 inches into the flask is loosely inserted into the neck of the flask. A stream of washed hydrogen sulphide is passed into the flask which is gently agitated. The steam released and the gas together expel any residual air and when the bung is firmly pressed into position the current of gas continues as the pressure in the flask falls with the gradual cooling of the solution. The sulphides separate in a clear liquid which after passage through a fritted glass filter gives a filtrate free from suspended sulphur. Any small residue of sulphides adhering to the flask is removed by boiling dilute nitric acid (30 ml. of conc. acid per 100 ml. of solution) added after hydrogen sulphide has been expelled by boiling a small amount of water in the flask.This hot acid is used to dissolve the main portion of the sulphides; the solution is evaporated and nitric acid is added to bring the free acid concentration to that recommended by Francis Harvey and B~chan.2~ The solution is electrolysed under the conditions of temperature and voltage which these workers recommend and the lead which separates as the peroxide is then determined gravimetrically (as sulphate) or colorimetrically (as sulphide) . Tests with t?zefiZter.-Tests of the ability of the filter medium itself to remove lead from water had already been made so that it remained only to test the operation of a filter charged with the medium. To operate satisfactorily the filter requires a considerable head of water.This head was obtained by placing a bell jar on a laboratory roof and passing a current of water from it through 56 ft. of glass tubing to the filter at ground level. To this water stream a solution of lead of known concentration was slowly added. Before each test water was allowed to run to waste for about 15 minutes. The water providing the head in the bell ja INGLESON DETERMINATION OF LEAD I N DRINKING WATER 557 was obtained from a rising lead main 1 inch in diameter supplying a storage tank ; though the pipe might be expected to influence the lead-content of the water, its replacement by an iron one was impracticable. The retention of the lead pipe, though causing some inconvenience during the tests was well justified by the discovery of a phenomenon hitherto unsuspected.The results obtained in these tests are shown in Table I. Lead added g. 0-1467 0-2052 0.0535 0.0618 (1) (2) (3) (4) TABLE I Lead recovered g* 0.1473 0.2038 0.0562 0.0637 Difference + 0*0006 - 0.0014 + 0.0027 + 0.0019 g. Investigation showed that no significant amounts of lead were obtained from (a) the filter medium (b) the apparatus (c) the water in the iron mains. The lead-content of the water from the rising leaden main was examined with the results shown in Table 11. TABLE I1 Blank Tests Lead found g -(1) Filter medium used as in Table I no lead added * * 0.0075 (2) Water from top of pipe after stagnation over-night . . 0.00064 (3) Repeat of No. 2 . .0.00061 (4) Water stagnant over-night in lead main removed after closing main valve from base of pipe . . . . . . 0.0056 (5) Repeat of No. 4 . . . . 0-0034 The results in Table 11 in which from 5 to 9 times as much lead is obtained from the base as from the top of the pipe may be explained on the assumption that slight disturbances due to vibration or thermal changes occur in the 56 ft. of vertical pipe. These disturbances cause small particles of “protective ” layer to become detached from the walls of the pipe and being specifically heavier to collect in the lower part of the pipe. These particles are unlikely to be removed from this pipe except by prolonged flushing. A second series of tests made after the pipe had been flushed out for four to five hours before each test gave the results in Table 111.TABLE I11 Lead added $5 (1) 0,1752 (2) 0-1562 (3) 0.2845 (4) 0.1383 Lead recovered Difference g* g . 0.1712 - 0.004 0.1583 + 0.0021 0.2886 + 0.004 0.1398 + 0-0015 In a blank test the same filter-bed was used on 5 successive days to collect the This resulted in the collectionof lead left after flushing out for 4 hours each day. 0.0074 g. of lead 558 INGLESON DETERMINATION OF LEAD IN DRINKING WATER Two conclusions can be drawn from these experiments: (1) that the filter removes dissolved and suspended lead quantitatively from the water ; (2) that in sampling drinking water for lead adequate attention should be paid not only to the length of the piping but also to its spatial disposition.Meters.-The meters used in the apparatus illustrated in Figs. 2 and 3 were of the displacement type and were tested under conditions of actual use i.e. by filtering a series of irregular amounts of water until 10 gallons had been recorded on the units dial. Some results are given below. Volume of water delivered Error Meter when 10 gallons recorded Per Cent. A 9974 + 2.6 B 9.82 + 1.8 C 9.68 + 3-2 D 9.78 + 2.9, E 9.74 + 2-6 Results of household tests.-The subjoined Table IV giving a selection of results obtained in tests made in households in different parts of the country is included to illustrate the amounts of lead found in practice. TABLE IV Site A B C D E l? G Period of year October Dec.- Jan. Jan .-Feb. March April-Ma y Aug.-Sept.Sept .-Oct. October Dec.-Feb. Feb.-April April- June Dec.-Feb. Feb.-April April- June Feb.-April April- June June-Sept. Sept .-Nov. 0ct.- Jan. Jan .-April April- June June- July Period of test (Days) 27 34 40 20 23 31 31 13 61 71 54 61 71 54 71 54 92 82 42 63 48 42 Volume of water passed through filter (Gallons) 36 35 43 21 21 33 29-5 33 100 182 150 245 278 144 22 114 70 72 42 148 65 82 PbSO, g. 0.0254 0.0210 0.0240 0.0082 0.0125 0.0194 0.0120 0.0090 0.0027 0.0150 0.0106 0.0150 0.0094 0.0148 0,0394 0,0968 0.0602 0,0362 0.0374 0.1050 0 *0926 0.1866 Pb p.p.m. 0.106 0.090 0.084 0.059 0.090 0.088 0.061 0.041 0.004 0.012 0-010 0.009 0.005 0.015 0.270 0.127 0.129 0-078 0.134 0.106 0.214 0.27 INGLESON DETERMINATION OF LEAD IN DRINKING WATER 559 TABLE IV-continwd Period Volume of water passed through filter of test f A t Site Period of year (Days) (Gallons) PbSO Pb g- p.p.m.H Jan.-April 69 168 0.0590 0.053 April- June 64 131 0.040 0.046 July-Nov. 69 111 0.043 0.058 June- July 46 184 0.1586 0.129 I Jan.-March 59 188 0.3978 0.317 March-June 74 71 0.2484 0.526 June-Aug . 72 277 0.5824 0.315 Aug.-Nov. 87 273 0.17’75 0.094 SUMMARY AND CONCLUSIONS.-A new method is described for finding the average lead-content of water used in a household for drinking and cooking. By means of the apparatus the housewife can draw water for these purposes at such times during the day as it is required.A filter abstracts the small amounts of lead from the water while a meter records the sum of the small volumes used over a period of several weeks. In this way it is not only possible to sample the water under actual conditions of use but also to eliminate the irregularities associated with The weight of lead to be determined is so very much greater than that present in snap samples that an analytical separation of the lead can be made. It is thus possible to avoid the use of analytical devices commonly employed to prevent interference with colour tests by other metals during the determination of minute amounts of lead. The assistance courteously extended by a number of water engineers has enabled tests to be made under a variety of conditions at pressures ranging from 40 to 100 lbs.per sq. in. and with waters of widely differing characteristics. The purpose of the experiments made has been to develop and improve a method of sampling for lead and thus facilitate the carrying out of the main objective-that of trying to correlate the attack 0x1 lead piping with the character of the water and other conditions of supply. snap” sampling. < c The investigation described in this paper was carried out as part of the programme of the Water Pollution Research Board. My thanks are due to Mr. A. Harrison for the analysis of numerous samples so far obtained from 18 sites. The investigation has also been materially helped by the assistance received from Mr.A. L. Davis and Mr. W. H. Sullivan during the testing and installation of apparatus. For permission to publish I am indebted to Sir Gilbert Morgan Director of the Chemical Research Laboratory and to the Department of Scientific and Industrial Research. REFERENCES 1. “ The Action of Water on Lead with Special Reference to the Supply of Drinking Water.’’ H.M. Stationery Office 1934, W. Kruse 2. Hyg. Infekt. Kr. 1936 118 143. Theo. (Water Pollution Research Technical Paper No. 4. price 2s.) Steinkopff Dresden and Leipzig 1938. 2. 2a. H. Fuchss H. Bruns and H. Haupt Die Bkiverg&ftungsgefahr durch Trinkwasser 560 DAUBNEY AND NICKOLLS AN INVESTIGATION INTO THE 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. J. F. Beale and E. V. Suckling Brit. Med. J. 1931 (2) 165. F. Weyrauch and H. Muller 2. Hyg. Infekt. Kr. 1935 117 196. H. V. Pedersen J . Amer. Wat. W k s . Ass. 1937 29 370. H. E. Magee J . Hyg. Camb. 1937 37 30. J. Brown Clinical and Chemical Observations on Plumbism. S. White Rept. on the Action of Shefield Water on Lead Pipes Jan. 21 1886. N. Porritt Menace of Eclampsia in EngZand and Wales Oxford Univ. Press 1934. P. Schmidt Int. 2. Wasservers. 1914 1 11. H. Slatter cf. J . SOC. Dyers 4% Col. 1889 5 61. J. C. Thresh ANALYST 1924 49 124. T. Campbell James Min. Agric. Fish Stand. Conim. Ricers PoEl. Serial No. 219 Repi. V. A. Elsden and J . F. Stansfield J . Chenz. SOC. 1924,60 485. B. A. Adams Water Eng. 1930 32 415. J. A. Pickard J . SOC. Chem. Ind. 1930 49 2 5 9 ~ . K. Holl Archiv Hyg. u. Bakt. 1935 113 288 296. M. Pleissner Arb. Gesundheitsarnt Berlin 1907 26 384. A. J. Cooper J . SOC. Chem. Ind. 1886 5 84. L. Dede and P. Bonin Ber. 1922 55 [B] 2327. N. L. Allport and G. H. Skrimshire ANALYST 1932 57 440. H. Fischer 2. angew. Chem. 1929 42 1025. C. Luckow 2. anal. Chem. 1880 19 1. A. G. Francis C . 0. Harvey and J. L. Buchan ANALYST 1929 54 725. Thesis 1889. No. 147 Jan. 1927. CHEMICAL RESEARCH LABORATORY (DEPARTMENT OF SCIENTIFIC AND INDUSTRIAL RESEARCH) TEDDINGTON MIDDLESEX May 193

ISSN:0003-2654    

DOI:10.1039/AN9386300546

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

560 DAUBNEY AND NICKOLLS AN INVESTIGATION INTO THE An Investigation into the Methods of Toxicological Analysis of Viscera BY C. G. DAUBNEY M.Sc. F.I.C. AND L. C. NICKOLLS M.Sc. F.I.C. PART 11. THE EXTRACTION OF ALKALOIDS FROM VISCERA INTRoDucTIoN.-In Part I of this investigation1 an attempt was made to separate and consider each point of difficulty in the classical Stas-Otto process of determining alkaloids in viscera and as a result a modified method was devised which tested on quinine and morphine gave satisfactory results. We have satisfied ourselves moreover that with the technique advocated as good results can be obtained from experiments on animals injected with the required alkaloid as from experiments involving the simple addition of alkaloid salts to minced viscera.Accordingly in this part of the investigation the experiments were confined to those involving this addition of alkaloid salts to minced viscera. This is more particularly advisable since the alkaloids now dealt with are easily hydrolysed and under the influence of metabolic processes undergo comparatively rapid degradation. Hence the percentage of such alkaloids recovered in an animal experiment would yield little information on the efficiency of a recovery process. Subsequent to the sending for publication of Part I of this investigation, Stewart Chatterji and Smith2 published some work on the subject on lines similar to those we have followed and to the experiments published by P e l t ~ e r . ~ Their method consists essentially in precipitation of the proteins with tri-chloroacetic acid adsorption of the alkaloid with kaolin and its elution wit METHODS OF TOXICOLOGICAL ANALYSIS OF VISCERA 561 chloroform or ether after being rendered alkaline with sodium carbonate and dried with sodium sulphate.During this process even easily hydrolysable alkaloids are subjected to the action of hot conc. sodium carbonate solution which does not appear to be very desirable. As stated previously we discarded adsorption and elution processes largely owing to the lack of adsorbing power of the materials used in the presence of comparatively large amounts of strong electrolytes. In addition moreover we found that under the conditions of quantitative adsorption and elution the resulting alkaloid was always contaminated with a by no means negligible amount of biological extractives which involved a subsequent rigorous purification of the isolated alkaloid.We have now tried trichloroacetic acid as a protein precipitant and find that the resulting coagulated protein has a somewhat rubbery consistence and so does not in our opinion lend itself to washing and draining as readily as the more granular precipitate produced by ammonium sulphate. In general in our hands trichloroacetic acid has given lower results than ammonium sulphate. We have endeavoured to prove the general applicability of the method described by us (Zoc. cit.) by determining the efficiency of the process for the recovery of strychnine cocaine atropine and aconitine. These together with quinine and morphine the two alkaloids originally selected provide a range of alkaloids varying in their behaviour with immiscible solvents and in degree of stability to hydrolytic agents.It will be agreed we think that this range is sufficiently general for it to be assumed that all other alkaloids will behave similarly. One of the fundamental requirements of the Stas-Otto process is the utilisa-tion of a series of solvents petroleum spirit ether chloroform etc. to effect a separation of any mixture of alkaloids. Such a separation may be necessary but we consider it essential to isolate in the first place the mixture of alkaloids in a fairly pure state. Accordingly we have modified our original process so as to be applicable to all alkaloids. Experiments with the further additional alkaloids specified have shown that morphine is unique in being extracted from tissue by saturated ammonium sulphate solution.All the other alkaloids are extracted to only a small extent by 1 to 1.5 litre of this solution. Accordingly in any general scheme the extraction medium must be acidulated water. Again chloroform is the only solvent which dissolves all alkaloids with the exception of morphine which requires special treatment. GENERAL METHoD.-The general method of extraction of alkaloids from viscera finally adopted is as follows: (i) Freeze the tissue overnight in the ice compartment of a refrigerator. (ii) Mince 400 g. or other suitable quantity into a tared casserole while the material is still frozen. (iii) Add 50 ml. of water and 10 ml. of glacial acetic acid and warm with stirring to about 50" C.(iv) Add sufficient ammonium sulphate (200-300 g.) to leave a small amount undissolved and warm with stirring to about 65" C.; the protein will then have coagulated and the thick gruel will have become quite fluid. Filter on a large Buchner funnel and wash with about 100 ml. of warm water (65" to 70" C.). (v 562 DAUBNEY AND NICKOLLS AN INVESTIGATION INTO THE (vi) Return the residue to the casserole macerate at about 65" to 70" C. with approximately 200 mi. of water containing 1 per cent. of acetic acid until the mixture has been stirred into a thin gruel free from lumps and filter. (vii) Repeat the maceration of the residue with hot acidulated water until approximately 1-5 litre of total filtrate has been obtained.* (viii) Transfer this filtrate to a 2-litre separating funnel and render alkaline with ammonia.Extract five times with 100-ml. portions of chloroform and filter the chloroform extracts. Unite the filtered chloroform extracts and extract successively with 25, 15 and 10-ml. portions of 3 N sulphuric acid followed by 25 ml. of water and filter the aqueous extracts in turn through a small filter. (x) Render the united aqueous liquors alkaline with ammonia extract with five 20-ml. portions of chloroform and filter into a small carbon dioxide flask. (xi) Evaporate the colourless extract to dryness in the flask and weigh. (xii) Dissolve the contents in a little dilute acid and filter the solution through a small paper. Wash the flask and paper well with dilute acid (if necessary the washings can be left separate from the main bulk of concentrated alkaloid solution).(xiii) Return any small amount of insoluble matter on the filter to the flask by dissolving it in acetone and then in chloroform. Dry and re-weigh the flask, the difference from the previous weight representing the weight of pure alkaloid. This general method serves for all alkaloids except morphine. For that alkaloid 200 ml. of alcohol are added after (vii) above and extraction is repeated with chloroform and alcohol. If morphine only is sought however it is preferable to adhere to the method given in Part I. (ix) There are a few points which require amplification: (a) The mixing of the ammonium sulphate extract with the water extracts causes a turbidity. No attempt must be made to filter this.The admixture must, however be made since with certain alkaloids the amount dissolved by the ammonium sulphate solution though small is not negligible. If the heating of the tissue with ammonium sulphate is not sufficiently thorough or if the material is abnormal in behaviour possibly owing to disease, persistent emulsions in the chloroform layer are formed on extraction. This need cause no delay. The chloroform emulsion is fdtered through a small Buchner funnel containing a filter-paper with a half-inch layer of acid-washed sand on top; this breaks the emulsion. Subsequent extractions are run through the same funnel so that only a small washing with chloroform is needed at the end to free the funnel and its contents from alkaloid. Filtration of the chloroform and aqueous liquors at every stage is essential for securing a high-grade product at the termination of the experiment even though it is not apparently necessary.With the technique as finally evolved it is unnecessary to remove any (b) (c) (d) traces of fat from the acid filtrate since the fat remains in the first chloroform extract. * For quantities of tissue weighing as little as 200 g. the volume of dilute acetic acid used for extraction may be reduced somewhat bur. we make it a practice to finish with not less than 1 litre of extract. The quantities of chloroform used subsequently are not altered METHODS OF TOXICOLOGICAL ANALYSIS OF VISCERA 563 (e) Too much stress cannot be laid on the necessity of confirming where possible the purity of the alkaloid returned as I‘ yield ” in quantitative work of this nature.EXPERIMENTAL.-h the experiments described below the alkaloid was added in aqueous slightly acidulated solution to the minced material prior to its acidifica-tion with acetic acid. The “standard method” referred to in the following tables is the general method described above : TABLE V QUININE EXPERIMENTS 50 mg.* of alkaloid used in each experiment Yield of quinine Expt . Material Crude Purified Per Cent. Per Cent. Remarks - 27 Liver,400g. 89 86 Standard method. Purified by precipitation as tartrate. + 50 ml. of blood 28 Liver,400g. 92 84 Standard method except that the material was macer- + 50 ml. of ated with water acidulated with trichloroacetic blood instead of acetic acid.29 Liver,4OOg. 86 56 Protein coagulated with 50 g . of trichloroacetic acid in + 50ml.of 50 ml. of water and material macerated with water blood acidulated with trichloroacetic acid. * Quinine weighed as quinine hydrate and dissolved in a minimum of hydrochloric acid N.B.-Compare the result in Expt. No. 27 with those in Expts. Nos. 7 and 11 (Pt. I). prior to addition to minced material. TABLE VI STRYCHNINE EXPERIMENTS 50 mg.* of alkaloid used in each experiment (except No. 35) Yield of strychnine Expt. Material Crude Punfied-f Per Cent. Per Cent. Remarks A 30 Liver 350g. 99 78 Mass extracted with a total of 1-5 litre of saturated ammonium sulphate solution (27 per cent. yield), followed by extraction with a total of 1.5 litre of acidulated water (51 per cent.yield). Emulsions not broken by sand. 31 Liver 340g. - 77 32 Liver 340g. - 89 33 Liver 400g. - 115 34 Liver 350g. - 93 35 Liver 350 g. + 5mg. of strychnine Standard procedure. N.B.-The chloroform extraction is incomplete if the emulsions are not broken. The loss naturally depends on the type of emulsion. As Expt. 30. Recovered strychnine crystalline but straw-coloured and slightly impure even after recovery as ferrocyanide. Standard procedure. Strychnine dry colourless crystals. Standard procedure. Strychnine dry pale straw-coloured crystals. Emulsion not broken by sand. I * , * IJ * Strychnine weighed as free base and dissolved in a minimum of hydrochloric acid prior to t The yield represents the amount of alkaloid recovered after purification by precipitation as addition to minced material.ferrocyanide (4) 564 DAUBNEY AND NICKOLLS AN INVESTIGATION INTO THE TABLE VII COCAINE EXPERIMENTS 60 mg.* of alkaloid used in each experiment (except Sos. 39 40) Yield of Expt . Material cocaine Remarks Per Cent. 36 Liver 350g. . . 56 37 Liver 400g. . . . . 84 38 Liver 350g. . . 85 mg-39 Liver 350g. + Cimg. of 2-7 40 Liver 400 g. + 5 mg. of 5.7 cocaine cocaine Liver somewhat diseased rendering filtration difficult. Filtrate made alkaline with NaHCO and cocaine extracted with petroleum spirit throughout. Final product readily and completeIy crystallisable. Petroleum spirit gave very persistent emulsions which rendered complete extraction impossible. As Expt.36 but ether used in place of petroleum spirit. Standard procedure; chloroform used as solvent and emulsion broken with sand. Cocaine solidified to mass of pale brownish crystals. As Expt. 37 ether used. Yield not satisfactory. Standard procedure. Cocaine solidified to thick pasty mass of crystals. * Cocaine weighed as cocaine hydrochloride and dissolved in acidulated water prior to addition to minced material. N.B.-The cocaine was extracted with petroleum spirit instead of chloroform from the final acid liquor to increase the purity of the product. In each experiment microscopical and chemical tests were made on the final product. TABLE VIII ATROPINE EXPERIMEXTS 50 mg.* of alkaloid used in each experiment (except No 44) Yield of atropine Expt .Material Crude Punfied Per Cent. Per Cent. -7 Remarks 41 Liver 400g. 72 L Standard method except filtrate made just alkaline with NaHCO,. The solution was probably not sufficiently alkaline to obtain complete extraction. 42 Liver 400g. 97 79 Standard method. 43 Liver,400g. 89 - Standard method except that material was macerated with water acidulated with 1 per cent. of trichloro-acetic instead of acetic acid. 44 Liver 400 g. 7 - Standard method. Product gave strong reactions in all the characteristic chemical tests. + 50 ml. of blood + 5mg. of atropine mg-*The atropine was weighed as sulphate and dissolved in water prior to addition to the * The atropine was purified by precipitation with Wagner’s reagent and weighed as minced material. hydriodide-octaiodide (5) METHODS OF TOXICOLOGICAL ANALYSIS OF VISCERA 565 Expt 45 46 47 TABLE IX ACONITINE EXPERIMENTS Yield of Per Cent.Material crude aconitine Remarks * Aconitine weighed out as hydrobromide and dissolved in acidulated water prior to addition + Cf. Discussion of Results infra. to minced material. Liver 400 g. + 50 mg. 75 Standard method. The yield was low as this was a of aconitine* preliminary experiment. Repetition not considered necessary owing to low poisonous dose. Final product formed resinous mass which crystallised from ether in fine colourless needles. mg-Liver 400 g. + 5 mg. 6.0 Standard method. Gave characteristic chemical and of aconitine* mg. physiological tests. Liver 400 g . + 0.5 mg. 4.2 Standard method. General chemical alkaloidal tests of aconitine* strongly positive.Physiological tests positive. The crude alkaloidt was dissolved in 10ml. of acidu-lated water. 1.0 ml. of the solution was diluted again to 10ml. with water and four rats were injected intraperitoneally with 1.5 2.5 2.5 3.5 ml. respec-tively of this solution. The rat receiving 3-5 ml. died in 14 hours; one of the 2.5 ml. rats died a t end of 24 hours. The other two recovered after severe symptoms of aconitine poisoning. DISCUSSION OF REsuLTs.--The results obtained show that the modified method of extraction of alkaloids from viscera described here is suitable for the determination of quinine strychnine cocaine atropine and aconitine. The method has throughout been treated as a routine method and the experiments quoted are typical experiments.Without doubt by taking extreme precautions such as extracting the minced material with further quantities of acidulated water from which the alkaloid is extracted separately and by further exhaustive extraction of all the solutions obtained the yield of alkaloid could be increased. It is doubtful, however whether this is worth while when the quantity of alkaloid present is small. In those experiments involving 5 mg. of alkaloid it will be noticed that the quantity of alkaloid weighed approximates closely to the actual quantity present, the amount of adventitious matter being only 1.5 to 2 mg. In Experiment 47 the adventitious matter is slightly more owing to an attempt to avoid loss of alkaloid. It may be pointed out that good yields depend on removing completely each chloroform extract before re-extraction.The advantage of breaking any residual emulsion by filtering through sand under suction is due to the complete separation of the chloroform layer obtained by this method. We have tested the method of Stewart Chatterji and Smith (Zoc. cit.) though not exhaustively because we have had so far greater success with our method. Work is in progress to adapt our method to the determination of synthetic drugs such as the barbiturates amidopyrin etc. SUMMARY.-A modified method of extraction of alkaloids from viscera is described which is suitable for any alkaloid. It has been demonstrated that good yields can be obtained and that the amount of contaminating adventitious material is small 566 TAYLOR-AUSTIN 1 THE DETERMINATION OF ALUMINIUM IN CAST IRON REFERENCES 1. C. G. Daubney and L. C. Nickolls ANALYST 1937 62 851. 2. C. P. Stewart S. K. Chatterji and S. Smith Brit. Med. J . 1937,11 790; Abst. ANALYST, 3. J. Peltzer Chem.-Ztg. 1933,57 816; Abst. ANALYST 1933,58 773. 4. C . Simmonds ANALYST 1914,39 81. 5. A. B Prescott and H. M. Gordin J . Amer. Chem. SOC. 1898 20 706; Abst. ANALYST, 1937 62 891. 1898,23,324. METROPOLITAN POLICE LABORATORY January lst 1938 HENDON LONDON N.W.

ISSN:0003-2654    

DOI:10.1039/AN9386300560

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

566 TAYLOR-AUSTIN 1 THE DETERMINATION OF ALUMINIUM I N CAST IRON The Determination of Aluminium in Cast Iron BY E. TAYLOR-AUSTIN A.I.C. INTRODUCTION.-At the present time no reliable method is available for the determination of aluminium in cast iron and ferro-alloys. The literature so far published on the subject is at variance as to the merits of existing methods. Thus, Lundell Hoffman and Bright1 regard the phosphate method as inaccurate but not sufficiently so to exclude it for routine analyses. Ibbotson2 gives this particular method as the best whilst Naish and ClennelP suggest its use for the determination of small amounts of aluminium only and recommend precipitation as hydroxide by means of sodium thiosulphate in reduced solution for larger amounts of the element.A survey of the published literature shows that the phosphate method which is probably that most widely used to-day was first introduced by Stead4 in 1889; a subsequent slight modification by Carnot5 appeared in 1891. In present-day literature this method differs only in very slight detail from the original and it is obvious from a study of the methods of 50 years ago that it was devised to deal with the small amounts of aluminium remaining in steel after the addition of the element as a deoxidant. In view of the rapidly increasing importance of aluminium as an alloy addition to cast iron and the consequent need for its accurate determination in com-paratively large amounts an investigation of the existing methods has been carried out. Special attention has been paid to the use of 8-hydroxyquinoline a reagent recently introduced into inorganic analysis by Berg6 and his co-workers.It has been found that this reagent may be used under certain conditions which are given in detail later (p. 587) for the determination of aluminium in cast iron and ferro-alloys and the results obtained are of a high standard of accuracy both for small (0.05 per cent.) and large (14 per cent.) amounts of the element. This investigation has occupied a period of more than twelve months and every detail of the method subsequently given has been carefully studied and the conditions of precipitation verified by numerous determinations carried out on both synthetic solutions and samples of cast iron containing aluminium. The effects of other alloying elements such as nickel copper chromium molybdenum, etc.as well as titanium and vanadium have also been examined in the same manner TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 567 THE SEPARATION OF ALUMINIUM FROM IRoN.-The primary difficulty encoun-tered during the determination of aluminium in ferrous materials is its separation from the large amounts of iron associated with it. There are numerous methods in use at the present time for achieving this object as follows:-(1) the Roth6 ether separation; (2) the sodium hydroxide separation; (3) the precipitation of aluminium as phosphate in reduced solution; (4) the precipitation of aluminium as hydroxide in reduced solution; (5) the removal of iron by precipitation with hydrogen sulphide in ammoniacal tartrate solution ; (6) the sodium bicarbonate separation ; (7) electrolysis with a mercury cathode.All these separations with the exception of the last have been studied in detail and it appears that none is entirely satisfactory when used singly for amounts of aluminium in excess of 1 per cent. A combination of the bicarbonate separation and the ammonium sulphide-tartrate separation has proved to be the most satisfactory. The Rothe ether separation introduced by Rothe' in 1882 is tedious and exacting. Its success depends upon the sp.gr. of the test solution since a solution of sp.gr. 1-100 to 1-105 saturated with ether retains the smallest amount of ferric chloride. It cannot be employed as an absolute separation as some iron is invariably left in the aqueous layer.Considerable experience is necessary to obtain satisfactory results with this separation. Precipitation of the aluminium as phosphate in reduced solution is regarded by many of the later writers as unreliable as previously stated and the present investigation has shown that this is undoubtedly correct. With very small amounts of aluminium the error may be negligible for routine purposes as stated by Naish and ClennellJ3 but for the determination of more than 0-5 per cent. of aluminium the method is very unreliable. Further it is subject to serious interference from such elements as titanium and vanadium which are frequently present in cast iron. Chromium also causes interference and no entirely satisfactory method of removing it from the precipitated phosphates except in very small quantity has yet been devised.Titanium and vanadium may be removed by fusion of the mixed phosphates with sodium carbonate but this procedure renders the method long and tedious and the final result even more uncertain. Precipitation as hydroxide in reduced solution may be accomplished by means of either sodium thiosulphate as recommended by Naish and Clennell? or phenylhydrazine recommended by Hess and Campbell.* Both these methods have been examined. The former yielded only about 75 per cent. of the aluminium present, whilst the latter failed to separate the large amount of iron present If however, the bulk of the iron present was removed by one of the other separations the latter method yielded as good results as could be expected from a hydroxide method.The hydrogen sulphide separation in ammoniacal tartrate solution28 shows greater promise. It has the disadvantage that samples of more than 1 g. cannot be conveniently handled owing to the unwieldly bulk of the iron sulphide precipi-tate obtained but for separating less than 1 g. of iron it is absolutely quantitative, though the statement of G a d e a ~ ~ that the iron sulphide does not adsorb aluminium, appears to be incorrect if more than 1 per cent. of that element is present. The same writerg states that the precipitated sulphide filters and washes well but thi 568 TAYLOR-AUSTIN THE DETERMINATION OF ALI?MINIT_;M I N CAST IRON statement also appears to be erroneous for although a great variety of filter-papers has been tried the iron precipitate was found to filter and wash very badly indeed unless not more than 20 mg.of iron was present. This difficulty was subsequently overcome by fractionation as will be seen later. This separation has the advantage of removing other elements such as manganese and nickel together with the iron. Lundell Hoffman and Brightlo state that the separation of manganese and nickel by this method is incomplete but the amount remaining in the solution after filtration is of the order of 0-1 mg. and is therefore negligible. The sodium hydroxide separation is stated by Lundell Hoffman and Brightll to give low results and Lundell and Knowles13 suggest that this is due to the retention of aluminium by the precipitated ferric hydroxide. This is rather to be expected as the adsorbing properties of ferric hydroxide are well known and the present work has confirmed this fact.The difficulty may be largely overcome by making a second sodium hydroxide separation after dissolving the first precipitate in acid but this is inadvisable since by so doing large quantities of sodium salts (probably more than 20 g.) are introduced into the final solution and their presence is objectionable. The sodium bicarbonate separation has proved to be valuable for separating the aluminium present from most of the iron prior to the use of one of the other separations. It removes about 90 per cent. of the iron present and a similar proportion of the manganese. The precipitate contains in addition to aluminium and a little iron and manganese any titanium vanadium chromium copper tin or phosphorus which may be present at the time of precipitation.In order to study the various methods for the precipitation of aluminium in synthetic solutions with or without the addition of iron or other elements (such as titanium chromium etc.) several standard solutions were prepared. Standard Aluminium Chloride Solution.-The basis of this solution was a sample of aluminium metal supplied by the Royal School of Mines and its chemical analysis was as follows:-aluminium 99-70; silicon 0.13; iron 0.15 ; total 99-98 per cent. The solution was prepared by dissolving 4.2000 g. of this aluminium metal in 40 ml. of 50 per cent. hydrochloric acid diluting filtering and making up the volume of the resulting solution to exactly 2000 ml.; 50 ml.of this solution were theoreti-cally equivalent to 10.50 mg. of aluminium. This value was repeatedly checked by several modifications of the oxine method and was found to be 10-30mg. per 50 ml. of solution. The filtrates from these check determinations were tested for aluminium by the colorimetric method of Lampitt and Sylvester,l4 but none could be detected thus proving the completeness of the precipitations. Iron.-For the additions of iron a sample of " Powdered Iron" was employed; its chemical analysis was as follows :-iron 99-65; carbon 0.10 ; manganese, 0.17 ; silicon trace ; phosphorus 0.01 ; sulphur 0.03 ; titanium nil ; vanadium, nil; chromium nil ; total 99.96 per cent. The preparation of other standard solutions is described later under the headings of the appropriate elements.THE PRECIPITATION OF ALUMINIUM AS HYDROXIDE AND SUBSEQUENT IGNITION TO OXIDE.-The aluminium having been successfully separated from th TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIU&I I N CAST IRON 569 iron associated with it it becomes necessary to decide by what method the solution shall be treated to obtain the final result. At the present time precipitation with ammonium hydroxide followed by the ignition of the precipitate to aluminium oxide is extensively employed. This method however is open to serious objection. In the first place precipitation with ammonium hydroxide needs careful manipula-tion if correct results are to be obtained. The $H of the solution is of great import-ance; the minimum solubility range of aluminium hydroxide lies between the limiting pH values 5.7 and 6 ~ 3 .l ~ Thus an excess of ammonium hydroxide must be avoided; the use of methyl red as an indicator for this purpose is to be strongly recommended since its yellow colour is developed at PH 6 4 . For the precipitation of small amounts of aluminium an excellent procedure has been evolved by Lampitt and Sylvesterl4 and is carried out as follows:-The solution containing the aluminium is treated with a very slight excess of ammonium hydroxide and it is then boiled until the vapours no longer smell of ammonia. These authors1* state that "this is the most important part of the procedure and is the only point where serious error can be made. The excess of ammonia must be removed otherwise an appreciable loss of aluminium will occur.On the other hand if the boiling is unduly prolonged the solution becomes distinctly acid and the aluminium re-dissolves." This procedure has been extensively used during the present investigation for the testing of precipitates for adsorbed aluminium and for testing filtrates to ensure that precipitation has been complete. Such tests were completed by the colorimetric method using aurintricarboxylic acid as described by Lampitt and Syl~ester.1~ Thus by careful control of the @H of the solution the whole of the aluminium may be precipitated with ammonium hydroxide. When the materials such as cast iron contain phosphorus this element is precipitated completely or in part, together with the aluminium. Titanium is similarly precipitated.It is therefore necessary to determine the amount of phosphorus in the precipitate and to deduct its equivalent of P,O from the weight of aluminium hydroxide,15 etc. recorded if ignition to oxide is chosen as the method for the final determination. The presence of titanium necessitates the fusion of the precipitate with anhydrous sodium carbonate and the removal of the titanic acid so obtained after extraction in water. The procedure of ignition to oxide is however by no means satisfactory, except for the determination of small amounts of aluminium (up to 0.5 per cent.). First it is very difficult to ignite the hydroxide completely to oxide since it retains its elements of water very tenaciously and further the ignited alumina is very hygroscopic and is therefore difficult to weigh accurately.The present investigation has shown that results obtained by this method are invariably high, quite apart from any interference due to phosphorus or other elements. Numerous precipitations were made from pure aluminium solutions with ammonium hydroxide and the precipitates so obtained were ignited in platinum crucibles at temperatures ranging from 1000" to 1450" C. The results are shown in Table I and it will be seen that the figures are still above the theoretical even after ignition at the highest temperature. Although the plus error is small being of the order of 0.6 mg. it is by no means negligible since it represents an error of 0.06 per cent. of aluminium if samples of 1 g. are employed. For larger amounts of aluminiu 570 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I X CAST IRON the error would become more and more serious especially if samples of 0.5 g.or less are employed. For this reason together with the fact that the procedure for estimating phosphorus or for removing titanium renders the process long and introduces greater possibilities of error the method is not one to be recommended for the determination of aluminium in cast iron or any other alloy. Weight of A&03 obtained and corresponding weight of A1 Aluminium in synthetic solution mg. 20.00 10.30 10.00 10.00 10.00 Ignition temp. 1000" C. * 4 0 3 A1 mg. mg. 38.9 20.58 21.0 11.11 20.0 10.58 20.4 10.79 20-4 10.79 Ignition temp. 1250" C. Ignition temp. 1450" d. * * A403 A1 A403 A1 mg.mg. mg- mg+ 38.7 20.48 - -20.9 11 -07 19.9 10.54 - -20.3 10.74 19.9 10.54 20.3 10.74 19.9 10.54 - .__ PRELIMINARY EXAMINATION OF HYDROXYQUINOLINE METHODS.-~-H~~~OX~-quinoline or oxine is a light buff-coloured crystalline solid m.p. 73" to 75" C. It forms insoluble complexes with certain metals by substitution of its phenolic hydrogen atom. These compounds are characterised by their definite constitution, their stability and their crystalline easily filterable form; they are believed to be either chelate or co-ordination compounds. Further owing to the high molecular weight of the organic part of the molecule these compounds are very voluminous and heavy; this means that they contain only a small percentage of the metal and thus results are of a high standard of accuracy and further small amounts of metal give rise to large precipitates so that traces may be readily detected and determined.These remarks are of special interest when considering aluminium which is a tervalent metal of comparatively low molecular weight and consequently combines with three mols. of oxine to give (C,H,ON),Al which contains only 5-88 per cent. of aluminium. Solvents.-The reagent is insoluble in cold water but comparatively soluble in hot water from which it may be recrystallised if necessary for purposes of purification. It is also soluble in alcohol acetone and acids. For general purposes an alcoholic solution is used but for the determination of aluminium this is impracticable since the aluminium oxine complex is itself slightly soluble in this solvent.The alcoholic solution will keep only for a few days but an acetic acid solution is stable for a period in excess of one month. 8-Hydroxyquinoline was introduced into quantitative analysis about 1927. Its use is described by Berg6 Hahn and Viewegl' and Robitshek.l* These early researches dealt with the determination of such elements as copper aluminium, cadmium bismuth zinc and magnesium in pure solutions Later workers12 have shown that the reagent may be employed to separate aluminium from such elements as phosphorus vanadium titanium arsenic molybdenum uranium fluorine, boron beryllium and tantalum provided that suitable conditions are established. Conditions of Preci$itation.-In acetic acid solution the following elements are precipitated completely or in part :-silver lead mercury (bivalent) copper TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON 57 1 bismuth cadmium antimony (tervalent or quinquevalent) uranium iron (bivalent or tervalent) zinc manganese cobalt nickel aluminium zirconium niobium, tantalum chromium and vanadium (quadrivalent or quinquevalent).In ammoniacal solution all the above-mentioned elements yield precipitates except silver and in addition beryllium magnesium calcium strontium barium and tin (quadrivalent or bivalent). Contrary to statements made in previous literature,lQ precipitation of small amounts of aluminium can be made in very dilute hydrochloric acid solutions containing ammonium chloride.27 Tartrates however must be absent. Under these conditions no precipitation of manganese occurs but chromium is partly precipi-tated.Amounts of aluminium in excess of about 5 mg. cannot be precipitated by this method because the amount of acid required to hold the aluminium in solution (in the absence of tartaric acid) renders the solution too acid to allow precipitation to take place. It will be seen that aluminium may be precipitated either in acetic acid or ammoniacal solution in the presence of tartrate. Since the reagent forms precipitates with so many metals it would at first appear that it would be useless for the present purpose. But most of these elements are either easily removed or are seldom encountered in ferrous materials in sufficient quantity to cause serious interference. Thus mercury lead bismuth copper, cadmium and antimony are readily removed by hydrogen sulphide in acid solution, whilst iron zinc manganese nickel and cobalt are removed by the same reagent in ammoniacal tartrate solution.The remaining elements with the exception of titanium vanadium and chromium which have received special attention in this investigation are as yet unlikely to be encountered in cast iron. Precipitation with oxine has the advantage that the metallic complexes may be dissolved in mineral acid and titrated with a solution of potassium bromate containing potassium bromide followed by the titration of the iodine liberated by adding potassium iodide with sodium thiosulphate. Provided that suitable precautions are taken this volumetric completion has been found to give results as accurate as the gravimetric method and it reduces the time required €or a determination very considerably.The final procedure is given later together with the reasons that led to its adoption. During the titration the oxine is quantitatively brominated to its 5 7-dibromo-derivative in accordance with the following equation : CQH~ON + 2Br2 -+ C,H,0NBr2 + 2HBr In the early part of this work the gravimetric method was first employed and then the precipitates were dissolved in acid and titrated. This proved a valuable check on the volumetric method since weighed amounts of the oxine complex were being determined. Later the volumetric ending was employed alone but from time to time a gravimetric result was obtained as a check. Drying Temperature of Oxine Precipitates.-The various writers on the gravi-metric oxine method are at variance as to the temperature at which the precipitate should be dried.Their recommendations vary from 100" to 160" C. In the course of the work which forms the subject of this paper the precipitates were dried at various temperatures between these limits without any appreciable change in the results obtained. Finally a temperature of 105" to 110" C. was adopted as bein 572 TAYLOR-AUSTIN THE DETERMINATION OF A4LUMINIUM Il\i CAST IRON most convenient and all precipitates have been dried at this temperature for one hour and then re-heated to constant weight at the same temperature. In every instance the weights were constant after the first hour except with precipitates weighing more than 0.7 g.when a period of 1.5 hours at 105” to 110” C. or 1 hour at 135” to 140” C. was necessary. The first method of precipitation to be studied was the hydroxyquinoline method of Gadeau9 : “A sample of 0.5 g. is dissolved in conc. hydrochloric acid (or sulphuric acid) and the solution is evaporated to dryness (or to fumes of sulphuric acid) to render silica insoluble. Without filtering off the silica etc. 200 ml. of water are added together with 5 g. of tartaric acid and sufficient ammonium hydroxide solution to render the odour of ammonia apparent. Then a rapid stream of hydrogen sulphide is passed through the solution for 20 minutes after which the liquid is filtered and the precipitate is washed carefully with water saturated with hydrogen sulphide. The filtrate contains the whole of the aluminium without a trace of iron.” After boiling for some time to expel the excess of hydrogen sulphide the aluminium is precipitated as follows : “After the solutionhasbeen made slightly acid with acetic acid it is warmed to 50” to 60” C.and 10 ml. of an acetic acid solution (5 per cent. oxine in 12 per cent. acetic acid) are added. Then a little ammonium acetate solution (154g. per litre) is added drop by drop until precipitation com-mences and then a further 20ml. of the ammonium acetate solution. Under these conditions (C,H,ON),Al is precipitated. After the solution has stood in a warm place for three or four hours it is filtered through a sintered glass filter and the precipitate is washed with water contain-ing a little ammonium acetate and dried for one hour a t 140” to 160” C.The aluminium is weighed as (CgH,ONj,Al.” Per cent. of A1 = (C,H,ON),Al x 0.0588 x 2. After the precipitates obtained by this method had been weighed they were dissolved in hydrochloric acid and titrated in accordance with the procedure given by Hopkin and Williams16 which is as follows: “The precipitate is dissolved in a little warm conc. hydrochloric acid (the compound is not readily soluble in weaker acid) the solution is diluted with about five times its volume of water, and a little indigo carmine or methyl red solution is added. The solution is then titrated with N/5 bromine solution (5.567 g. of potassium bromate and 50 g. of potassium bromide dissolved in water and diluted to one litre) to a yellow colour.A few g. of potassium iodide are added and the liberated iodine is titrated with N/10 sodium thiosulphate using starch solution as indicator (1 ml. N/5 bromine solution = 0.0004495 g. Al). . . . A simple direct titration with the bromine solution is not possible since the indicators do not show a sharp end-point.” In order to study this method varying quantities of the aluminium chloride solution already described (p. 568) were taken thus :-lo0 ml. = 20.6 mg. A1 ; 50 ml. E 10.3 mg. A l ; 25 ml. = 5.15 mg. Al. Some of the solutions containing only aluminium chloride were diluted to a volume of 300 ml. tartaric acid was added and the precipitation carried out exactly as described in the method under examination. To other solutions varying amounts of iron (p. 568) were added and the hydrogen sulphide separation was carried out ; determinations were completed exactly as prescribed.Some of the results for these two series are shown in Table 11. It will be noticed that in every instance the results were lower than the known amount of aluminium. In some the figure is much lower (Expts. Nos. 3 and 4). Further the volumetric results are occasionally much lower than the corresponding gravimetric figures (Nos. 2 4 and 7). After the weighed precipitates had been dissolved from the sintered glass crucibles a slight black residue remained and an attempt was made to weigh this with a view to explaining the lowness of some of the volumetric results. It was found that the substance was little more than a stain since no increase in weight coul TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 573 be detected after the crucibles had been dried to constant weight a t 105" to 110" C.The true explanation was not found until a much later stage in the investigation (see p. 584). TABLE I1 Expt. No. Aluminium added mg. 5.15 10.30 15-45 10.30 10.30 10.30 5.15 5.15 Iron added mg. nil 11 il nil 1000 500 500 500 500 Weight of oxine complex obtained mg. 85.2 169.0 247-2 165.7 169.6 168.9 84-5 83.5 Aluminium from Aluminium volumetric obtained determination mg* mg. 5.0 1 4.86 9.94 9.32 14-52 14.40 9-74 9.44 14-93 9.89 9.92 9.92 4-97 4.69 4-91 4.83 It was also noticed that contrary to the statement of Gadeau it was very difficult to filter off and wash the iron sulphide precipitate.In an attempt to over-come this difficulty a large variety of filter-papers was employed but without success. Finally the following procedure which avoids all washing of the precipi-tate was adopted :-One-gram samples were taken instead of the recommended 0.5 g. and dissolved in hydrochloric acid; the hydrogen sulphide separation was carried out as before and then the precipitate and solution were transferred to a 500-ml. graduated flask diluted to the mark and well mixed. After the precipitate had settled somewhat 250 ml. were filtered off into another graduated flask through a Whatman No. 1 filter paper (18.5 cm.). This 250-ml. portion (0.5 g. of the sample), was transferred to a beaker and after evaporation to 100 ml.it was acidified with hydrochloric acid and sufficient bromine water was added to colour the solution yellow. The excess of bromine was expelled by boiling the solution was rendered ammoniacal and then just acid with 30 per cent. acetic acid (litmus paper as indicator) and the aluminium was precipitated with oxine as before. The bromine was added because it had previously been noticed that when boiled to remove hydrogen sulphide the solution deposited sulphur and it was necessary to remove this before proceeding with the gravimetric determination. The sulphur can of course be filtered off after concentration of the liquid to a very small bulk but the use of bromine is more reliable and saves much time. The results obtained by this method were closely similar to those already given in Table 11 and therefore inconsistencies observed must have been due to some other factor.Precipitation irt Ammoniacal Sol.ution.-Since it is stated in the published works on the oxine methods that the element titanium is precipitated in acetic acid solution together with vanadium precipitation in ammoniacal solution seemed to offer greater promise than the previous methods for the determination of aluminium in cast iron. The procedure examined was that published by Lundell and Knowles,12 and is as follows: "TO the solution slightly acid with either sulphuric or hydrochloric acid and containing not more than 0.1 g. of aluminium in 100 ml. volume add an excess of 2.5 per cent. solution of 8-hydroxyquinoline in dilute acetic acid (made by triturating 2-5 g.of oxine with 5 ml. of glacia 574 TAYLOR-AUSTIN THE DETERMINATIOK OF ALUMINIUM I N CAST IRON acetic acid and pouring the solution obtained into 100 ml. of water a t 60" C.; after cooling the solution is filtered) and then dilute ammonium hydroxide solution until the solution is alkaline, and finally an excess of 5 ml. of strong ammonium hydroxide per 100 ml. of solution. Warm to 60" C . to 70" C. and digest a t this temperature until the precipitate becomes dense and crystalline. Cool preferably in iced water and filter washing with a cold dilute solution of ammonium hydroxide (1 40) containing 25 ml. of the reagent previously neutralised with ammonium hydroxide per litre. ' ' The above-mentioned authors complete their determinations by destroying the organic matter in the precipitates by treatment with conc.sulphuric and nitric acid and then precipitating the aluminium as hydroxide with ammonium hydroxide and igniting to oxide. In view of the difficulties and drawbacks of this latter procedure already mentioned the precipitates obtained during this investigation were filtered off on sintered glass crucibles dried and weighed. To do this success-fully it was necessary to omit the addition of the reagent to the washing medium. When titanium and vanadium are present Lundell and Knowles recommend the procedure described with the addition of 10 ml. of 20 vol. hydrogen peroxide prior to the addition of the oxine solution. This method was studied by using the same aluminium chloride solution as before and adding titanium as a standard solution prepared in the following manner :-Standard Titanium Solution.-Pure titanium oxide (0.8324 g.) was fused with 10 g.of potassium bisulphate a t a dull red heat. The fused mass was extracted in 200 xnl. of 5 per cent. sulphuric acid; it was found that sometimes the titanium salt tends to be hydrolysed during extraction; if this occurs the solution should be evaporated until fumes appear when the precipitated titanium will redissolve. The solution was then diluted to 1 litre. (1 ml. = 0.0005 g. of titanium). The results obtained by precipitation are shown in Table 111 and it will be seen that they are lower than those obtained in acetic acid and acetate solution. This may be due to filtration difficulties for it was found that whereas the precipi-tates obtained in acetic acid and acetate solution are bright yellow in colour and coagulate readily on standing and consequently are easy to separate those obtained in ammoniacal hydrogen peroxide solution turn a dark brown colour on standing and are difficult to filter off and wash.The solution also turned deep red and all attempts to coagulate the precipitate failed. Owing to these difficulties pre-cipitation in ammoniacal solutions containing hydrogen peroxide was abandoned after numerous experiments on solutions containing no iron. TABLE I11 Expt . No. 9 10 11 12 13 Aluminium added mg-5.15 5.15 5.15 5.15 5-15 Titanium added mg , nil nil nil 2.0 2.0 Aluminium found mg.4.84 4-93 4-76 4.84 4.74 In view of the fact that neither of the two oxine methods studied yielded satisfactory results it was decided to endeavour to correlate results for aluminium determinations obtained by the oxine method of Gadeau and the standard phos-phate method on samples of cast iron. The phosphate method is given in detail at a later stage TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 575 The results obtained by the phosphate method and the oxine method are shown in Table IV and it will be seen that they are far from satisfactory since both methods yield inconsistent figures. TABLE IV Sample Aluminium by phosphate method Aluminium by oxine method No. r A > I A \ Per Cent. Per Cent .- Per Cent. Per Cent. Per Cent. Per Cent.Per Cent. Per Cent. 6781 2.20 2.17 2-20 2.08 2.10 2.12 6782 4-84 4.76 4-82 4.93 4.72 4-82 6780 7-51 7-46 7-72 7.57 6.98 6.99 7-04 7.13 7-27 7.22 7-05 7.16 7.45 7-39 7.35 7-35 6797 6762 6761 6783 6758 8-85 9.31 5.90 5-85 5-76 5-90 6.94 6.62 3-88 3-76 8-44 8.07 8.30 8.29 4.72 4-96 5.12 5.84 5-76 4-20 4-87 6.28 6.28 3-92 3.91 3-93 For example with sample 6780 twelve oxine determinations were made, and only one pair agree among themselves (7.35 7-35) whilst the phosphate figures vary by 0.3 per cent. With 6761 the variations are even greater for the oxine method. For sample 6758 the oxine figures agree well but those by the phosphate method are much lower. The only instance where reasonable agreement is shown between the two methods is the 2 per cent.aluminium sample 6781 and it is a significant fact that the differences become greater as the aluminium-content of the samples increases. It was subsequently found that all the oxine figures given above were low. In view of these irregularities a complete study of both methods was begun. The investigations were carried out simultaneously but for the sake of clarity the phosphate method is dealt with first. THE PirosPHATE METHOD.-AS stated at the outset the phosphate method has altered little during the last fifty years. The details of the first method examined are as follows20: “Weigh out 2 g. of the sample dissolve in 20 ml. of conc. hydrochloric acid and keep the solution as free from oxidation as possible. Filter off the graphitic carbon silica etc.and to the filtrate add dilute ammonium hydroxide until a permanent precipitate is obtained then add 10 ml. of sulphurous acid and boil. Add 2 g. of sodium phosphate followed by ammonium carbonate solution until a faint permanent precipitate appears ; just redissolve this precipitate in conc. hydrochloric acid and add 2 ml. in excess. Then add 20 ml. of 30 per cent. acetic acid make the volume of the solution up to about 300 ml. and heat to boiling. Next cautiously add 10 g. of sodium thiosulphate a small amount a t a time until a precipitate of sulphur appears and then add the remainder. Boil the solutions until the vapour stream is free from sulphur dioxide (about 20 minutes) filter through a paper pulp filter and wash with hot water containing 1 per cent.of ammonium phosphate and nitrate. Transfer to a crucible and ignite a t the mouth of the muffle furnace. Dissolve the precipitate in a small amount of conc. hydrochloric acid heating if necessary and filter off the insoluble matter. Add 1 g. of sodium phosphate neutralise as before and make the final precipitation with 5 g. of sodium thiosulphate. Ignite the precipitate obtained a t the mouth of a muffle furnace and weigh as aluminium orthophosphate which contains 22.10 per cent. of aluminium.” This method was first applied to a series of pure aluminium chloride solutions, and later iron was added in varying amounts from 0-5 to 2-0 g. The results are shown in Table V in which each number represents a duplicate determination, for which each figure is given.It will be seen that all the results are below the know 576 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON amount of aluminium added. In spite of this the weights of aluminium phosphate obtained after two precipitations are in very close agreement; this is the most serious difficulty yet disclosed since in the normal course of routine analysis all the results given in Table V with the exception of No. 22 would be accepted as correct although they are actually in error to the extent of from 0.05 to 0.15 per cent. The results for Nos. 21 and 22 are examples of those in which the phosphate method failed to precipitate more than about 10 and.20 per cent. respectively of the aluminium present; no explanation for this could be found since as far as possible the conditions of precipitation were identical with those in the preceding experiments.It will be noticed of course that the weights of the first precipitates show a large increase as soon as iron is introduced but this is of no consequence, since it has long been known that some iron is invariably precipitated at this stage. The writers of literature published on this method are at variance as to the correct ignition temperature for the final precipitate. Thus Ibbotson2 states that a temperature of 600" C. should not be exceeded whilst Lundell Hoffman and Bright1 recommend 1000" C. as the correct temperature. In the present series of tests the precipitates were first ignited at 600" C. and then reheated at 1000" C. for 30 minutes and a study of the resultant figures given in Table V shows that no appreciable change in the weights occurs.Thus it appears that either temperature may be employed without vitiating the results. In explanation of the fact that the phosphate method gives inconsistent results several writers have suggested that the phosphate AlPO is formed only when a definite proportion of Al,O to P,O is present and Naish and ClennelP suggest that a blank determination with an amount of aluminium as nearly as possible equal to that in the assay and the same quantities of phosphate and other reagents should be carried out at the same time as the determination proper. The same writers state that their experiments show that the weight of the ignited precipitate obtained from the same weight of aluminium increases indefinitely as the ratio P205 A1,0 is increased.This is contrary to the statement of Ibbotson and AitchesonZ1 that the difficulty is overcome by using a large excess of phosphate. Composition of the Ignited Precipitates.-In order to determine whether or not the low results recorded in Table V were due to the fact that the ignited precipitates did not conform to the formula AlPO, a number of precipitates were examined and their aluminium and phosphorus contents determined as follows :-The precipi-tates were dissolved in 10 ml. of conc. hydrochloric acid. The residue of paper-ash etc. was filtered off and reserved (a). To the filtrates 5 g. of tartaric acid were added and the aluminium was precipitated by the final oxine method which is givenin detail on p.588. The reserved residue (a) insoluble in hydrochloric acid was fused with a little potassium bisulphate and after extraction with dilute acid the solution was tested for aluminium by the colorimetric method of Lampitt and Sylvester. The filtrates from the aluminium determinations were evaporated with nitric and sulphuric acids to destroy the organic matter and then 0.5 g. of powdered iron was added and the phosphorus was precipitated with ammonium hydroxide as ferric phosphate. The precipitate was dissolved in nitric acid and the phosphorus was precipitated first with ammonium nitro-molybdate and finally with lead acetate as lead "phosphomolybdate" in the usual manner. The addition of iro TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 577 and subsequent precipitation with ammonium hydroxide is necessary to separate the phosphorus from the sulphuric acid present.TABLE V Weight of AIPO precipitates obtained Expt No. 14 15 16 17 18 19 20 21 22 Aluminium added mg. 10.3 10.3 10.3 5.15 10.3 20.6 10.3 10.3 5.15 Iron added mg* nil nil 2000.0 2000*0 2000 *o 2000-0 500.0 500.0 2000-0 ignited at 600" C. & 1st pptn. 2nd pptn. g. g. 0.0426 0.0370 0.0420 0.0362 0.0435 0.0380 0.0440 0.0391 0.0526 0.0430 0.0577 0.0416 0.0339 0.0147 0.0286 0.0136 0.0683 0.0391 0.0708 0.0399 0.1034 0.0870 0.1023 0.0862 0.0586 0.0370 0.0675 0.0368 0.0303 0.0044 0.0478 0.0047 0.0313 0-0064 0.0398 0*0083 3 Reheated Aluminium a t 1000" C .found g. mg. 0.0370 8.17 0.0363 8.02 0.0379 8.38 0.0391 8.66 0-0432 9.54 0.0415 9-17 0.0145 3.20 0.0140 3.09 0,0385 8-51 0.0396 8.75 0.0870 19.23 0.0860 19.03 0.0367 8.1 1 0.0366 8.10 - 0.98 - 1 -04 0.0056 1 -24 0.0078 1.72 The results obtained are given in Table VI and it will be seen that they are in such close agreement with the theoretical figures that it appears unlikely that any change in the composition takes place under the conditions employed. The precipitates examined were chosen from two batches of samples taken on different days; precipitate A was from a double precipitation from a pure aluminium chloride solution; precipitate B was from a single precipitation from a pure solution. TABLE VI wt. of Wt. of Theoretical Actual Theoretical Actual AlPO residue wt.of A1 wt. of A1 wt. of P wt. of P ppt. insol. in HC1 in ppt. in ppt. in ppt. in ppt. g. g. g. g* g. g. A 0.0379 0.0022 0.0079 0-0077 0.0093 0.0095 B 0.0435 0.0035 0.0088 0.0088 0.0102 0.0098 Sample No aluminium could be detected by the aurintricarboxylate method in the residues insoluble in hydrochloric acid. Several modifications of the phosphate method were examined; e.g. the weak hydrochloric acid solution was boiled and the aluminium phosphate was precipi-tated with ammonium acetate instead of sodium thiosulphate ; also ammoniu 578 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON acetate solution was added in the method already studied after sulphur dioxide had been boiled off,22 but the results were not materially altered.Another modification consists in the removal of most of the iron present by means of a sodium bicarbonate ~ e p a r a t i o n ~ ~ followed by a normal phosphate precipitation. This procedure was not studied since the phosphate method gives low results whether iron is present or absent. Recently an attempt was made to apply the phosphate method to the routine analyses of a series of samples of iron containing approximately 7 per cent. of aluminium and 0.75 per cent. of chromium. The chromium was removed from the phosphate precipitate by oxidising it to chromate with potassium permanganate solution and precipitating the aluminium phosphate from the oxidised solution with ammonium acetate solution ; the precipitate was dissolved in hydrochloric acid and the aluminium phosphate was precipitated in the usual manner.Check analyses were made on some of the samples by the recommended oxine procedure given on p. 588 chromium being removed before the precipitation of the aluminium, as there described. It was found that under these conditions the results obtained by the phosphate method were higher than those recorded by the oxine method. This is the exact opposite of the results with chromium-free irons. I t was thought that possibly one treatment for the removal of chromium was insufficient although the phosphate precipitates obtained showed no trace of green colour. Accordingly a second re-precipitation was made and when this was done the figures fell below those obtained by the oxine method.The results obtained for a series of these tests in which numerous re-precipitations were made on the same samples are given in Table VII the last column of which shows the figures obtained for the same samples by the latest hydroxyquinoline method chromium being removed as described on p. 588. The figures for the first phosphate precipitation are not given since some contamination from iron is inevitable at this stage and thus the figures shown in the first column of Table VII (first re-precipitation) are those which would normally be considered correct. TABLE VII Sample First Second Third Fourth A1 (by oxine No. re-pptn. re-pptn . re-pptn. re-pptn. method) Per Cent. 8784 6-82 6-52 6.20 5.98 6-68 6.80 6.43 6.24 6.00 6.65 8786 7.54 6.95 6.77 6.23 7-20 7.55 6-98 6.77 6.25 7.21 The treatment for the removal of chromium was omitted after the second re-precipitation so that the figures shown in the third and fourth columns of the table were obtained by dissolving the precipitates from the previous treatment in hydrochloric acid and re-precipitating the aluminium phosphate in the usual manner.It will be noticed from the figures given that the agreement between the duplicates is again excellent but that the results obtained fall consistently as re-precipitations are made. This fact supports the previous contention that th TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 579 phosphate method does not precipitate the whole of the aluminium present. It was hoped to obtain a constant figure after the removal of all the chromium present but, as the results show this was not so.The investigation described above has shown that the phosphate method yields low results when applied to chromium-free materials but that in the presence of chromium results obtained by the standard procedure may be high. I t was therefore concluded that the phosphate method is unsuitable for the accurate determination of aluminium unless only very small quantities (1 per cent. max.) are present in the samples under examination. The most serious aspect of the matter is that the duplicate tests show excellent agreement but are nevertheless erroneous. THE HYDROXYQUINOLINE METHOD.-In view of the divergencies reported in the preliminary examination of this method (supra) it was decided to endeavour to discover a suitable indicator to enable the precipitations to be made in solutions at a constant p H value.The indicators in general use in the laboratory were first tried. To a solution containing 10.3 mg. of aluminium and free from iron 5 g. of tartaric acid were added; the volume of the solution was increased to 300 ml., and after the acidity had been adjusted to the change-point of the indicator chosen and the solution warmed to 60" C. precipitation was made with 15 ml. of a 2 per cent. oxine solution followed by 50 ml. of 4 N ammonium acetate solution. The beakers and their contents were allowed to stand at 60" C. until the precipitate coagulated and then for a further 45 minutes. The precipitates were collected on a sintered glass crucible (porosity No.4) washed six times with boiling water and finally dried to constant weight at 105" to 110" C. The oxine solution was prepared as recommended by Mitchell and Ward.24 A 2 per cent. solution of oxine in 2 N acetic acid is treated with ammonium hydroxide solution until a permanent white precipitate is produced. The solution is then gently warmed until the precipitate re-dissolves; it is then filtered into the storage bottle and is ready for use. This procedure reduces the amount of free acetic acid present to a minimum. The results of this stage in the investigation are shown in Table VIII and it will be seen that none of the indicators employed is really satisfactory. This, however is not surprising since both acetic acid and ammonium hydroxide are weak reagents.Methyl red showed the most promise and was accordingly subjected to further trials but after a number of determinations had been made its use was abandoned since the results obtained were no better than former figures. The figures obtained are shown in the lower portion of Table VIII. A search was next made for an indicator with a change-point nearer the neutral j5H value 7-07 and finally neutral red was chosen for trial. Neutral Red Indicator.-Neutral red or aminodimethylamino-toluphenazonium chloride has a $H range of 6.8 (red) to 8.0 (yellow). I t is soluble in water and a 0.5 per cent. aqueous solution was employed. With very weak acid solutions the indicator gives a pink colour but as the acid concentration increases the colour changes through red to greenish-purple.Ammonium acetate solution.-This solution was tested with both methyl red and neutral red indicator solutions. I t showed a distinct yellow colour to the former and red to the latter and thus its $H was between 6.4 and 6.8 580 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON Determinations were then made on a series of pure aluminium chloride solutions and it was found that the neutral red indicator was much more sensitive in hot than in cold solutions. Accordingly the solutions were first warmed to 60" C. and two drops of the indicator were added; then a 50 per cent. ammonium hydroxide solution was added until the solution was of a distinct yellow colour; Indicator Phenolphthalein TABLE VIII Aluminium pH range added Aluminium found mg.mg. 8.3- 10.0 10.3 No reaction in presence of tartaric acid f---h----\ Litmus (to first pink) . . 5.0- 8.0 10.3 8.80 8-88 Litmus (to red) . . 5.0- 8.0 10.3 8-69 8.57 Methyl orange . . . 3.0- 5.0 10-3 Xo pptn. when solution was red & 5.15 4.22 4-24: 10.3 8.93 9.27 20.6 19-84 20.50 Methyl red . . . . 4.7- 6.4 10.3 9-80 9.81 this was followed by 10 per cent. acetic acid added a few drops at a time until a faint pink colour was produced. The determinations were then completed exactly as in the trials of the other indicators; the results obtained are given in Table IX. It will be seen that they are by far the most satisfactory yet recorded being both accurate and consistent. TABLE IX AIuminium added " g . 10.3 10.3 10.3 10-3 5.15 5-15 5.15 5-15 5.15 5-15 Aluminium found : neutral red used as indicator mg.10.17 10.26 10-31 10.30 5.04 5.09 5.10 5.16 5.09 5.10 Several facts however were noted during the determinations. Thus it was found that if on adding the oxine solution a precipitate appeared immediately, or at least before all the oxine had been added the results obtained were of a high standard of accuracy; if on the other hand it was necessary to add ammonium acetate solution in order to produce a precipitate the results tended to be low. By carefully controlling the $H of the solutions i.e. by adding just sufficient 10 per cent. acetic acid to render the solutions pink but not red immediate precipitations could always be obtained TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 58 1 In this series of tests some of the precipitates were washed with hot 0.1 per cent.ammonium acetate solution and 0.1 per cent. acetic acid solution instead of hot water but as may be seen no appreciable change in results occurred and the use of hot water was adopted as a standard procedure. I t is essential that the washing water should be as near boiling-point as practicable since 8-hydroxy-quinoline is readily soluble in boiling but practically insoluble in cold water. Precieitation of Small Amounts of AZaminizcm.-Attempts were next made to precipitate much smaller amounts of aluminium and a fresh aluminium chloride solution was prepared by dilution of the original solution; 100 ml. of the aluminium chloride solution (= 20.6 mg.Al) were diluted to 1 litre so that 50 ml. of the new solution were equivalent to 2-06 mg. of aluminium. The determinations were carried out exactly as those just described neutral red being used as indicator; the results are shown in Table X. TABLE X Aluminium added Amount of 2 per cent. oxine soln. used for pptn. mg. ml. 2.06 5.0 2-06 5.0 1 -03 5.0 1.03 5.0 0.52 10.0 0.52 10.0 0.26 10.0 0.26 10.0 Aluminium found *g. 2.03 1 -98 0.97 0.98 0-47 0.49 0.28 0.25 It will be seen that they are highly satisfactory for aluminium-contents down to 0.2 per cent. and samples of 1 g. It was however noticed that with these small amounts of aluminium the pH of the solution a t the time of precipitation was even more important than for higher amounts.Ten per cent. acetic acid must be added until the solution is just tinged pink to the neutral red indicator. If sufficient acetic acid is added to render the solution red no precipitation of the aluminium takes place even after the addition of 40 to 50 g. of ammonium acetate. It was also found that better results were obtained by using a comparatively large excess of the reagent and hence the sudden increase in the amount of oxine solution added for the lower amounts shown in Table X. In all the experiments except the two lowest recorded (0.26 mg. Al) a precipitate was obtained before the slow addition of the reagent was completed. In the last two experiments it was necessary to add 10ml. of 4 N ammonium acetate solution to produce a precipitate.The precipitates eventually obtained coagulated readily on maintaining the solutions a t 60" C. for fifteen minutes and showed on filtration and subsequent washing, no tendency to pass through the filters. THE EFFECTS OF ALLOYING ELEMENTS.-The effects of various elements likely to be present in samples of plain and alloy cast-iron were next investigated, both synthetic solutions and those prepared from actual samples of cast-iron being used. Titanizm.-In all the published literature on the use of hydroxyquinoline in quantitative analysis mention is made of the fact that in acetic acid and acetat 582 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON solution the element titanium is precipitated. When this statement was checked, however it could not be substantiated.The procedure adopted was as follows :-To solutions containing known amounts of aluminium 5 mg. of titanium were added in the form of the standard solution already described (p. 574) and the determination was completed exactly as before neutral red indicator being used. The filtrates from the final precipitation of the aluminium as the oxine complex were evaporated with conc. sulphuric and nitric acids in order to destroy the organic matter present and the titanium was determined colorimetrically. Finally 10-3 mg. of aluminium were added to a sample of cast iron containing titanium and the separation of the iron was carried out by the hydrogen sulphide and tartrate method as described earlier followed by determination of the aluminium exactly as before.The results are shown in Table XI and from them it is evident that under the conditions of precipitation established in the course of this work titanium is not precipitated and does not, in any way interfere with the determination of aluminium since the results recorded for aluminium are not too high and the whole of the titanium added was recovered from the filtrates. TABLE XI Expt . No. 23 24 25 26 27 28 29 30 Aluminium added mg. 10.3 10.3 10.3 10.3 5-15 5.15 10.3 10-3 Titanium added mg. 5.0 5.0 5.0 5.0 5.0 5.0 4.6 4.6 Aluminium found mg* 10.31 10.2’7 10.27 10.30 5.08 5-16 10.15 10.16 Titanium found in filtrates mg-5.0 5.0 4.9 5.1 4.8 4.9 --In Table XI Expts.Nos. 29 and 30 represent the sample of cast iron containing titanium referred to above. It is noteworthy in view of subsequent results that these two results for aluminium are slightly lower than others for synthetic solutions recorded in the same table. Since the vast majority of pig irons contain titanium the fact that it has been established during this investigation that its presence causes no interference under the conditions described is of considerable importance. Nickel and Chromium.-In order to study the effects of these two alloying elements 10.3 mg. of aluminium were added to a sample of cast iron containing 2-52 per cent. nickel and 0-37 per cent. chromium and the determinations were carried out exactly as before. The precipitate obtained by hydrogen sulphide in ammoniacal tartrate solution was found to contain the whole of the nickel present, as well as the iron and manganese.The precipitate was dissolved in hydrochloric acid and the nickel determined in the usual manner with dimethylglyoxime. The results are shown in the first part of Table XII and it is evident that the nickel is completely removed by hydrogen sulphide whilst the chromium caused no perceptible interference. It was found however that some chromium is precipi-tated with the aluminium although a t the relative concentrations of aluminiu TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON 583 and chromium used in the previous tests this is insufficient to vitiate the results. The interference becomes more serious as the ratio of chromium to aluminium in the sample approaches 1 1.If the chromium-content exceeds that of the aluminium the figures recorded for the latter element may be as much as 30 per cent. too high depending upon the actual chromium and aluminium contents of the material. The results illustrating these facts are shown in the latter part of Table XII, and it will be seen that the amount of chromium precipitated is by no means con-stant. In all the tests carried out on samples containing chromium considerable difficulty was experienced in preventing its precipitation by ammonium hydroxide with tartaric acid prior to the removal of iron etc. For samples containing 1.0 per cent. of chromium it was found necessary to double the amount of tartaric acid present.If the chromium-content exceeded 2.0 per cent. it was impossible to prevent its precipitation. There are two ways in which the interference by chromium can be dealt with. If the chromium-content does not exceed about 1.0 per cent. precipitation may be carried out in the ordinary manner and the impure aluminium oxine complex dried and weighed the volumetric method being obviously inapplicable. The precipitate is then dissolved from the crucible in hot hydrochloric acid the organic matter is destroyed by evaporation with nitric and sulphuric acids and the chromium is determined by the usual colorimetric method employing diphenyl-carbazide. The chromium figure thus obtained is calculated to its oxine complex, and the weight of the latter is deducted from the original weight of aluminium-chromium complex.In order to make this correction the composition of the chromium oxine complex was determined. I t was found to conform to the formula Cr (C9H,0N), which contains 10.74 per cent. of chromium. This method was found to be sufficiently accurate for all practical purposes, but it increases very considerably the time required for the determination. It is far better therefore to separate the aluminium and chromium before the hydroxy-quinoline precipitation is made. The details of the procedure for this separation are given in the recommended method on pp. 588-589. TABLE XI1 Expt. No. 31 32 33 34 35 36 37 38 39 40 Aluminium Nickel added present Per Cent. Per Cent. 1 -03 1 -03 1.03 1 -03 1 -03 1-03 1 -50 1.50 1 -50 1 *50 2.52 2-52 2-52 2.52 2.52 2.52 nil nil nil nil Chromium Nickel present in FeS ppt.Per Cent. Per Cent. 0.37 2.50 0.37 2.48 0.37 2.49 0.37 2.48 0.37 2.51 0.37 2.51 1.75 -1.75 -1-75 -1-75 -Aluminium found Per Cent. 1 so0 1 -00 1.05 1 -02 1.07 1.01 1 -90 1.94 1 *63 1-60 In the course of these last experiments it was noticed that on maintaining the solutions and oxine precipitates at 60" C. the solutions acquired a deep red colour and that when this occurred the results were inclined to be high e.g. Expts 584 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON Nos. 33 and 35 in Table XII. For some time no explanation could be found for this phenomenon since experiments carried out on synthetic solutions containing chromium in various states of valency failed to produce the colour.It was thought that it might be due to some reaction with the neutral red indicator but in experi-ments made without the addition of the indicator the colour was still produced. The trouble was finally found to be due to the presence of traces of bromine, which appear to oxidise the hydroxyquinoline at about 60" C. giving a brown precipitate and producing a deep red colour in the solution. It was concluded that the colour and precipitate were not due to the production of bromine derivatives of the hydroxyquinoline since the effects could be obtained with other oxidising agents such as hydrogen peroxide. The action of the latter reagent had already been noticed when making precipitations in ammoniacal hydrogen peroxide solution.Incidentally this brown precipitate produced by oxidation undoubtedly accounts for the slight dark stain observed on the sintered glass crucibles after dissolving precipitates in hydrochloric acid for titration as mentioned on p. 572. Steps were therefore taken to ensure the complete removal of the bromine used to remove the precipitated sulphur produced after the passage of hydrogen sulphide gas. The slightly modified procedure was as follows :-After the removal of iron etc. by hydrogen sulphide in ammoniacal tartrate solution the filtrates were boiled down to a volume of about 100ml. as previously described. Then 20 ml. of conc. hydrochloric acid were added followed by sufficient bromine water to colour the solution yellow.Then the solution was boiled down to a volume of 50 ml. by which time the whole of the bromine was expelled. The method thus modified was employed for all subsequent determinations, and no further trouble was experienced. NOTE ON THE HYDROGEN SULPHIDE SEPARATION,--From time to time during the determinations on samples of cast iron so far described difficulty was experienced in removing the whole of the iron with hydrogen sulphide. It was found that if the solution was made distinctly ammoniacal prior to the passage of hydrogen sulphide, the iron sulphide precipitate was in a very fine state of division and tended to pass through the filters. If on the other hand the solution was made just ammoniacal, the solution became acid before the separation of the iron was complete.Theoreti-cally from the equation: FeC1 + H,S -+ FeS + 2HC1 56g. of Fe -+ 73 g. HCl 1 g. -+ 1.3 g. = 0.605 g. of NH,OH = 1-85 ml. of NH,OH (sp.gr. 0.880). Hence 1-85 ml. of ammonium hydroxide solution (sp.gr. 0.880) are necessary to neutralise the acid formed when removing 1 g. of iron by means of hydrogen sulphide. Accordingly an excess of 2.5 ml. of ammonium hydroxide was added in every instance and no further trouble was experienced (cf. p. 572). DETERMINATIONS ON CAST IRoN.-Since the standard of accuracy attained by the hydroxyquinoline method was now so high it was decided to apply it to a series of bars containing approximately 7.5 per cent. of aluminium. The deter-minations were carried out in precisely the same manner as for the syntheti TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 585 solutions the hydrogen sulphide and tartrate separation being used prior to precipitation with 2 per cent.oxine solution. The pH of each solution was carefully adjusted by means of neutral red indicator solution. Half-gram samples were used instead of the usual 1 g. and these on fractionation after the sulphide separation yielded a final sample weight of 0.25 g. It was first established by experiment that 25 ml. of 2 per cent. oxine solution were sufficient to precipitate 7 per cent. of aluminium when 0~25gram samples were used. The results obtained are shown in Table XIII and it will be seen that they are quite as inconsistent as any recorded for methods without rigid control of the precipitation conditions.TABLE XI11 Aluminium per cent. , Sample No. 1st Determn. 2nd Determn. 3rd Determn. 4th Determn. 5th Determn. - - e-7 - 7427 6.74 6.94 7.03 7-24 6.44 7.46 7.41 7.57 7.07 7-14 7428 6-61 6-25 6-03 6.21 6.10 6.23 6.51 6.37 6-42 6-26 7429 6-52 6.54 6.76 6.85 6.67 6.59 6.37 6.51 6.70 6.54 Since the modified method had shown such promise when used on synthetic solutions it was thought that these discrepancies were due to some factor other than the precipitation conditions. Accordingly a series of cast iron bars containing from 1 per cent. of aluminium upwards was examined with a view to discovering at what percentage of aluminium the method becomes inaccurate. The samples were treated exactly as before and the results are shown in Table XIV.One-gram samples were employed throughout. TABLE XIV Aluminium per cent. rA-7 r-+ /-+ - /-+ h 7- 7 Sample No. 1st Determn. 2nd Determn. 3rd Determn. 4th Determn. 5th Determn. 7239 1.15 1.14 1-16 1.15 1.14 1.16 1-15 1.16 1-14 1.15 6781 2.24 2.24 2-20 2.19 2.20 2.21 2.19 3-23 2.20 2.23 6758 3.89 3.99 3.92 3.98 3.83 3.97 4.01 3.95 4.02 4.03 It will be seen that the results for samples up to 2 per cent. of aluminium are well within the limits of analytical tolerance although those recorded for the 2.2 per cent. aluminium sample are not as good as they should be considering the small amount of aluminium contained by the oxine complex. Above 2 per cent. of aluminium the results become very erratic. A final attempt to overcome these difficulties was made by removing all the silica by baking prior to filtration into a 250-ml.measuring flask. It was realised that this procedure would probably result in slightly low figures owing to the removal of a little aluminium together with the silica. The residue of graphite silica etc. was therefore examined for aluminium by the colorimetric method and all the results are given in Table XV. It will be seen that the removal of silica has no material effect on the results obtained 586 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON TABLE XV Aluminium per cent. r -l Silicon removed A f A \ Aluminium in SiO Silicon not removed Sample No. residues * 7239 1.13 1.14 0.0 1 1.14 1.16 6758 4.0 1 4.03 0.01 3.83 3.97 6758 4-02 3-95 0.01 3.87 3.79 At this stage of the work the conclusion was reached that the statement of G a d e a ~ ~ that iron sulphide does not adsorb aluminium may not hold for amounts of the element in excess of 1 per cent.for 1-gram samples and a method was there-fore sought which would overcome this possible difficulty. The introduction of a re-precipitation of the iron sulphide was deemed impracticable owing to the difficulty of filtering it off and the ease with which it is oxidised if allowed to stand in contact with air for any length of time. A review of the other methods available for the separation of iron and aluminium has already been given and after numerous trials it was decided to employ the sodium bicarbonate separation followed by a double sodium hydroxide separation.Sodium Bicarbonate Sejbaration.-In order to carry out this separation which is described in detail at a later stage the samples were dissolved in 10 per cent. sulphuric acid instead of hydrochloric acid and a t first the solutions were evaporated until fumes appeared in order to remove silica but on 5 g. of sample this procedure led to a loss of about 3 mg. of aluminium which remained behind with the silica. This necessitated the fusion of the residues with potassium bisul-phate extraction of the melt with water and addition to the main solution. Deter-minations carried out without the removal of silica showed no appreciable difference in the results so that this procedure was omitted. No aluminium could be detected in the residues obtained from unevaporated solutions.When samples containing titanium and vanadium were dissolved in 10 per cent. sulphuric acid the whole of these two elements was found in the residue of silica etc. on filtration of the solutions into the 250-ml. measuring flask. For example in a bar containing 0.15 per cent. of titanium and 0.40 per cent. of vanadium the amounts of the two elements found in the silica residue were 0.14 and 0.38 per cent. respectively. This is a great advantage so far as vanadium is concerned, since unlike titanium it is precipitated together with the aluminium under the conditions established in this investigation. The behaviour of manganese was also investigated since this element is precipitated by oxine in acetic acid and acetate solution and is incompletely removed by hydrogen sulphide.The greater part of the manganese present is removed by the sodium bicarbonate separation and examination of the filtrates obtained from this operation on a sample containing 0.73 per cent. of manganese showed that 0.66 and 0-67 per cent. of manganese were present in a duplicate determination. The remainder appears to be removed successfully together with the iron since none could be detected in the final aluminium precipitates TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON 587 No aluminium could be detected in these bicarbonate filtrates by the colori-metric method thus showing the completeness of the precipitation of aluminium by sodium bicarbonate. Between 5 and 10 per cent. of iron remains associated with the aluminium, together with phosphorus and any chromium present.Phosphorus causes no interference but the iron and chromium must be removed. The removal of chromium is described on p. 588 and for the separation of iron a double sodium hydroxide separation was carried out but all the results so obtained were high, and precipitates were slightly contaminated with iron. It was found on further tests that it was exceedingly difficult to remove all traces of iron by the sodium hydroxide separation and after two precipitations by this reagent the solutions contained about 15 to 20 g. of sodium salts which proved a great disadvantage. Finally a return was made to the ammonium sulphide and tartrate separation, which proved highly satisfactory as subsequent results will show.The procedure, given in detail below was thus evolved and extensively tested. I t will be noted that the residue obtained by the sodium bicarbonate separation is not ignited and fused as recommended by many writers. This was deemed unnecessary since the alternative procedure given proved very satisfactory and no aluminium could be detected in the residue of filter-paper pulp obtained after the treatment of paper and precipitate with hydrochloric acid. CHROMIUM.-DiSSOlVe 5g. of drillings in 100ml. of 10 per cent. sulphuric acid, filter off graphitic carbon silica etc. on a paper-pulp filter and wash well with hot 5 per cent. sulphuric acid and hot water alternately. Allow the filtrate to cool, dilute to a volume of exactly 250 ml. in a graduated flask and mix well.For samples containing 1 to 2 per cent. of aluminium remove 50 ml. of the solution (= 1 g. of the sample) by means of a pipette; (for samples containing 2 to 8 per cent. of aluminium take 25 ml. 5 0-5 g. of the sample) and place it in a 400-ml. beaker. Dilute to approximately 100 ml. with hot water boil run in 8 per cent. sodium bicarbonate solution from a burette until a permanent precipitate appears, and then add 4 ml. excess. Boil for one minute and filter rapidly through a 9-cm. Whatman No. 41 filter-paper and wash three times with hot water. The precipitate so obtained will contain all the aluminium present together with about 5 to 10 per cent. of iron and usually all the phosphorus. During this separation care must be taken to avoid excessive oxidation of the ferrous sulphate to add no more bicarbonate than is necessary and to filter and wash the precipitate as rapidly as possible.Neglect of these precautions leads to excessive precipitation of iron. For amounts of aluminium in excess of 8 per cent. a larger excess of bicarbonate is necessary; thus for 15 per cent. of aluminium 6 ml. are required. Transfer the paper and precipitate containing the aluminium etc. back into the original 400-ml. beaker add 10 ml. of conc. hydrochloric acid dilute to 50 ml. with water and boil stirring continually to break up the filter-paper and prevent bumping. Allow the beaker and its contents to stand on a boiling water-bath for ten minutes and then filter off the paper washing well with hot water and occasionally putting drops of conc.hydrochloric acid round the edges of the filter. RECOMMENDED 8-HYDROXYQUINOLINE METHOD I N THE ABSENCE O 588 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON Dilute the filtrate to about 200 ml. add 5 g. of tartaric acid and pass a rapid stream of hydrogen sulphide gas for twenty minutes. Filter off any precipitate which may be obtained at this stage and which will contain any copper molyb-denum etc. present in the iron. After washing thoroughly with 1 per cent. hydro-chloric acid saturated with hydrogen sulphide render the filtrate just ammoniacal to litmus and add 2 ml. excess of ammonium hydroxide (sp.gr. 0.800). Pass a rapid stream of hydrogen sulphide through the solution for ten to fifteen minutes; allow the precipitate to settle and filter through a 12.5-cm.Whatman No. 1 filter-paper, washing twice with water containing a little tartaric acid and ammonium hydroxide and saturated with hydrogen sulphide. Dissolve the iron sulphide off the paper into the original beaker with conc. hydrochloric acid and hot water add 2.5 g. of tartaric acid make the solution ammoniacal and pass hydrogen sulphide as before. Filter through a Whatman No. 1 filter-paper wash three times with the previous washing medium and combine the filtrate with that from the first precipitation of the iron. Boil the combined filtrates until all the ammonium sulphides are decomposed, and the volume is about 100 ml. and then add 20 ml. of conc. hydrochloric acid. Cool somewhat add sufficient bromine or bromine water to colour the solution yellow and boil down to about 50 ml.Filter through a Whatman No. 54 filter-paper into a 400-ml. beaker and wash with hot water. To the filtrate add three drops of a 0.05 per cent. aqueous solution of neutral red indicator and then add ammonium hydroxide solution from a dropping-bottle until a yellow colour is produced; wash down the sides of the beaker and add 10 per cent. acetic acid, drop by drop until the solution is just pink. Warm to 70" C. and add slowly and with constant stirring 15 ml. of 2 per cent. oxine solution for every 10 mg. of aluminium present followed by 50 ml. of 4 N ammonium acetate solution. Maintain the solution at 60" to 70" C. until the precipitate coagulates (about 15 minutes), and then filter through a weighed sintered glass crucible (No.4 porosity) and wash six times with boiling water. Dry the precipitate to constant weight a t 105" to 110" C. The precipitate so obtained contains 5.87 per cent. of aluminium. The need for adhering to the above-described conditions of precipitation cannot be too strongly emphasised for upon them depends the whole success of the method. The procedure has been applied to a great variety of samples and some of the results obtained are given in Table XVI. A number of the samples have been previously examined by other oxine methods and these results will be found in Tables XI11 and XIV. The greater part of Table XVI shows results obtained on a series of samples with an aluminium-content of 1 to 10 per cent. The first sample and the last two have already been included in Tables XI11 and XIV.I t will be seen that the results are of a high standard of accuracy and consistency. Compared with those previously obtained they are considerably higher except for the 1 per cent. aluminium sample, and therefore support the suggestion that aluminium is adsorbed on the iron sulphide precipitate if more than 1 per cent. of aluminium is present. of drillings in 100 ml. of 10 per cent. sulphuric acid and treat the solution exactly as set out in the preceding method until the bicarbonate separation is to be made. Run in 8 per cent. sodium bicarbonate solution from a burette until a slight RECOMMENDED METHOD I N THE PRESENCE OF CHROMIUM.--DiSSOlve 5 g TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON 589 permanent precipitate appears and then add 6 ml.in excess. Boil for one minute, filter rapidly through a Whatman No. 41 filter-paper and wash three times with hot water. TABLE XVI Sample No. 7239 7596 7597 7598 7524 7525 6780 7601 7527 7603 7428 7429 Aluminium per cent. 1st Determination 2nd Determination -7 & 1.14 1.15 1.14 1-16 2-37 2.36 2.36 2-36 2.88 2-88 2.86 2.87 3.73 3.74 3-73 3.73 4.23 4.21 4.25 4.23 6.61 6.61 6-63 6.62 7.12 7.13 7.1 1 7-12 7-47 7.48 7.46 7.47 7.84 7.86 7.85 7.84 9-46 9.46 9-47 9.46 6-64 6.64 6.64 6-65 6.96 6.96 6-97 6-97 r A 3 Dissolve the precipitate from the filter-paper with 50 ml. of hot 10 per cent. sulphuric acid dilute to 300 ml. and boil. To the boiling solution add 10 ml. of silver nitrate solution (17 g.per litre) followed by 10 to 15 ml. of freshly-made ammonium persulphate solution (30 g. in 55 ml. of water ; saturated solution) and continue boiling for three minutes. Precipitate iron and aluminium by adding excess of ammonium hydroxide and boiling the solution until the issuing vapour no longer smells of ammonia (Lampitt and Sylvester methodI4). Filter off the precipitate on a Whatman No. 54 filter-paper and wash well with hot 1 per cent. ammonium nitrate solution. At this stage the filtrate is generally turbid owing to the silver present. Dissolve the precipitate in hydrochloric acid exactly as described for the bicarbonate precipitate in the previous method and complete the determina-tion exactly as in the absence of chromium the silver which has been added to the solution being removed during the passage of hydrogen sulphide through the acid solution.Application of the above procedure to samples to which known amounts of aluminium and chromium had been added yielded the very satisfactory results given in Table XVII. Experiments Nos. 41 to 43 were on solutions similar to those given in Table XI1 (Expts. Nos. 37 to 40) and show that the interference due to chromium has been prevented. The aluminium precipitates were tested for chromium by the diphenylcarbazide method and amounts of the order of 0.00005 g. were found. This shows that the separation of aluminium and chromium described is sufficiently complete for the present purpose. has been mentioned previously that the oxine method may be readily adapted for a volumetric finish and one method of procedure has already been outlined (p.572). In most of the determinations hitherto recorded the aluminium oxine complex, after being dried to constant weight in order to obtain the gravimetric figures given was dissolved in hydrochloric acid and the amount of aluminium present THE BROMOMETRIC DETERMINATION OF THE ALUMINIUM OXINE COMPLEX.-I 590 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM I N CAST IRON was determined by the volumetric method but for the sake of clarity it was deemed better to deal with this branch of work in a separate section. A few of the results so obtained have already been given in Table 11 and it will be seen that they are generally speaking unreliable. TABLE XVII Amount of Expt.No. A1 added Per Cent. 41 1 -50 42 1 -50 43 1 40 44 1 -25 45 1.25 46 1 -25 Amount of Cr added Per Cent. 1.75 1.76 1.75 4-00 4 *OO 4-00 Amount of A1 found Per Cent. 1 -50 1.51 1 -50 1 -24 1.23 1 -26 The bromination of 8-hydroxyquinoline has however been carefully studied by Fleck Greenane and WardJ25 and their work has been substantially verified during the present investigation. It was found that in many instances the end-points of the titrations were variable and indefinite. In all the titrations a certain amount of white solid matter was precipitated but in many of them a slate-blue solid appeared which was only slowly decomposed by the sodium thiosulphate solution and thus caused an unsatisfactory end-point.In the titration of small amounts of 8-hydroxyquinoline this difficulty rarely occurs and results are con-sistent and reliable but when dealing with amounts of aluminium above 1 per cent., steps must be taken to avoid the formation of this slate-blue solid. The authors previously mentioned25 have also observed these facts and suggest the addition of carbon disulphide after bromination as a means of overcoming the difficulty. They further point out that the bromination of hydroxyquinoline by the potassium bromate-bromide solution is not instantaneous and have experimentally established the fact that from three to five minutes are necessary for this reaction to proceed to completion. The use of carbon disulphide was therefore introduced into the volumetric method under examination and it was found to be very satisfactory, although contrary to the statement of Fleck Greenane and Ward the slate-blue solid was observed on several isolated occasions when dealing with weights of aluminium oxine complex in excess of 0.6 g .I t was further found that chloroform could be substituted for carbon disulphide under certain conditions but generally speaking carbon disulphide is by far the more satisfactory. Carbon tetrachloride is quite ineffective. The following procedure was therefore devised and extensively used both on weighed amounts of oxine complex and as a direct volumetric finish to the method already given. SOLUTIONS REQUIRED Potassium Bromate-Bromide.-For aluminium-contents up to 2 per cent. it is most convenient to use N/5 solution which contains 5.567 g.of pure potassium bromate and 50 g. of potassium bromide per litre. (1 ml. = 0.0004495 g. Al). For amounts of aluminium in excess of 2 per cent. a solution containing 11.134g. of potassium bromate and 50 g. of potassium bromide should be used. (1 ml. = 0.0008990 g. Al) TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IX CAST IRON 591 If AnalaR potassium bromate is used and the solution carefully prepared, no standardisation is necessary. If desired the solution may be standardised by titration against the standard thiosulphate solution. Sodium thiosu@hate.-A solution of the same normality as the potassium bromide is employed i.e. either 50-00 or 100.00 g. of sodium thiosulphate per litre. The stability of this solution will be greatly increased if 0.1 g.of anhydrous sodium carbonate per litre is added. The thiosulphate solution may be standardised, but if the potassium bromate solution is prepared carefully this is unnecessary. Indigo carmine i.zzdicator.-l-O per cent. solution in water. Starch.-A starch indicator which will remain unchanged for several months may be prepared as follows26:-Dissolve 1 g. of salicylic acid in 100 ml. of boiling water and add 1 g. of potato starch in a small amount of water. Continue boiling until a clear solution is obtained and then dilute to 1 litre after allowing the solution to cool to room temperature. PROCEDURE.-The method given on p. 588 is followed in every detail as far as the final filtration of the precipitated aluminium oxine complex.Instead of this being filtered off on a sintered glass crucible a No. 40 Whatman filter-paper is employed and the precipitate is washed six times with boiling water; it is then dissolved off the paper with 50 ml. of hot conc. hydrochloric acid (the precipitate is not readily soluble in dilute acid) added in small portions with alternate washings with hot water into a 500-ml. conical flask. The solution is diluted to 250 ml. with water and cooled to room temperature. (The acid concentration is now 20 per cent.) Next four drops of indigo carmine solution are added and the standard potassium bromate-bromide solution is run in in a thin stream with constant stirring until the colour of the solution is pure yellow. During this part of the titration the colour of the solution passes from blue to green and finally to yellow.A further two drops of indigo carmine solution are added and if the solution shows a green tint bromate-bromide solution is added until the pure yellow colour is restored. It was found that indigo carmine was a much better indicator than methyl red since the latter is rapidly decolorised during the titration and continual additions are therefore necessary. When an excess of bromate-bromide solution has been added the flask is fitted with a rubber bung or glass stopper well shaken, and allowed to stand for five minutes. Then 10 to 15 ml. of carbon disulphide are added the mixture is shaken and 10 ml. of potassium iodide solution are introduced slowly and with constant shaking. Ten ml. of starch solution are added and the standard solution of sodium thiosulphate is run in until after vigorous shaking, no trace of blue colour remains in the carbon disulphide layer.A t the end of the titration the carbon disulphide is usually pale brown. With regard to the titration solutions it should be remembered that the potassium bromate is the true standard and the sodium thiosulphate solution should be titrated against it exactly as described above with the addition of indigo carmine carbon disulphide etc. In carrying out this titration it is essential to use the same amount (six drops) of indigo carmine solution as that solution requires about 0.2 ml. of bromate solution to discharge its colour. CoNcLusroN.-This investigation has shown that the 8-hydroxyquinoline method is capable of yielding results of a very high standard of accuracy for a wid 592 TAYLOR-AUSTIN THE DETERMINATION OF ALUMINIUM IN CAST IRON range of aluminium-contents provided that the conditions and modifications given are strictly adhered to.It is applicable to all present-day compositions of alloy cast iron but when chromium is present special precautions must be taken especially when the proportions of aluminium and chromium present are of the same order or when chromium is in excess. Normally chromium is present along with aluminium only in irons intended for very particular applications although it is likely that in future chromium will be added as a special addition to irons containing aluminium. Contrary to published literature titanium is not precipitated under the conditions given by 8-hydroxyquinoline.The volumetric finish with its recommended modifications is quite as accurate as the gravimetric one and greatly reduces the time required for a determination, After extensive trials it has been shown that the recommended method described is the most accurate yet devised for the determination of aluminium in cast iron and ferrous materials. REFERENCES 1. G. E. F. Lundell J. I. Hoffman and H. A. Bright Chemical Analysis of Iron and Steel. John Wiley (New York) 1931 p. 348. 2. F. Ibbotson The Chemical Analysis of Steel Works Materials. Longmans Green (London), 1920 p. 103. 3. W. A. Naish and J . E. Clennell Select Methods of n/IetaZlurgical Analysis. Chapman & Hall (London) 1929 pp. 82 and 83. 4. J. E. Stead J .SOC. Chem. Ind. 1889 8 965. 5. A. Carnot hfoniteur Scienti$que 1891 14; A. A. Blair Chemical Analysis of Iron and Steel. Lippincott (Philadelphia). 8th Ed. 1918 p. 191. 6. R. Berg 2. anal. Chem. 1927,70 341; 1927,71 23 171 321 369; 1928,72 177; 1929, 76 191; Abst. ANALYST 1927,52 302 431 494 611; 1928,53 58; 1930 55 596, 6 ~ . G. W. Monier-Williams Reports on Public Health and Medical Subjects No. 78 1935; Abst. ANALYST 1935 60 823. 7. Part 111. A. Carnot Methods d’Analyse des Fontes des Fers et des ,4ciers 1895 p. 125 et seq. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. J . W. Rothe Mitt. aus den Konigl. Tech. Versuchsanstalten z u Berlin 1892. W. H. Hess and E. D. Campbell .J. Awzer. Chem. SOC 1899 21 776. R. Gadeau Rev. de Metallurgie Sept. 1935 p. 398. G. E. F. Lundell J . I. Hoffman and H. A. Bright loc. cit. p. 70. G. E. F. Lundell J . I. Hoffman and H. A. Bright ibid. p. 71. G. E. F. Lundell and H. B. Knowles Bureau of Standards J . Res. July 1929 3 No. 1, R. Hatfield I n d . Eng. Chem. 1924 16 233; Abst. ANALYST 1924 49 243. L. H. Lampitt and N. D. Sylvester “Determination of Small Amounts of Aluminium in British Cast Iron Research Assoc. Research Report No. 105 p. 27. Hopkin and Williams Organic Reagents f o r Metals 2nd Ed. 1934 p. 56. F. L. Hahn and K. Vieweg 2. anal. Chem. 1927,71 122; Abst. ANALYST 1927,52,431. J. Robitschek J . Amer. Ceramic SOC. 1928 11 587. G. E. F. Lundell J. I. Hoffman and H. A. Bright loc. cit. p. 82. B.C.I.R.A. Special Publication No. 1 (Research Report No. 105) 1933 p. 27. F. Ibbotson and L. Aitcheson Analysis of Non-Ferrous Alloys 1915 p. 140. A. A. Blair Chemical Analysis of Iron. United Steel Companies Ltd. Standard Methods of Analysis 1936 p. 52. A . D. Mitchell and A. M. Ward Modern Methods in Quantitative Chemical Analysis H. R. Fleck F. J. Greenane and A. M. Ward ANALYST 1934 59 325. A. D. Mitchell Sutton’s Volumetric Analysis 12th Ed. 1935 p. 127. T. E. Rooney W. W. Stevenson and T. Raine Seventh Report on the Hetwogeneity of J. Haslam “The Quantitative Separation of Aluminium from Iron,” ANALYST 1933, p. 95. Foods,” ANALYST 1932 57 148. Lippincott Co. (Philadelphia) 1918 p. 190. (Longmans) 1932 p. 26. Steel Ingots Section IV p. 120. Iron and Steel Institute London 1937. 58 270. I wish to express my thanks to the Director and Council of the British Cast Iron Research Association for permission to publish this work. BRITISH CAST IRON RESEARCH ASSOCIATION 21-23 ST. PAUL’S SQUARE BIRMINGHAM 3 February 193

ISSN:0003-2654    

DOI:10.1039/AN9386300566

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate.There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international.The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr. Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively.Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies. Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited.The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation.Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE.By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time.The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice.Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years.The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr. Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion.The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies. Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on.Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300593

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate.There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international.The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr. Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively.Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies. Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited.The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation.Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE.By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time.The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300596

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate.There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international.The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr. Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively.Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies. Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited.The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300598

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate.There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international.The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr. Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively.Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies. Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited.The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation.Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE.By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time.The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents. It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300599

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300602

出版商:RSC    

年代:1938

数据来源: RSC

摘要:

426 REVIEWS INKS : THEIR COMPOSITION AND MANUFACTURE. By C. AINSWORTH MITCHELL, D.Sc., F.I.C. Fourth Edition. Pp. xi + 408. London: Charles Grihn tt Co., Ltd. 1937. Price 12s. 6d. net. This, the fourth edition of the standard and, indeed, so far as the reviewer’s knowledge goes, the only text-book on the subject in the language, bridges L gap of 13 years. The author, pre-eminent in his particular sphere, needs little more introduction to the world of technical industry than he does in his official capicity to readers of THE ANALYST, while his reputation in forensic science in all that appertains to handwriting is international. The chemistry of ink, difficult as it is and at times not a little obscure, hcl- riot developed markedly in the interval since 1924; but what progress has been made is covered by Dr.Mitchell in this edition in a very thorough manner. He has found it necessary to enlarge his work to the extent of some 20 per cent. and, in addition, to rewrite a large portion. The arrangement of the book follows the lines of previous editions. After a comprehensive historical introduction, the work is divided into three sections dealing with writing inks, printing inks, and inks for miscellaneous purposes, respectively. Under Section 1 are considered the chemical nature and treatment of the various raw materials used for writing inks from lcmp black to galls, the composition of finished iron-gall, logwood, vanadium, aniline black, and coloured inks, as well as a comprehensive scheme €or the tech~ical examination of inks, handwriting specimens and the identification of forge:-ies.Section 2 deals with the manufacture and examination of printing inks. ,tnd Section 3 with the miscellaneous materials entering into the compositilxx of copying, marking, safety, sympathetic, typewriter inks and so on. Amongst new matter may be noted references to the use of lignone sulphni--,ites in connection with writing ink, a scheme for the identification of individual con- stituents in inks in the form of writing, and the application of filtered ultra-.& if )let light and of infra-red photography in the elucidation of those problems to which such methods are suited. The British Government Standard Specificatior:s for Writing Inks, revised in 1928, are included for the first time. The avaihble evidence upon the constitution of gallotannin is brought up to date and <tbly reviewed, and there is a Comprehensive list of British patents.It is as difficult to withhold admiration of the encyclopaedic scope cjf the matter and references in this book as it is of the erudition and industry displiiyed in its compilation. Practically nothing that comes to mind has escaped atterition, and it is with rather impish glee that the reviewer, after careful search, asserts that he finds no specific reference to the type of alkaline (ammoniacal) gallotannate- iron ink, said t o find favour in the United States, although the di-ammonium hydroxyferrigallate compound of Silbermann and Ozorovitz receives notice. Nor is there mention of that class of quick-drying writing fluids which depend for their efficiency upon partial destruction of the paper sizing by caustic alk 1.5 or sodium silicate. There is no evidence that lignone sulphonate inks have proved se-rious competitors to iron-gall writing inks (pp. 15 and 175). Apart from the unkttmwn quantity of permanence, the principal failing of this type lies in their liability to contain traces of free sulphurous acid to which suspicion attaches in connt-ction

ISSN:0003-2654    

DOI:10.1039/AN9386300603

出版商:RSC    

年代:1938

数据来源: RSC