摘要:

150 GARDINER AND KEYTE: A STUDY OF SOME METHODS FOR [Vol. 85 A Study of Some Methods for Determining Water in Refined Sugars, Including the Newly Devised Cobaltous Bromide Method BY S. D. GARDINER AND H. J. KEYTE (Tale & Lyle Research Laboratory, Westerham Road, Keston, Kent) The adsorption of cobaltous bromide reagent solution by refined sugar crystals, with and without grinding, is used as a method of determining total water and surface water, respectively. Comparison is made with water determined by oven-drying methods and a vacuum-distillation method. TOTAL water in the sugar crystal may be considered to include (a) free water, called moisture, that is readily driven off by normal methods of drying in an oven, and (b) water bonded to the crystal structure, which is released only after very fine grinding.Syrup is known to be entrapped during the physical build up of 1:he crystal. This “building in” of syrup, i.e., water, has been reported by Powers,l and illustrated by his descriptive photomicrographs and electron micrographs. The normally accepted method for determining moisture in pure sucrose and refined sugars is by drying in an oven at 105” C for 4 hour,2 and the results are used as a guide to the true water content of the crystal. When pure sucrose crystals are used as a primary analytical standard, i.e., for polarimetry, it is usual to grind the crystals carefully before drying to a specific surface of approximately 3500 sq. cm per g, so that total water is exposed. The true water content of the crystals is not required, but only that which remains, so that a correction can be made to the weight of powdered sucrose used for calibration purposes.During grinding water may be lost or gained. Normal drying methods based on the measure- ment of percentage loss of weight have three serious disadvantages, namely, (a) incomplete cooling due to inefficient desiccator design, (b) the formation of degradation products due to heat, and (c) inability to measure total water content. These can be overcome, first by cooling for a long time or by using a water-cooled copper block in the desiccator, secondly by preparing a graph of percentage loss of weight against time of drying and applying a correction for destruction3 and thirdly by grinding the crystals under specific conditions before heating. A method in which heat is not used is advantageous.Two newly devised chemical methods are described, in which cobaltous bromide dissolved in dry chloroform, a reagent that has very great affinity for water, is used. They are (a) chloroform extraction method using cobaltous bromide reagent, (b) cobaltous bromide reagent adsorption method. The first is only very briefly (described because chloroform extracts only “free” or surface moisture from refined sugar crystals, ground or unground, and therefore is no better than the normal conventional methods already in use. In the second method, the forces retaining water in the crystal structure are overcome by allowing the cobaltous bromide reagent to react with this water in sit%, i.e , after adsorption by the crushed crystals; it has revealed a water content greater than was expected.Briefly described, chloroform is used in the first method to extract water from the crushed crystals during the grinding process in a ball mill, and, after separation, the “wet” chloroform is treated under dry air conditions with cobaltous bromide reagent to precipitate the water as a cobaltous bromide hydrate (or complex). The precipitate is washed, dried and weighed as anhydrous cobaltous bromide. Standardisation of the method is by determining added water. The results given in Table I, p. 153, show that the water extracted by chloroform is surface moisture. The second method is described below. METHOD FOR THE DETERMINATION O F WATER IN REFINED SUGAKS BY ADSORPTIOK OF COBALTOUS BROMIDE REAGENT ON THE GROUND SUGAR APPARATUS- flasks.” less-steel balls.Fig. 1 shows the two-piece grinding and filtration apparatus and one of four “pipette The grinding vessel is 4 inches in diameter and holds six $-inch diameter stain- The top filtration section is fitted with a No. 2 sintered-glass filter disc.March, 19581 DETERMINING WATER IN REFINED SUGARS 151 The €329 joint must be fitted with a sleeve. The rubber bung and polythene disc, 4 inches in diameter, fit over the top part to support the apparatus and to facilitate rotation on rollers during grinding. The rubber covered rollers are 16 inches in diameter and are spaced 32 inches apart between centres; they are inclined a t 20" to the horizontal and rotate at 160 r.p.m. m' If B I 9 Joint I I n I B29 joint G r i n d K v e ss e t Scale inches O f 2 Fig.1 . Apparatus for determining water by the cobaltous bromide method There are four pipette flasks and one is seen on the right-hand side of Fig. 1. One delivers 5 ml of dry chloroform, the second 5 in1 of standard "wet" chloroform, the third SO ml of cobaltous bromide reagent dissolved in chloroform and the fourth 12.5 ml of dry carbon tetrachloride. The last reagent is kept in contact with anhydrous cobaltous bromide. This flask is fitted with a medium-porosity glass sinter to separate the liquid from the solid. Volumes are delivered to within + O - 1 ml. The pipette section of the flask is filled automati- cally by blowing dry air through the top of the apparatus, and later, by manipulation of the three-way stopcock, the specified amount of reagent is blown into the grinding vessel, which is attached in readiness a t the spherical joint, J.Strict precautions are taken at a11 times to prevent ingress of moist air. Caps are used to seal joint J when it is not in use. Glassware is dried at 105" C overnight, except the grinding vessel, which, because of its thick- ness, is dried by using absolute ethanol. The apparatus is flushed out with dry chloroform beforc use. REAGENTS- chloroform, containing 150 mg of cobaltous bromide per 20 ml. Cobaltous bromide reagent solution-A solution of recrystallised cobaltous bromide in Carbon tetrachloride-Analytical-reagent grade. Chloroform-Analytical-reagent grade. Standard "wet" chloroform-Chloroform containing 5 mg of water per 5 ml.These reagents are stable. Recrystallised cobaltous bromide is obtained directly from 1" C for a few hours the manufacturer. The crystals are partly dried in a mortar at 105".I 52 GARDINER AND KEYTE: A STUDY OF SOME METHODS FOR [Vol. 83 and the bluish product is ground to facilitate complete dehydration on further heating. The finely powdered cobaltous bromide keeps excellently in its anhydrous state and is conveniently stored in an oven at 105°C. The dark-magenta hydrated crystals are pale green when anhydrous and are extremely hygroscopic. Because of the hygroscopicity, the cobaltous bromide reagent solution is prepared from an approximately saturated solution of the salt (1.1 per cent. w/v) in chloroform at room temperature. As the concentration of the saturated solution can be determined (see below, "Determination of cobaltous bromide in the filtrate"), the cobaltous bromide reagent solution is prepared by proportional dilution with chloroform.The reagent solution is kept in the presence of dry air; its strength will decrease slowly with time, owing to the inevitable pick-up of water from the air and the subsequent precipitation of a few reddish crystals of hydrated cobaltous bromide. Re- determination is needed every few days; a variation of $10mg of cobaltous bromide per 20 ml will not seriously affect the slopes of the calibration curves for water described later. The reagent solution is checked by drying an aliquot overnight at 105" 1" C and weighing the residue as cobaltous bromide, CoBr,. Chloroform can be obtained with very little con- tamination from water, if any, and blank tests can be made to determine these unknown amounts, Cobaltous bromide was tested for metallic impurities by Gross,* who used his high-voltage electrophoresis technique, and showed that the reagent was free from nickel, iron, copper, manganese and lead.The standard. "wet" chloroform is prepared by adding water from an Agla micrometer-syringe pipette to a. known amount of analytical-reagent grade chloroform. Carbon tetrachloride, which normally contains some water, is dried by the addition of freshly dried cobaltous bromide powder, excess of which remains in the flask. PROCEDURE FOR PREPARING THE CALIBRATION GRAPH FOR TOTAL WATER- Initial preparaiion of powdered sucrose-The six stainless-steel balls are placed in the grinding vessel and 30 f 0.5 g of sucrose crystals are ground for 30 minutes -t 10 seconds in the presence of 32-5 k 0.5 ml of undried carbon tetrachloride.The powder is separated by filtration using dry air technique to prevent condensation of moisture and agglomeration of particles, and partly dried at 105" C for 15 minutes. Drying the powdered sucrose-The powdered sucrose is dried immediately before a test point on the curve is required. The aim is to dry the powder irrespective of the slight formation of degradation products. It is dried at 105" 2 1" C for 2 hours in a flat-bottomed metal dish that has a spout at one end, so that when dry the hot powder can be quickly transferred to the grinding vessel and allowed to cool under dry air conditions.Dry air is used to expel moist air when required. Procedure-Cobaltous bromide is adsorbed by the exposed active surfaces of the dry ground sucrose crystals. A blank test determines this amount together with the cobaltous bromide removed by traces of water in the chloroform, ingress of moist air, etc. Points on the calibration graph for total water (curve B, Fig. 2) are determined by the procedure described below, excluding the initial grinding phase, and substituting standard "wet" chloroform for dry chloroform. In this way, cob6altous bromide removed is correlated with milligrams of water in the presence of dry sucro:je powder. PROCEDURE FOR DETERMINING TOTAL WATER IN REFINED SUGARS- Initial grinding-Four 5-ml portions of chloroform, 12-5 ml of carbon tetrachloride and 30 k 0.5 g of refined sugar crystals are transferred to the dry grinding vessel.The polythene disc is placed in position and the flask is rotated for 30 minutes 10 seconds. Large crystals may need longer time and very :;mall crystals less time. The criterion is a specific surface of 3500 2 200 sq. cm per g. Addition of cobaltous bromide reagent solution-The grinding assembly is removed from the rollers and 20 ml of the cobaltous bromide reagent solution of previously determined concentration are added. The flask is rotated for a further 10 minutes 10 seconds. During this period, the reagent is given ample time to react with the bonded water exposed by grinding and there is a slight increase in specific surface. Immediately the second grinding phase is over, the remaining cobaltous bromide is filtered into the top flask by inverting the assembly and applying a pressure of dry air of 2 to 4 lb per sq.inch to speed up filtration. Determination of cobaltous bromide in the3Ztraie-A 10-ml aliquot is heated at 65" to 70" C to volatilise the chloroform and carbon tetrachloride and dried at 105" f 1" C for 2 hours,March, 19581 DETERMINING WATER IN REFINED SUGARS 1 53 or overnight, to produce anhydrous cobaltous bromide, which is weighed when cold. A circular silica capsule with ground lid5 is used. CALCULATION OF RESULTS- Amount of cobaltous bromide added to test . . = 150mg Volume of solution . . .. .. .. .. = 52.5ml Amount of cobaltous bromide in 10 ml of filtrate . . = W mg 52.5 Therefore amount of cobaltous bromide in 52.5 ml = W - mg 10 Therefore amount of cobaltous bromide removed .. = 150 - W - mg 52-6 10 This result, when referred to the calibration curve for total water (curve B, Fig. 2) gives the result for the total water in 30 g of refined sugar. Cobaltous bromide removed, mg Fig. 2. Calibration curves: curve A, surface moisture; curve B, total water PROCEDURE FOR DETERMINING SURFACE MOISTURE IN REFINED SUGARS- Determinations are carried out in a similar way to the total-water determinations, except that the initial grinding period of 30 minutes is omitted, and the six stainless-steel balls are not present during the 10 minutes' mixing with the cobaltous bromide reagent solution. A calibration graph for surface moisture (curve A, Fig.2) is prepared in a similar manner to the calibration graph for total water. TABLE I DETERMINATION OF WATER IN REFINED SUGARS Water found by drying in an oven at 105' C for- & 1 3 % Yo Sample hour, hours, Refined sugar No. 1 . . 0.011 0.012 Refined sugar No. 2 . . 0.012 0-017 Powdered sucrose . . 0-008 0.017 Sucrose crystals . . 0,002 0.003 Refined sugar No. 3 . . 0.026 0-039 * Results supplied by Water found by .drying in a vacuum oven at 70" C for 20 hours, % - 0.016 0,003 0.016 S. Hill and Water found by adsorption on cobaltous bromide method- W- grinding, grinding, % % 0.010 0.033 0.013 0.047 - 0.013 <0-002 0.018 0.012, 0.066, 0.014 0.055, 0.056 A. G. R. Dobbs.% Water found by chloroforni extraction of cobaltous bromide reagent with grinding, 0.010 0.006 0.006 Yo - - Water found by vacuum dis tilla- tion with grinding,* Yo - 0.0414 0.01 32 0.0195 0.0395154 several methods.revision, or possibly rejection, of normally accepted methods is to be advocated. GARDINER AND KEYTE: A STUDY OF SOME METHODS FOR [Vol. 83 Table I gives the results of determinations of water in refined sugars and sucrose by Comparison is made to show that serious differences exist, and that DISCUSSION OF RESULTS Heat, whether by thermal conduction or radiation (by infra-red rays) or heated air, causes degradation products, which interfere with the determination of water. Table I1 shows the high results produced by incomplete cooling of one aluminium dish with lid and 20g of refined sugar, which indicate that drying in an oven is not suitable as a routine method for the determination of moisture.TABLE 1:I HIGH RESULTS CAUSED BY INCOMPLETE COOLING IN THE DETERMINATION Water was determined by heating for 1 hour at 105" C OF MOISTURE BY DRYING IN AN OVEN Water found after cooling for 20 minutes in-- (-..-.---h__-----.- 7 < No. 1* No. 27 Sucrose finely ground . . 0.028 0-0:I 3 Sucrose crystals . . 0.025 0.002 Caster sugar . . . . 0.029 0.0 I. 2 Refined sugar . . . . 0.032 0.01!4 Sample desiccator, yo desiccator, yo Water found after cooling for 90 minutes in- No. 1* No. 27 desiccator, % desiccator, yL 0~000 0.009 0.002 0.001 0.016 0.01 1 0*016 0.014 A 7 I__--_ * Containing barium oxide. 7 Containing barium oxide and a water-cooled copper block, the cooling water being at the balance Table I shows that, for accurate water determinations, sugar crystals must be ground to a specific surface of 3500 sq.cm per g to expose all the bonded water. In the cobaltous bromide method the increase of invert sugar is less than 0.005 per cent. Dry methods of grinding can cause serious local increase of temperature at the many points of contact, and also greater formation of invert sugar. The determination of water in sugar crystals by drying in a vacuum oven at 70" C is satisfactory in that the results agree well with surface-moisture determinations by the cobaltous bromide method (see curve A, Fig. 2). Drying in an oven at 105" C for 1 hour gives a higher result for surface moisture becaiise of partial decomposition and possible breakdown of the sugar crystal by excessive heat.Further experiments, in which the crystals were dissolved in distilled water and dried on Celite powder, were tried exhaustively, but the results were erratic. temperature. The very small losses in weight made the method too difficult. CONCLUSIO~U'S There is no doubt that, as suggested by Powers, the sugar crystal conceals more internal water than is generally assumed. The water can be measured by the adsorption of cobaltous bromide reagent on the ground crystal, provided the specific surface is at least 3500 sq. cm per g, and this is considered to be suitable as a routine method for total water. Surface rnois- ture can also be speedily determined. The cobaltous bromide method is not absolute, as it requires standardisation by water in the presence of dry sucrose powder of specific surface of 3500 sq. cm per g. The adsorption on cobaltous bromide method may be simplified by using only the first and third reagents. A slight loss of sensitivity is caused by not using carbon tetrachloride. Instead of the use of "wet" chloroform, water is added directly to the grinding vessel from an Agla micrometer-syringe pipette. The few drops of water are added through the side-arm and washed down by chloroform and cobaltous bromide reagent solution. Time is reduced considerably if the gravimetric method for the determination of cobaltous bromide is replaced by a comparative electrical conductivity method. We express our thanks to the Directors of Tate &. Lyle Limited for permission to publish this work, and to D. Buxton for the many pieces of glass apparatus made by him. We also thank F. J. Gardiner (%be F. J. Farmiloe) and R. Runeckles for their co-operation.March, 19581 DETERMINING WATER I N REFINED SUGARS REFERENCES 1 . 2 . 3. 4. 5. British Standard 420: 1931. 6. Powers, €4. E. C., h'ature, 1956, 178, 139. PYOC. In&. Comm. Uniform Meth. Sugar Anal., Tenth Session, 1949, p. 31. Gardiner, S. D., and Farmiloe, li. J., Analyst, 1954, 79, 447. Gross, D., Nature, 1937, 180, 596. Hill, S., and Dobbs, ,4. G. R., Analyst, 1!158, 83, 143. 155 licceived A icgzasf S2nd, I857

ISSN:0003-2654    

DOI:10.1039/AN9588300150

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

March, 19581 DETERMINING WATER I N REFINED SUGARS 155 The Determination of Monobromamine and Monochloramine in Water BY J. K. JOHANNESSON (Wellington City Couizcil Laboratory, Wellington, New Zealand) Chlorination of waters containing both bromide and ammonia, such as obtains in sea-water swimming pools, leads to the formation of both monochloramine and monobromamine. The amount of these substances present may be determined (a) by reaction of the monobromamine with neutral o-tolidine and titration with ammonium ferrous sulphate-addition of potassium iodide then permits the determinatiop cf monochloramine- and ( b ) in neutral solution at zero applied voltage, monobromamine alone is titrated a t the rotating platinum electrode with phenyl arsenoxide. Mono- chloramine is then determined by addition of potassium iodide and further titration with the same reagent.I-r was shown by Johannesson in a previous publication1 that, when chlorine is added to a solution containing excess of bromide together with ammonia, either monobromamine or monochloramine, or both, are formed according to the reaction conditions. When the chlorine is added in neutralised form, i.e., as sodium hypochlorite, to alkaline solutions a t pH 8.0 containing an excess of bromide and ammonium ions, only monochloramine is formed. Conversely, under acid conditions of reaction for both reactants, only monobromamine is formed. Addition of an acid solution of chlorine to excess of bromide and ammonium ions, in a solution containing a reserve of alkalinity, such that the final mixture is not acid, results in a mixture of the amines, the proportion of each present depending upon, among other factors, the rate of mixing, i.e., formation of local acidity.Previously, Houghton2 had made some investigations on the chlorination of water containing bromide and showed that there was some evidence of the formation of mono- bromaniine in the presence of ammonia, but he was uncertain as to what actually existed in the solutions. He further found that solutions containing an excess of bromide and ammonia were, when chlorinated, more bactericidal than pure monochloramine solutions. I have found that solutions of monobromamine are much more strongly bactericidal than monochloramine solutions of the same equivalent oxidising strength and, further, that the bactericidal effects are complicated by the following reaction- NH,Br + R2NH -+ NH, + R,NBr, the R,NH arising by diffusion from the bacterial cells.This type of reaction has been demonstrated with dimethylamine in amperometric titration apparatus. This effect is shown by the break in the curve when the logarithm of the number of survivors is plotted against time. As can be seen from Fig. 1, no such effect is observed with monochloramine. Farkas and Lewin3 have shown that at pH values of 10 to 12 the reaction- OC1- + Br- -+ OBr- + C1- is very slow and not complete even after many hours; at pH 8 to 10 the reaction is still slow and requires some minutes for completion. On the other hand, the reaction- is one of the most rapid reactions known. Laitinen and Woerner4 titrated ammonia ampero- metrically with hypochlorite in the presence of bromide at pH 8.2 and presumably the excess Only below pH 8 is the reaction rapid.NH, + OC1- -+ N€I,Cl + OH-156 JOHANNESSON: THE DETERMINATION OF [Vol. 83 of hypochlorite over that required to form monochloramine reacts with the bromide to give hypobromite, which will react in turn with the inonochIoramine and cause a rapid over-all decomposition. 0 10 20 Time, minutes Fig. 1. Death rate of Escherichia coli in the presence of 1 0 - 3 M ammonium sulphate at pH 8.2 with sodium hydrogen carbonate buffer solution a t 15" C on exposure to : curve A, monochloramine ; curve B, monobromamine. The con- centration of each bactericide was 0.15 p.p.m., as chlorine The chlorination of sea-water swimming pools containing ammonia from the bathing load or of sea water containing sewage usually involves the addition of a solution of free chlorine, which gives rise to both monochloramine and monobromamine.The study and control of the chlorination in these circumstances requires the use of methods of analysis that will differentiate between monochloramine and monobromamine. Although a spectrophotometric method permits the determinations to be made when the chlorine dose is about 10 p.p.m. or higher, the residuals encountered in practice are usually between 0.2 and 0.8 p.p.m., expressed as chlorine, and require the use of much more sensitive met hods of cleterminat ion. TABLE I EFFECT OF MONOBROMAMINE AND MONOCHLORAMINE ON VARIOUS METHODS OF DETERMINING, RESIDUALS Me thotl I'otassium iodide with starch as indicator in--- ( u ) acid solution .. a . .. . . Monobromamine ( b ) neutral solution . . . . . . Monobromamine (a) acid solution . . .. .. . . Monobromamine ( b ) neutral solution .. . , . . Monobromamine Methyl orange in acid solution . . . . . . Monobromamine Amperometric at 0.2 volt measured against the saturated-calomel electrode (pH 7.2) . . Monobromamine o-Tolidine in- Remarks and monochloramine react and monochloramine react and monochloramine react only reacts only reacts only reacts When bromide is present, free chlorine and halogen amines can have only a momentary co-existence. A method mentioned later (see p. 157) for the determination of mono- bromamine and monochloramine, i.e., the F.A.S.method, will not differentiate between free bromine, combined bromine and free chlorine. The proposed method, amperometric titration, will, however, permit this distinction to be made.March, 19581 MONOBROMAMINE AND MONOCHLORAMINE I N WATER 157 SURVEY OF POSSIBLE METHODS FOR DIFFERENTIATING BETWEEN MONOBROMAMINE AND MONOCHLORAMINE Solutions of monobromamine were prepared by adding freshly prepared bromine water to 0.1 M ammonium sulphate containing 300 p.p.m. of sodium hydrogen carbonate. Mono- chloramine was similarly prepared, but with chlorine water instead of bromine water. The concentrations were made to about 1.0 p.p.m., expressed as chlorine. The solutions were tested by using the usual methods and reagents for determining chlorine, and the results of these tests are given in Table I.Three of the methods appeared to be suitable and they were examined in greater detail; they were the methyl orange, the neutral o- t olidine and the amperome t ric t itrat ion met hods. METHYL ORANGE METHOD- Methyl orange was introduced by Tarass for determining free chlorine. it reacts instantaneously with free chlorine, but not with monochloramine. monobromamine reacts rapidly, apparently owing to appreciable hydrolysis, as follows- At pH 3 or less, At this pH, NH,Br+ + H,O --+ NH,+ + HOBr. However, in the presence of bromide at pH 3, monochloramine reacts rapidly with methyl Methyl orange and so renders this method impracticable for the purpose of this investigation. red behaves similarly. U-TOLIDINE METHOD- With an acidified solution of o-tolidine, both monochloramine rind monobromamine react, the latter rapidly.Palin6 introduced a neutral o-tolidine reagent mixed with sodium hexametaphosphate to act as a combined buffer and sequestering agent. This reagent forms a blue quinonoid compound with free chlorine, but does not react with monochloraminc. The blue colour can be determined colorimetrically or, better, titrated with ammonium ferrous sulphate to the disappearance of the colour. At this stage, addition of iodide causes any monochloramine that may be present to react, producing a further blue compound, which can then be titrated with ammonium ferrous sulphate. As it was found that monobromamine behaves with this reagent as if it were free chlorine and, further, that monochloramine in the presence of bromide does not react, it was possible to evolve a suitable method for the determination of the concentration of both monochloraminc and monobromamine when they are present together.It was found that the F.A.S. method, as described by Palin,7 would differentiate between monobromamine and monochloramine, the monobromamine reacting as if it were free chlorine. Free bromine will, of course, also react as if it were free chlorine and consequently the method will not distinguish between these forms, but, after an excess of an ammonium salt has been added, chlorine will no longer react. This method is not, however, suitable for use with sea water, as the phosphate buffer solutions cause precipitation of the magnesium salts and the resulting turbidity interferes with the titration.It is convenient to express all results in parts per million, as “chlorine.” titre, ml x volume of test solution, ml 100 p-p-m., “Chlorine” present = where 1 ml of standard ammonium ferrous sulphate solution = 0.1 mg of chlorine. AMPEROMETRIC METHOD- The initial experiments were made with a large-area rotating electrode, which was made by coating a glass bulb at the end of a hollow glass rotor with a conducting film of platinum deposited by applications of saturated platinic chloride solution and heating. Electrical contact was established by means of a platinum wire sealed through the glass. This has been described elsewhere.* At a later date, a Wallace and Tiernan apparatus became available. Marks and Bannisters found that at neutral pH values the chlorine wave commenced near + 0-7 volt measured against the saturated-calomel electrode, the monochloramine wave being much more negative.Further, at these pH values sodium arsenite and phenyl arsenoxide reduce only free chlorine, but, when iodide is added, the monochloramine liberates158 JOIIANNESSON THE DE;TERMINATION OF [Vol. 83 free iodine, which yields a reduction wave and at this applied potential is also reduced by the a bove-mentioned reagents. I have found that the monobromamine wave commences at approximately +O-5 volt measured against the saturated-calomel electrode and that it is reducible with both arsenite m d phenyl arsenoxide. It was therefore possible to use this method for the determination o f monobromamine and, further, by adding iodide, to determine monochloramine when present at the same time.In the presence of sodium hexametaphosphate it is possible to titrate the monobromamine with ammonium ferrous sulphate, but the rate 'of reaction is somewhat slow and, further, the graph of the volume of titrant against current is not linear, as it is when the arsenic compounds are used. As mentioned previously, free chlorine will not co-exist , other than momentarily, with the halogen amines when bromide is present, nor will bromine in the presence of mono- bromamine or monochloramine. If free bromine or chlorine is present, addition oi an am- monium salt will cause a large and rapid decrease of diffusion current to a quarter or less of the original value. If there is no reduction of current, then the combined forms are present and of these only monobromamine will be titrated with phenyl arsenoxide.Subsequent ;{&lition of potassium iodide permits the monochloramine to be titrated. Titration with phenyl a rsenoxide is recommended. Free brominl ', ' \ ; I ' " I I 0 0.4 0.8 1-2 1.6 2.0 2.4 2.8 Chlorine dose, p.p.m. Fig. 2. Chlorination. of sea water, con- taining natural bromide with 0.26 p.p.m. of ammonia added, by treating it with increasing amounts of sodium hypochlorite and then determining the nature of the residue after 30 minutes METHOD REAGENTS- Pjzenyl arsenoxide solution-Dissolve 0.4 g of phenyl arsenoxidelo in a small amount of dilute alkali solution, dilute to nearly 1 litre with distilled water and adjust the pH to between 6 and 7 with dilute hydrochloric acid.Standardise against standard iodine, using starch as indicator. 1 ml = (factor from standardisation) x 0.2 mg of chlorine. Sodium hydrogen carbonate-Analytical-reagent grade. Ammonium sulplzate solution, N. Potassiwn iodide solution, 1 per cent. aqueous. PROCEDURE- With the amperometric titration apparatus set at zero applied volts, 200 ml of the test solution are placed in the apparatus and approximately 0.2 g of sodium hydrogen carbonateMarch, 19581 MONOBROMAMINE AND MONOCHLORAMINE I N WATER 159 is added. The current reading is observed and then 1 ml of ammonium sulphate solution is added. A large reduction of current indicates that free bromine or chlorine is present, and 110 reduction of current indicates that the combined forms are present.The solution is now titrated with the phenyl arsenoxide solution until no further change of current occurs. This titration represents either bromine or monobromamine, according to whether free or combined bromine is present. One millilitre of potassium iodide solution is now added and the titration is continued until once again there is no further reduction of current. This additional titration represents either chlorine or monochloramine, according to whether free or combined chlorine is present. As mentioned before, it is convenient to calculate the results as p.p.m. of chlorine. Chlorine, p.p.m. = volume of phenyl arsenoxide solution used, ml. RESULTS Table I1 shows a comparison between results by the F.A.S. method and the proposed Fig.2 shows the results obtained from the chlorination of sea water containing method. an excess of ammonia, the determinations being made by the amperometric method. TABLE I1 COMPARISON OF RESULTS BY DIFFERENT METHODS The test solutions were prepared as described on p. 157 Found by F.A.S. method A c 7 Monobromamine, Monochloramine, Total, as chlorine, as chlorine, as chlorine, p.p.m. p.p.m. p.p.m. 0.95 1.00 1-95 0.65 0.65 1.30 0.63 0.50 1-13 0.35 Nil 0.35 0-50 Nil 0.50 Nil 0.72 0-72 0.30 0.65 0.95 Found by amperometric method Monobromamine, as chlorine, p.p.m. 0.93 0.64 0.65 0.41 0.49 Nil 0.25 Monochloramine, Total, as chlorine, as chlorine, p.p.m. p.p.m. 0.95 1-88 0.68 1-32 0.45 1.10 Nil 0.41 Nil 0-49 0.72 0.72 0.69 0.94 1 . 3. 4. 5. 6. 7. 8. 9. 10. 3 d . REFERENCES Johannesson, J . I<., N.Z. J . Sci. Tech., 1955, 36, 600. Hougliton, G. U., J. SOC. Chem. I n d . , 1946, 65, 324. Farkas, L., and Lewin, M., Anal. Chcm., 1947, 19, 865. Laitinen, H. A., and Woerner, 13. E., Ibid., 1955, 27, 215. Tarss, RI., I b i d . , 1947, 19, 342. Palin, A. T., J . I n s t . Wat. E n g . , 1947, 3, 100. American Public Health Association, “Standard Methods for the Examination of Water, Sewage Johannesson, J. K., Chem. & I n d . , 1956, 1141. Marks, H. C., anti Bannister, G. L., AnaZ. Chem., 1947, 19, 200. “Organic Reactions,” John Wiley & Sons Inc., New Yorlc, 1944, Volume 11, p. 424. and Industrial Wastes,” Tenth Edition, New York, 1955. Received December 19th, 1956

ISSN:0003-2654    

DOI:10.1039/AN9588300155

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

160 VERMA, BHUCHAR, THERATTIL AND SHARMA: DETERMINATION OF [VOl. 83 Determination of Free Lime in Lime and Silicate Products Part I. Extraction of Freshly Ignited Lime with Non-aqueous Solvents and Determination of the Calcium Oxide Content of the Extracts BY M. R. VERMA, V. M. BHUCHAR, MISS K. J. THERATTIL AND S. S. SHARMA (National Physical Laboratory of India, Division of A nalytical Chemistry, Hillside Road, New Delhi) An examination of methods of extracting calcium oxide from freshly ignited lime by various non-aqueous extracting agents and its determination by potentiometric, volumetric and compkxometric methods has been carried out. New indicators have been recomniended for the visual alkalimetric titration of the extracts. The use of these indicators gives values consistent with those obtained by potentiometric titrations.Various factors, such as the ratio of the extractant to alcohol in the titration medium, the percentage of water that can be tolerated therein and the possibility of using industrial methylated spirit, have been examined. These factors can be a source of error in :such determinations. DETERMINATION of “free lime” in lime and silicate products, such as cements, pozzolanas, mortars and concrete, is very often required in the evaluation of these materials. In the past, many methods have been developed and have also been critically examined for their accuracy. The chemical methods for the deterrnination of “free lime” involve essentially the extraction of “free lime” by aqueous or non-aqueous solvents and then the determination of the calcium oxide content of the extract by a suitable titration method.Several aqueous solvents, such as lime watt:r,l dilute and weak acids, solutions of an ammonium salt2 and sugar,3 have been used. 14s with these extracting agents there is a likelihood of increase in the extracted calcium oxide content caused by hydrolysis of the silicate: the use of non-aqueous solvents has been preferred to that of aqueous solvents. Of the several non-aqueous extracting agents, glycerol was suggested by Emley5 and its use was re-examined by a number of workers. The extracted lime was titrated against am- monium acetate by Lerch and Bogue,6 against benzoic acid by Bessey7 and against tartaric acid by Rathke.8 Konarzewski and Lukaszweicsg used phenol for extracting lime and titrated the extract with standard hydrochloric acid.The method was extended to pozzolanic mixtures by Sestini and Santarelli3 and by Wittekindt.lo Schlapfer and Bukowskill $12 used ethylenc glycol for extracting lime, and this method has been extended to hydrated cement mixes by Forsen13 and Rodt,14 and to clinkers and Portland cements by MacPherson and Forbrick.Ij The Associated Cement Company of India16 also use ethylene glycol for extracting “frec lime” from cement. Franke17Js used a mixture of acetoacetic ester and isobutyl alcohol for extracting “free lime” from cements. Assarsson and B ~ k s t r o r n ~ ~ separated the free calcium oxide in lime and silicate products by using some of the previously mentioned extracting agents and carried out the determina- tion by titrating the extract against standard hydrochloric acid or benzoic acid, For visual titrations these investigators used the same indicators as previous workers.They also carried out potentiometric and conductimetric titrations on the extracts. On analysis of their results, it appears that, although the method suggested- titration of the extracted calcium oxide with a strong acid-is an improvement over the earlier method of titration with ammonium acetate, the results when different indicators are used are unreliable and there is no correlation between the results by visual and potentiometric titrations. From the original paper of Konarzewski and Lukaszweicsg it was noted that the factor used by them for the calcium oxide determination is arbitrary.Normally, 1 ml of N/15 hydrochloric acid should be equivalent to 0.00187 g of calcium oxide, whereas these authors found that it is equal to 0-00202 g. Similarly, when reference was made to the work of MacPherson and Forbrick15 on the determination of “free lime” in cements, it was noted that a rather unconventional procedure was adopted. Pre-treatment of the solvent, viz. ethylene glycol, was carried out with calcium oxide and it was then titrated against an acid, a mixture of phenol- phthalein and naphtholphthalein being used as the indicator; the pre-treated ethylene glycol%arch, 19581 FREE LIME IN LIME AND SILICATE PRODUCTS. PART I 161 was used for extracting calcium oxide from cement, the titration again being carried out in the same manner.In a paper presented at the symposium on the Chemistry of Cements held at Stockholm in 1938, Bessey4 drew attention to the views held by various authors on the inadequacy of reliable procedures for assessing the accuracy of "free lime" determinations, and the position does not seem to have improved since. In our view, an important point that requires consideration in calcium oxide - acid titrations is the choice of indicators. I t is well known that both the sensitivity and pH range of response of indicators in non-aqueous media are not20,21 necessarily the same as in aqueous media. In view of these facts and also the lack of a well defined and agreed procedure of extrac- tion, the uncertainty of conversion factors and the use of indicators that may not be suitable for titration in non-aqueous media, it was considered desirable to re-investigate the problem.* Particular care was taken during the study of the various factors involved in the extraction of calcium oxide from freshly ignited lime and its determination by electrometric and ordinary volumetric titrations.Recently developed complexometric titration procedures were also used in these determinations, and it is hoped that applications of these procedures to the investigation of lime and silicate products will be discussed in a separate paper. The method of Emley and of Lerch and Bogue-titration of the extract with ammonium acetate-was not used, as it was too time-consuming, EXPERIMENTAL This fact seems to have been largely overlooked by the previous investigators.The experimental work was divided into the following steps- (i) Procedure for extracting calcium oxide from freshly ignited lime with various non-aqueous solvents, viz., glycerol, ethylene glycol and phenol, and also the procedure for diluting the extracts to standard volumes. (ii) Plotting of potentiometric titration curves of these extracts by titrating with standard hydrochloric acid and determining the titre from the maximum d(pH)/dv value. (iii) Determination of the extracted calcium oxide by titrating with hydrochloric acid, use being made of both the established and the newer indicators. (iv) Complexometric titration of the extracts from lime. (v) A study of the effect of adding ethanol or water, or ethanol plus water, to the calcium oxide extracts on the shift of end-point in potentiometric determination and the change in titre when visual indicators are used.( v i ) Practicability of the use of distilled rectified spirit or industrial methylated spirit (without drying) for the final dilution of the extracts. -IPPA4RATUS AND REAGENTS- A Beckman pH meter, model H2, with glass calomel electrodes was used for the potentio- The reagents used in the experimental work were as follows- Calcium oxide-This was prepared by fully igniting AnalaR calcium carbonate at 850" to 900" C for 2 hours. The loss 011 ignition for different batches varied from 43.5 to 13.9 per cent, Glycerol-The B.P. grade was further purified by collecting the fraction boiling at 177" to 178" C under a pressure of 7 to 8 mm of mercury.Ethylene glycol-The C.P. grade was further purified by distillation, and the fraction boiling at 195" C was collected. Phenol-The C.P. grade was further purified by distillation, and the fraction boiling between 178" and 180" C was collected. Ethanol--Rectified spirit was purified by distillation over lime and finally by treatment with anhydrous copper sulphate. The distillate when tested was free from water, lime, copper and so on, and was neutralised, when necessary, before it was used for dilution purposes. metric titrations. Water, dou ble-distilled-Free from calcium ions. Ethanolic hydrochloric acid-An approximately 0.05 N solution. 1 ml z 1.4 mg of calcium oxide. * Since the preparation of this paper, Pressler, Brunauer and Kantrote have investigated Franke's method of extraction with acetoacetic ester.162 VERMA, BHUCHAR, THERATTIL AND SHARMA: DETERMINATION OF [Vol.83 Ethanolic sulphuric acid-An approximatelly 0.05 N solution. Methyl orange indicator solution-A 0.1 pier cent. aqueous solution. Mixed methyl red - methylene blue indicator solution-A solution containing 0.1 per cent. Alizarin S indicator solution-A 0.1 per cent. aqueous solution. Bromocresol green indicator solution-A solution obtained from the British Drug Houses Limited was used. Disodium ethylenediaminetetra-acetate solution-An aqueous solution standardised against the standard calcium solution. Standard calcium solution-A known amount of dried AiialaR calcium carbonate wa\ dissolved in hydrochloric acid and diluted to a definite volume. Ammonium chloride - ammonium hydroxide buffcr solution-This was prepared by clis- solving 67.5 g of ammonium chloride in 570 ml of ammonia solution, sp.gr. 0-880, and diluting to 1 litre.of each. Eriochrome black T indicator solution--A 1 per cent. solution in ethanol. EXPERIMENTS WITH GLYCEROL AS THE EXTRACTING AGENT- Preparation of solution of calcium oxide in glycerol-A known weight of freshly ignited lime (0.05 to 0.200 g) was transferred rapidly to a conical flask fitted with a B24 ground-glass joint and containing about 10 ml of anhydrous glycerol; 40 ml of anhydrous glycerol were added and the flask was stoppered and set aside overnight. At the end of this period, a con- denser having a calcium chloride tube at its other end was fitted to the flask. The flask was heated on a water bath for 6 to 9 hours.During the heating, any speck of undissolved lime was broken with a glass rod. This solution was cooled and diluted to a known volume (250 or 500 ml) with dried neutralised ethanol. The possibilitj- of water entering the extracting solvent during dissolution of the lime was reduced by using a reflux flask fitted with the cone of a standard joint, the condenser being fitted with the socket of the joint. The solution was heated until clear. Volume of hydrochloric acid added, ml Fig. 1. Titration of calcium oxide extracted with glycerol: curve A, extract with ratio of ethanol to glycerol of 4 to 1 ; curve B, extract with ratio of ethanol to glycerol of 50 to 1 ; curve C , extract with ratio of ethanol to glycerol of 4 to 1, containing 85 per cent.of water. The arrows correspond to the points at which do are maxima dv Potentiometric titration-Aliquots of the prepared solution (generally 25 ml) were diluted with different amounts of purified anhydrous ethanol and potentiometric titrations were carried out with use of the Beckman pH meter. Curves were plotted of pH against theMarch, 19581 FREE LIME IN LIME AND SILICATE PRODUCTS. PART I 163 volume of acid added, and, from each curve, the amount of acid corresponding to the maximum value of d(pH)/dv was found. f i g . 1 shows the titration curves of a typical solution of calcium oxide extracted with glycerol. Curvc A is for a solution of calcium oxide in which the ratio of ethanol to glycerol is 4 to 1. The pH values were recorded about 5 minutes after the addition of the acid, and this time interval was adhered to in all the titrations.If, however, the solution is shaken for a longer time, the flat portion of the curve is extended towards the pH-axis. The curve plotted when this procedure was used showed pronounced kinks, as did those obtained by Assarsson and BokstrOm.l9 It was also noted that electrodes that had been in use with non-aqueous media responded sluggishly, but not erroneously, when used with aqueous media and vice versa. For the sake of uniformity of results, the same set of electrodes was used throughout the investigation for non-aqueous titrations. Volzcmetric determination of calcium oxide-From the titration curves shown in Fig. 1 , it is clear that an indicator that changes colour between pH 4 and 4-5 in the described ethanolic medium could be used for determining the end-point.The choice of indicator, however, remains a matter of trial. The colour changes of methyl red and phenolphthalein have been shown to be not too satisfactory in ethanolic media. Bromocresol green has been said to give satisfactory colour changes when titrations of calcium oxide extracted with ethylene glycol are carried out in an ethanolic medium.l6 In this instance, in addition to these indicators, alizarin S was tried and was found to be suitable. This indicator gave a perceptible colour change from pink through colourless to pale yellow. The mixed alizarin S - bromocresol green indicator gives a colour change from pink to pale yellow and would seem to be preferable.These colour changes are coincident with the neutralisation point as deduced from the data from the electrometric tit r at ions. For volumetric determinations, an aliqiiot (generally 25 ml) of the prepared solution was taken and 3 drops of alizarin S solution were added. Ethanolic hydrochloric acid was added to the solution until the colour changed from pink to colourless by way of a rather faint pink. The titration was continued until the appearance of a yellow colour. The end-point was taken as being just 1 drop before this point, i.e,, the point at which the solution was just colourless without any tinge of pink. When the mixed indicator, made by mixing 3 drops of alizarin S solution with 2 drops of bromocresol green solution, was used, the colour passed through violet, pink and faint pink to clear yellow, the end-point being taken as the point at which the colour changed from pink to clear yellow.The solution was stirred throughout the titration. The results obtained when this procedure was used are given in Table I. TABLE I EXTRACTION OF CALCIUM OXIDE FROM FRESHLY IGNITED LIME WITH GLYCEROL Calcium oxide found by-- Weight of calcium oxide taken for extraction, g 0.1202 0.0782 0.0956 0.1258 0.3213 7 Final volume potentio- of metric solution, titration, 250 98.7 250 500 97-1 500 98-1 ml % - 500t - titration with titration with titration with mixed alizarin S - alizarin S as bromocresol green bromocresol green indicator, as indicator, as indicator, % % 98.6 96.8" - 99.1 97.3* - 97.1 - 96.6 98.5 - 98.9 - - - complexo - metric titration, 97.9 98.8 98.6 100~00 % - * Colour change was graded.With regard to the use of other indicators, it may be noted that, in a particular deter- mination, the recovery of calcium oxide with phenolphthalein as the indicator was as low as about 60 per cent., and in a determination with methyl red as the indicator, the recovery was about 80 per cent.; even the colour change was not sharp. Other indicators, such as bromothymol blue, thymol blue, bromocresol purple, Congo red and rosolic acid were tried, but were found to be unsuitable, as no clear end-point was obtained. When bromocresol t Industrial methylated spirit was used for the dilution164 VERMA, BHUCHAR, THERATTIL AND SHARMA: DETERMINATION OF [VOl. 83 green was used as the indicator, the change of colour was graded and extended over 0.2 ml of 0-05 N hydrochloric acid.Complexometric titration-A 10-ml aliquot of the prepared solution was titrated against a standard solution of disodium ethylenediarninetetra-acetate in the usual manner in presence of an ammonium chloride - ammonium hydroxide buffer and with Eriochrome black T as the indicator. Titrations were also carried out after limited dilutions of the extract with double-distilled water free from calcium ions. The complexometric titrations do not present any difficulty, and the results are in fair agreement with those obtained potentiometrically or alkalimetrically (see Table I). Efect of addition of ethanol or water on the shi,ft of the end-points-Aliquots of the prepared solution were diluted to 100 ml with ethanol.I n another titration, the aliquot was diluted with water. Potentiometric titration curves for these solutions were obtained by titrating with ethanolic hydrochloric acid. A potentiometric titration curve for one such solution containing ethanol and glycerol in the ratio of 50 to 1, and another for a solution containing 85 per cent. of water are shown in Fig. 1 (curves B and C, respectively). It can be seen that there is a difference between curve B and curve A, but the end-point falls at the same titre. However, when a large amount of water is present (see Fig, 1, curve C), the end-point is rather uncertain. Titrations were also carried out with alizarin S, bromocresol green, and mixed alizarin S - bromocresol green as the indicator.Table I1 shows some typical results of the titration of the calcium oxide solution on progressive dilution with water. TABLE I1 EFFECT OF VARIOUS AMOUNTS OF ~VATEK Weight of calcium oxide taken for titration, g 0*00629 0.00475 0.00456 Amount of water present at the end-point, Nil % r 1: 15 I 31 Nil I : : ( Nil 10 20 Method o f determination Volumetric with alizarin S as indicator Potentiometric Voluinetric with mixed alizarin S - bronio cresol green as indicator Potentiometric Volumetric with alizarin S as indicator Volume of ethanolic hydrochloric acid required, ml 5.47 5.48 5-48 5.47 5-45 3.80 4.07 4.00 3.90 4.00 3.90 3.80 3.60 Use of aTt ethanol - water mixture for the j h a l dilution of the extracts-Ethanol mixed with different amounts of water was used for the final dilutions of glycerol extracts, andit was found that no turbidity developed when a mixture containing up to 10 per cent. of water by volume was used.The most uniform results are obtained if the lime is soaked overnight in glycerol, digested for 6 to 9 hours and then diluted with at least 90 per cent. ethanol. The determinations can be carried out by titration potentiometrically with et hanolic hydrochloric acid, or volumetrically with alizarin S as the indicator. The colour of the indicator changes from pink to colourless and immediately afterwards to pale yellow. The end-point is taken as the colourless point just before the appearance of the pale yellow. The results are somewhat lower when bromo- cresol green is used as the indicator. The colour change of bromocresol green is not sharp,March, 19581 FREE LIME I N LIME AND SILICATE PRODUCTS.PART I 165 but extends over a definite range. Complexometric titration can be carried out without difficulty and with the same degree of precision as the alkalimetric titration. KXPERTMENTS WITH ETHYLENE GLYCOL AS THE EXTRACTING AGENT- Preparation of solution of calcium oxide in ethylene gJycol-A known weight of freshly ignited lime was placed in a conical flask with about 10 ml of anhydrous ethylene glycol. A 40-ml portion of anhydrous ethylene glycol was added and the mixture was allowed to digest on a water bath for 30 to 60 minutes. For complete dissolution, this last step is essential. When cool, the solution was diluted to the required volume with anhydrous ethanol.Potentiometric and volumetric titration of the extract-Aliquots of this solution were titrated potentiometrically against ethanolic hydrochloric acid by using the Beckman pH meter and titration curves were plotted. In this instance also, kinks were observed in the titration curve (see Fig. 2 , curve A). From the shape of the titration curve it was expected that an indicator that changes colour at about pH 4.5 in ethanolic media would be suitable, Rromocresol green and alizarin S were tried, and both showed colour changes coincident with the point of maximum d(pH)/dv. 4.0 I I 1 1 I 1 Volume of hydrochloric acid added, m\ Fig. 3. Titration of calcium 0 I 2 3 4 5 6 7 8 9 10 I1 12 Volume of hydrochloric acid added, ml 4 5 6 7 8 9 10 1 Fig.2. Titration of calcium oxide extracted with ethylene glycol: curve A, extract with ratio of ethanol to ethylene glycol ---:J. -_-I--_ -I-= ---:IL -L _.__ 1 . _--I-_ _ L r n . 1 - 1 1 - 1 1 A . .- - 1 7 - 2 1 ~ - 1 1 - 1 - 01 z to I; curve D, excract witn ratio 01 etnanoi to etnyiene oxiue exrraccea wizn pnenoi ; excract glycol of 20 to 1; curve C, extract with ratio of ethanol to with ratio of ethanol to phenol of ethylene glycol of 2 to 1 containing 90 per cent. of water. 1 to 1 by volume. The arrow corresponds to the point a t which The arrows correspond to the points at which do are maxima dv dv do is a maximum Co.mpZexometric titrations-Complexometric titrations could be carried out without any difficulty. Solutions of freshly ignited lime were prepared in the same manner as before and aliquots were titrated separately against a solution of disodium ethylenediaminetetra- acetate and ethanolic hydrochloric acid.The results of alkalimetric and complexometric determinations are shown in Table 111. E$ect of dilution with ethanol, water and ethanol - water mixtures-Aliquots of the prepared solution were diluted with large volumes of ethanol and water, and titration curves were plotted. I t can be seen that, although the position of the plot is different on curve B, the maximum value of d(pH)/dv, occurs at the same point as on curve A. Also, although the end-point can be correctly deter- mined by potentiometric titration, it becomes less distinct if alizarin S or bromocresol green is used as the indicator when the ratio of ethanol to ethylene glycol in the titration medium exceeds 4 to 1.The end-point can be best judged when the ratio of ethanol to ethylene glycol is between 2 to 1 and 4 to 1, These curves are shown, in Fig. 2 (curves B and C, respectively).166 VEKMA, BHUCHAR, THERATTIL AND SHARMA: DETERMINATION OF [VOl. 83 TABLE 111 IhTIIACTION OF CALCIUM OXIDE FROM FRESHLY IGNITED LIME WITH ETHYLENE GLYCOL Weight of calcium oxide taken for cxtraction, g 0.2516 0.2546 0.1382 0.2820 7- ___- Final volume potentio- solution, titration, 500 99.2 500 95.1 250 500* - of metric ml % - Calcium oxide found by- _-~-.---.--.---- titration with titration with titration with mixed alizarin S - alizarin S as Ln-omocresol green bromocresol green indicator, as indicator, as indicator, O / % % / O 99.2 98.3 - 95.1 04.5 I 100.0 99.5 - - - 100.0 complexo- metric titration, % - 98.9 100.4 * Industrial methylated spirit was used for the dilution.When the solution is diluted with a large a-mount of water, the shape of the curve is entirely different and the end-point is uncertain (see Fig. 2, curve C). In another experiment, an aliquot of the prepared solution was diluted with ethanol containing different proportions of water so that the ratio of ethanol to ethylene glycol did not exceed 4 to 1, and the end-point wits determined with alizarin S as the indicator. The results of these experiments are shown in Table IV. TABLE IV EFFECT OF VARIOUS AMOUNTS OF WATER \%'eight of Amount calcium oxide of water taken for present at the titration, end-poin t, F: % Nil Nil 12 20 30 Nil 12 r I I 0.01 2.5 { I I Volume of ethanolic hydrochloric acid ml Potentionictric 11.0 11.0 11.0 10.93 Volumetric with alizarin S as indicator 10.90 } indicator { 30.85 Method of deterniination required, r J Volumetric with bromocresol green as It can be seen from the results that visual titrations can be satisfactorily carried out The results when the water content of the titration medium does not exceed 10 per cent.when ethanol of greater dilution is used are not accurate. The following conclusions can be drawn- (i) Calcium oxide can be extracted readily if ethylene glycol is used as the extracting agent. (ii) The extract can be diluted with commercial rectified spirit or industrial methylated spirit, the permissible ratio of rectified spirit to ethylene glycol being between 2 to 1 and 4 to I, and this solution can be titrated with ethanolic hydrochloric acid potentiometrically, or visually with bromocresol green or alizarin S as the indicator. (iiz) The determination of calcium oxide in these solutions can be carried out complexo- metrically without any difficulty.(i'o) If the solution is diluted excessively with ethanol, the end-point of the titration can be found potentiometrically, but not volumetrically. This difficulty can be obviated by diluting with a mixture of ethanol and ethylene glycol so that thc ratio of 4 to 1 is not exceeded. (v) If the solution is diluted excessively with water, calcium oxide tends to be pre- cipitated, and no proper titrations can be carried out potentiometrically or visually.(ti) Only a limited amount of water can be tolerated in the ethanol used for preparing the solution; 90 per cent. ethanol appears to be suitable. EXPERIMENTS WITH PHENOL AS THE EXTRACTING AGENT- Preparation of solution of calcium oxide in j)henol-A known weight of freshly ignited lime was transferred t o a conical flask containing about 10 ml of a 1 + 1 mixture of phenol andMarch, 19581 FREE LIME IN LIME AND SILICATE PRODUCTS. PART I 167 ethanol. A further 40-ml portion of this mixture was added and the whole was digested 011 a water bath for 30 to 60 minutes. When cool, the solution was diluted with ethanol to a known volume. Pofentiometric and volumetric titrations-An aliquot of the prepared solution was titrated with ethanolic hydrochloric acid and the titration curve was plotted.A very smooth and regular curve was obtained (see Fig. 3) with a maximum value of d(pH)/dv at pH 5.5 to 5.6. This point corresponds to the pH at which methyl red changes colour. Methyl orange has been used as the indicator for titrations of phenol-extracted calcium oxide, but the results were rather high; in one test the recovery was about 116 per cent. The colour change was not sharp and was, on the whole, unsatisfactory. Table V gives typical results of the titration of phenol-extracted calcium oxide with ethanolic hydrochloric acid. TABLE V EXTRACTION OF CALCIUM OXIDE FROM FRESHLY IGNITED LIME WITH PHENOL The final volume of each solution was 250nd Calcium oxide found by- Weight of calcium oxide taken for extraction , g 0.0588 0.0922 0.1 190 0.1214 r 3 titration with mixed potentiometric methyl red - methylene complexometric titration, blue as indicator, titration, 98.0 98.0 99.4 100.5 100.5 100.8 98.5 98.8 99.1 - 98.9 98-1 % % % From the titration data it is clear that accurate values of calcium oxide content can be found by using a mixed methyl red-methylene blue indicator.The end-point with this indicator is coincident with the end-point found by potentiometric titration. Complexometric titration-Complexometric titrations were carried out in the manner previously described, and the results have been included in Table V. It was observed that the complexometric and alkalimetric titrations could be easily carried out and did not present any unusual difficulty. I t was also noticed that, in the phenolic solution, a much larger amount of water could be tolerated than in the glycerol and ethylene glycol solutions.COMPARATIVE TESTS ON THE SPEED AND PRECISION OF DIFFERENT METHODS OF EXTRACTING Comparative tests were made on the speed and performance of the different extracting solutions and the titration procedures, all the tests from the initial extractions to the final titrations being carried out on the same batch of freshly ignited lime. The results are shown in Table VI. CALCIUM OXIDE- COMPAKISON OF RESULTS WITH DIFFERENT EXTRACTING SOLVENTS The final volume of each solution was 250ml Calcium oxide found by- Weight of calcium oxide EX- taken tracting for ex- solvent traction, g Glycerol . . 0-1202 Ethylene glycol . . 0-1382 Phenol . .0.1160 r titration Ratio of titration with ethanol with bromo- Time to ex- potentio- alizarin S cresoI of tracting metric as green as :ligestion, solvent titration, indicator, indicator, 6 t o 8 5 t o 2 98-7 98-56 96.79 hours % % “/o 1 t o 2 4 t o 1 - 100.02 99-50 * t o 1 4 t o 1 - - - 1 titration with mixed methyl red - methylene complexo- blue as metric indicator, titration, Yo % - 97.92 - 98.74 98-34 97.82168 VERMA, BHUCHAR, THEB’ATTIL AND SHARMA [Vol. 83 CONCLUSIONS From the preliminary experiments, it was noted that the course of titration of the extracted calcium oxide with an acid in an ethanolic medium was entirely different from that in an aqueous medium. Since not more than a limited amount of water could be tolerated in the ethylene glycol - ethanol or glycerol - ethanol titration media containing the extracted calcium oxide, an ethanolic solution of acid was used in preference to an aqueous solution.This eliminated the possibility of exceeding the 10 per cent. limit of water in these media, above which the end-point could not be satisfactorily detected. The choice of the acid depended upon the fact that, whereas hydrochloric acid gave a clean potentiometric titration, sulphuric acid caused precipitation of calcium sulphate, which impaired the potentiometric titration. It was further noted that the concentration of the ethanolic hydrochloric acid remained fairly constant. In view of these considerations, ethanolic hydrochloric acid has been used in the proposed procedure. Another point that can be emphasised is that the solvents used for extraction, vix,, glycerol, ethylene glycol, phenol and ethanol, slnould be as free from water as possible.If filtration becomes necessary, as when silicate products are being tested, washing should also be carried out with dry ethanol. Ethanol containing not more than 10 per cent. of water, or industrial methylated spirit, can, however, be used for the final dilution of the solutions without any effect on the results. We have used ethanolic hydrochloric acid for the alkalimetric titrations. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 39. 20. 21. 22. REFERENCES Bakewell, B., and Bessey, G. E., D.S.I.R. Building Research S$ecial Report No. 17, London, 1931. Breslau, T. H., “Diplomarbeit Radecker,” 1925; Brandenburg, Chem.Ztg., 1909, 99, 880s; see also Diepschlag, E., and Matting, A., Centralblatt der Huttens und Walzwerke, 1927, 31, 363, 377 and 394. Sestini, Q., and Santarelli, L., Ann. Chim. ABPl., 1936, 26, 202 and 533. Bessey, G. E., “Symposium on the Chemistry Iof Cements,” Ingeniorsvetenskpsakademien, Stock- holm, 1938, pp. 485 to 488. Emley, W. E., Trans. Amer. Ceram. Soc., 1915, 17, 720. Lerch, W., and Bogue, R. H., I n d . Eng. Chenz., Anal. Ed., 1930, 2, 296; see also Scott, W. W., and Furman, N. A., “Standard Methods of Chemical Analysis,” D. Van Nostrand Co. Inc., New York, 1948, Volume 11, p. 1618. Bessey, G. E., J . SOC. Chem. Ind., 1933,52, 219; D.S.I.R. Building Research Technical Paper No. 9, 1930. Rathke, H., TonindustrZtg., 1928, 52, 1318. Konarzewski, J., and Lukaszweics, W., Prezelnysl Chena., 1932, 16, 62. Wittekindt, W., TonindustrZtg., 1935, 59, 139. Schlapfer, P., and Bukowski, R. , “Eidgenossisrhe Material plpifungsanstalt, ara der E. T . H . Zuvich,” Bukowski, R., TonindustrZtg, 1935, 59, 616; Chem. Abstr., 1935, 29, 6723. Forsen, L., Rodt, V., Zement, 1935, 24, 94; 1936, 25, 161. MacPherson, D. R., and Forbrick, L. R., I n d Eng. Chetn., Anal. Ed., 1937, 9, 451. Associated Cement Company of India Ltd., Bombay, India, personal communication. Franke, B., 2. anorg. Chem., 1941, 247, 180. Bogue, R. H., “The Chemistry of Portland Cement,” Reinhold Publishing Corporation, New York, Assarsson, G. O., and Bokstrom, J. M., Anal. Chem., 1953, 25, 1844. Kolthoff, I. M., and Rosenblum, C., “Acid Bast: Indicators,” The Macmillan Company, New York, Palit, S. R., Das, M. N., and Somayajulu, G. R., “Non-aqueous Titrations,” Indian Association Pressler, E. E., Brunauer, S., and Kantro, D. L., Anal. Chem., 1956, 28, 896. Report No. 63, Zurich, 1933; Chem. Abstr., 1935, 29, 5621. Proceedings of Second Congress on 1,arge Dams,” Washington, 1936, Question 111. 1947, pp. 71 to 75. 1937, p. 206. for Cultivation of Science, Calcutta, India, 1954, p. 111. First received September loth, 1956 Amended, September 2nd, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300160

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

March, 19581 COLSON 169 The Ebullioscopic Micro-determination of Molecular Weight: An Improved Micro Form of the Menzies = Wright Ebulliometer BY A. F. COLSON (Imperial Chemical Industries Ltd., Research Department, Alkali Division, Northwich, Cheshire) The construction and operation of an improved apparatus for the micro- determination of molecular weight by the Menzies - Wright method is des- cribed, and some of the results of the determinations of molecular weights ranging from about 200 to 1000, with benzene, carbon tetrachloride and ethanol as solvents, are given. The volume of solvent used is about 3ml and the weight of sample varies from about 3 to 16 mg, according to the molecular weight of the substance. Expressed as coefficients of variation, the accuracy and precision of the method for molecular weights in the region of 200 are 1.8 and 1.3 per cent., respectively, with benzene as the solvent.The corresponding values with carbon tetrachloride as the solvent are 2.3 and 2.1 per cent. The ebulliometer is unsuitable for use with solvents of higher boiling- point, e.g., dioxan and toluene, because they give unstable readings on the differential thermometer. IN an earlier publication,l a modified Menzies - Wright2 ebulliometer incorporating an internal electric heater and a water-filled differential thermometer was described. This apparatus is suitable for the determination of molecular weights up to about 1000, by using 6ml of solvent and about 50 mg of sample. If about 20 mg of a suitable substance, e.g., anthracene, are first added to the solvent in the ebulliometer, the sample weight can be reduced to about 10mg for the determination of molecular weights in the region of 200, but recovery of the sample for other work is then difficult.So that various determinations, in addition to molecular weight, could be carried out when the amount of sample available was small, e.g., 20 to 30mg, we required an ebulliometer capable of giving results accurate to about 2 per cent. on about 10 mg of sample, which could be recovered by simple evaporation of the solvent. A smaller version of the ebulliometer just described was constructed for use with about 3 ml of solvent, but this apparatus failed to give stable thermometer readings. The results were similar with an ebulliometer designed for use with an external electric heater.This apparatus was similar to the ebulliometer described by Smith and Milne~-,~ in which smooth boiling of the solvent is promoted by a tungsten wire sealed through the base of the boiler. In place of the small gas flame used by Smith and Milner, a tubular electric heater, which formed a close fit on the protruding tungsten wire, was used to boil the solvent. This heater was later replaced by one designed to heat the tungsten wire and also a portion of the base of the boiler, but with neither apparatus was satisfactory stability of the thermometer readings obtained. Finally, experiments were carried out with an ebulliometer heated by the carefully controlled flame of a microburner, and the results were satisfactory when benzene, carbon tetrachloride or ethanol was used as the solvent.The failure of all attempts to use solvents of higher boiling-point, such as toluene or dioxan, suggests that ebulliometers having a small capacity, of the type described in this paper and in an earlier publication,l cannot be operated with solvents that boil above about 80°C. EXPERIMENTAL APPARATV S- The assembled apparatus is shown in Fig. 1 and the various components are shown in Figs. 2 to 6. The ebulliometer, supported on a Perspex pillar (see Fig. l), is enclosed in a case of the same material 22 inches high, 13 inches wide and 10 inches from back to front. The top of the case is open and the front is fitted with a sliding panel that gives access to the lower part of the apparatus. A constant-level device, not shown in the diagrams, is con- nected between the tap-water supply and the condenser.A one-mark pipette suitable for the accurate delivery of a fixed volume of solvent is shown in Fig. 8. The mouth of the170 COLSON : THE EBULLIOSCOPIC MICRO-DETE:RMINATION OF MOLECULAR WEIGHT : [VOI. 83 pipette should be closed by a small drying tube to prevent the entry of moisture. A suitable device for the transference of viscous or mobile liquid samples to the ebulliometer is shown in Fig. 7. Solid samples are usually introduced i n the form of pellets prepared in an Orthofee tablet press. ~ 16 I 1;- inch x8- inc Perspex base \ 7 -Terry clip 2 - 23- inch x 16 -inch Perspex support with Perspex strut at rear -Adjustable glass micro burner , pi inch 1-1 Iiinches-4 - 85 mm I B.S.B I B.S. D14. (Nearest B.S. joint Ext. dia. 28 mm B.S. BIO i 19 B.S.W.G. Total length 10 mm A Tungsten wire. c;#T 9 1’16. I X L C j . Y Fig. 1. Assembled micro-ebulliometer Fig, 2. Detail of micro-ebulliometer Much of the preliminary work undertaken to determine the best operating conditions for the gas-heated ebulliometer was concerned with the effect of variation of the volume of solvent, the height of the gas flame and the relative positions of the Cottrell pump and the differential thermometer. In the course of this work, it was noted that when the ebullio- meter was in operation, the temperature inside the Perspex case gradually increased during a period of about 1 hour and then remained nearly constant at about 1” C above the initial temper at ure .In order to determine what effect, if any, this temperature change might have on the “zero” reading of the differential thermometer, the following experiment was carried out. The capillary tube above the tap, A (Fig. l), was filled with mercury and a fixed volume of carbon tetrachloride was transferred to the ebulliometer from the one-mark pipette. Tap- water was then circulated through the condenser at a rate of about 50 ml per minute and theMarch, 19581 AN IMPROVED MICRO FORM OF THE MENZIES - WRIGHT EBULLIOMETER 171 gas flame under the boiler was adjusted to enclose about one-half of the length of the pro- jecting tungsten wire, B (Fig. 1). Twenty minutes later the position of the meniscus in the longer limb of the thermometer, C (Fig. l), was observed with a cathetometer, and further readings were taken at intervals of 2 to 10 minutes over a period of about 100 minutes.The results are shown in Fig. 9, in which the measured changes in height of the “zero” reading of the thermometer have been converted into the corresponding temperature differences by applying the appropriate factor from the table given by Men~ies.~ The curve shows that, although prolonged boiling of the solvent tends to raise the “zero” reading of the thermometer by a significant amount, the change during short periods, e.g., 10 minutes, is too small to cause Fig. 4 Bore 2 mm - Int. dia.4.5 mm Fig. 3 Fig. 5 mi. Bore 2 mm 4 Fig. 6 Ext. dia. Fig. 7 m , Fig. 8 Fig. 3. Cottrell pump Fig. 4. Differential thermometer Fig. 5. Fig. 6. Fig.7. Fig. 8. Detail of differential thermometer Detail of glass microburner Device for adding liquid samples Pipette for dispensing solvents172 COLSON THE EBULLIOSCOPIC MICRO-DETERMINATION OF MOLECULAR WEIGHT: [VOl. 83 serious error in the determination of molecular weight. It is also apparent from the curve that the 10-minute period beginning 20 minutes after lighting the burner is a suitable interval in which to complete a determination, if the sample can be dissolved and the necessary measurements made within this time. This 10-minute period has been adopted in the normal procedure described in this paper, because the complete dissolution of the sample does not usually require more than 5 minutes. Time, minutes Fig. 9. Variation of temperature difference with time For substances requiring between about 5 and 30 minutes for complete dissolution, a modified procedure has been developed.In this modified method, the “zero” reading of the thermometer is determined in the usual manner, and the solvent is then allowed to cool to room temperature. The sample is added to the cold solvent, the thermometer reading is recorded 20 minutes after lighting the microburner, and the determination is completed in the usual manner. The “zero” reading already obtained can be used in further determinations of molecular weight with fresh portions of solvent. I t is self-evident that the results by this modified method cannot be accurate unless the “zero” thermometer reading is reproducible to within narrow limits. The results in the following experiments show that this condition is in fact satisfied. A fixed volume of carbon tetrachloride was transferred to the ebulliometer from the one-mark pipette, water was circulated at it rate of about 50ml per minute through the condenser and the flame of the microburner was adjusted to enclose about one-h.alf of the length of the protruding portion of the tungsten wire. The “zero” reading of the thermo- meter was recorded after 20 minutes, and further measurements were made at intervals of 2 minutes over a period of 10 minutes.The solvent was allowed to cool to room temperature and then a second set of readings was taken as before. This sequence of operations was repeated until four series of measurements had been made. In the four series, the final “zero” readings showed satisfactory agreement with each other.The observed values, expressed as cathetometer readings, ranged from 779.40 to 779-44 mm. In a further experi- ment, the reproducibility of the “zero” readings with successive equal volumes of carbon tetrachloride was determined. The measuremeiits were made at intervals of 2 minutes over a period of 10 minutes, as in the first experiment. Before the introduction of any fresh portion of solvent, a stream of warm dry air was passed through the ebulliometer for about 15 minutes. The final “zero” readings in these four series of measurements were again in close agreement with each other, the ran& of the observed values being from 779.38 to 779.41 mm. METHOD APPARATUS- The various pieces of apparatus, shown in Figs.1 to 8, have already been described, p. 169.March, 19581 AN IMPROVED MICRO FORM OF THE MENZIES - WRIGHT EBULLIOMETER SOLVENTS- Benzene-“Benzene for molecular weight determination,” obtainable from the British Drug Houses Limited, dried and stored over sodium wire. Carbon tetrachtloride-Pure carbon tetrachloride, obtainable from Hopkin and Williams Limited, used without further treatment. Ethanol, absolute-Purified as follows. To 500 ml of absolute ethanol add a mixture of 3 ml of concentrated sulphuric acid and 10 ml of distilled water. Distil off the ethanol and boil it under reflux with 5 g of silver nitrate and 0-5 g of potassium hydroxide for about 1 hour. Distil as before and boil the distillate under reflux with about 250 g of lump quicklime for several hours.Distil off the ethanol and transfer it, in portions of about 8ml each, to a series of glass ampoules previously dried in an oven and cooled with a stream of dry nitrogen passing through them. NORMAL PROCEDURE FOR DETERMINATION OF THE CONSTANT, K- By using the one-mark pipette, transfer a fixed volume of the selected solvent to the ebulliometer. Adjust the rate of flow of water through the condenser to about 50 ml per minute, light the microburner and adjust the flame to enclose half the length of the tungsten wire projecting from the base of the ebulliometer; any convenient type of regulator capable of fine control can be used. After 20 minutes, determine the position of the meniscus in the longer limb of the differential thermometer by using a cathetometer, and repeat the measurement at intervals of 1 minute until two successive readings agree to within about 0.02 mm.Not more than about 5 minutes should be required to complete the determination of the “zero” thermometer reading. Add a weighed pellet (about 8 mg) of pure dry benzil and record the thermometer reading at intervals of 1 minute, as before. Finally, determine the boiling-point of the solvent to the nearest 0.1” C by using a short-stem thermometer sus- pended in the vapour of the boiling solution. It is not advisable to determine the boiling- point before adding the benzil, because this addition cannot afterwards be made without first withdrawing the thermometer and with it a small amount of the solvent. Calculate the value of the constant, K , from the formula- 173 Seal each ampoule and store them in the dark.210.22 x (R - 2) x F W K = J where 210.22 = molecular weight of benzil, 2 = “zero” thermometer reading in mm, R = thermometer reading after addition of benzil in mm, F = factor for conversion of (R - 2) mm to elevation of boiling-point in O C, and W = weight of benzil in mg. To clean the apparatus in readiness for further determinations, drain off the solution through tap A, add about 3ml of pure benzene to the ebulliometer, boil until the vapour reaches the uncooled condenser and drain off as before. Repeat these operations twice more and then pass warm dry air through the apparatus for about 15 minutes. NORMAL PROCEDURE FOR THE MICRO-DETERMINATION OF MOLECULAR WEIGHT- weighed amount of the sample to the boiling solvent. and then determine the boiling-point of the solvent.sample from the formula- K x W Determine the “zero” thermometer reading, as already described, and add an accurately Record the new thermometer reading Calculate the molecular weight of the M = (R - 2) x F where K , R, 2 and F have the same meaning as before, and W is the weight of sample in mg. If the sample is a viscous or mobile liquid, weigh it in the small tube shown in Fig. 7 ; attach the tube to the end of a thin platinum wire, and suspend it from the platinum hook, D (Fig. 2), so that the sample and tube are immersed in the solvent. Withdraw the tube after a few minutes and complete the determination as described. MODIFIED PROCEDURE FOR DETERMINATION OF THE CONSTANT, K- Determine the “zero” reading of the thermometer in the usual way, and then allow the solvent to cool to room temperature. Add an accurately weighed pellet of benzil to the cold174 COLSON : THE EBULLIOSCOPIC MICRO-DETERMINATION OF MOLECULAR WEIGHT : [VOl.83 solvent, light the microburner under the ebullilometer and complete the determination as in the normal procedure. MODIFIED PROCEDURE FOR THE MICRO-DETERM1:YATION O F MOLECULAR WEIGHT- Determine the “zero” reading of the thermometer, allow the solvent to cool to room Light the microburner temperature and add an accurately weighed amount of the sample. TABLE I DETERMINATION OF THE CONSTANT, I<, BY THE NORMAL PROCEDURE benzil, solvent, (R--Z), Weight of Boiling-point of mg “ C F x 1103 mm K W i t h benzene as solvent- 7.955 80.5 4.884 8.10 1.05 7.747 80.5 4-884 8.07 1.04 6.982 80-5 4.884 7.03 1.03 7.200 80.6 4.867 7.20 1-02 6-751 80.4 4.901 6.72 1.03 Mean value of K = 1.03 4.407 76.7 5,578 4.90 1.30 5.171 77.0 5.520 5.80 1.30 5.118 77.0 5.520 5-82 1.32 4.632 76.6 5.598 5-26 1.34 4.809 76.6 5.598 5.4 1 1-32 4.528 76.6 5.598 5.17 1-34 Mean value of K = 1-32 7-698 78.6 6.218 3.37 0.48 9.130 78.6 5.218 4.16 0.50 8.430 78.6 5-218 3-77 0.49 7.864 78.0 5-2 18 3.44 0.48 Mean value of K = 0.49 W i t h carbon tetrachloride as solvent- W i t h ethanol as solvent- TABLE 111 ACCURACY AND PRECISION OF THE RESULTS OF THE MICRO-DETERMINATION OF THE MOLECULAR WEIGHT OF ANTHRACENE BY THE NORMAL PROCEDURE WITH BENZENE AS SOLVENT Molecular Molecular Molecular Weight of weight Weight of weight Weight of weight anthracene, found anthracene, found anthracene, found mg mg mg 8.322 179 6-276 174 3.007 175 7.838 175 6-389 177 2.650 179 8.124 171 6-216 175 2-215 172 7.5 14 178 5-934 176 2.172 177 7.740 176 6-310 177 2.650 178 6-041 177 True molecular weight = 178.22 Coefficient of variation from true value = 1.8% Coefficient of variation from mean value = 1.3% Mean molecular weight found = 176.0 TABLE I:[I ACCURACY AND PRECISION OF THE RESULTS OF THE MICRO-DETERMINATION OF THE MOLECULAR WEIGHT OF ANTHRACENE BY THE NORMAL PROCEDURE WITH CARBON TETRACHLORIDE A S SOLVENT Molecular Molecular Molecular Weight of weight Weight of weight Weight of weight anthracene, found anthracene, found anthracene, found mg mg mi3 9.255 182 6.344 183 3.48 1 176 5.820 180 6.108 178 8.855 181 6.274 184 3.590 173 8.341 183 3.228 176 True molecular weight = 178.22 Coe:mcient of variation from true value = 2.3% Coefficient of variation from mean value = 2.1 yo Mean molecular weight found = 179.6March, 19581 AN IMPROVED MICRO FORM OF THE MENZIES - WRIGHT EBULLIOMETER 175 TABLE IV MICRO-DETERMINATION OF MOLECULAR WEIGHT BY THE NORMAL PROCEDURE WITH BENZENE AS SOLVENT Compound used Sulphonal .... .. .. Cholesterol .. .. f . P@-Bis-(3 : 5-dibromo-kmethoxy- 6 : 6 : 7 : 5‘ : 6’ : 7’-Hexa-acetoxy- 3 : 3 : 3’ : 3’-tetramethylbis- pheny1)propane .. .. 1 : 1’-spirohydrindine . . .. 5 : 6 : 7 : 5’ : 6’ : 7’-Hexabenzoxy- 3 : 3 : 3’ : 3’-tetramethylbis- 1 : 1’-sflirohydrindine . . .. Weight taken, mg 7.892 7.730 7.726 7.237 8.153 7-835 9.32 1 8.760 9.724 7.103 8.360 8-653 7.922 7.377 7.210 7.366 14.479 15.530 16.222 Molecular weight found 229 229 234 396 399 402 539 559 565 607 615 634 1052 1059 1035 1035 1021 1042 1031 Molecular weight calculated 228.32 228.32 228.32 386.64 386-64 386.64 571-96 571.96 571-96 624.6 624.6 624-6 997 997 897 997 997 997 997 Error, * + 0.5 + 0.5 + 2-5 + 2.5 + 3-0 + 4.0 % - 6.0 - 2.5 - 1.0 - 3.0 - 1.5 + 1.5 + 5.5 + 6.0 + 4.0 + 4.0 + 2.5 + 4.5 + 3.5 * The figures in this column are given t o the nearest 0.5 per cent.TABLE V MICRO-DETERMINATION OF MOLECULAR WEIGHT BY THE NORMAL PROCEDURE WITH CARBON TETRACHLORIDE AS SOLVENT Compound used Benzophenone . . Ethyl phthalate . . Sulphonal . . * . Octadecane . . Cholesterol I . Hexatriacontane . . .. .. .. .. .. .. pP-Bis-(3 : 5-dibromo-4-methoxy- pheny1)propane .. .. Weight taken, mg 7.277 7.058 7.367 8.236 7-263 7.224 7.117 8.884 8.342 8.726 0.707 0.815 1.070 0.898 6-464 7-632 8-215 8.600 8.092 8.292 9.148 9.332 9.201 8.584 8.676 9-679 Molecular weight found 186 187 226 233 226 224 227 229 228 231 226 235 230 226 250 262 251 394 386 394 502 516 557 558 565 487 Molecular weight calculated 182.21 182-2 1 222.23 222-23 222.23 222.23 222.23 228.32 228.32 228.32 228.32 228.32 228.32 228.32 254.49 254-49 254.49 386-64 386.64 386.64 506.95 506.95 506.95 57 1-96 571.96 57 1.96 Error, * Yo + 2.0 + 2.1 + 1-5 + 5.0 + 1.5 + 1.0 + 2.0 + 0.5 + 0.0 + 1.0 - 1.0 + 3.0 + 1.0 - 1.0 - 2.0 + 3.0 - 1.5 + 2.0 0.0 + 2.0 - 4.0 - 1.0 + 2-0 - 2.5 - 2.5 - 1.0 The figures in this column are given t o the nearest 0.5 per cent.176 COLSON [Vol.83 and complete the determination as in the norrnal procedure. After the ebulliometer has been cleaned and dried, further determinations of molecular weight can be carried out with fresh portions of solvent without repeating the determination of the “zero” thermometer reading. RESULTS- The results by the normal procedure for the determination of the constant, K , with benzil being used as the solute and with benzene, carbon tetrachloride and ethanol as solvents, are given in Table I. The precision and accuracy of the normal procedure for the determina- tion of molecular weights in the region of 200 is shown by the results given in Tables I1 and I11 for anthracene, with benzene and carbon tetrachloride as solvents. Further results for the determination of molecular weights ranging from about 200 to 1000, with benzene, carbon tetrachloride and ethanol as solvents, are given in Tables IV, V and VI.TABLE VI MICRO-DETERMINATION OF MOLECULAR WEIGHT BY THE NORMAL PROCEDURE WITH ETHANOL AS SOLVENT Molecular Molecular Weight weight weight taken, found calculated Error, * Compound used mg % (- 4.373 180 179.21 + 0.5 4.101 171 179.21 - 4.5 ’ ’ { 4.216 174 179.21 - 3.0 Phenacetin .. . . 1 4.347 174 179.21 - 3.0 7.695 219 228.32 - 4.0 7-956 226 228.32 - 1.0 7-140 234 228.32 + 2.5 8.210 222 228.32 - 3.0 8.914 538 571.96 - 6.0 9-822 545 57 1-96 - 4.5 9.318 580 57 1.96 + 1.5 8.964 552 57 1.96 - 3.5 c Sulphonal . . .. .. .. p p-Bis-(3 : 5-dibromo-kmethoxy- pheny1)propane . . . . .. 8.125 623 624.60 - 0.5 8.547 586 624.60 - 6.0 8-528 598 624.60 - 4.5 8.262 598 624.60 - 4.5 5 : 6 : 7 : 5’ : 6‘ : ‘I’-Hexa-acetoxy- 3 : 3 : 3’ : 3’-tetramethylbis- 1 : 1’-s$irohydrindine . . .. * The figures in this column are given to the nearest 0.5 per cent. The constant, K , was determined by the modified procedure with carbon tetrachloride, b.p. 76.8” C, as the solvent, and the value of F x lo3 was 5.558; the results were as follows- Weight of benzil, mg . . . . 4.506 5.308 4.869 4.996 (R - Z), mm .. .. . . 4-91 5.76 5-28 5.48 K .. .. .. .. . . 1.27 1.27 1.27 1.28 These results give a mean value for K of 1.27. Some determinations of the molecular weight of cholesterol, molecular weight 386.64, were caxried out by the modified procedure with carbon tetrachloride as the solvent ; the results were as follows- Weight of cholesterol taken, mg 8.504, 8.167 8.068 7.914 Molecular weight found . . .. 383 392 391 396 Error, yo . . .. .. . . - 1.0 + 1.5 + 1.0 +2*5 REFERENCES 1. 2. 3. 4. Pregl, F., “Quantitative Organic Micro-analysis,” Fourth Edition, J. and A. Churchill Ltd., 6. Received May 27th, 1967 Colson, A. F., Analyst, 1955, 80, 690. Menzies, A. W. C., and Wright, S. L., J. Amer. Chem. Soc., 1921,43, 2314. Smith, J. H. C., and Milner, H. W., Mikrochemi8, 1931, 9, 117. London, 1945, p. 194. Menzies, A. W. C., J. Amer. Chem. Soc., 1921,421, 2311.

ISSN:0003-2654    

DOI:10.1039/AN9588300169

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

March, 19581 NOTES 177 Notes SPECTROPHOTOMETRIC DETERMINATION OF URANIUM I N ORGANIC-SOLVENT SOLUTIONS WHEN uranium, either from natural sources or irradiated material, is extracted by a liquid - liquid extraction method, the need arises for an easy and rapid method of determining it in the organic phase. Usually, uranium is extracted from the organic phase by one of the numerous known methods. As small volumes of solution are generally used, the possibility of error is great, since the complete separation of phases is seldom possible in practice. Another method consists in the decomposition of the organic phase with a mixture of sulphuric and perchloric acids before the determination, but this is time-consuming and is seldom convenient. Very few workers have dealt with the direct determination of uranium in organic solutions.Fisher and Thomasonl have proposed a polarographic method for the direct determination of uranium. This method is limited to uranium present in a solution of tri-n-butyl phosphate diluted with isopropyl ether, the concentration of tri-n-butyl phosphate being between 5 and 30 per cent. The general application of this method is prevented, since the concentrations of tri- n-butyl phosphate used often exceed these limits and also many other substances are used as diluents, the most frequent being different varieties of kerosine. Among the proposed spectrophotometric methods, the acetone - ammonium thiocyanate method2 is used for the direct determination of uranium in organic solutions. The possibilities of adapting this method are limited.Nietzel and De Sesa3 have applied it to solutions of uranium in ethyl acetate, Sinyakova and Novikov4 to solutions of uranium in ethyl methyl ketone and Kimball and Rein6 to solutions of uranium in a mixture of tri-n-butyl phosphate and carbon tetrachloride. Kimball and Rein have modified the original method by substituting ethanol for acetone, so as to attain a uniform single phase of the organic solution and the aqueous solution of the reagents, This could not be done with tri-n-butyl phosphate diluted with kerosine. Nietzel and De Sesas have recently, in a similar manner, substituted butyl Cellosolve for acetone, so as to avoid many of the disadvantages of the acetone - ammonium thiocyanate method. They applied the method to the determination of uranium in isobutyl methyl ketone solutions.Francois' produced a coloured complex by adding a solution of uranium in tri-n-butyl phosphate and isooctane to a solution of dibenzoylmethane in acetone and water. EXPERIMENTAL We have attempted to develop a method that is rapid and sensitive under different conditions. The idea was to combine two of the steps of the analytical procedure that are usually performed separately, viz., the extraction of the uranium from the organic phase with a complexing agent and the development of the colour in the aqueous phase with the same complexing agent to permit the absorptiometric determination. As this was achieved, there was no necessity for finding a reagent that would form an homogeneous phase with many different organic solvents. We then investigated the different spectrophotometric methods available for the determination of uranium in aqueous solutions and arrived at the conclusion that the ammonium thioglycolate method of Davenport and Thomasons could be used directly without loss of any of its advantages.Owing to the very slight solubility of inorganic substances in organic solvents, the interference caused by some cations and anions in aqueous solutions need scarcely be considered. METHOD ~ A G E N T S - A mmonium thioglycolate solution-Prepared by diluting 10 ml of thioglycolic acid with about 60ml of water and then neutralising the solution with diluted ammonia solution and adjusting the volume to 100 ml by adding more water. Ammonia solution, diluted (1 + 1)-Prepared by diluting analytical-reagent grade ammonia solution, sp.gr.0.880, with water. PROCEDURE- An aliquot of the organic solution is transferred to a 25-ml calibrated flask by using a micro- pipette, which is then washed with the pure organic solvent, the washings being added to the flask, and 2 ml of diluted ammonia solution (1 + 1) and 2 ml of ammonium thioglycolate178 NOTIES [Vol. 83 solution are added. The flask is then shaken well and the aqueous layer is made up to the mark by adding water. On standing for a few minutes, the two phases separate completely, the organic layer is removed either by using a pipette or by decantation. The aqueous layer is examined spectrophotonietrically in the usual way. If the separation of the phases requires a longer time, as when solvents having a specific gravity nearly the same as that of water are used, e.g., tri-n-butyl phosphate, the formation of the complex can be achieved in a centrifuge tube by adding the reagents quantitatively so that the final volume of the aqueous phase will be exactly 25 ml.Then, after spinning in a centrifuge, the clear aqueous solution is used for the spectropliotornetric determination. A similar procedure can be adopted for solvents heavier than water, e.g., tri-n-butyl phosphate diluted with carbon tetrachloride. Because of the very low acidity of the organic phase, e.g., tri-n-butyl phosphate in kerosine,B as well as the acidity in general, it is not necessary to neutralise the organic phase before the addition of the reagents. RESULTS The proposed method was applied to solutions of uranium in tri-n-butyl phosphate, isobutyl The organic methyl ketone, ethyl acetate and diethyl ether.phases were analysed after re-extraction as well as directly by the proposed method. 'The results are shown in Table I. TABLE 1 DETERMINATION OF URANIUM IN VARIOUS SOLVENTS Uranium found Uranium found Solvent after re-extra.ction, by proposed method, Difference, mg per nil mg per ml YO Tri-n-butyl phosphate . . . . 1.632 1.620 - 0.7 isoButyl methyl ketone . . . . 13.100 13.480 + 2.9 Ethyl acetate . . . . . . 1.407 1.390 - 1.2 Since the use of tri-n-butyl phosphate was of special interest, it was examined in greater detail and in the presence of different diluents. The diluents used were kerosine, diethyl ether, dibutyl ether, isopropyl ether, hexane and caribon tetrachloride.The solutions of uranium were prepared by extracting a solution of uranium nitrate with mixtures of tri-n-butyl phosphate and the diluent, each mixture containing 30 per cent. of tri-n-butyl phosphate. The solutions were analysed after re-extraction as well as directly by the proposed method; the results are shown in Table 11. Diethyl ether . . . . .. 9.100 9-250 + 1.6 TABLE I1 DETERMINATION OF URANIUM IN TRI-~Z-BUTYL PHOSPHATE DILUTED WITH VARIOUS SOLVENTS Uranium found Uranium found Diluent for after re-extraction, by proposed method, Difference, mg per ml YO tri-n-butyl phosphate mg per ml Kerosine . . . . . . . . 2-370 2-400 + 1.3 Diethyl ether . . . . .. 4.865 4.880 + 0.3 isoPropy1 ether . ... .. 5.965 5.840 - 2.1 Dibutyl ether . . .. . . 7.155 6.980 - 2.4 Hexane . . . . . . . . 18.950 18.900 - 0.3 Carbon tetrachloride . . .. 12.130 11.890 - 2.0 All the results given in Tables I and I1 are an average of two determinations. In order to determine the influence of impurities that can be present in the solvent phase, we analysed the solvent phases obtained by the extraction of solutions of uranium chloride in presence of hydrochloric acid and other chlorides. Extraction was carried out with 30 per cent. of tri-n-butyl phosphate in di-n-butyl ether. The !solvent phase was analysed after decomposition with sulphuric and perchloric acids as well as directly by the proposed method; the results are shown in Table 111. The differences are within the limits of error for the proposed method, thereby indicating that small amounts of the inorganic salts that can be present in the solvent phase do not affect the accuracy of determination.March, 19581 NOTES TABLE I11 DETERMINATION OF URANIUM IN PRESENCE OF VARIOUS CHLORIDES Each solution contained M hydrochloric acid Uranium found Concentration with sulphuric and by proposed after decomposition Uranium found Chloride present of chloride, perchloric acids, * method, A’M mg per ml mg per ml - - 0.95 0-94 Sodium .. . . saturated 18.00 18-04 Potassium . . saturated 6-65 6.74 Ammonium . _ 5 10.05 9.98 Lithium . . .. 5 11.50 11.68 Calcium . . .. 2.5 20.30 20.84 Magnesium . . 2.5 22.40 22-66 Aluminium .. 1.67 23-33 23-44 * We thank Dr. I. Gal, who kindly provided these results.179 Difference, % - 1.0 + 0.2 -t 1.3 - 0.7 + 1-6 + 2.7 + 1.2 + 0-5 The concentration range investigated was from 0.384 to 1-460 mg of uranium per 25 ml, and in this range Beer’s law is obeyed. The colour of the complex formed was stable for 1 hour. Blank values were determined after shaking a solution of the reagents with the same volume of the solvent as was used for the determinations as well as on a solution of the reagents alone. No difference between the blank values was observed when 0-1 to 1 ml of the solvent was shaken with 25 ml of the reagent solution. This indicates that traces of solvent that go into the aqueous phase do not affect the characteristics of the method. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. Fisher, D. J., and Thomason, P. F., Anal. Chem., 1956, 28, 1285.Crouthamel, C., and Johnson, C., Ibid., 1952, 24, 1780. Nietzel, 0. A., and De Sesa, M. A., “Peaceful Uses of Atomic Energy,” Proceedings of the Inter- national Conference in Geneva, August, 1955, United Nations, 1956, Volume VIII, p. 320. Sinyakova, S. J., and Novikov, G. J., quoted by Paley, P. N., op. cit., p. 225. Kimball, R. B., and Rein, J. E., U.S. Atomic Energy Commission Report, IDO-14380, 1956. Nietzel, 0. A., and De Sesa, M. A., Anal. Chem., 1957, 29, 756. Francois, C. A., U.S. Atomic Energy Commission Report, DOW-150, 1956. Davenport, W. H., jun., and Thomason, P. F., Anal. Chem., 1949, 21, 1093. Alcock, K., Grimley, S. S., Healy, T. V., Kennedy, J., and McKay, H. A. C., Trans. Faraday Soc., 1956, 52, 39. LABORATORY OF PHYSICAL CHEMISTRY 2.I. DIZDAR INSTITUTE OF NUCLEAR SCIENCES “BORIS KIDRICH” MISS I. D. OBRENOVIC BELGRADE, YUGOSLAVIA Received June 25th, 195’7 THE DETERMINATION OF CALCIUM I N MILK AND WHEY THE use of metastannic acid in the preparation of a phosphate-free serum permits a rapid deter- mination of calcium in milk and whey by the ethylenediaminetetra-acetic acid (EDTA) titration method. The sera are free from colloidal matter and lend themselves admirably to the EDTA procedure. With an increase in the appreciation of the importance of calcium in milk and other biological material] the need has arisen for more rapid and convenient methods for its determination, particu- larly when many samples have to be examined. Several alternatives to the standard oxalate method have been advanced. In the turbidimetric method of Marier and Boulet,l the solution, containing 40 to 100 pg of calcium, is treated with a phosphate - citrate reagent and then with an alkaline gelatin - oleate solution.The resulting turbidity is determined by measuring the optical density at 420 mp. Chloranilic acid has been used as a precipitant for calcium in plant- ash2 and in soil extractsI3 and the method has been applied to milk-ash by Breyer and M~Phillips.~ In their method, an excess of reagent is added to a hydrochloric acid extract of milk-ash and the excess is determined by measuring the polarographic diffusion current. Two further develop- ments have been the flame-photometric and EDTA titration methods. In both, however, difficulties have been experienced through the presence of orthophosphates.For the flarne- photometric determination of calcium, phosphates have been removed by passing the solution through a cation-exchange resin, from which the cations were subsequently released by elution with acid.6ys Several workers have dealt with the problem of phosphate interference in the EDTA titration method, some finding it sufficiently troublesome to necessitate special procedures,180 NOTES [Vol. 83 whereas others considered it to be very slight. When the solutions were sufficiently dilute to bring the solubility product below that of calcium phospha.te, Kenny and Toverud' found no interference. It would seem, therefore, that for each level of calcium there should be a concentration of ortho- phosphate beyond which interference with the EDTA titration can be expected.However, the work of Colliers does not entirely substantiate this hypothesis. There was no clear-cut line of demarcation between phosphate concentrations that adversely affected the titration and those without effect, but rather there was a broad band of gradually increasing interference. In this work, the photometric titration apparatus of Fricker9 was used. In view of these uncertainties, which exist in very dilute solutions, other methods of avoiding phosphate interference have been sought. The back-titration procedure has been used by MalkkilO and Stephens,ll in which the excess of EDTA added to the solution is titrated with a standard calcium solution. At the end-point, addition of 1 drop of calcium solution causes an immediate change in the murexide indicator.As there is a slow formation of calcium phosphate, it is essential that the end-point should be judged by the immediate change in colour of the murexide indicator. It is pointed out by Malkki that, when the back-titration procedure is used, it is no longer necessary to use carbonate-free sodium hydroxide to adjust the pH of the solution to about 12. Masonla con- sidered that this procedure was not altogether sa,tisfactory for routine analysis. His method consists in passing an acid extract of plant-ash through a column of Zeo-Karb 215, eluting with hydrochloric acid, evaporating the effluent to dryness, dissolving the residue in water and titrating with EDTA in the customary manner. JennesslJ extracted phosphates from an acid milk-ash extract by passing it through the anion-exchange resin, Duolite A4.The solution, after adjust- ment of pH, can then be titrated satisfactorily with EDTA. Gehrke, Affsprung and Lee14 used the chloride form of another anion-exchange resin, Amberlite IR-4B, for the same purpose. Collierls extracted phosphates by means of a butanol - chloroform - sodium molybdate mixture. The elimination of phosphates by means of staiinic oxide (or metastannic acid) is a well tried method of qualitative analysis. Although the precise mechanism of the method is not fully understood, there are no doubts about its efficiency. Although a search of the literature failed to discover any quantitative application of this reaction, it seemed reasonable that, under the right conditions, metastannic acid could be used to give a phosphate-free milk serum admirably suitable for the EDTA titration method.EXPERIMENTAL In a number of trials, milk sera were produced tinder a variety of conditions and were titrated with EDTA after adjustment of the pH to about 12. After a correction had been made for the volume of the precipitate, the resultant values for calcium content were compared with those given by the standard oxalate procedure. In the latter, 25 ml of milk or whey were evaporated to dryness, ignited and ashed at 700" C . The ash was dissolved in hydrochloric acid, and the solution was neutralised and then rendered slightly acid by adding acetic acid. From this solution, calcium was precipitated as oxalate and determined by titration with potassium permanganate.I t was soon apparent that, unless a fairly higlh level of acidity was maintained throughout the metastannic acid treatment, there was a danger of incomplete recovery of calcium. In these early trials a suspension of metastannic acid was used. Pure tin was treated with the minimum amount of concentrated nitric acid, and the resulting metastannic acid was washed by decantation with water and finally made into an aqueous suspeiision containing 18 per cent. of stannic oxide. I t was evident that the amount of metastannic acid needed more precise measurement than was possible with the aqueous suspension. It was preferable, therefore, to add measured volumes of potassium metastannate solution to milk containing the requisite amount of nitric acid.In this manner the two critical factors, acidity and amount of metastannic acid present, were brought under more effective control. By trial and error the procedure described below was finally adopted, and, as shown by the results in Table I, the method agrees satisfactori:ly with the standard oxalate procedure, over which it has the advantages of rapidity and convenience. METHOD REAGENTS- Potassium metastannate soZution-In an 800-ml beaker, just cover 16 g of pure granulated tin with distilled water and add concentrated nitric acid in small amounts until no more metallic tin remains (about 65 ml). Wash the metastannic acid by decantation, using about 1700 ml of distilled water; pass the wash liquor through an 113-cm Whatman No. 2 filter-paper.Wash theMarch, 19581 NOTES 181 residue back into the beaker and dissolve the whole by the addition of 5 ml of potassium hydroxide solution (0-71 g of potassium hydroxide per m1) and make the solution up to 250 ml by adding water. The alkalinity of this solution, determined by titration against 0.1 A; hydrochloric acid with phenolphthalein as indicator, was 11.6 ml per 10 ml of metastannate solution. TABLE I DETERMINATION OF CALCIUM OXIDE IN MILK AND WHEY . . Sample f 0.194 0.188 - 0.006 0.164 0.159 - 0.005 0-161 0.163 + 0.002 0-176 0.176 0.0 0.171 0.170 - 0.001 0.171, 0.171 - - 0.230 0.226 - 0.004 0.156 0.155 - 0.00 1 0-143 0.142 - 0.00 1 0.163 0.164 - 0.00 1 0.185, 0.185 0.185 0-0, 0.0 0.224 0.223 - 0.00 1 L Individual milk Mid lactation milk Late lactation milk Individual milk Bulk raw milk .. Separated milk. . Bulk milk . . Individual milk Whey . . .. Potassium hydroxide soZution-Dissolve 56 g of analytical-reagent grade potassium hydroxide pellets in 44 g of distilled water. By titration against 0.1 N hydrochloric acid this solution was found to contain 0-71 g of potassium hydroxide per ml. Nitric acid, diluted (1 + Z)-Dilute 1 volume of concentrated analytical-reagent grade nitric acid with 2 volumes of distilled water. Sodium hydroxide solution, 0-5 N-Prepare by diluting a 50 per cent. w/w carbonate-free sodium hydroxide solution. EDTA solution-Dissolve 4 g of disodium dihydrogen ethylenediaminetetra-acetate in water and make u p to 1 litre. Mztrexide indicator-Grind 0.1 g of murexide in a mortar with 20 g of analytical-reagent grade potassium sulphate.Store the mixture in a stoppered tube and keep it in a dark cupboard when not in use. Standard colour solution-Add 0-15 ml of cresol red indicator solution (0.1 g of cresol red dissolved in 20 ml ethanol and madc up to 100 ml with distilled water) to 60 ml of saturated sodium tetraborate solution. Weigh accurately about 2 g into a beaker and dissolve i t in a minimum amount of dilute hydrochloric acid. Dilute this solution with water and boil to expel carbon dioxide; cool and make up to 1 litre with water. Standardise with potassium hydrogen phthalate. Standard calcium solution-Dry analytical-reagent grade calcium carbonate a t 100" C.182 NOTES [Vol. 83 PROCEDURE- Add about 30 ml of distilled water and then 1 ml of diluted nitric acid (1 + 2).Mix gently, and, with a pipette, add 10 ml of potassium metastannate solution with constant rotation of the flask. Dilute to the mark with distilled water and shake briskly. After a few minutes, pass the solution through an ll-cm Whatman No. 40 filter-paper, rejecting the first few millilitres of filtrate. The filtrate should be perfectly clear, but a slight turbidity has no adverse effect on the titration. With a pipette transfer 50 ml of filtrate to a 150-ml flask, add a small fragment of pH test-paper and iieutralise with 0.5 N sodium hydroxide solution. Add a further 2 ml of the sodium hydroxide solution, and, by means of a small glass tube or a suitably marked narrow spatula, add 0.02 to 0.03 g of murexide indicator.Using a 10-ml burette graduated in 0.02 ml, titrate with EDTA solution until the addition of 1 drop changes the colour frlom pink to the violet shade of the standard. Standardise the EDTA solution by measuring 10ml of standard calcium solution into a flask containing about 20 ml of distilled water, neutralise with 0.5 N sodium hydroxide solution, add a further 2 ml and titrate with EDTA exactly as described. A blank determination is essential and is carried out exactly as indicated, but with distilled water instead of the sample. The correction for the volume of precipitate can be determined by comparing the titre for milk with the titre for an identical sample to which a measured volume of standard calcium solution has been added. Measure 5 ml of milk or whey into a 100-ml calibrated flask. A typical example is as follows- For milk sample- For 5 m1 of milk plus 5 ml of standard calcium solution- Difference equivalent to 2.5 ml of standard calcium solution = 4.88 nil. 5 ml of standard calcium solution titrated directly required 9.51 ml of EDTA solution. 50 ml of filtrate required 7.38 ml of EDTA solution. 50 rnl of filtrate required 12-26 ml of EDTA solution. 9-5 1 2 X 4.88 Correction for volume of precipitate = -____ x 100 = 97.4 per cent. For the samples examined the correction factor varied from 97.3 to 97-7 per cent. REFERENCES 1. 2. 3. 4. 5. 6, 7. 8. 9. 10. 11. 12. 13. 14. 15. Marier, J . R., and Boulet, M. A., J . Agric. Food Chem., 1956, 4, 720. Tyner, E. H., Anal. Chem., 1948, 20, 76. Gammon, N., and Forbes, R. B., Ibid., 1949, 21, 1391. Breyer, B., and McPhillips, J., Analyst, 1953, 78, 666. Sutton, W. J. L., and Almey, E. F., J . Dairy Sci., 1953, 36, 1248. Leyton, L., Analyst, 1954, 79, 497. Kenny, A. D., and Toverud, S. W., Anal. Chem., 1954, 26, 1059. Collier, R. E., Chem. G. Ind., 1955, 74, 587. Fricker, D. J., Ibid., 1955, 74, 426. Malkki, Y., Finnish J . Dairy Sci., 1953, 1. Stephens, R. L., J . Pharm. Pharmacol., 1953, 5, 709. Mason, A. C., Analyst, 1952, 77, 529. Jenness, R., Anal. Chem., 1953, 25, 966. Gehrke, C. W., Affsprung, H. E., and Lee, Y. C., Ibid., 1954, 26, 1944. Collier, R. E., Chemist Analyst, 1954, 43, 41. NOTTINGHAM UNIVERSITY SCHOOL OF AGRICULTURE SUTTON BONINGTON LOUGHBOROUGH E. R. LING Received September 131h, 1957

ISSN:0003-2654    

DOI:10.1039/AN9588300177

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

182 NOTES [Vol. 83 Ministry of Agriculture, Fisheries and Food FOOD STANDARDS COMMITTEE REPORT ON THE ICE C:REAM STANDARD THE Minister of Agriculture, Fisheries and Food, the Minister of Health and the Secretary of State for Scotland have approved for publication a Report of the Food Standards Committee on the standard of composition for ice cream. The report examines the need for revising the Food Standards (Ice Cream) Order, 1953, in the light of present-day conditions and current commercial practice. Copies of the report may be obtained from H.M. Stationery Office, or from any bookseller, price 8d. (plus postage).March, 19581 BOOK REVIEWS 183 REPORT ON FLUORINE : REVISED RECOMMENDATIONS FOR LIMITS FOR FLUORINE CONTENT OF FOODS THE Minister of Agriculture, Fisheries and Food, the Minister of Health and the Secretary of State for Scotland have approved for publication a Revised Report of the Food Standards Committee’s Metallic Contamination Sub-committee on the fluorine content of foods. An earlier report was published in June, 1953. Copies of the Revised Report may be obtained from H.M. Stationery Office, or from any bookseller, price 6d. (plus postage).

ISSN:0003-2654    

DOI:10.1039/AN9588300182

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

March, 19581 BOOK REVIEWS 183 Book Reviews COMPLEXOMETRIC TITRATIONS. By GEROLD SCHWARZENBACH. Translated and revised in Pp. xviii I t is important to realise that this book is not just a translation of Schwarzenbach’s original work (reviewed in AnaZyst, 1957, 82, 294), which was first published in 1955. The opportunity has been taken to bring the subject matter up to date, at any rate up to August, 1956, and to embody in it many corrections and additions to the original text. Literature references now number 305, as compared with 181 in the original work. A certain amount of new material has been included, and many new diagrams have been provided. The text has been re-grouped into seven chapters dealing, respectively, with Polyamines and Complexones, Basic Theory of Complexometric Titrations, Indicators Used in Complexometry, Titration of Mixtures, Solutions used in Complexometric Titrations, Details of Procedures and Concluding Remarks.In reading the book, one has the feeling that, on occasion, the translation has been too literal, e.g., one meets rather unfamiliar phrases such as “. . . the clear liquid standing over the precipitated magnesium hydroxide . . .” p. 71. Then again, though tremendous efforts have been made to effect adequate cross-referencing, this object has not always been achieved, for example, the reader will find adequate cross-references to a type IV reaction in one part of the book, but not in others. The reference to the precipitation of calcium as sulphite on p. 62 should include reference 83 a.s well as 304. On p.50, barium has been omitted from the list of metal cations that can be determined complexometrically by direct titration with EDTA. Some of the diagrams could have been improved by much bolder lettering. Really Dr. Irving has presented us with an up to date translation of Schwarzenbach’s work, which has had such a profound effect on analytical chemistry during the last few years. In doing so, he will, without doubt, enable greater numbers of analytical chemists to take advantage of the many procedures that the introduction of EDTA has provided for the up to date analyst. This book will have a further function; it will help English readers to establish more firmly the stoicheiometry of some of the reactions involved, just as in the 1930’s it was necessary to digest a great deal of information about 8-hydroxyquinoline and to place this substance in its correct place as an analytical reagent.collaboration with the author by HARRY IRVING, M.A., D.Phil., D.Sc., F.R.I.C. + 132. London: Methuen & Co. Ltd. 1957. Price 21s. All these are minor criticisms. J . HASLAM FUSION METHODS IN CHEMICAL MICROSCOPY. By WALTER C. MCCRONE, JUN. Pp. viii + 307. In 1891, Otto Lehmann observed that the microscopical appearance exhibited on crystal- lisation of an organic compound from its own melt is usually characteristic. The results of his researches were published in a book entitled “Die KrystnZZanaZyse,” and he listed a number of New York and London: Interscience Publishers Inc. 1957. Price $6.75; 52s.184 PUBLICATIONS RECEIVED properties that could be determined on crystals from the melt and discussed the value of this technique in studying the phase diagrams of systems of one and two components. Lehmann’s work received little attention until, nearly 60 years later, the three Koflers-Ludwig, Adelheid and Walter-along with Marie Brandstatter in Austria and Walter C.RlcCrone, jun. and his fellow workers in the United States expanded the method. This book describes the technique and gives analytical data for the identification of many organic compounds. Besides a polarising microscope, the essential accessary is a hot stage, including ineans for controlling and measuring temperatures over a wide range. Various designs are now available commercially and are described and illustrated in the earlier pages of this book.Then comes a, discussion of the various optical properties to be observed during heating, after the compound has melted, during cooling and crystallisation, after cooling is complete and, finally, during re- heating. The refractive index of the melt is the most useful character, and this is determined by the use of a standard set of 24 glass powders covering a range from 1.300 to 1.6877. The last hundred pages of this book comprises four analytical tables giving the data for the identification of 1189 organic compounds. Table I lists the cornpounds alphabetically with their melting-points and with an identifying code number. Table 11 lists the compounds by code number in the order of increasing melting-point. Table I11 gives the eutectic melting-points of each compound with several standard compounds. Table IV gives the temperature at which the melt of the compounds has the same refractive index as a c‘hosen glass standard; for a few compounds this temperature coincides with the melting-point, but for most it is a temperature slightly higher than that at which the compound first melts. I t is stated that, by this method, fusible compounds may be rapidly identified on very small samples and that relatively little specialised training is needed to acquire the technique. However this may be, the author is to be congratulated upon a well planned, fully documented and carefully written exposition of an interesting analytical method. The book, which includes a number of remarkably fine coloured plates, has been excellently produced, and it should have a special appeal to those engaged in research work in organic chemistry. N. L. ALLPORT

ISSN:0003-2654    

DOI:10.1039/AN958830183b

出版商:RSC    

年代:1958

数据来源: RSC

摘要:

184 PUBLICATIONS RECEIVED Publications Received ORGANIC SYNTHESES. An Annual Publication of Satisfactory Methods for the Preparation of Organic Chemicals. Volume 37. Editor-in-Chief: JAMES CASOX. Pp. viii -i- 109. New York: John Wiley & Sons Inc.; London: Chapman & Hall Ltd. 1957. Price $4.00; 32s. ENCYCLOPEDIA OF CHEMICAL TECHNOLOGY. First Supplement Volume. Edited by RAYMOND E. KIRK and DONALD F. OTHMER. Pp. xviii + 974. New York: The Interscience Encyclo- pedia Inc. ; London: Interscience Publishers Ltd. 1957. Price $25.00; 176s. THE ANALYTICAL USES OF ETHYLENEDIAMINETETRAACETIC ACID. By FRANK J . WELCHER. Pp. xviii + 366. Princeton, N.J., New ‘Ilork, Toronto and London: D. Van Nostrand Co. Inc. 1958. Price $8.50; 64s. INTRODUCTION TO PROTEIN CHEMISTRY. By SIDNEY JV.Fox and JOSEPH F. FOSTER. Pp. viii + 459. New York: John Wiley & Sons Inc.; London: Chapman Si Hall Ltd. 1957. Price $9.50; 76s. ORGANIC ELECTRGDE PROCESSES. By MILTON J. ALLEN. Pp. xiv + 174. London: Chapman & Hall Ltd. 1958. Price 32s. IBERT MELLAN. Pp. iv + 267. New York: Reinhold Publishing Corporation; London: Chapman & Hall Ltd. 1957. Price $7.00; 56s. Volume I of this Series was published under the title “Handbook of Solvents” (see Pubtications Second New York: John Wiley & Sons Inc.; London: Chapman and Hall CRYSTAL STRUCTURES. Supplement 111. By RALPH W. G. WYCKOFF. Loose-leaf, 441 sheets. Supplement 111 consists of punched leaves for insertion into the binders of Volumes I and I I ; Pp. 239. SOURCE BOOK OF INDUSTRIAL SOLVENTS. Volume 11 HALOGENATED HYDROCARBONS. By Received, Analyst, 1957, 82, 536). INDUSTRIAL CHEMICALS. By W. L. FAITH, DONALD B. KEYES and RONALD L. CLARK. Edition. Pp. x + 844. Ltd. 1957. Price $16.00; 128s. New York and London: Interscience Publishers Ltd. 1958. Price $20.00; 150s. i t comprises additions to chapters I to V I I I . THE DEVELOPMENT OF TITRIMETRIC ANALYSIS ‘TILL 1806. By E. RANCKE MADSEN. Copenhagen: G.E.C. Gads Forlag. 1958. Price Dan. kr. 20.00.

ISSN:0003-2654    

DOI:10.1039/AN9588300184

出版商:RSC    

年代:1958

数据来源: RSC