摘要:

JANUARY, 1975 The Analyst Vol. 100, No. 1186 The Determination of Molybdenum in Plants by an Automated Catalytic Method E. G. Bradfield and Jennifer F. Stickland Long Ashton Research Station, University of Bristol, Bristol Molybdenum is determined by its catalytic effect on the liberation of iodine from the reaction between potassium iodide and hydrogen peroxide. The detection limit is 0.003 p g ml-1 of molybdenum. Interference from iron and tungsten can be overcome by addition of ammonium fluoride, but for the greatest precision and accuracy a preliminary separation of molybdenum as its benzoin a-monoxime complex is recommended. Molybdenum concentra- tions determined by the proposed method in some reference plant materials are presented. The molybdenum concentration in plants may range from 0.05 p.p.m.to greater than 2 p.p.m. in dry matter and its determination in samples of low molybdenum content involves handling solutions containing 0.01 pg ml-l of molybdenum (assuming 2 g of dry matter is digested and diluted to a volume of 10ml). In the past, methods that involve the use of dithiol have most frequently been applied but more recently some attention has been given to the determination of small amounts of molybdenum by measuring its catalytic effect on the liberation of iodine from potassium iodide by hydrogen peroxide. A method based on this reaction was published by Yatsimirskii and Afanas’eval and was subsequently studied by Yatsimirskii and Alek~eeva,~,~ Svehla and Erdey4 and Hadjiioannou.6 Fuges noted that the reaction was time dependent, sensitive to various reagent concentrations and difficult to perform manually.He overcame these difficulties by adapting the reaction for use with an AutoAnalyzer and described a method for the determination of molybdenum in geological and biological samples. In attempting to use this method to determine molybdenum concentrations in a range of plants, considerable interference from other elements, particularly iron, was encountered and high results were obtained. Experiments were therefore carried out to study the effect of elements present in dried plant material on the molybdenum-catalysed potassium iodide - hydrogen peroxide reaction and to devise a rapid and sensitive method for the determination of this element in the range 0-01-0.1 pg ml-1.Experimental Apparatus Auto Analyzer The manifold used is shown in Fig. 1. A Technicon sampler plate and proportioning pump were used in conjunction with a Cecil CE 212 ultraviolet monitor and a Servoscribe RE 511 recorder. Glass tubing must be used to make connections from the double mixing coil to the delay coil, and from the delay coil to the flow cell, as iodine is adsorbed on the surface of plastic tubing and this causes unsteady base lines. Glassware Glassware was thoroughly cleaned by refluxing with a nitric acid - perchloric acid mixture and used only for molybdenum determinations; Jena 50-ml conical flasks were found to give the lowest blank values. 12 Reagents still and then passed through an Elgastat ion-exchange unit. and dilute to 11. BRADFIELD AND STICKLAND : THE DETERMINATION OF MOLYBDENUM Analyst, VoZ.100 Use analytical-reagent grade chemicals. Water should be distilled from an all-glass Potassium iodide solution, 0-5 per cent. m/V. Dissolve 5 g of potassium iodide in water Hydrogen peroxide solution. Dilute 6.5 ml of 10-volume hydrogen peroxide to 1 1. Ammonitvm @ o d e soltvtion, 0.25 per cent. m/V. Dissolve 2.5 g of ammonium fluoride Wash solution. 0.125 N hydrochloric acid. Benzoin a-monoxime solution, 2 per cent. m/V. Dissolve 2 g of benzoin a-monoxirne in Chloroform. Nitric acid - perchloric acid mixtwe. Mix nitric and perchloric acids in the proportions 3 + 1. Standard molybdenum solution, 100 pg ml-l. Dissolve 150 mg of molybdenuni(V1) oxide in 10ml of 0.2 N sodium hydroxide solution, dilute to about 200m1, add 25ml of 0.1 N hydrochloric acid and make up to 11.Prepare a working standard by diluting 1 ml of the concentrated standard to 1OOml with distilled water and from this prepare solutions containing 0.02410 pg ml-l of molybdenum in 0.125 N hydrochloric acid. in water and dilute to 11. 1OOml of absolute ethanol. Store in a refrigerator. Waste @ 2-9 0.125~ HCI @ 0.8 HZOz I Recorder Colorimeter 364 nm, 10 mm @ 0.8 K1 @ 0-8 NH4F @ 2.5 Sample Time @ delay -= coil Double Single mixing mixing I @ 1.2 Air coil coil Fig. 1. Flow diagram for AutoAnalyzer. Preparation of Sample Wash the fresh sample of plant material by immersion in 0.1 per cent. Teepol solution for 15 s with gentle wiping of the surface, rinse rapidly in distilled water and dry at 65 to 70 "C for 24 h.Grind the material to a powder in an agate mortar and then store it. For molybdenum concentrations of 0.05-0.5 p.p.m. weigh 2 g, for concentrations between 0.5-1 p.p.m. weigh 1 g and for concentrations greater than 1 p.p.m. weigh 0.5 g of plant material (previously oven-dried at 105 "C for 1 h) into a 50-ml conical flask. Ash the material overnight in a muffle furnace at 450 "C. To the cooled ash add 2 ml of nitric acid - perchloric acid mixture. Place the digestion reflux funnel (as described by Bradfield7) in the flask and digest on a hot-plate until dense white fumes of perchloric acid are evolved. Remove the reflux funnel and evaporate the contents of the flask to dryness. Prepare blank digestion flasks in a similar manner. Method I Add 10 ml of 0-25 N hydrochloric acid to the residue and boil under reflux for a few minutes.Materials with a high manganese content may yield pink or brown solutions at this stage. If this occurs add very dilute hydrogen peroxide solution (see Reagents) drop- wise until the colour disappears. Allow the digest to cool and dilute to 20 ml. Determine the molybdenum content of the samples by the catalytic method on the AutoAnalyzer.January, 1975 IN PLANTS BY AN AUTOMATED CATALYTIC METHOD 3 Method 11 Add 5 ml of 5.5 N hydrochloric acid to the residue and boil under reflux for a few minutes. Cool and dilute to 50ml in a 100-ml separating funnel. Add 4ml of benzoin a-monoxime and extract with two 5-ml portions of chloroform. Combine the extracts in the original conical flask.Evaporate the chloroform, add 1 ml of nitric acid - perchloric acid mixture and heat under reflux until fumes of perchloric acid are evolved. Remove the reflux funnel and evaporate the contents of the flask to dryness. Dissolve the residue by warming in 10 ml of 0.25 N hydrochloric acid and dilute to 20 ml. Determine the molybdenum content of the samples by the catalytic method. Determination of Molybdenum Switch on the ultraviolet monitor and the recorder 1 h before use. Set the absorbance scale of the monitor in the range 0-2 and the recorder at 20 mV. Prepare standard molyb- denum solutions in 0.125 N hydrochloric acid in the range 0-02-0*1 pg ml-l of molybdenum. Pump the reagents and wash solution through the system until a steady base line is obtained (about 30 min).Set the base line at 0.2 absorbance unit on the ultraviolet monitor and 10 divisions on the recorder by appropriate adjustment of the absorbance zero control on the ultraviolet monitor and the zero shift on the recorder. Load the sampler plate, set at the rate of 40 samples per hour with equal sample and wash uptake times, with a set of stan- dards, one wash, sixteen samples, one wash, sixteen samples, one wash. Read the peak heights on the recorder, construct a calibration graph and calculate the molybdenum concentration in the sample. Results and Discussion Sensitivity and Detection Limits The sensitivity of the reaction for the determination of molybdenum depends on the difference in the rate of the catalysed reaction and that of the uncatalysed reaction.These rates are functions of time and the concentrations of acid, iodide and peroxide. Maximum sensitivity is achieved by using the manif old assembly and reagent concentrations shown in Fig. 1. Fig. 2 shows the relation between molybdenum concentration and absorbance 0.02 0-04 0.06 0.08 0.10 Molybdenum/pg ml-' Fig. 2. Sensitivity of the reaction in the determination of molybdenum : A, at 364nm with CE 212 ultraviolet monitor; B, at 420nm with Chemlab colorimeter; and C, at 420 nm with Technicon colorimeter.4 Analyst, VoE. 100 when measured in cells of 15-mm path length with either a Technicon colorimeter, using a filter whose peak transmission was at 420nm, a Chemlab multi-channel colorimeter with a 420-nm filter, or a Cecil CE 212 variable wavelength ultraviolet monitor at a wavelength of 364 nm, near to the absorbance peak of the triiodide ion; the figure shows that molybdenum can easily be determined in the range 0.005-0.1 pg ml-l.Most of the work subsequently reported here was carried out with the Cecil CE 212 ultraviolet monitor. A value of twice the standard deviation of the blank was used as a measure of the detec- tion limit. Twelve blank digests were carried through Method 11 procedure and a mean value of 0.006 pg ml-l of molybdenum was obtained with a standard deviation of 0.0014; the detection limit for the method is therefore 0.003 pg ml-l of molybdenum. No absorbance readings were obtained for the blank solutions if the peroxide line was replaced with water. Thus the whole of the blank value can be considered to be due to molybdenum contamination and none due to any residual yellow colour following a nitric acid - perchloric acid digest of the chloroform - benzoin a-monoxime extract.Little difference was noted in the blank value obtained with amounts of nitric acid - perchloric acid from 1-10 ml and of chloroform from 5-30 ml. Therefore it is concluded that the major portion of the blank is derived from the borosilicate glass flasks in which the initial digestion is carried out. For this reason it is recommended that a set of flasks should be thoroughly cleaned and thereafter used only for molybdenum determinations. BRADFIELD AND STICKLAND : THE DETERMINATION OF MOLYBDENUM Interferences Ionic strength of solutions Hadjiioannou5 reported no effect on the rate of the reaction when the ionic strength was increased from 0.013 to 0.230 by the addition of zinc chloride, but Fuge6 found that salt concentrations of greater than 0.2 per cent.m/m of sodium or potassium chloride or sulphate had a marked effect on the rate of reaction. The salt concentration (calculated as the sum of the concentrations of calcium, potassium and magnesium chlorides) in a solution prepared from 2 g of plant material in a volume of 20 ml, may lie between 0.5 and 2.5 per cent. m/V and the ionic strength could be greater than 0.2. Table I shows the effect of increasing the concentrations of potassium, calcium and magnesium chlorides, both singly and in combination, on the measured values obtained for molybdenum in solutions containing 0.02 and 0.08 pg ml-1 of molybdenum.TABLE I EFFECT OF SALT CONCENTRATION ON THE DETERMINATION OF MOLYBDENUM Salt (as chloride)/mg ml-l A I 1 7 Potassium Calcium Magnesium Added 0.8 - - 0.020 4 8 - 1.2 - - 2.4 - - 12 I_ - - 0.1 - - 0-6 0.8 1.2 0.1 4 2.4 0-6 8 12 1 - - - - 1 - - Molybdenum/pg ml-l Found Added 0.020 0.080 0.019 0.018 0.020 0.020 0.016 0.019 0.019 0.01 7 0.019 0.019 0.016 1 Found 0.079 0.079 0.078 0-080 0.078 0.070 0.079 0.078 0.076 0.077 0.07 7 0.070 No interference is noted from a solution containing 4mgml-l of potassium plus 2.4 mg ml-l of calcium plus 0.5 mg ml-l of magnesium; this gives a salt concentration (calculated for chlorides) of 1.6 per cent. m/V. These concentrations are equivalent to 4 per cent. m/V of potassium, 2-4 per cent. m/V of calcium and 0-5 per cent.m/V of magnet.' .mm in dry matter if 2 g of plant material is digested and the residue dissolved in a volume of 20 ml. It is concluded that the ionic strength of solutions derived from the digestion of most materials is unlikely to cause interference in the direct determination of molybdenum by Method I but that low results may be obtained with plants that have a high salt contentJanHary, 19 75 IN PLANTS BY AN AUTOMATED CATALYTIC METHOD 5 (e.g., a sample of tomato leaf that contains 4 per cent. of potassium, 6 pel cent. of calcium and 0.6 per cent. of magnesium in dry matter). ,Tro;pz. Iron catalyses the reaction in a manner similar to molybdenum and a solution containing 20 pg ml-1 of iron(II1) gave an absorbance equivalent to 0.1 pg ml-1 of molyb- denum. Thus plant material containing 200 p.p.m, of iron would show an apparent molyb- denum content of 1 p.p.m.Initially, trans-l,2-diaminocyclohexane-NNN’N’-tetraacetic acid (DCTA) (13.84 g dissolved in sodium hydroxide solution and diluted to 1 1, at pH 6.2) was used to reduce interference from iron by a factor of 10 (Fig. 3) but over a period of time it was found the the free acid precipitated in the tube leading from the flow-through cell and eventually blocked the system. However, ammonium fluoride, used to mask tungsten, was found to be equally effective in reducing interference from iron. Phosphate. Phosphate reduced the catalytic effect of both molybdenum and iron (Fig. 4). The phosphate concentrations shown in the figure are equivalent to those which would result from digestion, in a volume of 20m1, of 2 g of dry plant material containing phosphorus in the range 0-1.8 per cent.as P. Twzgstert. Tungsten catalyses the reaction in a similar way to molybdenum (Fig. 3) but its effect can be reduced five-fold in the presence of ammonium fluoride. 10 20 30 40 Fe3+/pg ml-’ 1 I I I I 0.05 0.1 0.15 0-2 W/pg mi-’ Fig. 3. Effect of iron and tungsten on the reaction: x, iron; 0, iron plus DCTA; A, iron plus fluoride; @, tungsten; A, tungsten plus fluoride. Phosphorus (as NaH,PO,)/mg ml-1 Fig. 4. Effect of phosphate on the reaction: x , 0.06 pg 1111-1 of molybdenum ; 0 , 2 5 pg ml-l of iron. Determination of Molybdenum in Plant Samples Because of the possibility of interference in solutions of high salt content and from solutions containing phosphate, a preliminary separation of molybdenum was carried out by extraction from 0.5 N hydrochloric acid solution with benzoin a-monoxime and chloro- form,* as described in Experimental, Method I I .Tungsten is also extracted under these conditions and plants that had been grown in a nutrient solution containing tungstate gave high values for apparent molybdenum content. Bablto, Lisetskaya and Tsarenko9 reported that interference from tungsten could be overcome by carrying out the extraction withBRADFIELD AND STICKLAND TABLE I1 EFFECT OF BENZOIN CE-MONOXIME EXTRACTION AND AMMONIUM FLUORIDE ON TUNGSTEN INTERFERENCE Apparent recovery of molybdenum ( p g ) * from addition of , 3 1 pgof 10 pg of molybdenum tungsten No extraction ; fluoride solution replaced by water in AutoAnalyzer Extraction in presence of 5 per cent.of ammonium fluoride; fluoride Extraction in absence of fluoride; fluoride solution present in Auto- 1.00 1-36 solution replaced by water in AutoAnalyzer . . . . ” . 1.02 0.97 Analyzer . . .. .. .. .. .. .. .. 1.09 0.14 * Mean of six determinations. benzoin a-monoxime and chloroform in the presence of ammonium fluoride. In the present work it was found that tungsten was still partially extracted in the presence of fluoride but its effect on the catalytic reaction was almost completely eliminated by using 0.25 per cent. ammonium fluoride solution in the AutoAnalyzer manifold (Table 11). Molybdenum was determined in some reference plant materials by Method I and Method I I and the results are shown in Table 111. TABLE I11 MOLYBDENUM CONCENTRATION (p.p.m.) IN REFERENCE SAMPLES O F PLANT MATERIAL Extraction (Method 11) r- Standard Material Mean deviation! Lucerne . ... . . 0.18 0*023(6)1 Oat . . .. . . 0.18 0*016(8) Kale . . .. . . 1.10 0*176(8)1 Tomato leaf . . . . 0.45 0*035(8) Kale . . .. . . 2-48 0-205(8) Strawberry leaf . . 1.11 0*155(8) Direct (Method I) Reported values r--7 r--- Mean deviation: Mean Range Standard 0.20 0.043(14) 0.20 0*19-0.23* 0.20 0-043(15) 0.16 0.12-0.18* 0.90 0*126(13) 0.79 0.76-0.81 * 0.28 0.074(7) 0.42 0*36-0*46* 2.31 0.165( 10) 2.33 1.86-2.807 1.08 0*147(14) - - * Agricultural Research Council Report.lo 7 Standard plant material.ll 1 Figures in parentheses are the number of determinations made.Results obtained by the extraction method were more precise than those obtained by the direct method but the agreement between the mean values of molybdenum concentration determined by the two methods and the previously reported values was fairly good. Tomato leaf yielded low results by the direct method., which may be ascribed to the high salt content of this leaf and consequent depression of the rate of the catalytic reaction in a solution of high ionic strength and high phosphate concentration. It is concluded that for the most accurate results a preliminary separation of molybdenum as its benzoin a-monoxime complex is desirable but that for materials of low salt content more rapid, although somewhat less precise, results can be obtained directly on an acid digest. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Yatsimirskil, K. B., and Afanas’eva, L. I)., Zh. Analit. Khim., 1956, 11, 319. Yatsimirskii, K. B., and Alekseeva, I. I., Zh. Neorg. Khim., 1956, 1, 962. Yatsimirskii, K. B., and Alekseeva, I. I., Zav. Lab., 1958, 24, 1427. Svehla, G., and Erdey, L., Microchem. J.) 1963, 7 , 221. Hadjiioannou, T. P., Analytica Chim. Acta, 1966, 35, 360 Fuge, R., Analyst, 1970, 95, 171. Bradfield, E. G., J . Sci. Fd Agric., 1964, 15, 469. Jones, G. B., Analytica Chim. Acta, 1954, 10, 684. Babko, A. K., Lisetskaya, G. S., and Tsarenko, G. F., Zh. Analit. Khim., 1968, 23, 1342. “Report of a Group on Comparison of Methods of Analysis of Mineral Elements in Plants,” Bowen, H. J. M., AnaZyst, 1967, 92, 124. Received June 19th, 1974 Accepted September 2nd, 1974 Agricultural Research Council, London, 1963, p. 48.

ISSN:0003-2654    

DOI:10.1039/AN9750000001

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, January, 1975, Vol. 100, PP. 7-12 A Rapid Method of Determining Total Iodine in Bovine Milk M. M. Joerin" Waikato Dairy Laboratory, Ministry of Agriculture and Fisheries, Hamilton, New Zealand One millilitre of milk is wet ashed with chromic acid to free all bound iodine and oxidise it to iodate. The iodate is then reduced to free iodine by the addition of orthophosphorous acid. The iodine is distilled and is trapped in dilute thiosulphate solution. After dilution, the iodide ion is determined by a simple colorimetric method, based on the action of tetrabase and chlor- amine T. The assay range is 6200 pg of iodine per 100 ml and the colori- metric assay itself is 0-0.04 pg ml-l of iodine, with a sensitivity of 0.001 pg ml-l of iodine. 7 Milk and milk products are consumed universally.Since the adoption of iodophor as a disinfecting agent by the dairy industry in the last decade, there has been a rise in the level of iodine in milk.lp2 It has been suggested that this rise is associated with the rise in the incidence of thyrotoxicosis.1 The determination of iodine in micro-amounts is lengthy and complicated. The method of Ellis and D ~ n c a n , ~ as amended by Rogers and P o ~ l e , ~ consists in wet ashing with chromic acid followed by reduction of the digest by orthophosphorous acid, distillation of the liberated iodine and determination of the trapped iodide ion. However, this method has the disadvantages that two heatings of the digest are required, and the chromiurn(II1) oxide is often contaminated with iodine. Moreover, the determination of the iodide concen- tration is by the cerium(1V) - cerium(1II) ion reduction technique which is very sensitive to temperature variations. There is need of a quick, robust method for routine screening purposes.This paper describes a rapid digestion procedure, together with a quick, simple colorimetric method of determining liberated iodide ion. Apparatus The oxidation with chromic acid, the reduction of iodate to iodine and its subsequent distillation are based on the methods of other authorssy4 but are described in detail in this paper because of the variations adopted. Experimental T Air - ~ 2 ---t To water pump Fig. 1. Schematic diagram of the apparatus used for the iodine distillation: 1 = 50 ml Kjeldahl flask (B19); 2 = trap (B19 test- tubes, 14 cm long) ; 3 = gas wash bottle; 4 = water-bath (100 "C) ; T = three-way tap. A schematic diagram (Fig.1) shows the main apparatus used. The oxidation with chromic acid was carried out in the Kjeldahl flask. the policy of the Ministry. * All opinions and recommendations expressed are those of the author and do not necessarily reflect8 JOERIN: A RAPID METHOD OF DETERMINING Analyst, Vol. 100 Reagents Reagents of analytical-reagent grade were used unless otherwise specified. Water. Distilled and de-ionised. Sulphuric acid. Iodine free. Potassium dickromate. Hydrogen peroxide, 10 per cent. m/V solution. Orthophosphorom acid (H,PO,). Laboratory-reagent grade, 50 per cent. w/V solution. Orthophosphorous acid (1 kg) was added to 2 1 of water. The solution was boiled for 2 h.cooled and brought to the original volume with water. Sodiztm thiosul@hate, 0.050 per cent. m/V solution. Fresh solutions were prepared weekly. Glacial acetic acid. Sodiztm acetate trihydrate. Tetrabase reagent (4,4'-tetramethyldiaminodi~henylmethane). Spot-test reagent. Tetra- base (0.8 g) was boiled gently with 40 ml of water and 4 ml of glacial acetic acid for 2 min, stirring occasionally with a glass rod. This solution was then cooled in ice - water for 6 min, some tetrabase precipitating out. Then 0 4 g of sodium acetate trihydrate was added, the solution was stirred well and allowed to remain in the ice - water for at least a further 10 min. The solution was then filtered through a Whatman No. 1 filter-paper into a dropping bottle.This solution was stored in the dark at 4 "C. When stored in this manner the reagent was stable for 2 months. Chloramine T. Laboratory-reagent grade, 0.040 per cent. m/ V solution. Fresh solutions were prepared daily. Potassium iodide. Standard solution of 1.0 pg ml-l as iodide ion. Potassium hydroxide solution, 10 per cent. m/V. All apparatus was washed in dilute chromic acid and rinsed thoroughly with water to Procedure ensure the absence of any iodine. Digest i o n Potassium dichromate (1.5 g) was added to the Kjeldahl flask (Fig. 1) followed by 15 ml of water and 8 ml of sulphuric acid. The flask was cooled, then 1 ml of milk was added. A thermometer (0 to 220 "C) was then placed in the flask. The flask was placed on a depressed asbestos wire gauze and the contents were heated to boiling with a Bunsen burner.The contents of the flask were gently boiled for approximately 15 min until the thermometer registered 185 "C. The flask was then allowed to cool in air until the thermometer registered about 100 "C and then cooled to room temperature under running water. The thennometer was removed and washed with 10 ml of water, the washings being added to the digest. Distillation The apparatus was assembled as shown in Fig. 1. The trap contained 5 ml of sodium thiosulphate solution and the gas wash-bottle contained 10 per cent. potassium hydroxide solution. Air was drawn through the apparatus at about 1-14 1 min-1, the three-way tap being open only to the gas wash-bottle. The three-way tap was then opened to admit 3 ml of orthophosphorous acid solution, followed by 0.1 ml of hydrogen peroxide in about 3 ml of water.At 5-min intervals, two further additions of 3 rnl of orthophosphorous acid solution were made, and each washed in with about 3 ml of water. Distillation of the liberated iodine was continued for a further 5 min, after which the assembly was disconnected. Unexpected bumping of the contents of the flask during this stage is unlikely, because of the air being drawn through the reactants and the low temperature (about 98 "C). The contents of the thiosulphate trap were finally washed quantitatively into a 50-ml calibrated flask and diluted to volume. Determination Five millilitres of the diluted sodium thiosulphate solution were measured into a test-tube (140 x 20 mm), two drops (0.06 ml) of tetrabase reagent were added and the contents swirled to mix.One millilitre of chloramine T solution was added rapidy and immediately mixed by swirling the test-tube. Exactly 40 s after the addition of the chloramine T solution, the absorbance was measured in a Unicam SP600 spectrophotometer at 600 nni using a 10-mmJawayy, 1975 TOTAL IODINE IN BOVINE MILK 9 cell and a water blank. The absorbance was compared with the standard curve (Method 2) in order to find the concentration of iodine in milk, as micrograms per 100 ml. Reagent blank acid, especially, is liable to contain trace amounts of iodine. A blank value for all reagents was obtained by following the above procedure, as sulphuric Results Colour Measurement in the Determination Stage The blue colour developed in the determination stage has a sharp maximum at 600 nm.Unfortunately, the colour is transient. The maximum absorbance is developed between 30 and 40 s after adding the chloramine T solution and lasts for only 10 to 15 s (see Table I). TABLE I TIME AFTER ADDITION OF CHLORAMINE T SOLUTION veysz4s ABSORBANCE Time/s Absorbance Time/§ Absorbance 0 0.00 40 0-50 10 0.00 50 0.47 20 0.18 60 0.42 30 0.50 90 0.19 The standard deviation of replicate readings of the same solution is of the order of 0.01 absorbance unit. There was no variation in intensity of the blue colour with temperature in the range 15 to 25 "C. The only variation was in the timing of the formation and decay of the blue colour, which with the higher temperature was advanced by about 10 s.Standard Graph using three methods. The results are shown in Fig. 2. The relationship between absorbance and iodine concentration was investigated by 50 100 150 200 lodine/pg per 100 ml of milk Fig. 2. Relationship between absorbance and iodine ion concentration: A, Method 1 ; and B, Method 2. Method 1 Various aliquots of standard potassium iodide solution were each diluted to 50 ml with water and 5 ml of sodium thiosulphate solution. Five-millilitre aliquots of these solutions were treated as described under Determination. The results are shown by the solid line in Fig. 2. Method 2 As for Reagent blank but with various portions of standard potassium iodide solution added to the digest before heating. The results are shown by the broken line in Fig.2. Method 3 As for Method 2, but including 1 ml of milk of known iodine content. Recoveries of Iodine These recoveries were determined at four different concentrations in the range 50-200 pg of iodine per 100 ml of milk and are shown in Fig. 2 Iodine recoveries were 85 per cent.10 JOERIN: A RAPID METHOD OF DETERMINING Analyst, Vol. 100 by comparing the standard curve for Method 1 against that for Method 2, Method 3 giving the same recovery as Method 2. In Method 3, a milk of known, low iodine content was used. The iodine content was measured as given under Procedure before recoveries were determined by Method 3. Interference from Foreign Ions The tetrabase reagent was tested in the presence of chloride and bromide ions, To 1 ml of standard potassium iodide solution were added 100 pg of sodium chloride, 100 pg of sodium bromide and 5 ml of sodium thiosulphate solution, and the mixture was diluted to 50 ml. Five millilitres were treated as described under Determination.There was no alteration in the intensity of the blue colour. Variation of Reagent Concentration in the Determination Stage Tetrabase reagent prepared exactly as directed is stable for up to 2 months if kept cool and in the dark. There is little variation between batches. Chloramine T solution must be prepared daily and thiosulphate solution weekly, as these solutions are less stable. Vari- ation in their concentrations has a considerable effect on the absorbance due to the blue colour for a given iodide concentration (Tables I1 and 111). TAELE I1 EFFECT OF VARIATION IN CONCENTRATION OF THIOSULPHATE SOLUTION ON ABSORBANCE Concentration of thiosulphate, per cent.* Absorbance 0.0040 0.m 0,0045 0.78 0.0050 0.65 0.0065 0.48 0.0060 0-32 * In the final solution.Discussion The method of Rogers and Poole4 involved the use of chromium(II1) oxide as an oxidising agent in the wet-ashing stage to oxidise all forms of iodine to the non-volatile iodate. Chromium( 111) oxide usually contains trace amounts of iodine. Potassium dichromate can be used as an alternative, the analytical-reagent grade salt being iodine free. In the method described here a slight excess of potassium dichromate was used, in proportion to the amount of chromium(II1) oxide used by Rogers and Poole, in order to avoid the risk of a secondary side reaction between iodide and iodate to yield iodine.TABLE I11 EFFECT OF VARIATION IN CONCENTRATION OF CHLORAMINE T SOLUTION ON ABSORBANCE chloramine T, per cent. Absorbance Concentration of 0.030 0.01 0.035 0-37 0.040 0.66 0.045 0.82 0.05 1.01 Initial experiments on the digestion stage using 10ml of water gave a shorter heating time of about 5-8 min, but led to a reduced and variable iodine recovery (50-70 per cent.). Increasing the water content to 15 ml lengthened the heating time of continuous boiling to about 15 min and increased recoveries (greater than 85 per cent.). Any further increase in the amount of water or time of heating did not increase recoveries. The distillation stage closely follows the method of Rogers and P o ~ l e , ~ with three varia- tions. The trap contained thiosulphate instead of sodium arsenite, and only one trap was found to be necessary. In several experiments two traps were used but, as no iodine was found in the second trap, its use was discontinued.January, 1975 TOTAL IODINE I N BOVINE MILK 11 Investigation confirmed the necessity of adding orthophosphorous acid in three separate stages, as described by Rogers and P ~ o l e . ~ The air stream must be sufficient to carry all the iodine vapour over into the trap.In order to induce the reduction of the chromic acid by the orthophosphorous acid a small amount of hydrogen peroxide must be added,4 but if added in excess it may reduce iodine recoveries. The determination stage is based on Feigl’s reaction6 between iodide ions and tetrabase, in which a blue colour is produced.In very dilute solutions, the transient blue colour vaned with the iodide concentration. Chlorine and bromine did not interfere in the formation of the blue colour, but quenching occurred with reducing agents and acids. Therefore, the thiosulphate solution used to trap the iodine vapour must be accurately prepared and must be diluted before the blue colour is formed. Moreover, sodium acetate is added to the tetrabase reagent to buffer the acetic acid used in its preparation. This addition raises the pH of the reagent from 3.5 to 3.9. Excess of sodium acetate does not raise the pH further but precipitates the base, changes the colour of the reagent from yellow to blue - green and lowers the sensitivity of the reagent. The method described is a rapid and sensitive determination of the total iodine in milk. The repeatability is good if the method is followed diligently and in a standard manner. I thank the Director, Dairy Division, Ministry of Agriculture and Fisheries, New Zealand, for permission to publish this paper. References 1. 2. 3. 4. 5. Connelly, R. J., Med. J . Aust., 1971, 1268. Joerin, M. M., and Bowering, A. F., N.Z. JZ Dairy Sci. Technol., 1972, 7, 118. Ellis, G. H., and Duncan, G. D., Analyt. Chem., 1953, 25, 1558. Rogers, K., and Poole, D. B., Biochem. J., 1958, 70, 463. Feigl, F., “Spot Tests in Organic Analysis,” Elsevier Publishing Company, Amsterdam, Seventh Received May 16th, 1974 Accepted J d y 2nd, 1974 Edition, 1966.

ISSN:0003-2654    

DOI:10.1039/AN9750000007

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

12 Analyst, January, 1975, VvoL 100, pp. 12-15, Micro-determination of Xylose in Plasma P. Trinder Biochemistry Department, Central Laboratoyy, Royal Infirmary, Sunderland, Co. Durham A very sensitive method for the determination of xylose in plasma has been developed, which involves condensation of phloroglucinol with the furfural released from xylose by the action of hot hydrochloric acid - acetic acid. In the standard method duplicate analyses of plasma samples containing 0.667-2.667 mmol 1-1 of xylose can be achieved by using 0.2-ml samples. A modified method using 0-1 ml of whole blood can also be used. The absorbance obtained on analysing solutions containing 1.333 mrnol l-1 of xylose is 0.372 at 554nm using cuvettes of 10-mm light path. This value corresponds to a molar absorption of approximately 33 000.By using hydro- bromic acid - acetic acid an even higher molar absorption of a,pproximately 39 000 is obtained. Glucose gives only a weak reaction under the test condi- tions, 417 nmol of glucose giving an absorbance equivalent to 7-6 nmol of xylose. Since its introduction in 1973 by Helmer and Foutsl the xylose test for intestinal absorption has been widely used. Recently, Sladen and Kumar2 have considered that jejunal intubation will replace the xylose test, which will continue to be used when the intubation method is not available. The analysis is usually performed on udne collected over a 5-h period, using dosages of 0-6 or 0.1 g per kilogram of body mass, the lower dosage being used for adults owing to the unpleasant effects on the bowel of the larger dose.The 0.5 g k g l dose has been used for infants, but the collection of urine samples is often unsatisfactory and analysis of blood samples would be an advantage. The method most often used for urine analysis involves the condensation of xylose with 4-bromoaniline in glacial acetic acid - thiourea; this met hod has also been used for ~ l a s m a ~ , ~ in which xylose levels using the 0.5 g k g l dose should exceed 2 mmol l-1, but the sensitivity of the pink colour is very low and the absorbance obtained on analysing a sample containing 2 mmol 1-1 of xylose is only 0.12 when employing 0.1 ml of serum in a final volume of 6 ml. Xylose levels in plasma of above 0.4 mmol l-1 concentration might be expected using the 0.1 g k g l dose, but the 4-bromoaniline method is not sufficiently sensitive to enable such levels to be measured.A method with at least five times the sensitivity of the 4-bromoaniline method is required for the determination of plasma xylose and the method described below is approximately ten times more sensitive than the 4-bromoaniline method. Experimental The proposed method is based on the observation that in glacial acetic acid solution, xylose is rapidly converted into furfural in the presence of hydrochloric or hydrobromic acid. Glucose under these conditions is not converted into the corresponding methylfurfural to any great extent. Furfural absorbs strongly at 278 nm but owing to the presence in blood of other absorbing substances, this absorption cannot be used as a measure of the xyllose present.Phloroglucinol has been used as a test for pentoses and galactose. This test, described by Oshima and Tollens5 in 1901, is known as Tollens' test.e The carbohydrate solution is heated at 100 "C with an equal volume of concentrated hydrochloric acid containing phloro- glucinol and an unstable red colour is given by pentoses and galactose. The final concentra- tion of hydrochloric acid in Tollens' test is about 6 M, but as furfural can be released from xylose by using a reagent containing 5 volumes of glacial acetic acid plus IL volume of 6 M hydrochloric acid, phloroglucinol was added to this mixture in the expectation that the colour developed in Tollens' test would be produced. This reagent, which has a final concentration of hydrochloric acid of only 1 M, gives an intense stable red colour on heating with xylose for 6 min at 100 "C.Fructose, glucose and galactose give very weak colours under identical conditions. Hydrobromic acid also gives an intense red colour, but sulphuric acid gives only a weak colourTRINDER 13 and phosphoric and perchloric acids give no colour at similar final concentrations. The molar absorbance of the colour developed is approximately 33 000 when using hydrochloric acid and approximately 39000 when using hydrobromic acid. A method for the deter- mination of plasma xylose based on this reaction is described below. Method Reagents Use reagents of analytical grade when available. Concentrated hydrochloric acid, approximately 11-5 M. Comentrated hydro bromic acid, approximately 8.5 M . Zinc suljbhate (ZnS0,.7H20) solution, 10 per cefit.m/V. Sodium hydroxide solution, 0.5 M. Stock xylose solution, 33.33 mmol l-1 in 0.2 per cent. m/V benzoic acid solution. Dissolve 0.5 g of xylose in sufficient 0.2 per cent. m/V benzoic acid solution to make 100 ml of solution. Standard xylose solution, 1.333 mmoll-l. Dilute 4 ml of stock xylose solution with 0.2 per cent. m/V benzoic acid solution to a final volume of 100 ml. Colow reagent. Just before use, dissolve 0.5 g of phloroglucinol in 100 ml of glacial acetic acid. Procedure Into tubes labelled S (standard) and T (test) place 1.2-ml volumes of water; into a tube labelled B (blank) place 1.4 ml of water. To tube S add 0.2 ml of standard xylose solution, 1.333 To all tubes add 0-3 ml of 10 per cent.m/V zinc sulphate solution, mix and add 0.3 ml of 0 . 5 ~ sodium hydroxide solution. Mix the solutions and centrifuge the tubes. Transfer 0.5 ml of the clear super- natant fluid to 6 x #-in tubes, appropriately labelled. Add 0.5 ml of concentrated hydro- chloric acid (11.5 M) or concentrated hydrobromic acid (8.5 M) to each tube, mix and add 5 ml of colour reagent. Mix the solutions well and place the tubes in a bath of vigorously boiling water for exactly 5 min, then cool them rapidly. Set a spectrophotometer to zero absorbance with B and read the absorbances of S and T, using a wavelength setting of 554nm and cuvettes of 10-mm light path. to tube T add 0.2 ml of heparinised plasma. x 1.333 Absorbance T Absorbance S Plasma xylose, mmoll-1 = The colour produced is reasonably stable, fading occurring at a rate of 5-7 per cent.h-l. Analysis of whole blood obtained from a heel stab Use 0.1 ml of blood and half-volumes of water, protein precipitant, supernatant fluid and colour reagent. Analysis of urine Dilute the urine so that it contains 0.1-0.3 mmoll-1 of xylose. Treat 0.5 ml of diluted urine in the same manner as for 0.5 ml of supernatant fluid as described in the method for plasma. Use 0.5 ml of xylose solution (0.133 mmoll-l) as a standard. Results The recovery of added xylose is quantitative over the range 0.667-2.667 mmol 1-1 (Table I) and interference from glucose is low; 8-33 mmol 1-1 of glucose gives a colour equivalent to 0.15 mmol 1-1 of xylose. The time of colour development has been chosen so TABLE I RECOVERY OF ADDED XYLOSE No.of Pre-formed Added Xylose Xylose observa- xylose/ xylosel found/ recovered/ Recovery, per cent. (mean) (mean) Range Mean Plasma . . . . 10 0.08 0.667 0.752 0-672 97-107 100.7 Plasma . . . . 10 0.08 1.333 1-408 1.328 96.5-104.6 99.6 Whole blood . . 4 0-135 1.333 1.461 1.326 955-104 99.5 Plasma . . . . 10 0.08 2.667 2-736 2.656 96.5-103.8 99.8 tions mmol 1-1 mmol 1-1 mmol 1-I mmoll-1 &14 TRINDER : MICRO-DETERMINATION OF Analyst, Pol. 100 that interference from glucose is minimal, consistent with maximum colour development. Twenty successive plasma samples containing no xylose but 2.44-5-28 mrnol l-1 of glucose, mean 4-57 mmoll-1 of glucose, gave a mean apparent xylose content of 0.082 mmol l-1; range 0-040-0.127 mmoll-l.As a standard glucose solution containing 4-57 mmoll-1 gives a colour intensity equivalent to that of a xylose solution containing 0.082 mmoll-1, j t is evident that all the colour developed from these plasma samples is due to glucose. This conclusion is confirmed by analysing plasma samples containing no glucose, when no collour develops. Table I1 shows that Beer's law is obeyed up to an absorbance of 0.7. (Table I11 gives the rate of colour development for various sugars.) The precision of the method was determined from 20 analyses of plasma, using a Cecil C.E. 272 spectrophotometer; the coefficient of variation, expressed as a percentage, is 1.9 at a mean plasma xylose level of 1-82 mmoll-1 (range 1-75-1.89 mmol I-l), the standard deviation being 0.036.These results were obtained by using hydrochloric acid. Xylose in solution/ mmol 1-1 0.667 1.333 2.000 2.687 1.333 2.667 TABLE I1 RESULTS USING STANDARD SOLUTIONS Final volume, 6 ml; wavelength setting, 554 nm; cuvettes, 10-mm light path. Xylose in 0.5 ml of supernatant fluid/nmol 33.3 66.7 100.0 133.3 66.7 133.3 Acid used HCl HCl HC1 HC1 HBr HBr Absorbance Molar 0.182, 0.182, 0.183, 0.184, 0.180 . . Mean 0.182 32 800 0.374, 0.368, 0.374, 0.374, 0.368 . . Mean 0-372 33 500 0.546, 0.557, 0.555, 0.551, 0.556 . . Mean 0.553 33 200 0.726, 0.737, 0.732, 0.733, 0.744 . . Mean 0-734 33 000 0.441 39 600 0.868 39 100 A 7 I absorbance A higher sensitivity (molar absorbance of the coloured product approximately 39 000) can be achieved by substituting concentrated hydrobromic acid for concentrated hydrochloric acid, but the latter acid was preferred as it is more generally available and is more stable on storage. TABLE I11 RATE OF COLOUR DEVELOPMENT Final volume, 6 ml; wavelength setting, 654 nm; cuvettes, 10-mm light path, Time a t 100 "C/min Sugar (nmol in 0.5 ml supernatant fluid) Xylose (133) Glucose (417) Glucose (833) Glucose (2778) Glucose (5556) Fructose (278) Galactose (278) k 5 6 i of Absorbance w \ 0.711 0.735 0.73 1 0.734 0.049 0.084 0.092 0.114 0.042 0.140 0.265 0.055 0.072 Discussion and Conclusions The main advantage of the proposed method over the 4-bromoaniline method is its very high sensitivity with low interference from glucose. The method is also very rapid as colour development requires only 5 min compared with 80 min in the 4-bromoaniline method.A determination can be completed in 15 min. The reagent blank absorbance is very low and phloroglucinol has the advantage over 4-bromoaniline in that it can be obtained in a high state of purity. The colour development is carried out at 100 "C, thus eliminating the use of a water-bath thermostatically controlled at 70 "C, required in the 4-bromoaniline method. The method is ideally suited to the determination of plasma xylose using the 0.5 g k g 1 dose when the critical diagnostic level of 2 mmol 1-1 of xylose produces an absorbance of about 0.55. Plasma levels of xylose after using the 0.1 g kg-l dose have not been investigatedJawary, 1975 XYLOSE I N PLASMA 15 owing to the insensitivity of existing methods, but at an expected critical diagnostic level of 0*4mmo11-1 of xylose an absorbance of 0.11 would be obtained by using the proposed method. The absorbance contributed by glucose in non-diabetic fasting subjects is equivalent to approximately 0.08 mmoll-1 of xylose and this level would form an appreciable proportion of the total. This interference can be minimised by taking a blood sample before adminis- tration of xylose in order to establish an appropriate base-line, but a method that eliminated the interference of glucose and gave a higher final absorbance would be an advantage. The development of such a method is being investigated. References 1. 2. 3. 4. 5. 6. Helmer, 0. M., and Fouts, P. J., J . Clin. Invest., 1937, 16, 343. Sladen, G. E., and Kumar, P. J., Brit. Med., 1973, 3, 223. Varley, H., “Practical Clinical Biochemistry,” Fourth Edition, William Heinemann, London, 1967, Roe, J. H., and Rice, E. W., J . Biol. Chem., 1948, 173, 607. Oshima, K., and Tollens, B., Ber. Dt. Chem. Ges., 1901, 34, 1425. Varley, H., “Practical Clinical Biochemistry,” Third Edition, William Heinemann, London, 1962, p. 425. p. 71. Received April 22nd, 1974 Accepted A q p s t 6th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000012

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

16 Analyst, January, 1975, Vol. 100, pp. 16-18 Selective Colorimetric Determination of Paracetamol by Means of an lndophenol Reaction C. T. H. Ellcock and A. G. Fogg Chzrnistry Department, University of Technology, Loughborowgh, Leicestershzire, LE11 3T U The spontaneous oxidation of alkaline mixtures of p-aminophenol and phenol with molecular oxygen to form indophenol has been made the basis of a colorimetric procedure for the determination of paracetamol via its hydrolysis product, 9-aminophenol. The hydrolysis product of phenacetin, p-pheneti- dine, does not undergo this reaction. Several other active ingredients with which paracetamol is often formulated have been shown not to interfere. Total paracetamol and phenacetin can be determined by an indophenol reaction after hydrolysis and oxidation with acidified hypochlorite.Both $-aminophenol and p-phenetidine react with hypochlorite in acidic solution to give 9-quinonechlorimide, which can be made to react with phenol to give indophenol. The molar absorptivity of indophenol based on its complete formation in these reactions was found to be 2.85 x lo4 1 mol-l cm-l at the wavelength of maximum absorption (625 nm).l This value is slightly lower than that obtained by Corbett,2 who found the molar absorptivity of indophenol to be 3-16 x lo4 1 mol-l cm-1 at 637 nm. Corbett’s value was determined by use of a sample of indophenol prepared by oxidising 9-aminophenol in alkaline solution with potassium hexacyanoferrate(II1). Paracetamol and phenacetin have been determined, after hydrolytic: oxidation with acidified hypochlorite solution, by reaction with phenol to form indophenol.1 The present work was undertaken in order to determine whether hydrolytic oxidation of paracetamol and phenacetin could be effected in alkaline solution, possibly with potassium hexacyano- ferrate(II1).Attempts to develop such a method were unsuccessful, but it was observed that p-aminophenol was oxidised spontaneously in alkaline solution in the presence of phenol. The oxidising agent was shown subsequently by means of tests on de-oxygenated solutions to be dissolved molecular oxygen. In the present paper the application of this reaction to the determination of paracetamol is described. Experimental Yamaguchi3 has also noted this reaction recently. Absorbance measurements were made with a Uvispek 700 spectrophotometer using 1 -cm cells.Wavelength calibration was made with a didymium glass filter and the absorbance scale was checked by means of standard neutral filters. The procedure outlined below was developed by using a freshly prepared solution of 9-aminophenol hydrochloride M) as the standard. The effect of pH on the intensity of the indophenol colour developed is indicated in Table I; maximum colour is developed TABLE I EFFECT OF pH ON THE APPARENT MOLAR ABSORPTIVITY OF THE 1NDOPHE;NOL FORMED pH . . .. .. . . 8.98 9.46 10.02 10.48 10.82 11-32 11.62 12.02 lo4 1 mol-l cm-l , . . . 2-50 2-92 3.16 3-15 3.15 3-15 3-00 2-60 Apparent molar absorptivity/ for solutions with pH values in the range 20-11.3. The effect of phenol concentration is shown in Table 11; a final concentration of 0.6 per cent.m/V of phenol in the final solution was selected. The phenol was conveniently included in a buffer solution that contained TABLE I1 EFFECT OF PHENOL CONCENTRATION ON THE APPARENT MOLAR ABSORPTIVITY OF THE INDOPHENOL FORMED Phenol concentration, per cent. m/V 0.01 0.02 0.05 0-1 0.2 0.4 0.6 0-8 Apparent molar absorptivity/ lo4 1 mol-l cm-l . . . . . . 2.51 2.80 3.07 3.10 3.12 3-24 3.24 3.26ELLCOCK AND FOGG 17 trisodium orthophosphate and sodium hydroxide , the final measured solution having a pH of 10.5. Colour development was completed within 5 min of mixing at 20 “C. Paracetamol, phenacetin and p-phenetidine were shown not to react with phenol under these conditions. Conditions for the complete hydrolysis of paracetamol have been described previous1y.l Reagent AZkaZine solution of phenol. Dissolve 2.0 g of trisodium orthophosphate dodecahydrate plus 2.0 g of sodium hydroxide plus 6-0 g of phenol in water and dilute the mixture to 100 ml in a calibrated flask.Procedure for Determination of Paracetamol Weigh an amount of sample containing from 0.4 to 0.6 g of paracetamol into a 100-ml conical flask, fit a reflux condenser, add 15 ml of concentrated hydrochloric acid and reflux the mixture for 30 min. After allowing the solution to cool slightly, wash the condenser down with 10 ml of water. Transfer the solution at room temperature into a 100-ml calibrated flask and dilute to volume. Pipette 25ml of this solution into a 250-ml calibrated flask and dilute to volume.Finally, pipette 10 ml of this solution into a second 250-ml calibrated flask and dilute to volume. Transfer, by pipette, 5ml of this solution into a 50-ml calibrated flask, add 5ml of alkaline solution of phenol, mix by swirling and allow the solution to stand for 10 min. Dilute to volume and measure the absorbance of the solution at 625 nm in l-cm cells against water. A calibration graph can be obtained either by using known amounts of paracetamol in the above procedure or by using a 10-4 M solution of 9-aminophenol hydrochloride beginning at “Transfer, by pipette, . . .,’ and using 0-10-ml aliquots of solution. The calibration graph obtained with $-aminophenol is consistent with a molar absorp- tivity of 3.07 x lo4 1 mol-l cm-1 for indophenol.The coefficient of variation (12 deter- minations) was less than 2 per cent. The absorbance of the blank determination was 0.002. In order to study the recovery of paracetamol and the effect of phenacetin, the procedure was applied to mixtures of paracetamol and phenacetin. The results, which indicate complete recovery of paracetamol and a very slight apparent interference (about 3 per cent.) from phenacetin, are shown in Table 111. These mixtures were also analysed by the method given Deduct the absorbance recorded in a blank determination. TABLE I11 ANALYSIS OF MIXTURES OF PARACETAMOL AND PHENACETIN 0-502 0.50, Nil - 0.508 0.51, Nil - 0.503 0.50, Nil - Nil 0.01, 0.499 - Nil 0.01, 0.505 - Paracetamol takenlg Paracetamol found/g Phenacetin takenlg Phenacetin found/g* 0.433 0.45, 0.438 0.42, 0.435 0.45, 0-437 0-42, 0.433 0.46, 0.438 0.42, 0.435 0.45, 0.437 0.42, 0-43 1 0-45, 0.435 0.42, 0.430 0.46, 0.436 0.41, 0.43 1 0.45, 0.435 0.42, 0.430 0-46, 0.436 0.42, * As the total amount of paracetamol and phenacetin taken was twice that recommended for the hypochlorite procedure, 5 ml of the diluted hydrolysed solution was taken for reaction with hypochlorite instead of 10 ml.previously by which total paracetamol and phenacetin are determined1 ; phenacetin was then determined by difference. The hydrolysis procedure is the same in the two methods. The results obtained are also shown in Table 111. The amounts of paracetamol found in the presence of approximately equal amounts of phenacetin are about 5 per cent. too high.For the determination of paracetamol in formulations that also contain phenacetin, the blank due to the phenacetin should be subtracted.18 ELLCOCK AND FOGG The possible interference of other active ingredients with which paracetamol is commonly formulated was also considered. Aspirin Tablets B.P., A.P.C. Tablets B.P.C. and Compound Codeine Tablets R.P. were treated by the above procedure, including the hydrolysis step, taking 1 g of ground tablets in each instance. The absorbance values obtained were less than 0-015, which indicates that aspirin, caffeine and codeine do not interfere in the procedure. The British Pharmacopoeia monograph for paracetamnl includes a test for $-aminophenol impurity. The present procedure can be readily adapted for the quantitative determination of $-aminophenol in paracetamol: triturate 1 g of finely ground sample with 5 ml of 1 M hydrochloric acid, dilute the mixture to 100 ml with water and filter a portion of the solution.Take 5ml of the filtrate and use the procedure given above beginning at “Transfer, by pipette, 5 ml of this solution. . . .” When this procedure was applied to a sample of para- cetamol tablets (0.922 g), the absorbance obtained was 0*018, indicating that there was less than 0.01 per cent. m/m of $-aminophenol in the tablets. Discussion The recommended procedure involving spontaneous oxidation of a mixture of $-amino- phenol and phenol with molecular oxygen at an optimum pH is simpler than that given previously in which hypochlorite at a pH of less than 6 is used as the oxidant.1 In this latter reaction, oxidation of p-aminophenol to p-quinonechlorimide, and its subseq,uent reaction with phenol, is carried out in discrete steps, whereas in the recommended procedure a single reaction mixture is used, and the oxidising agent requires no preparation.Nevertheless, the two procedures can be readily combined in order to determine para- cetamol and phenacetin in mixtures. Phenacetin can only be determined by the hypochlorite method, as its hydrolysis product, +-phenetidine, does not react with phenol and molecular oxygen. Thus, paracetamol can be determined by the recommended procedure, total paracetamol and phenacetin by the hypochlorite procedure, and the phenacetin by difference. In practice, paracetamol and phenacetin are not usually dispensed together in pharma- ceutical formulations.Martindale4 lists only one product (Adwin tablets) that contains both of these drugs. Nevertheless, the two procedures should prove useful in the routine determination of the two drugs and in trace analysis of drug substances. The advantages of the selective procedure for paracetamol are its simplicity and the fact that it is specific for determining paracetamol in the presence of other drugs with which it is commonly formulated. $-Aminophenol interferes in the procedure, and, indeed, can be determined separately in paracetamol tablets by omitting the hydrolysis procedure. This procedure should be useful in carrying out stability trials on products containing paracetamol. The value for paracetamol should be corrected for the 9-aminophenol found to be present. References 1. 2. 3. 4. Blacow, N. W., Editor, “The Extra Pharmacopoeia (Martindale),” Twenty-sixth Edition, The Received January 14th, 1974 Amended May 31st, 1974 Accepted Aacgust 2nd, 1974 Davis, D. R., Fogg, A. G., Bums, D. T., and Wragg, J. S., Analyst, 1974, 99, 12. Corbett, J. F., J . Chem. SOC. (B), 1970, 1602. Yamaguchi, R., Hoshi Yakka Daigaku Kiyo, 1972, 14, 98. Pharmaceutical Press, London, 1972.

ISSN:0003-2654    

DOI:10.1039/AN9750000016

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Aiaalyst, Janwry, 1976, Vol. 100, PP. 25-28 25 A Continuous Flow Cell for Use with the Bendix = NPL Automatic Polarimeter. Application to Neomycin Analysis P. de Rossi Division of Antibiotics, National Institute for Biological Standards and Control, Holly Hili, Hampstead, London, N W3 6RB The design of a micro-cell (1 10-pl capacity) for use with the Bendix - NPL polarimeter is described. The cell has been used for the continuous flow analysis of neomycin and other aminoglycoside antibiotics using an ion- exchange column of small dimensions. Eluted material with an optical rotation down to 0-006° can be measured. In the course of choosing material suitable for use as the Second International Reference Preparation for Neomycin,l it was necessary to examine samples of neomycin obtained from different sources.The main purpose of the examination was to determine the degree of variation that existed in the content of neomycin C and of neamine in different source materials. Themethod used was basically that describedin the R.1'. 19732for the determination of neomycin C in Framycetin Sulphate B.P. (also of kanamycin B in Kanamycin Sulphate B.P.) using column chromatography on an ion-exchange resin. The recommended B. P. procedure requires the collection of the eluate in separate tubes followed by reaction with ninhydrin, and spectrophotometric measurement of the resulting colour. The procedure is time con- suming and tedious when numerous samples have to be examined and alternative procedures were investigated. Continuous development and monitoring of the colour produced with ninhydrin was effected using an AutoAnalyzer technique but the results obtained were difficult to evaluate quantitatively as the amplitude of a peak was related to the logarithm of concentration. In addition, poor resolution, due in part to the length of the flow system between the end of the column and the detector, is inherent in this method of analysis. Continuous flow with measurement of electrical conductivity has been used successfully for the analysis of aminoglycoside antibiotics by Gillet et aL3 In practice, this method requires a conductivity measuring system with a wide range, capable of producing a stable base line at the very low conductivity levels experienced using glass-distilled, carbon dioxide free water as the eluting agent.In our hands, a small but significant leakage of ions from the column resin contributed to base-line instability that could not be controlled when using the LKB Conductolyzer. Experimental and Results The Bendix - NPL automatic polariineter was designed for continuous monitoring of optical rotation during industrial sugar processing and, as both neomycin B and C are optically active, the possibility of using this technique was investigated. The flow-through cell supplied with the instrument has a retention volume of 2-4 nil with an optical path length of 20 mm. This cell proved unsuitable for our purpose and was successfully modified in order to reduce the volume contained within the light path to 110 pl without significant loss of sensitivity.The output of the polarimeter was coupled to a pen recorder (Smith Servoscribe RE 541.120) and shunted with a four-decade resistance box of 0.1-7000 IR (Sullivan). The shunt values were determined by using standard sucrose solutions to give the required range of response. A sensitivity corresponding to a full-scale deflection of 0-06" rotation could be used satis- factorily. Under these conditions, the noise level, with the eluent flowing through the cell, was 2 per cent. of full-scale deflection. A further increase in the sensitivity resulted in a disproportionate rise in noise, e.g., at a full-scale deflection of 0.01" rotation, the noise level was 25 per cent. of full-scale deflection. The recorder was equipped with a multi-range switch that permitted measurements of between 100 V and 0-5 mV full-scale value and a facility for voltage suppression of 100400 per cent.in four ranges. In normal use, an input setting of 50 mV together with a shunt value of 23 C2 gave a full-scale sensitivity of 0.125" rotation,26 Analyst, Vol. 100 which was sufficient for most purposes. In addition, the voltage suppression facility was valuable when the column was inadvertently overloaded, or when the ratios of the minor components to the main component were large. Fig. 1 shows the adaptation to the standard cell that was necessary. The modification consists of a cylinder of non-magnetic 880 or EN 68 D grade stainless steel axially drilled to 0.052 in (1.3208 mm) diameter. This size was adopted as there was no advantage in using a hole with a diameter significantly different from that of the tube connecting the cell to the column.In order to reduce mixing of the column effluent, modifications were made to the ends of the cell so as to reduce the dead space to a minimum. The modified cell fits snugly into the existing jacketed cell holder. As the optical path length of the cell is unchanged, the calibration of the instrument was virtually unaffected. The time required for the system to reach a steady state (response time), for the cell described, was 5.6 s and it was found experimentally that the diameter of the optical path could be further reduced to 0.0280in (0.7112 mm), with a light path volume of 32 p1, without an appreciable increase in either instrument noise or response time.Fig. 2 shows the recording of a separation of a sample of neomycin complex using this device. Peaks corresponding to neomycins B and C and neamine can be seen; the fourth peak is probably due to paromamine.4 DE ROSSI: APPLICATION TO NEOMYCIN ANALYSIS OF A - Micro-cell - Spacer end seal washer Micro-cel I 'internal surfaces finely finished I Polarimeter cell assembly. Washers, cover,slips and additional spacers omitted for clarity Fig. 1. Polarimeter micro-cell and cell assembly. In order to interpret the analysis on a quantitative mass basis, specific rotations for Areas were neomycins B and C and neamine as free base, given in Table I, were used. measured using a planimeter.January, 1975 CONTINUOUS FLOW CELL FOR USE WITH AN AUTOMATIC POLARIMETER 27 0.0750 0-0625 i $ 0.0500 9 0,0375 0 .- w e - m a -2 0.0250 0 0.0125 0 7 6 5 4 3 2 1 0 Time/h Fig.2. Analysis of a commercial sample of neomycin complex; conditions as for Table 11. Peak A, neamine; peak B, neomycin B; and peak C, neomycin C. The reproducibility of the method is demonstrated in Table I1 and results of two experiments using a fraction collector and ninhydrin development are included for compari- son. Results obtained when artificial mixtures of known composition were examined are shown in Table I11 and indicate the accuracy of recovery that is possible. TABLE I VALUES OF SPECIFIC ROTATION FOR NEOMYCIN B, C AND NEAMINE AS FREE BASE Mean values based on reports6.a and own observations. Compound Specific rotation Neomycin B . . .. .. + 80" Neomycin C . . .... + 120" Neamine .. .. .. +112" TABLE I1 REPLICATE DETERMINATIONS OF COMPOSITION OF THE SECOND INTERNATIONAL REFERENCE PREPARATION OF NEOMYCIN 8-0 ml of Dowex AG-1x2 (hydroxyl form), column, Pharmacia k9/15 (0.9 x 15 cm). Load 75 mg (all experiments). Elution with C0,-free distilled water. Elution rate: 4.5 ml h-1. Composition of total base, per cent. Experiment Detection 1 Optical rotation 2 Optical rotation 3 Optical rotation 4 Optical rotation 5 Optical rotation 6 Ninhydrin 7 Ninhydrin Neaniine <0.5 < 0.5 < 0.5 < 0.5 (0.5 < 0.5 < 0.5 Neomycin C 8.3 8-5 7.8 8.5 8.4 8.3 8.1 Neomycin B 91.7 91.5 92-2 9 1.5 91-6 91.7 91.9 The only problem that has arisen when using the micro-cell resulted from the formation of gas bubbles within the optical path; this problem was avoided by using freshly boiled, carbon dioxide free distilled water and maintaining the temperature.of the cell at several degrees below ambient by circulating cold water through the jacket. The sensitivity of the system is, to a large extent, dependent on the specific rotation of the material under examination; a neamine content of 0-5-1-0 per cent. in a 75-mg sample28 D E ROSS1 TABLE I11 RECOVERY OF NEOMYCIN COMPONENTS FROM ARTIFICIAL MIXTURES OF NEOMYCINS Known composition, per cent. Amount recovered, per cent. Neomycin B Neomycin C A t > 95.0 5.0 95.1 4.9 *&Z&K-i?FZ&ZC 90.0 80.0 70.0 60.0 10.0 91.9 8.1 20.0 82.2 17.8 30.0 69.8 30-2 40.0 60-3 39.7 of neomycin was easily detected. Current work in this laboratory is directed to the possibility of using the micro-cell as a non-destructive flow analyser in the high-pressure liquid chromato- graphy of optically active materials. The skilful technical assistance given by Mr. B. E. Watts during this work is gratefully acknowledged . References 1. 2. 3. WHO Expert Committee on Biological Standardisation, 1970, Tech. Refl. Ser. WZd HZdh Org., 1971, No. 463, p. 11. “British Pharmacopoeia 1973,” H.M. Stationery Office, London, 1973, p. A98. Gillet, A., Vanderhaeghe, H., Bogaerts, R., Boudru, I., Brouckaert, A., Coucke, G., Dony, (2.. Drion, P., Dumont, P., Haemers, A., Pijck, J., and Van Kerchove, C., J . Phavm. Belg., 1972, No. 27, 381. Hessler, E. J., Jahnke, H. K., Robertson, J. H., Tsuji, K., Rinehart, K. L., and Shier, W. T., J . Antibiot., Tokyo, 1970, 23, 464. Ford, J. H., Bergy, M. E., Brooks, A. A., Garratt, E. R., Alberti, J., Dyer, J . R., and Carter, H. E., J . Amer. Chem. SOC., 1955, 77, 5311. “The Merck Index,” Eighth Edition, Merck and Co. Inc., Rahtvay, N.J., 1968, p. 721. 4. 5. 6. Received Muy 31st, 1974 Accepted Septembw 2nd. 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000025

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, January, 1975, Vol. 100, $9. 29-32 29 Identification of C.I. Solvent Red 24 in Hydrocarbon Oil in Oil Pollution Investigations Mixtures as an Aid P. J. Matthews" Scimtific Branch, Greater London Council, CouMty Hall, I-ondon, S.E. 1 Mixtures of diesel fuel oil and other hydrocarbon oils are often found in sewers. When such oil mixtures are dirty, and contain used lubricating oils and gas oils, the identification of rebate markers is difficult. A method has been developed in which the gas oil red marker dye C.I. Solvent Red 24 is separated by dry column Chromatography using nylon tubing. The dye is identified subsequently by thin-layer chromatography. Quinizarin, furfuraldehyde and C.I. Solvent Red 24 are used as markers in certain oils exempted from H.M. Custonis and Excise Duty, such as gas oi1.l A premium paraffin oil also contains C.I. Solvent Red 24.2 Often, diesel fuel oils are identified in oil samples taken from water-pollution incident^,^ and it is of interest to the pollution control chemist to identify accurately the source of contamination.The demonstration of the presence of markers can indicate whether a diesel fuel is a derv oil (not exempt from duty) or a gas oil, which can be of significant help in locating the origin of a discharge. The current methods of identification of hydrocarbon oils by added markers have been reviewed by Field.* Unfortunately, as diesel oils causing pollution are often waste materials, they can be mixed with other oils, particularly lubricating oik3 Spillages of fresh diesel oil may become contaminated with other oils and natural materials.A method has been described4 in which a solution of contaminated gas oil in light petroleum is run down a column of sodium car- bonate. The dye is retained at the toy and the hydrocarbon oils washed off with light petroleum, the quinizarin is removed by chloroform, the solution evaporated and the residue dissolved in cyclohexane. The quinizarin is then identified spectrophot~metrica~ly.~ If the diesel fuel contains significant amounts of used lubricating oil, degradation products of the latter interfere in the spectrophotometric determination of quinizarin. Alkaline extraction5 is unsatisfactory when detergents are present in the mixture, as they give rise to emulsions. The identification of furfuraldehyde is not satisfactory, as it is relatively unstable and may be lost.Experimental As column chromatography using a sodium carbonate solid phase has been described el~ewhere,~ it was decided to investigate a1 ternative column-chromatographic systems. Choice of System Several solid phases were tried in dry glass columns. These phases included silica gel, aluminium oxide, cellulose, starch, Ekoperl, boric acid, sodium sulphate, sodium carbonate and activated charcoal. Each system was tested with mixtures of gas oil and used lubricating oil. A 1OW-30 lubricating oil was used wliich had been reinoved from an automobile engine afte? 3000 miles running. The mixtures were introduced on to the column and then eluted with light petroleum (boiling range 40-60 "C); an orange band at the top of the column was removed with ethanol.Each experiment was repeated with tlie use of chloroform instead of ethanol. The best results were obtained with silica gel. The dye MYIS retained at the top of the column after washing with light petroleum (boiling range 40-60 "C). The colour of the dye band was orange, but the absorption spectra of the ethanol and chloroform eluates were not suitable for quantitative use. Similarly, the spectra of cyclohexane solutions of evaporation residues of such eluates were unsatisfactory, because the spectra of the degrada- tion products of the lubricating oil interfered with the spectrum of the quinizarin. Walk, Huntingdon, PElS 6NZ. * Present address : Scientific Directorate, Anglian Water Authority, Diploma House, Grammar School30 Analyst, Vol.100 Loev and Goodman6 suggested the use of nylon film tubing for columns, so that fractions could be cut out of the column. A 6 x 1-in nylon column of adsorption silica gel was therefore prepared, and about 5 ml of 25 per cent. lubricating oil in gas oil were placed on the medium. The column was eluted thoroughly with light petroleum (boiling range 40-60 "C). The pink - orange band at the top of the column was cut off and placed in a beaker. The silica gel was washed with a small volume of acetone, which was decanted off, evaporated to dryness and a small volume of light petroleum (boiling range 40-GO "C) added. The mixture was then poured on to another 6 x 1-in nylon column of silica gel and the system eluted with light petroleum (boiling range 40-60 "C).An orange band was observed at the top of the column. The initial procedure with the first column was repeated, but the acetone extract was evaporated to a small volume and the resulting solution was then chromatographed using the system of Hausser' with a silica gel G thin layer and benzene. The pink - purple colour of the marker showed the clearest separation from the yellow interfering components when the solvent was run for a distance of 6 cm. The uncontaminated marked oil gave rise to a pink-purple spot at approximately the same RF value as the mixture and C.I. Solvent Red 24. Quinizarin gave a different result (see Table I). Several mixtures were investigated and the results are shown in Table I. MATTHEWS: IDENTIFICATION OF C.I.SOLVENT RED 24 TABLE I THIN-LAYER CHROMATOGRAPHIC CHARACTERISTICS OF THE RED DYE ISOLATED FROM CERTAIN HYDROCARBON MIXTURES Sample Marked gas oil .. .. .. .. .. .. Marked (pink) kerosene . . . . .. .. .. 25 per cent. used lubricating oil in marked gas oil . . 26 per cent. used lubricating oil in unmarked derv oil . . 26 per cent. used lubricating oil in kerosene (unmarked). . 26 per cent. used lubricating oil in kerosene (marked pink) 26 per cent. unused lubricating oil in marked gas oil 26 per cent. unused lubricating oil in unmarked derv oil 76 per cent. used lubricating oil in marked gas oil . . . . 26 per cent. light fuel oil in marked gas oil . . .. Quinizarin . . .. .. .. .. .. .. C.I. Solvent Red 24 . . .. .. .. * . .. Observation Purple - pink spot .. .. Faint blue spot . . . . Purple - pink spot . . .. Purple - pink spot . . .. Orange spot . . .. .. Yellow spot . . .. .. Orange spot . . .. .. Faint blue spot . . .. Orange spot . . .. .. Purple - pink spot . . .. Purple - pink spot . . .. Yellow spot . . .. .. Faint purple - pink spot Purple - pink spot . . .. Continuous yellow streaking to solvent front . . .. Orange spot . . .. .. Purple - pink spot . . . . . . RF 0.47 0.66 0.47 0.63 0.68 0.60 0.66 0.86 0.66 0.47 0.50 0.47 0.63 0.60 - 0.61 0.47 Reagents and Materials Solvents Light petroleum, boiling range 40-60 "C. Acetone. Benzene or toluene. Solid phases Silica gel, 30-120 mesh, as supplied by BDH Chemicals. Plates of dimensions 10 x 20 cm with a 250-pm layer using 30 g of silica gel G and 60 ml of water slurry.Standards Gas oil (marked with C.I. Solvent Red 24). Used motor lubricating oil (3000-5000 miles wear). C.I. Solven,t Red 24. Prepare a 0.002 per cent. m/V solution in acetone. Nylon tubing (1 in diameter) to be used for dry column chromatography, as supplied by Tubing Walter Coles, 48 Tanner Street, London, S.E.I.January, 19 75 IN HYDROCARBON OIL MIXTURES 31 Procedure Cut off 7 in of the nylon tubing and seal one end with a match flame. Pack the column tightly with the 30-120-mesh silica gel up to a height of 6 in. Make a small nick as an air hole at the sealed end. Pour 5 ml of oil (or less, if only small amounts are available) on to the column and allow it to percolate into the silica gel. Elute the column with about 10-15ml (or as much as is needed to remove the hydrocarbon oils) of light petroleum (boiling range 40-60 "C).Cut the top 1 in off the silica gel column and place it in a beaker, then wash the silica gel with about 30 ml of acetone. Decant or filter off the acetone and evaporate it to about 0.5 ml. This procedure should be followed for the sample and for standard mixtures of used lubricating oil in marked gas oil for 10,25,50,75 and 90 per cent. solutions. When the sample proportions are known from other techniques, a similarly proportioned standard should be prepared. For qualitative identification of a mixture, a comparison with a 25 per cent. used lubricating oil in gas oil standard should be employed. Spot the same volume of the acetone solutions of the residues from samples and standards on to a silica gel G thin-layer plate, which has been scored 6cm from the starting line.Also, spot the plate with a 0.002 per cent. m/V solution of C.I. Solvent Red 24 in acetone. Place the plate in benzene contained in a standard thin-layer chromatographic tank and allow the solvent to ascend 6 cm. Remove the plate, dry it and measure the RF values of any purple - pink spots that may have developed in samples and standards. There has been some discussion on the toxicity of benzene and the hazards of its use in the laboratory; for this reason, it is recommended that toluene should be used in place of benzene (RF values for Solvent Red 24: benzene, 0.49; toluene, 0 ~ 4 3 ) . ~ Stability of C.I. Solvent Red 24 A 1 + 1 mixture of used lubricating oil and gas oil was allowed to stand for 21 days.No difference was observed in the appearance or intensity of the thin-layer chromatographic spot. A 50-ml volume of gas oil was allowed to percolate down a 20 cm x 3 cm diameter column of a sandy garden soil; no detectable adsorption occurred. However, 1.0 g of powdered activated charcoal readily adsorbed the red dye from 50ml of gas oil. Applicability to environmental sam$des The results in Tables I and I1 suggest that C.I. Solvent Red 24 may be used to give further information on the source of a diesel-type fuel. The results compare well. The RF values of the dye in certain samples containing used lubricating oil are slightly higher than those for uncontaminated gas oil and the dye itself, owing to the effect of the inter- fering fractions, with higher RB values, dragging the dye spot forward very slightly. TABLE I1 ANALYSIS OF DIESEL FUEL OIL - USED LUBRICATING OIL MIXTURES TAKEN FROM SEWERS* Results and Discussion Sample Observation R F Nature of the diesel fuel oil 1 2 3 4 5 6 7, 8, 9 10 11 12 Purple - pink spot 0.47 Purple - pink spot 0.48 Very faint purple - pink spot 0-47 Purple - pink spot 0.47 Purple - pink spot 0.45 Purple - pink spots 0.50 Purple - pink spot 0.47 No obvious spot - No obvious spot - No obvious spot - Marked oil Marked oil Unmarked oil with a small amount of marked oil present Marked oil Unmarked oil Marked oil Marked oils Unmarked oil Marked oil Unmarked oil * Identified by thin-layer chromatographya and gas chromatographyg ; only small amounts of used lubricating oil present (all < 25 per cent.).When an oil has been identified as kerosene by thin-layer and gas chromatography, the presence of the pink dye would indicate a particular brand. However, in the presence of a mixture of diesel fuel oil and kerosene, it is not possible to assign the dye to either component. This is one specific instance, and has not been encountered in the author's32 MATTHE WS experience. Gas oil has been isolated many times (see Table 11). C.I. Solvent Red 24 appears to be sufficiently stable to persist in samples taken under field conditions, and does not appear to be affected when gas oil is mixed with used lubricating oil. In such samples, C.I. Sol- vent Red 24 would appear to be more easily 'identified than quinizarin.It is significant that it is still possible to identify the red dye even in the presence of a light fuel oil, although 50ml of light petroleum (boiling range 40-60 "C) were needed in order to wash the silica gel column. Other red dyes are used in a small number of specialised oils, such as Garia H cutting oil. In the opinion of the author, the possibility of their occurring as waste materials, particularly mixed with derv oil, causing water pollution, is remote. The presence of C.I. Solvent Red 24, therefore, can be taken as indicating in almost all instances that a gas oil is present. Conclusion Marked gas oils that contain C.I. Solvent Red 24 can be identified readily in the presence of heavier hydrocarbon oils by chromatography of the dye. The method has been tried and used successfully by the GLC Scientific Branchlo and by the Area Laboratory of the Research and Development Division of the British Railways Board. The author thanks the Scientific Adviser to the Greater London Council, Mr. R. T. Kelly, for permission to publish this paper. The views expressed are not necessarily those of the Scientific Adviser, the Greater London Council, or any of its Departments. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. References Tucker, K. B. E., Sawyer, R., and Stockwell, P. B., Analyst, 1970, 95, 730. Fudge, R., and Nicholas, P. V., J . Ass. Publ. Analysts, 1970, 8, 35. Matthews, P. J., J . APFl. Chem., Lond., 1970, 20, 87 and 228. Field, K., PYOC. SOC. Analyt. Chem., 1970, 7, 10. Field, K., and Godly, E. W., Analyst, 1966, 91, 287. Loev, B., and Goodman, M. M., Cherny Ind., 1967, 48, 2026. Hawser, H., Arch. Kriminol., 1960, 125, 72. Adam, J., Chemy Britain, 1973, 3, 133. Ellerker, R., Dee, H. J., Lax, P. G. I., and Sargent, D. A., Wat. Pollut. Control, 1968, 67, 542. A . Rep. Scient. Advis. Greater London Council, 1972, 56. Received July 29th, 1974 Accepted September 2nd 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000029

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, Janztary, 1975, Vol. 100, $$. 33-38 33 Di-2-pyridyl Ketone Azine as an Analytical Reagent and its Application to the Determination of Iron(ll) M. Valcarcel, M. P. Martinez and F. Pino I.efiarttnend of Analytical ChemistYy, Univevsitv of .Seville, Sevilie, Spain The synthesis, characteristics and analytical applications of di-2-pyridyl ketone azine are described. This compound reacts with iron(I1) to produce a red 3: 1 complex (Am,,. 460 nm, E = 9.3 x lo3 1 mol-1 cm-l in aqueous ethanolic solution) and a green 3: 1 complex (Ama,. 750 mi, E = 4.5 x lo3 1 mol-l cm-l in aqueous ethanolic solution and 6.4 x l o 3 1 mol-l cin-l in chloroform). The green complex, extracted into chloroform, has been used for the spectrophotometric determination of trace amounts of iron. The method is selective and has been applied to the determination of iron in some industrial waste waters.Oxinies have often been used as analytical reagents. In particular, di-2-pyridyl ketone oxime has been proposed by Holland and co-workers for the photomctric detection of cobalt,l iron2 and palladium,3 and has also been used as a gravimetric reagent for the determination of palladi~m.~ Thiosemicarbazone, hydrazone and acyclic azine derivatives of the same ketone, however, have not been used as reagents. Studies on the synthesis, properties and application of di-2-pyridyl ketone azine (DPKA) as a reagent for the determination of iron(I1) are now reported. Di -2- pyridyl ketone azine Experimental Synthesis of Di-2-pyridyl Ketone Azine A 0-27-ml volume of 99.9 per cent.mlm hydrazinc hydrate was added to 2 g of di-2- pyridyl ketone (Aldrich) dissolved in 12 rnl of hot absolute ethanol. Two drops of concen- trated hydrochloric acid were added and the mixture was refluxed for 1 hour. The mixture was then cooled to 0 "C and the yellow crystals that separated out were filtered off and re- crystallised twice from hot ethanol - water (1 + 1). The crystals obtained (yield 55 per cent.} melted at 192 "C. On analysis their elemental content was found to be C 72.4, €1 4.5 and N 23.3 per cent.; calculated for C22H,6N6: C 72.53, H 4-39 and N 23.08 per cent. Reagents All solutions were prepared with analytical-reagent grade chemicals using distilled water. Di-2-pyridyl ketone axine reagent solution, 0-1 per c e d . m/V in chloroform.This solution is stable for 1 week. Standard iron(11) solution, 2-006 g k - l of ivon(1I). Dissolve the appropriate amount of ammonium iron(I1) sulphate, (NH4),[Fe(S04),].6H,0, in dilute sulphuric acid and standardise it gravimetrically. From this solution prepare standards containing 100.3 p.p.m. of iron(I1). Ascwbic acid solution, 2 $er c m t . mjV. Prepare this solution daily.34 VALCARCEL et UZ. : DI-8-PYRIDYL KETONE AZINE AS AN Analyst, VOZ. 108 Bufer solution, pH 3.7. Prepare by dissolving 30.6 g of potassium hydrogen phthalate in water, adding 2 ml of concentrated hydrochloric acid and diluting the mixture to 1 litre. Apparatus (digital) spectrophotometer, equipped with 1-cm glass or quartz cells. Sj5ectrophotometers. Unicam SPSOO and SP600 spectrophotonieters and a Coleman 55 Digital pH meter.Philips PW9408, with glass - calomel electrodes. Procedure for the Determination of Iron( 11) To 25 to 200 ml of sample solution, containing up to 120 pg of iron, in a separating funnel, add 3 g of sodium perchlorate, 2 ml of 2 per cent. m/V ascorbic acid solution, 10 ml of buffer solution and 10 ml of 0.2 M potassium nitrate solution and extract the mixture with two 10-ml volumes of DPKA reagent solution. Shake the funnel vigorously each time for 2 minutes, allow the phases to separate and transfer the lower layers into a 25-nil flask contain- ing anhydrous sodium sulphate. Measure the absorbance of the combined green chloroform extracts in 1-cm cells against water at 750 nm. The calibration graph is prepared by using standard solutions containing from 0-2 to 1.2 ml of a POO-3 p.p.m.solution of iron(I1) (3. to 6 p.p.m. in 20 rnl of chloroform phase) treated in the same way. Results and Discussion Di-2-pyridyl Ketone Azine Reagent An aqueous reagent solution of 5.4 x 10-5 M concentration shows maximum absorption at 280 nm, with a molar absorptivity of 2.2 x lo5 1 mol-l cm-l. A simultaneous potentiometric photometric methods was used for the determination of the ionisation constant; the average pK value was found to be 3.45. This behaviour may be caused by protonation of one pyridinic nitrogen atom. The conventional photometric method cannot be used owing to hydrolysis of the reagent. DPKA has solubilities in chloroform, methanol, ethanol, nitrobenzene, acetone, 4-methylpentan-2-one and water of 90.1, 11.4, 6.8, 5.6, 5-0, 1.0 and 0.5 g 1-l, respectively.Solutions of the reagent at 0.1 per cent. m/V concentration in several solvents, e.g., water - ethanol, ethanol and chloroform, were stable for at least 1 week. Hydrolysis of the reagent to di-2-pyridyl ketone and hydrazine occurred in aqueous solution at pH values below 3. Cu(ll) {ye! tow-green) Fe(ll) Co(ll) Ni(ll) (brown), ,(yellow) (yellow) I 1 1 - 5 4 pH 9.5 , I , j l buffer 7 6 Cu( I I) C;(II) Ag(I) (brown -yellow) (ye1 low) (ye1 low} Fig. 1. pH values of DPKA reactions with metal ions. The reaction of the reagent in aqueous ethanolic solution with thirty cations at different pH values was investigated; it reacts with iron(II), copper(II), cobalt(II), nickel(II), paUa- dium(II), gold(III), silver(1) and zinc(II), as shown qualitatively in Fig.1.Janzcary, 1975 ANALYTICAL REAGENT FOR THE DETERMINATION OF IRON (11) 35 Formation and Study of Iron Complexes When dilute iron(I1) solutions and a 0.1 per cent. solution of DPKA are mixed, two different complexes, one green and the other red, are formed according to the pH. AbsGrption spectra of both complexes are shown in Fig. 2. The formation, interconversion and decom- position of these complexes are influenced strongly by the experimental conditions. Wavelengt h/n rn Fig. 2. Absorption spectra of iron complexes. Concentration of iron(II), 6 p.p.m. I, Red complex in aqueous ethanolic medium at pH 7, after 5 hours; 11, green complex in aqueous ethanolic medium a t pH 4.5; 111, green complex extracted into chloroform a t pH 2.9 ; and IV, reagent blank.Oxidation state of iron. From experimental evidence it was concluded that the reagent DPKA reacts with iron only in the bivalent state to give both complexes. In order to ensure that the iron is present as iron(II), ascorbic acid was selected for use as a reducing agent. Hydroxylammonium chloride or hydrazinium sulphate cannot be recommended because of the possible formation of di-2-pyridyl ketone oxime or hydrazone by interchange of >C=N- groups. InJuence of PH. Ten millilitres of a 0.1 per cent. solution of DPKA in 20 per cent. ethanol were allowed to react with 3-6 ml of a 100 p.p.m. iron(I1) solution at different pH values in a series of 50-ml calibrated flasks. The absorbances were recorded at 500 and 750 nm after 0, 30 and 60 minutes and are shown in Fig.3 (a and b). The optimum pH range for formation of the green complex is 4.2 to 5-1 and for the red complex 6.5 to 7.5. The reagent itself does not absorb at the wavelength of maximum absorption of its iron complexes. Formation and stability of the iron(II) complexes in aqueous ethanolic sohtion. The green complex is formed immediately at pH 4.5 in aqueous ethanolic solution, but the absorbance at 750 nm decreases with time, while the absorbance at 480 to 500 nm increases [see Fig. 4 (a)], the colour of the solution changing from green to orange. The absorbance of the red complex formed at pH 7 decreases by 10 per cent. after 1 to 5 hours and then remains stable. At temperatures of 50 to 70 "C the stability of the solution of the red complex at pH 7 is improved, but the instability of that of the green complex at pH 4.5 remains unchanged. It is concluded that in aqueous media the two complexes of iron(I1) with DPKA are not of great analytical interest.Extraction of the complexes. When a solution of DPKA reagent in an organic solvent is shaken with a weakly acidic aqueous solution of iron(II), the green complex is formed immediately in the organic phase. When nitrobenzene is used the resulting complex is, however, unstable, displaying the characteristics of the complex in aqueous ethanolic solution [Fig. 4 ( b ) ] . Solutions of the complex in chloroform are stable except in the presence of ethanol. The optimum pH range for the extraction of the green complex is 2.5 to 3.8 [Fig.The presence of potassium nitrate in the aqueous phase increases the stability of the green chloroform solutions (the absorbance at 750 nm is, however, constant for 1 hour in the 3 (43.36 VALCARCEL et d. : DI-2-PYRIDYL KETONE AZINE AS AN A%t&Zyst, VOZ. IQO absence of potassium nitrate) and sodium perchlorate (not less than 3 g> acts as a “salting out” agent. Two 2-minute extractions with 10 ml of the 0.1 per cent. m/V solution of DPKA in chloroform are necessary for the complete extraction of iron(I1) from aqueous solution. Solutions of the red complex formed in aqueous media at pH 7.0 are also extracted into organic solvents. When chloroform is used, however, six extractions are necessary although this complex is stable in the organic phase; the complex when extracted into nitrobenzene is unstable. 0% 0.7 0.6 0.5 0 c m 0.4 II 4 0-3 0.2 0.1 0 2 4 6 8 2 4 6 8 PH 2 4 6 8 Fig. 3.Influence of pH on the formation of iron complexes. (a) In aqueous etlnanolic solution (!) In aqueous ethanolic solution, at 750 nrn : at 800 nm: A, 0 hours; B, 0.5 hour; and C, 1 hour. A, 0 hours; B, 0.5 hour; and C, 1 hour. (c) Extracted into chloroform, at- 750 nm. SCoicheionietry of the complexes. Job’s curves were plotted for both complexes. ‘The absorbance of the green complex obtained at pH 3.6 was measured at 750 nm15 minutes after preparation. In orlder to determine the composition of the red complex, the samples, at pH 7-0, were heated for 15 minutes at 60 O C , and the absorbance was measured at 460 nm; the same molar ratio (3:l) was found (Fig.5, a). The complexes are possibly two isomeric forms. The red complex, formed completely at a higher temperature, is sterically hindered during its formation. The molar extinctions were evaluated by a statistical method. The red complex in aqueous ethanolic solution (pH 7.0) gave a value for E of 9.299 x lo3 1 mol-l cm-l (460 nm) and the green complex of 4.496 x lo3 I rno1-I cm-l (750 nm) [in the chlorofoi-m phase (20 m1) E = 6.367 x lo3 1 mol-l cm-l (750 nm)]. Results showed (Fig. 5, A) a stoicheiometric ratio of 3:l. Spectrophotometric Determination of Iron with DPKA Based on the foregoing experimental findings a method is proposed for the determination of trace amounts of iron involving the formation of the 3:l green complex with DPKA and its extraction into chloroform.January, 1975 ANALYTICAL REAGENT FOR THE DETERMINATION OF IRON (11) 37 The sensitivity of the method, according to Sandell, is 0.008 pg cm-2.The optimum concentration range, evaluated by Ringborn's method, is 2 to 6 p.p.m. of iron and the relative erra of the method is +0-S per cent. Beer's law is obeyed between 1 and 6 p.p.m. of iron(I1). I I I I I I Time/hours W C m -2 s1 -0 Q Fig. 4. Stability of coloured solutions. [Fe(ll)] + [DPKAI (a), 6 p.p.in. of iron(I1) at pH 4-5: A, h = 750 nrn ; and B, X = 500 nm. ( b ) , Extraction of coloured Composition of DPKA - iron(I1) solution (6 p.p.m.1 a t pH 4-5 into nitrobenzene: complexes by the continuous variations A, h = 750 nm and B, h == 500 nm; and into method: A, green complex (pH = 3.6, chloroform: C, h = 750 n m and D, h = 500 inn.X ;--1 750 nm) : and B, red complex (pH :-= 7.0, A = 460 nm). Fig. 5. For the determination of 3 p.p.m. of iron by this method, the ioreigii ions can be tolerated at the levels given in Table I. Copper(II), reduced by ascorbic acid to copper(I), interferes at an equal concentration, but when the mass ratio of iron to copper is 2, the latter does not interfere. EDTA also strongly interferes. TABLE I INFLUENCE OF FOREIGK IONS IN THE DETERMINATION OF IRON Ions that interfere at levels of ~~- ______A I_ 6p.p.m. 10 p.p.m. 50 ppm. Cu2+ EDTA Tartra-te 7 100 pp.m Niz+ Bi3+ SbS+ HEr;+ Ag+ Au3+ Pt4+ Ions that do not interfere up to 100 p.p.m. 1 Ca2 f SCN- As3+ Citrate Ce4+ C20,2- Mo6+ 17- V6+ P0,S- La3+ B610,2- Cr6+ s 2 0 3 2 - u022+ The method has been applied satisfactorily to some natural waters and industrial effluents from a sulphuric acid plant (pyrites process) anid fertiliser plants (phosphates).The results are compared in Table I1 with those obtained by the spectrophotometric 1, 10-phenanthroline method.38 VALCARCEL, MARTINEZ AND P I N 0 TABLE TI COMPARISON O F RESULTS FOR THE DETERMINATION OF IRON IN INDUSTRIAL WASTE WATERS BY THE DPKA AND ~,~O-PHENANTHROLINE METHODS Iron, p.p.m.* I River water I DPKA 1 , 10-Phenanthroline method method 0.0 1 0.0 1 0-03 0.06 0.0 1 0.01 0.0 1 0.0 1 0.005 0.0 1 0.005 0.02 0.02 0.02 Water from sulphuric acid plant -7 DPKA 1,lO-Phenanthroline method method 59.0 59.8 49.0 49.4 48-5 48.4 59.5 60.6 65.0 65.2 56.0 56.0 55.5 56.0 _ _ Water from fertiliser plant r-----A---7 DPKA 1,lO-Phenanthroline method method 0.005 0-01 0.43 0.44 0.02 0.03 0.02 0.02 0-005 0.02 0-03 0-03 0-15 0.14 * Results are the means of three determinations. References 1. 2. 3. 4. 5. 6. Holland, W. J., and Bozic, J., TaZanta, 1968, 15, 843. Holland, W. J., Bozic, J., and Gerard, J., Analpica Chim. Acta, 1968, 43, 417. Holland, W. J., and Bozic, J., AnaZyt. Chem., 1968, 40, 433. Holland, W. J., Bozic, J., and Gerard, J., Analyst, 1968, 93, 490. Muiioz-Leyva, J. A., and Pino, F., Infcidn. Quinz. Analit., 1973, 27, 67. Valcarcel, M., and Pino, F., TaZanta, 1973, 20, 224. Received March 25th, 1974 Accepted May lst, 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000033

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, January, 1975, Vol. 100, pp. 39-45 39 raniurn and a inophenol Shoji Motomizu Department of Chemistry, Faculty of Science, Okayawta University, Tsushima, Okayanza-shi, Japan Micro-amounts and trace amounts of cobalt in uranium oxide (U,O,), uranium and a uranyl salt were determined spectrophotometrically with 2-nitroso-5-dimethylaminophenol (nitroso-DMAP) reagent. According to the cobalt content (above 0.01 p.p.m.) of the sample, four methods for deter- mining cobalt are recommended. By use of one method, cobalt in a sample containing more than 5 p.p.m. of cobalt was determined in aqueous solution, while by use of the other three methods, cobalt in samples containing more than 0.5, 0.1 and 0.01 p.p.m., was determined by a solvent-extraction - spectrophotometric method.Cobalt is one of the most undesirable impurities in the fuel and materials of nuclear reactors because of its large cross-section of neutron absorption. Accordingly, its determination in uranium is very important. The cobalt in a sample of uranium has been determined spectro- photometrically by a variety of methods, such as spectrophotometric determination after the extraction of cobalt thiocyanate with tetraphenylarsonium chloride into chloroform,l after extraction of the cobalt complex with diethyldithiocarbamate,2-4 and with the nitroso-R salt after extraction of the cobalt complex.5~6 With these methods, however, it is not possible to determine micro-amounts of cobalt in uranium samples because of the time-consuming procedures, the instability of the cobalt complex and poor sensitivities.Recently, Hashitani, Yoshida and Adachi' determined cobalt above a level of 1 p.p.m. in a uranium sample by use of a solvent extraction - spectrophotometric method with l-nitroso-2-naphthol reagent. Previously, the author reported a method for determining micro-amounts of cobalt with 2-nitroso-5-dimethylaminophenol (nitroso-DMAP) and 2-nitroso-5-diethylaminophenol re- agents, and determined micro-amounts and trace amounts of cobalt in nickel salt^,^,^ iron and steel,*JO pure chemical reagent+ and sea water.12 The spectrophotometric methods for determining cobalt with these two reagents constituted a new, sensitive, rapid and very simple method. In this work, the author intended to develop a method by which micro-amounts and trace amounts of cobalt in uranium, uranium oxide (U,O,) and a uranyl salt could be determined spectrophotometrically.Therefore, he improved the previously reported method making use of nitroso-DMAP, and applied this improved method to the determination of micro- amounts and trace amounts of cobalt in uranium samples. By use of this method, which is described in this paper, 0.01 p.p.m. of cobalt in uranium samples was determined. Experimental Reagents All of the reagents used were of analytical-reagent grade. 2-Nitvoso-5-dimethylam~nopkenol (nitroso-DMAP), 0.2 per cent. 2-Nitroso-5-dimethyl- aminophenol hydrochloride, obtained from the Tokyo Kasei Kogyo Co. Ltd., was used. Dissolve 0-2 g of the nitroso-DRIAP hydrochloride in 0.1 N hydrochloric acid. This nitroso- DMAP solution can be kept for at least 1 month.Prepare a stock solution (containing 0.3 mg ml-l of cobalt) from cobalt chloride hexahydrate, and standardise the solution by titration with EDTA. Before use, dilute this solution accurately with 0.01 N nitric acid to the concentrations required for the various procedures. Standard cobalt solution.40 Analyst, VoZ. 100 B u f e r solution A . Dissolve 42 g of citric acid monohydrate and 20 g of sodiuni hydroxide in water, and dilute the solution to 100 ml with water. Bz@er solution B. Dissolve 200 g of trisoclium citrate dihydrate in water. To this solution, add 20 ml of 0.2 per cent. nitroso-DMAP solution and dilute the mixture to 500 ml with water. After 1 hour, remove any complexed cobalt and the excess of reagent by ex- tracting three times with 20 ml of 1,2-dichloroethane and twice with 20 ml of benzene.Then filter the aqueous phase through a dry filter-paper. This filtrate will now contain only trace amounts of cobalt (below 0.01 pg per 10 ml). MOTOMIZU : SPECTROPHOTOMETRIC DETERMINATION OF COBALT IN Extraction solvent. 1 ,Z-Dichloroethane was used without further purification. Apparatus A Hitaclii - Perkin-Elmer, Model 139, spectrophotometer, equipped with standard cell holder (10-mm path length) and a micro-scale cell holder (50-mm path length, 0.7-ml capacity), was used for measuring absorbances. An Iwaki, niodcl KM, shaker was also used. Preparation of Sample Solution To this were added 4 ml of concentrated nitric acid and the solid was dissolved by heating the acid.After dissolution, the liquid was evaporated nearly to dryness. The residue was again dissolved and f he solution diluted to the required volume with distilled water. In order to determine the amount of cobalt in the concentrated nitric acid used in this work, 10 ml of the concentrated nitric acid were transferred by pipette into a beaker and evaporated nearly to dryness, then the cobalt in the acid was determined by use of Procedure l IIb (mentioned below). The absorbance of the cobalt complex measured against the solvent was 0.009, which was almost the same value as that of the reagent blank. Thus, the amount of cobalt in this acid was not detectable. The volume of the acid used in dissolving the samples was also examined; when 4 to 10 ml of the acid were used, no difference in the cobalt content of the sample determined was observed.The required amount of sample was weighed into a 50-ml beaker. Procedures According to the cobalt content of the sample, one of the following three procedures was used. Procedure I , for samples containing more than 5 p.9.m. of cobalt. This method was estab- lished by improving a previous method,8 and was applied to the determination of cobalt in a uranium sample eontaining cobalt at a level above 5 p.p.m. When 3 g of sample containing 5 p.p.rn. of cobalt were dissolved in 100 ml of water and 15 ml of the solution (i.e., a solution containing 0.15 pg ml-1 of cobalt) were used, the absorbance of the cobalt complex was about 0-05 Zterws the reagent blank. Transfer up to 15 ml of the sample solution (less than 40 pg of cobalt) by pipette into a 25-ml calibrated flask.To this, add 2-5 ml of buffer solution A, mix the solutions thoroughly and allow the mixture to stand for 10 minutes. Add 2 ml of nitroso-DMAP solution, again mix thoroughly and allow to stand for 10 minutes. Next, add 5 ml of 1 + 1 hydrochloric acid solution and dilute to the mark with water. Rleasure the absorbance at 530 nm in a glass cell of 10 mm path length against a reagent blank prepared as follows. Transfer equal volumes of sample solution, buffer solution A and 1 + 1 hydrochloric acid into a 25-ml calibrated flask and mix them thoroughly. To this mixture, add 2 ml of nitroso-DMAP solution and dilute to the mark with water. The calibration graph was prepared by use of the same procedure and was linear, passing through the origin, in the range from 7 to 40 pg of cobalt.The slope of the calibration graph was 0.218 absorbance unit per 10 pg of cobalt. Procedwe I I , for samples containing moYe than 0.5 j5.p.m. of cobalt. This is the most used procedure, and it was established by improving the previous meth~d.~sl~ The procedure was applied to the determination of cobalt in a sample containing cobalt at a level above 0.5 p.p.m. When 5 g of sample containing 0-5 p.p.m. of cobalt were dissolved in water to yield 100 nil of solution and 10 ml of the solution, which contained 0.025 pg ml-1 of cobalt, were taken, the absorbance of the cobalt complex was about 0.05 against the reagent blank. The method is as follows. The method is as follows.January, 1975 URANIUM COMPOUNDS WITH 2-NITROSO-&DIMETHYLAMINOPHENOL 41 Transfer up to 10 ml of the sample solution (less than 4 pg of cobalt) by pipette into a 25-ml, glass-stoppered test-tube, add distilled water and dilute to 10 ml.To this mixture add 2 ml of buffer solution A, mix thoroughly and allow the mixture to stand for 10 minutes. Add 1 ml of nitroso-DMAP solution, again mix and allow to stand for 10 minutes. Transfer 5 ml of 1,2-dichloroethane into this tube with a pipette. Then shake the tube for 1 minute by hand. After the two phases have separated, discard the aqueous phase and wash the organic phase with 5 ml of 1 + 2 hydrochloric acid by shaking for 30 s. Discard the aqueous phase, wash the inside of the tube with 5 ml of water and then again discard the aqueous phase.Finally, wash the organic phase with 5 ml of 1 N sodium hydroxide solution by shaking for 30 s, filter the organic phase through a dry filter-paper (about 5 cm in diameter) and measure the absorbance at 456 nm in a glass cell of 10-mm path length against the reagent blank. The calibration graph was prepared as above and was linear, passing through the origin, in the range 0.2 to 4 pg of cobalt. Its slope was 0.230 absorbance unit per 1 pg of cobalt. The absorbance of the reagent blank against the solvent was about 0.007. Procedure I I I , fGr samples with more than 0-1 or 0.01 p.9.m. of cobdt. As the distribution ratio between water and 1,2-dichloroethane of the cobalt - nitroso-DMAP complex was very high,13J4 the concentration of the metallic impurity by solvent extraction could be carried out very easily.The absorbances of the cobalt complex in l,%dichloroethane were measured in a cell of either 10-mm path length (Procedure IIIa) or of 50-mm path length (Procedure IIIb). When 2.5 g of a sample containing 0.1 p.p.m. (Procedure IIIa) and 0-01 p.p.m. (Procedure IIIb) of cobalt were used (Le., 50 ml of solutions containing 0.005 and 0.0005 pg ml-l of cobalt, respectively), the absorbance of the cobalt complex was about 0.05 by Procedure IIIa and about 0.03 by Procedure IIIb, in both instances versus the reagent blank. The method is as follows. Transfer up to 50 ml of the sample solution (less than 3.5 pg of cobalt in Procedure IIIa and less than 0.7 pg in Procedure IIIb) by pipette into a lOO-ml, glass-stoppered test-tube or separating funnel, add distilled water and dilute to 50 ml.To this solution add 10 ml of buffer solution B, mix thoroughly and allow the mixture to stand for 10 minutes. Add 5 ml of nitroso-DMAP solution, again mix and allow to stand for 10 minutes. Next add, by means of a pipette, 5 ml of 1,2-dichloroethane to the mixture, and shake it for 10 minutes with a mechanical shaker. After the two phases have separated, discard the aqueous phase and wash the organic phase with 5 ml of 1 + 2 hydrochloric acid solution by shaking for 30 s. Then discard the aqueous phase, and wash the inside of the vessel with 5 ml of water, discarding the aqueous phase once more. Repeat these washings with hydrochloric acid solution and water.Finally, wash the organic phase with 5 ml of 1 N sodium hydroxide solution by shaking for 30 s; filter the organic phase through a dry filter-paper. Measure the absorbance of the filtrate at 456 nm in a glass cell of either 10-mm path length (Procedure IIIa) or 50-mm path length (Procedure IIIb) against the reagent blank. The calibration graph was prepared by use of the same procedure and was linear, passing through the origin, in the range from 0.2 to 4 pg (IIIa) and 0.05 to 0.7 pg (IIIb) of cobalt. The slope of the calibration graph obtained was 0.250 absorbance unit per 1 pg of cobalt in Procedure IIIa, and 0.113 absorbance unit per 0.1 pg of cobalt in Procedure IIIb. The absorbance of the reagent blank against the solvent was about 0.003 in Procedure IIIa and about 0.008 in Procedure IIIb.When 10 ml of the other buffer solution, A, and trisodium citrate solution (1.5 M), the cobalt in which had not been removed, were used in Procedure IIIb, the absorbances of the reagent blank against the solvent were 0.016 and 0.042, respec- tively. From these results, the use of buffer solution B, from which cobalt had been removed before use, was considered to be preferable in determining trace amounts of cobalt by use of Procedure IIIb. In determining cobalt by means of Procedure IIIa, buffer solution A can be used, as the absorbance of the reagent blank against the solvent is about 0-005. In this work, the pro- cedures were carried out at pH 4 to 8 as a precaution, and the excess of acid, which is added in dissolving the sample, must be removed by evaporation almost to dryness.When the pH of the sample solution is very low (below about pH 1) and it is difficult to adjust the pH by adding the above-mentioned volume of the buffer solution, a two-fold amount of the buffer solution can be added to the sample solution and the reagent blank. When relatively large amounts of cobalt are to be determined the adjustment of the pH can be carried out by adding small amounts of ammonia solution. Large amounts of ammonia solution, however, Thus in this method, ten-fold concentration was achieved. In all of the above procedures the optimal pH region is 3 to 8.42 MOTOMIZU : SPECTROPHOTOMETRIC DETERMINATION OF COBALT IN Analyst, vol. 100 must not be added, especially in Procedure IIIb, because commercially available ammonia solution contains about 7 x lo-' per cent.m/V of cobalt.ll Results and Discussion Effect of Co-existing Ions In considering the results of determinations of the impurities in uranium oxide that are listed in JAERI 4053,6 the effect of metal ions was examined. When silver, cadmium, lead and vanadium(V) ions at a concentration of 5 x 10-5 mol l-l, chromium(III), copper, mag- nesium, calcium, manganese(II), molybdenurn(V1) and zinc ions a t a concentration of l o 4 moll-1, aluminium and iron(II1) ions at a concentration of 5 x 10-4 mol l-l, nickel ions at a concentration of moll-1 and uranyl ions at a concentration of 0.2 mol 1-1 were co-existing with cobalt, these metal ions did not interfere in the determination of cobalt by Procedures I, I1 and 111.However, when tin(I1) ions at a concentration of 1 x moll-1 were co- existing with cobalt, they caused a small negative error of less than 3 per cent. As the amounts of each metal contained in uranium samples were much smaller than those examined, and as any tin in the sample was oxidised by dissolution in nitric acid to metastannic acid, co-existing metals did not interfere. The uranyl ion, at concentrations below 0.2 M, was also found not to interfere. Accordingly, in each procedure, about 5 g of sample per 100 ml of solution can be taken. Determination of Cobalt in Uranium Oxide (U,O,) The results obtained by use of the proposed methods, I, 11, IIIa and IIIb, for micro- amounts of cobalt in a JAERI-U2 sample are given in Table I. From Table I, the results for the cobalt content of JAERI-U2 when determined by the various procedures were in good agreement with each other and were 6.5 p.p.m.in U,O, or 7.7 p.p.m. in U on average. TABLE I DETERMINATION OF COBALT IN JAERI-U2 Procedure I I1 IIIa IIIb Sample solution Sample solution 1.912 g per 60 ml 5 (Two determinations 10 (Two determinations) concentration taken/ml Absorbance* 0.030 f 0.001 0.054 f O*OOO 0-147 &-0*003 0.290 f 0.006 0.9940 g per 60 ml 5 (Three determinations) 10 (Three determinations) 0.4910 g per 100 ml 10 (Three determinations) 20 (Three determinations) 0.5012 g per 100 d 5 (Three determinations) 10 (Three determinations) 15 0.554 0*078f0.001 0*152f0*000 0.183 f 0.003 0-367f0.001 * Reference : reagent blank. Cobalt, p.p.m. c In U,O, In U 7 .l f 0 . 0 8-4&-0.0 6.6f0.0 7.8f0.0 6.4f0-1 7.6f0.1 6-4f0.1 7*5&0.1 6.4 f 0.0 7.6 f 0.0 6.4 f. 0-0 7.5 f 0.0 6.4f0.1 7.5f0-1 6.4fO-1 7*5&-0*1 6.5 7.6 According to the results listed in JAERI 4053,6 the cobalt content of JAERI-U2 was 7.0 to 10.5 p.p.m. (8.7 p.p.m. in U on average). In the method used in reference 6, the sample was dissolved in concentrated nitric acid, which was then neutralised with ammonium citrate solution (30 per cent.) and ammonia solution. To this solution, 2 in1 of 2 per cent. sodium diethyldithiocarbamate were added, and the cobalt was extracted into 5 ml of carbon tetra- chloride. After evaporating off the organic solvent, the cobalt was determined by use of the nitroso-R salt. The result obtained by using the extraction with tributyl phosphate and an atomic-absorption method in the same reference was 7-0 p.p.m.The result obtained by using l-nitroso-2-naphthol' was 7.9 p.p.m. of cobalt in U and was in good agreement withJanuary, 1975 URANIUM COMPOUNDS WITH 2-NITROSO-5-DIMETHYLAMINOPHENOL 43 the author’s result. Accordingly, the results obtained by the proposed methods are con- sidered to be reasonable. The results obtained by use of proposed methods 11, IIIa and IIIb, for micro-amounts of cobalt in JAERI-U1 (No. 4) sample, are given in Table 11. From this table the cobalt content of JAERI-U1 (No. 4) determined by each of the three procedures can be seen to be in good agreement and was 0.85 p.p.m. in U,O, or 1.0 p.p.m. in U on average. According to the results listed in JAERI 4053, the cobalt content of JAERI-Ul (No.4) was 1.1 to 2.0 p.p.m. (1.5 p.p.m. in U on average). These results were obtained by use of the method mentioned above, that is, extraction with sodium diethyldithiocarbamate and spectro- photometric determination with nitroso-R salt. The result obtained by using l-nitroso-2- naphthol7 was 1 p.p.m. of cobalt in U and was in good agreement with the author’s result. TABLE I1 DETERMINATION OF COBALT IN JAERI-U1 (No. 4) Procedure Sample solution Sample solution Absorbance* r Cobalt, p.p.m. concentration takenlml In U,O, In U I1 4-935 g per 100 ml 6 (Five determinations) 10 (Five determinations) 0*049f0*001 0*85&0.01 0.99&0*01 0.099 f 0.001 0.87f0.01 1.02f0.01 0.109 f 0.001 0.84f 0.00 0.98 & 0.00 0.2 16 f 0.004 0.85 f 0.01 0*99& 0.01 0.265 f 0,001 0*89&0*00 1*04&0*00 0.518f0.005 0.91 k0.01 1*06&0*02 5 (Three determinations) 10 (Three determinations) 6 (Four determinations) 10 (Four determinations) * Reference : reagent blank.IIIa IIIb 6.132 g per 50 ml 5.039 g per 100 ml In Table 111, the results obtained by using the proposed method, IIIb, for trace amounts of cobalt in JAERI-U1 (No. 5) sample are given. As the cobalt content of this sample was very small, cobalt could be determined only by use of method IIIb. From Table 111, the cobalt content of JAERI-U1 (No. 5) was 0.03 p.p.m. in U,O, or 0.04 p.p.m. in U on average. The cobalt content of JAERI-U1 (No. 5) could not be determined by the previous methods.697 TABLE I11 DETERMINATION OF COBALT IN JAERI-U1 (No.5) (PROCEDURE IIIb) Cobalt, p.p.m. Sample solution Sample solution Absorbance* 1 concentration takenlml In U,O, In U 6.062 g per 100 ml 25 0.054 0.04 0.04 6.118 g per 100 ml 20 (Two determinations) 50 0.076 0.03 0-03 0.042 f 0.003 0*04&0.01 0*04&0*01 4.766 g per 100 ml 20 (Two determinations) 0.033 0.002 0.03 f 0.00 0.04f 0.01 60 0.069 0.03 0-03 6.336 g per 100 mlt 26 0.040 0.03 0.03 * Reference: reagent blank. t Dissolved in 10 ml of concentrated nitric acid. Determination of Trace Amounts of Cobalt in Pure Uranium Dissolution of the metallic uranium was more time consuming than that of uranium oxide. The pure uranium sample examined in this work contained trace amounts of cobalt, so that cobalt could be determined only by proposed method IIIb.The results obtained are given in TabIe IV. From this table it can be seen that the absorbance of the cobalt complex obtained was very small. As the limit of determination of cobalt by method IIIb was about 0.01 p.p.m. of cobalt, the error in the results obtained was relatively large. The cobalt content of JAERI-U4 (pure uranium) was found to be below 0.01 p.p.m.44 MOTOMIZU : SPECTROPHOTOMETRIC DETERMINATION OF COBALT IN TABLE IV DETERMINATION OF COBALT IN JAERI-U4 (PURE URANIUM) (PROCEDURE IIIb) 0.933 0.010 0.009 0.904 0.006 0.007 1.754 0.013 0.007 * Reference : reagent blank. Analyst, VoZ. 100 Sample takenlg Absorbance” Cobalt, p.p.m. in U Determination of Trace Amounts of Cobalt in a Uranyl Salt The amount of cobalt in a uranyl salt (uranyl nitrate) was also determined by the proposed method after dissolving it in 0.01 N nitric acid solution. The results obtained by use of method IIIb for trace amounts of cobalt in commercially available uranyl nitrate are given in Table V.From Table VI Table VI shows the recoveries of cobalt added to uranyl nitrate. TABLE V DETERMINATION OF COBALT IN WRANYL NITRATE (PROCEDURE IIIb) Supplier Grade* Taken/g Absorbancet Cobalt, p.p.m. A a 3.754 0.005 0*001 5.005 0.008 0.001 B a 2.513 0*080 0.028 2.485 0-080 0.029 5.026 0.151 0.026 4.966 0-155 0.028 * a = Analytical-reagent grade. t Reference: reagent blank. it can be seen that uranyl nitrate at levels below 0-2 M did not interfere in the determination of trace amounts of cobalt and that the recovery of added cobalt was about 100 per cent.TABLE VI RECOVERY OF ADDED COBALT (PROCEDURE IIIb) UO, (NO,) ,.6H,O (from supplier B) Cobalt added/pg Absorbance’ r--’-------, takenlg pg per cent. Cobalt recovery in 4.966 0 0.155 5.026 0 0.151 0.587 0.828 0-588 100 0.294 0.485 0.297 101 * Reference : reagent blank. Conclusions The spectrophotometric determination of cobalt with 2-nitroso-5-dimethylaminophenol By use of the proposed methods, micro-amounts and trace amounts of cobalt in The procedures 13y use of method is proposed. pure uranium, uranium oxide and a uranyl salt have been determined. recommended are simple, rapid, selective for cobalt and very sensitive. IIIb, 0.01 p.p.m. of cobalt in a sample could be determined. The author is greatly indebted to Professor Kyoji T8ei of Okayama University and Drs. Hiroshi Onishi and Hiroshi Hashitani of the Japan Atomic Energy Research Institute for their valuable advice and discussion, and also to JAERI for supplying the JAERI samples and for permission to publish this paper. References 1. 2. 3. Bane, R. W., in Roddon, C. J., Editor, “Analytical Chemistry of the Manhattan Project,” McGraw- Chilton, J . M., Rep. Congr. Atom. Enevgy Comma U.S., AECD-3607, 1963. Bane, K. W., and Grimes, W. R., in Roddon, C. J., Editor, op. cis., p. 415. Hill, New York, 1950, p. 430.January, 1975 URANIUM COMPOUNDS WITH 2-NITROSO-~-DIMETHYLAMINOPHENOL 45 5. 6. 7 . 8. 9. 10. 11. 12. 13. 14. Suzuki, M., and Takeuchi, T., Japan Analyst, 1960, 9, 179. “Analysis of Uranium Dioxide,” J AERI 4053, The Committee on Analytical Chemistry of Nuclear Hashitani, H., Yoshida, H., and Adachi, T., in “The Twenty-second Annual Meeting of the Japan Motomizu, S., Japan Analyst, 1973, 22, 695. -, Nipfion Kagaku Zasshi, 1971, 92, 726. -, Japan Analyst, 1972, 20, 1507. -, A%alyst, 1972, 97, 986. -, Analytica Chirn. Acta, 1973, 64, 217. -, Talanta, in the press. Fuels and Reactor Material, Japan Atomic Energy Research Institute, 1970, p. 90. Society for Analytical Chemistry,” 1973, p. B389. -, Ibid., 1971, 56, 415. Received May 16th, 1974 Accepted June 12th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000039

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

46 Analyst, January, 1975, VoI. 100, pfi. 46-50 Quantitative Evaluation of Peroxides, Hydroperoxides and Alcohols of m-Diisopropylbenzene by Nuclear Magnetic Resonance Spectroscopy L. Cavalli and G. Cancellieri Societd Italiana Resine, Centro Richerche Analisi, Sean, Via Trento 106, 20099 Sesto S. Giovanni. Milan, Italy -1he nuclear magnetic resonance spectroscopic analysis of the reaction mixtures obtained in the catalytic oxidation of m-diisopropylbenzene is reported. The solvent effects on the methyl resonances due to -C(CH,),OO-, -C(CH,),OOH, -C(CH,),OH and -CH(CH,), groups in the oxidation products have been studied, using as reference compounds the main reaction products, which had been synthesised for this purpose. The best results for the resolu- tions of the different resonances and their assignments were obtained by using deuteroacetone-d, as solvent.The quantitative evaluation of the various components present in the crude reaction mixtures is shown to be possible. Peroxides and hydroperoxides can be studied successfully by nuclear magnetic resonance spectroscopy. A few examples of the analysis of mixtures of peroxides, hydroperoxides and alcohols by this method have appeared in the Literature which show that the nuclear magnetic resonance technique can be a rapid and non-destructive method of analysis.192 It has been shown that protons in the a-position to ROO-, €300- and HO- groups, and to a lesser extent also those in the P-position, absorb at different chemical shift values, depending on the different magnetic environment of these protons, owing to stereochemical and inductive effects of functional groups.a-Protons in hydroperoxides and peroxides are strongly de-shielded (not less than 0.4 p.p.m.) with respect to the corresponding protons in alcohols. P-Protons are definitively less well differentiated, and at 60 MHz these protons in hydroperoxides and peroxides are separated by about 5 Hz from the corresponding protons in alcohols. It has been generally noted that the shielding of p-protons increases in the order: -OH > -0OH > -OOR, even though exceptions have been rep~rted.~ As far as the -0OH protons are concerned the corresponding resonances are considerably shifted to a low field (8 value about 9) and are well differentiated from the -OH protons (6 value about 5).This effect can only be observed in dilute solution in anhydrous solvents, whereas in complex mixtures a broad and analytically useless band is generally displayed by these protons. In this paper a brief study of the analytical application of nuclear magnetic resonance spectroscopy to some crude reaction mixtures obtained from the catalytic oxidation of m-diisopropylbenzene (m-DIB) is described. Data for a series of new compounds are given and examples of their quantitative analysis are presented. Experimental All of the compounds examined were provided by our research group. The products and the solvents used were dried before nuclear magnetic resonance measurements were made. Spectra were recorded on dilute solutions, the over-all concentration of the solutes being maintained at less than 15 per cent.A srnall amount of tetramethylsilane was added for reference and field-frequency locking purposes. All the samples were de-gassed and sealed into 5 mm 0.d. nuclear magnetic resonance tubes. The 6O-MHz proton spectra were obtained by using a Jeol J NM-C-6OHL nuclear magnetic resonance spectrometer operating in the field- frequency mode and all spectra were calibrated with a Hewlett-Packard 5300 frequency counter. Spectra were recorded with a sweep scale of 0.75 and 1.5 Hz cm-1 at a sweep rate of 0.027 and 0.108 Hz s-l. All frequency measurements were averages from at least three separate spectra and all experiments were performed at ambient probe temperature (about 30 "C).CAVALLI AND CANCELLIERI 47 Results The catalytic oxidation of m-diisopropylbenzene is expected to give as reaction products the following compounds: Y C(CH,),OOH Dihydropeiroxide 1 C(CH3 ),00H (DHP) Mono hydroperoxide C(CH3 ),OH Hydroperoxy-alcohol (HPOH) The reaction mixtures are also complicated by the the likely formation of peroxides.C(CH3),0H Monoalcohol (MOH) WHP) 1 C(CH3 12 0 H Dialcohol (DOH) presence of unreacted m-DIB and by The- -0OH and -02 protons were tentatively investigated in dilute solution using anhydrous solvents with the aim of readily distinguishing analytically between hydroperoxides and alcohols. These protons, however, are affected so much by the concentration and com- position of the reaction mixtures that in most instances only broad and overlapping bands are obtained. Even so, the resonances of protons in a /3-position relative to -OOR,-OOH and -OH groups were used with success to analyse the different components of the crude reaction mixtures.The P-protons, however, resonate over a small range (about 6 to 8 Hz at 60 MHz) and, in addition, are strongly influenced by the nature of the solvent and by the concentration. Consequently, a preliminary investigation was undertaken in order to ascertain the best conditions for the analytical separation of the different /&resonances and to determine the appropriate assignment. The three main reaction products of the catalytic oxidation of m-DIB, namely, DHP, MHP and MOH, were prepared in a pure condition and the solvent and concentration de- pendence of their chemical shifts were investigated.The solvent effects on the resonances of protons in a p-position relative to the -0OH and -OH groups of DHP, MHP and MOH in three typical solvents are reported in Table I, the over-all concentration of DHP, MHP and MOH, together with some m-DIB, being kept at about 15 per cent. The assignments TABLE I CHEMICAL SHIFTS OF METHYL PROTONS FOR DHP, MHP, MOH AND ~-DIISOPROPYLBENZENE*~ -C(CH&OOH - -C(CH&OH -CH(CH& Solvent DHP MHP MOH m-DIB Deuterochloroform . . . . 94-0 94.0 91.0 74.6 Benzene .. .. . . 90.1 90.1 84.1 70.9 Deuteroacetone-d, . . . . 93-1 92.2 90-7 72.9 * Chemical shifts are given in Hz relative to tetramethylsilane. t The concentration of the single solutes is less than 5 per cent. reported in Table I were obtained by incremental addition of the single compounds.The nuclear magnetic resonance spectrum of a typical mixture of these compounds in deutero- acetone-d, is shown in Fig. 1. Another compound, namely HPOH, was prepared but could not be obtained with a degree of purity higher than about 70 per cent. The impurities in crude HPOH were mainly due to the presence of DHP and of peroxide-type material. The nuclear magnetic resonance spectrum in deuteroacetone-d, of a mixture similar to that used for the spectrum shown in Fig. 1, to which a known amount of the crude HPOH had been added, is shown in Fig. 2.48 CAVALLI AND CANCELLIERI : QUANTITATIVE EVALUATION OF OXIDATION Analyst, 'vd. 100 B 1 C 100 90 80 70 Hz (relative to tetramethylsilane) Fig. 1. Nuclear magnetic resonance (60 MHz) spectrum in CD,COCD, of a syn- thetic mixture prepared with DHP, MHP, MOH and m-DIB with molar ratios 12.2: 46.0: 5.7 : 36.1.See Table I1 for key to peaks. A comparison with Fig. 1 shows two additional peaks, E and F (v = 94.3 and 91.5 Hz relative to tetramethylsilane, respectively), in Fig. 2 (for key to peaks, see Table 11). These new peaks were reasonably assigned to peroxide, -C(CHJ200C(CH,),, protons (peak E) and to the alcoholic, -C(CH,),OH, methyl protons of HPOH (peak I?). The corresponding hydro- peroxide, -C(CH,),OOH, methyl protons of HPOH must be overlapped by the resonances B I I I I I I I 100 90 80 70 Hz (relative to tetramethylsilane) Fig. 2. Nuclear magnetic resonance (60 MHz) spectrum in CDJOCD, of a syn- thetic mixture prepared with peroxides, DHP, MHP, HPOH, M0I-I and m-DIB with molar ratios 1.4 : 16.9 : 41.8 : 6.9 : 8-6 : 24.4.See Table I1 for key to peaks.Jnnzwy, 1975 PRODUCTS OF m-DIISOPROPYLBENZENE BY NMR SPECTROSCOPY 49 A and B of DHP and MHP. The nuclear magnetic resonance spectrum in deuteroacetone-d, of a reaction mixture obtained from the catalytic oxidation of m-DIB is shown in Fig. 3. As can be seen, the spectrum in Fig. 3 is similar to that of the synthetic mixture in Fig. 2. 100 90 80 70 Hz (relative to tetramethylsilane) Fig. 3. Nuclear magnetic resonance (60 MHz) spectrum in CD,COCD, of a crude See Table I1 for reaction mixture from the catalytic oxidation of m-DIB at 110 "C. key t o peaks. Discussion Solvents other than those shown in Table I were tried, but none gave better results than benzene and deuteroacetone-d,.The former is the most suitable solvent for separating the hydroperoxide, -C(CH,),OOH, methyl protons from the alcoholic, -C(CH,),OH, methyl protons, whereas we found that deuteroacetone-d, is the only solvent that enables the hydro- peroxide methyl protons of DHP to be differentiated from those of MHP. The latter solvent was therefore chosen for the analysis of the crude reaction mixtures. The definitive assign- ments in deuterOaCetOne-d, of the different resonances observed in the crude reaction mixtures can be stated with confidence to be as given in Table 11. TABLE I1 ASSIGNMENTS OF THE METHYL PROTON CHEMICAL SHIFTS OF THE CRUDE MIXTURES OBTAINED FROM THE CATALYTIC OXIDATION OF m-DIB Chemical shift (relative to Assignment tetramethylsilane) at 60 MHz/Hz Peroxides -C( CH,),oO-C( CH,) 2- Peak E 94.3 A 93.1 -C(CH,),OOH €3 92-2 &Lj HPOH MOH Peroxides hlHP MOH lil-DIB - F 91.5 92 t o 93 C 90-7 -C(CH*),OH -CH (CHJ 2 D 73 (approx.) 1 The chemical shift values given in Table I1 are concentration dependent.However, if the over-all concentration of the solutes does not exceed 10 to 15 per cent., no significant50 CAVALLI AND CANCELLIERI changes in chemical shift can be expected. The methyl doublets of the isopropyl, -CH(CH,),, groups for the different compounds could not be resolved. In deuteroacetone-d, (Figs. 1 to 3), as in any other solvent tried, the corresponding band broadens, showing only a slight shoulder in some instances. It was not possible to ascertain the presence of DOH in our reaction mixtures, and, even if it cannot be excluded, the formation of DOH is unlikely.The quantitative determination of the components present in the crude reaction mixtures was performed by evaluating the relative intensities of the different methyl resonances E, A, B, F, C and D, which was achieved in two steps: (1) evaluation of the relative intensities of E, A, B, F and C at v = 90 to 94 Hz, using expanded and well resolved spectra (the direct measurement of peak heights at maximum resolution gave the best results); (2) evalu- ation of the intensity of D at v = about 73 Hz from the electronic integration of the ratio between methyl resonances at v = 90 to 94 (peaks E, A, B, F and C ) and methyl resonances of isopropyl groups, D, at v = about 73, followed by norrnalisation according to the measure- ment €or step (1).Taking into account a norrnalisation factor of 2 for B, F and C, and a correction factor of 1/2 for A and B (which allows for the overlapping of hydroperoxide methyl resonances of HPOH by A and B resonances) one can obtain the molar percentage of the different compounds from the following equations : Peroxides, per cent., = E/S x 100 DHP, per cent. = (A - 1/2 F)/S x 100 MHP, per cent. = (2B - 1/2 F)/S x 100 MOH, per cent. = 2 C/S x 100 HPOH, per cent. = 2 F/S x 100 D - (B + C) 100 m-DIB, per cent. = -- S where S = E + A + B + F + C + D. Experiments were carried out in order to determine the accuracy and precision obtainable by using the above method of analysis. As a model system, mixtures of DHP, MHP, MOH and m-DIB were prepared in deuteroacetone, using component concentrations of the order of those given in Fig.1. The resolution obtained was similar to that shown in Fig. 1. Results of repetitive analysis of a typical mixture are given in Table 111. Quantitative agreement for the known composition of the mixture, as shown by these results, is reasonably good for such closely spaced peaks. The precision of the analysis is also good. TABLE I11 QUANTITATIVE ANALYSIS OF A KNOWN MIXTURE OF DHP, MHP, MOH AND VZ-DIISOPROPYLBENZENE Theore tical, Found, mol per cent. mol per cent. Standard deviation* DHP .. .. .. * . 11.42 11.51 0.07 MHP .. .. .. .. 40.47 41.22 0-35 MOH .. . . . . .. 5.74 6.18 0.08 m-DIB . . . . . . * . 42-37 41-08 0.36 * Calculated on eight repeat analyses. The lower limit of detection of a minor component in the presence of larger amounts of the other components was also determined. I t appears that 1 to 2 per cent. of any minor component would still give a detectable peak. Based on these results nuclear magnetic resonance spectroscopy is the only analytical tool, as far as can be ascertained, which is able to resolve such mixtures of peroxides, hydroperoxides and alcohols. In fact, it is difficult to imagine how these systems can be analysed by chemical and other instrumental techniques. We thank Dr. V. F. Ticozzi, who provided a1.l of the compounds used, and Mr. L. Lanzini for technical assistance. References 1. 2. 3. Ward, G. A., and Mair, R. D., Analyt. Chem., 1969, 41, 538. Swern, D., Clements, A. H., and Luong, T. M., Ibid., 1969, 41, 412. Smith, W. B., Ibid., 1972, 44, 881. Received April 19th, 1974 Accepted July 2nd, 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000046

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, January, 1975, Vol. 100, $@. 51-53 51 Influence of Ions on the Electrical Characteristics of the Deoxyribonucleic Acid = Water Interface V. K. Srivastava and Prabir K. Banerjee Department of Chemistry, University of Gorakhpur, Gorakhpur ( U.P.), Iwdia The sedimentation potentials of native and denatured DNA adsorbed on negatively charged Pyrex particles were measured at fixed ionic strength and pH. The zeta potentials were calculated from these sedimentation values and it was observed that metal ions such as 2n2+, Cu2+, Ag+, Hga+ and Mg2+ decrease the zeta potentials in a manner that is dependent on the nature of the metal ions. Deoxyribonucleic acid (DNA) is negatively charged because of the presence of fixed phosphate in the molecule and it is to be expected that an electrical double layer would be formed in aqueous solutions of this acid.The stability of the structure of native DNA in aqueous solution depends strongly on the character and ionic strength of added counter ions. The purpose of this paper is to show that an interesting relationship exists between the double helical structure of DNA, in both its native and denatured states, and its sedimentation potential when adsorbed on Pyrex particles as in the commonly employed particle - metal ions - water system. As the magnitude of the Sedimentation potential is generally of the order of several millivolts, only a few measurements of this potential have so far been reported.1-3 The technique used for its measurement is an improvement on that described earlier by Rastogi and M i ~ r a .~ Method The sedimentation potentials were measured by the method described below. the values thus obtained the zeta (t) potentials were calculated from the relationship- From E x (9 x lo8) mV fr- 4777 m ( l - &)g (which was obtained by appropriate substitution in the classical equation for evaluating sedimentation potential3) where 7 P is the viscosity of the medium, E mV the potential developed across the electrodes, m the mass per unit volume of the particle suspended between the two electrodes, D the dielectric constant and p1 and p2 are the densities of the particle and medium, respectively. Measurements were made at a fixed ionic strength (p = lov4 M) and a pH of 5.6. It was observed that the pH value of the system did not vary appreciably with the addition of solutions of metal ions.Measurement of Sedimentation Potential Because of the high resistance of the system (several megohms), measurement of sedi- mentation potentials is complicated. As direct amplification of the voltage cannot readily be effected, a pre-amplifier (cathode follower) was used in the circuit, the output of which was subjected to a single-stage amplifier. A sensitive voltmeter was also incorporated in the plate circuit of the amplifier in a position parallel to the plate resistor (Fig. 1). The sedimentation column used is shown in Fig. 2 and consisted of a Pyrex tube 20 cm long and 0.4 cm in diameter with standard joints at the ends of the tube. The electrodes were placed so that they did not make contact with the particles falling in the column.Simple platinum electrodes were found to be unsuitable for this purpose because of their large asymmetrical potential (10 to 30 mV) but silver - silver chloride electrodes proved t o be satisfactory.52 SRIVASTAVA AND BANERJEE: INFLUENCE OF IONS ON THE Analyst, VoZ. 100 t t Signal Fig. 1. Circuit for measurement of the sedimentation potential. Valves: V, = 5Y3GT and V, and V, = 6SF6; inductance: L, = 1OH; capacitors: C, and C , = 16 pF; and resistors: R, = 10 MR, R, = 1.5 MQ, & and R, = 10 kfl, R, and R, = 0.47 MQ, R, = 100 kQ and R, = 68 kQ; S = switch. The column was kept in an atmosphere thermostatically maintained at 36-8 5 0.1 "C and the electrodes were connected to the input of the cathode follower through a mercury contact.At the commencement of the experiment the resistor R, was adjusted so as to give an initial meter reading of zero. The DNA-coated particles were then allowed to fall from the top of the column. When the particles reached the lower electrode a meter reading was taken, from which the value for the sedimentation potential was obtained. B10 /I\\ Fig. 2. Column for measurement of the sedimentation potential. Results and Discussion As a check on the apparatus, the zeta potential of the Pyrex particle-water system was calculated. The value of -0-164 V obtained agreed fairly well with the values of -0.13 and -0.16 V reported by Quist and Washbaml and Butler,2 respectively, the slight variation being due to the contribution of parameters included in the final equation.The experimental values obtained for the sedimentation potentials and the calculated values for the zeta potentials are given in Tables I and 11. From the typical results given in Tables I and 11, it is evident that the zeta potential of -149.0 mV for native DNA adsorbed on Pyrex particles is higher than that for the adsorbed denatured DNA (- 118.0 rnv). This decrease in zeta potential can be explained on the basis of the change in structure of the double layer that occurs during denaturation. As thermal denaturation is accompanied by a decrease in charge density5$G such a changeJanzcary, 1975 ELECTRICAL CHARACTERISTICS OF THE DNA - WATER INTERFACE 53 can be attributed to a marked increase in the average distance between ionised phosphate groups and the disordered molecules, in accordance with the equation k = 47~0~2 (where u is the charge per square centimetre, d cm is the thickness and D is the dielectric constant).D TABLE I SEDIMENTATION POTENTIALS OF NATIVE DNA-COATED PYREX PARTICLES M; concentration of metal-ion I N THE PRESENCE OF METAL IONS Concentration of DNA solution, 5.2 x solution, 1.0 x loF4 M ; density of particle, 2.5; 7, 0.0063 P; dielectric constant, 80; g, 981 cm s - ~ System Native DNA coated on Pyrex - Mass per unit volume of the particle/mg ml-l water 23.6 + Zn2+ 25.7 + cu2+ 25.5 + Ag+ 25-7 + Hg2+ 24.9 + Mg2+ 25.3 Sedimentation potential/mV Zeta potential/mV - 2.32 -149.0 f 2 - 2.72 -160.0 f 2 - 3.03 -179.6 & 2 - 2.59 -152.4 f 2 - 2.53 -153.3 f 2 - 2.80 -167.0 f 2 Because u decreases on thermal denaturation, 5 will also decrease. The zeta potential values for native and denatured DNA compare well with the values of -149 mV and -108 mV obtained by Costantino, Liquori and Vitagliano.’ The influence of metal ions can also be described in terms of apparent changes in zeta potentials, the differences in their values varying with the nature of the metal ions.Further work is in progress in an attempt to account for the metal ion - DNA interactions on the basis of their zeta potential values. TABLE I1 SEDIMENTATION POTENTIALS OF DENATURED DNA-COATED PYREX PARTICLES I N THE PRESENCE OF METAL IONS Mass per unit volume of the Sedimentation System particle/mg ml-l potential/mV Zeta potential/mV Denatured DNA coated on Pyrex - water 25-4 + Zn2+ 24.7 + cu2+ 22-1 + Ag+ 24.5 + Hg2+ 24.9 + Mg2+ 25-3 - 1.98 -118.0 f 2 - 1.82 -111.4 f 2 - 1.98 -135.4 f 2 - 1.9 -117.1 f 2 -2.13 -128.6 f 2 - 2.47 -147.6 f 2 The authors thank Professor R. P. Rastogi, Head of the Chemistry Department, Gorakhpur Wniversity, for providing all necessary facilities. One of us (P.K.B.) is grateful to the C.S.I.R., Delhi, for financial support. References 1. 2. 3. 4. 5 . 6. 7. Quist, J. D., and Washbam, E. R., J . Amer. Chem. SOC., 1940, 62, 3169. Butler, J. A. V., “Electrical Phenomena at Interfaces,” Methuen & Co. Ltd., London, 1961. Nysels, K. J., “Introduction to Colloid Chemistry,’’ Academic Press, New York, 1959. Rastogi, R. P., and Misra, B. M., Trans. Faraday SOC., 1967, 63, 584. Liquori, A. M., Ascoli, F., Botre, C., Crescenzi, V., and Mele, A.. Nature, Lond., 1967, 63, 584. Ascoli, F., Botre, C., and Liquori, A. M., J . Molec. Biol., 1959, 184, 1482. Costantino, I,., Liquori, A. M., and Vitagliano, V., Biopolymery, 1964, 2, 1. Received July loth, 1973 Amended April 30th, 1974 Accepted May 21~8, 1974

ISSN:0003-2654    

DOI:10.1039/AN9750000051

出版商:RSC    

年代:1975

数据来源: RSC