摘要:

Analyst, August, 1974, Vol. 99, p$. 491497 491 Colorimetric Determination of Small Amounts of Cs to Clo Alcohols in Their Phthalate Esters BY S. HARRISON, H. HINCHCLIFFE AND G. L. WOODROFFE (Research and Development Defiartment, Imperial Chemical Industries Limited, Petrochemicals Division, Billingham, Teesside, TS23 1 J B ) Methods were examined for the determination of small amounts of C, to C,, alcohols in their phthalate esters involving the use of vanadium 8- hydroxyquinolinate in benzene or toluene and 3,5-dinitrobenzoyl chloride in pyridine as colorimetric reagents. Two modified procedures are presented that will reliably determine free alcohols in the 0.01 to 0.4 per cent. m/m range. PHTHALATE esters are widely used as plasticisers and the presence of even small amounts of free alcohols can lead to undesirable odours in the finished plastic. Reliable and sensitive analytical methods are therefore required in order to monitor the residual alcohol content at levels of 0.01 to 0.5 per cent.The alcohols commonly used are both straight- and branched- chain alcohols of carbon number 8 to 10. Buscarons, Marin and Claverl were the first to use vanadium 8-hydroxyquinolinate (oxinate) for detecting small amounts of alcohols. Feigl and Stark2 used the reagent as a spot test for primary, secondary and tertiary alcohols. It is suggested that the colour complex formed is 8: o-v- OH---- R(CH.LOH), Following this work, vanadium 8-hydroxyquinolinate has been widely used for determining alcohols. 3-17 Poor reproducibility and sensitivity and in some instances the unpleasant or toxic nature of the solvents used have made the published methods unsuitable for the above application.An alternative reagent for the determination of alcohols is 3,5-dinitrobenzoyl chloride, the use of which was developed by Berezin18 and extended by Johnson and Critchfield. l9 The reaction of 3,5-dinitrobenzoyl chloride with alcohols in the presence of pyridine gives products that are highly coloured in some non-aqueous media. This reaction has been applied to both the identification and the determination of alcohols,20-22 including C, to C,, alcohols in their fumaric esters. l4 The purpose of this paper is to give the experimental background and the details of analytical procedures involving the use of vanadium 8-hydroxyquinolinate and 3,5-dinitro- benzoyl chloride that have been found to be satisfactory for the determination of free alcohols (C8 to Clo) down to 100p.p.m.in phthalate esters. A standard procedure for preparing vanadium 8-hydroxyquinolinate is also described. VANADIUM 8-HYDROXYQUINOLINATE METHOD REAGENTS- Glacial acetic acid, 2.5 per cent. VlV solution in toluene. Dichloroacetic acid, 10 per cent. VlV solution in glacial acetic acid. Toluene, alcohol-free-Wash 1 litre of toluene with 100 ml of 1 N sulphuric acid containing 1 g of potassium chromate, followed by 100 ml of 1 N sodium hydroxide solution, then wash the toluene with four 100-ml volumes of distilled water and dry it over anhydrous sodium sulphate. Sodium hydroxide solutions, 1 N and 0.1 N. @ SAC and the authors.492 HARRISON et al.: COLORIMETRIC DETERMINATION OF SMALL [Analyst, Vol.99 Dimethylformamide. Standard solution of alcohol in toluene- 1 ml of solution = 0.2 mg of alcohol (e.g., “isodecanol”* for diisodecyl phthalate). Sodiwz salt of vanadium 8-hydroxyquinolinate-Dissolve 1.35 g of ammonium meta- vanadate in 35 ml of l N sodium hydroxide solution in a 250-ml beaker. Add 15 ml of dis- tilled water and boil the mixture for 10 minutes in order to remove all the ammonia, adding water as required to keep the volume at 50 ml. This solution (solution A) is maintained at its boiling-point. Dissolve 5 g of 8-hydroxyquinoline in 35 ml of hot 1 N sodium hydroxide solution in another 250-ml beaker. Add 40-0 ml of distilled water, bring the solution to its boiling-point, then place the beaker in a boiling water bath.Add slowly, with stirring, the hot solution A and allow the beaker to stand in the bath for 15 to 20 minutes. The volume of the solution should be kept constant at 125 ml by adding hot distilled water. Remove the beaker from the bath, place it on top of the bath and add 50 per cent. V/V acetic acid dropwise from a burette, stirring the solution vigorously during the addition. A black precipitate forms on addition of acetic acid, which readily coagulates and also partly re-dissolves in the alkaline solution. The supernatant solution remains clear after the addition of each drop of acetic acid and coagulation of the black precipitate. Continue to add acetic acid dropwise until the solution becomes cloudy and remains cloudy after stirring without further addition of acetic acid (approximately 6 ml of 50 per cent. V/V acetic acid are required to neutralise the excess alkalinity; the pH of the solution should be 7 to 8).Quickly filter the hot solution through a Whatman No. 541 filter-paper and cool the filtrate in an ice-bath until a copious yellow precipitate of the sodium salt of vanadium 8-hydroxyquinolinate is deposited. Filter off the yellow precipitate using a vacuum filter. If necessary, wash out the flask with not more than 5 ml of distilled water in order to transfer all the yellow precipitate to the filter. (Do not use more than this volume of water to wash the precipitate as it is very soluble in water.) Continue vacuum filtration until all the excess of liquor has been sucked from the precipitate. Remove the yellow precipitate and the filter- paper from the funnel and place it on a large watch-glass.Dry the precipitate in an oven at 80 “C for 4 hours; the yellow solid will then be sufficiently dry to remove the filter-paper and also to be broken down into a coarse powder. Replace the precipitate in the oven for a further 24 hours at 80 “C. Grind the precipitate to a fine powder and store it in a glass container until required. Vanadium 8-hydroxyquinolinate reagent-Weigh 0.2 g of the sodium salt of vanadium 8-hydroxyquinolinate into a 150-ml conical flask, add 15 ml of dimethylformamide and shake the flask so as to dissolve the solid. Add 100 ml of alcohol-free toluene, mix, then add 10 ml of 2.5 per cent. V/V acetic acid in toluene, and mix again.This procedure should give a dark solution. Prepare the reagent freshly as required; it should be used the same day. APPARATUS- A Unicam SP600 spectrophotometer or another suitable colorimeter is used. PROCEDURE- Weigh a suitable amount of sample into a 25-ml calibrated flask, dilute to the mark with alcohol-free toluene and mix well. Into A, and B, transfer by pipette 5-ml aliquots of diluted sample and to A, add 2 ml of standard alcohol solution. To B, add 2 m1 and to C, add 7 ml of alcohol-free toluene; the volume of liquid in each cylinder is now 7 ml. Add 10 ml of vanadium 8-hydroxyquinolinate reagent to each cylinder, stopper, mix, then heat them to 60 & 1 “C for 20 minutes in a water-bath. Cool the cylinders in cold water for 2 minutes, then add 10 ml of 0.1 N sodium hydroxide solution to each and stopper and shake them vigorously for 1 minute. If phenolic antioxidants are present, the dark colour remains, otherwise the solution becomes pink to red.All reagents and the sample must be dry (k, give clear solutions in toluene) up to the stage of the addition of the 0.1 N sodium hydroxide solution. Allow the layers to separate, then transfer by pipette 5 ml of the upper toluene layer from each cylinder into each of three 10-ml stoppered measuring cylinders, A,, B, and C,. Add 1 ml * “Isodecanol” is a mixture of branched-chain decyl alcohols. Take three 25-ml measuring cylinders, A,, B, and C,.August, 19741 AMOUNTS OF cs TO clo ALCOHOLS IN THEIR PHTHALATE ESTERS 49s Measure the absorbance at 620 nm of dichloroacetic acid reagent to each and shake to mix.in 10-mm cells using the solution in C, as a blank. CALCULATION- Let M g be the mass of sample, a mg the mass of alcohol added as internal standard, i.e., the mass of alcohol added to cylinder A,, A A the absorbance of the solution in A, against that in C, and A B the absorbance of the solution in B, against that in C,. Then, A g x a x 25 x 100 Free alcohol*, per cent. m/m = (AA-AB) x M x 5 x 1000 DINITROBENZOYL CHLORIDE METHOD REAGENTS- 3,5-DinitrobenzoyZ chloride reagent-Dissolve 1 g of 3,5-dinitrobenzoyl chloride in 10 ml of dry pyridine (containing less than 0-05 per cent. m/V of water), warming the mixture on a steam-bath to assist solution. Prepare the reagent immediately before use; it will keep for approximately 1 hour.The water content must be less than 0.05 per cent. m/V (it must be specially dried). Pyridine-Analytical-reagent grade. CycZohexane or n-hexane-Analytical-reagent grade. Hydrochloric acid-AnalaR grade. Sodium carbonate-AnalaR grade. 1,2-Diaminopropane. Dimethylformamide-Analytical-reagent grade. The water content must be less than Standard alcohol solution-Prepare a 4 per cent. m/V solution in dry pyridine. Prepare a 5 per cent. m/V solution. 0.5 per cent. m/V. APPARATUS- Hamilton syringe is required. A Unicam SP600 spectrophotometer or another suitable colorimeter is used. A 10-pl PROCEDURE- Weigh a suitable amount of sample into a 25-ml calibrated flask, dilute to the mark with dry pyridine and mix well. Take three 150-ml Erlenmeyer flasks, A,, B, and C,.Into A, and B, transfer by pipette 2-ml aliquots of the sample solution and into C, 2 ml of dry pyridine. Into A, inject 10 p1 of the standard alcohol solution with the Hamilton syringe and then into all three flasks transfer by pipette 1-ml aliquots of the 3,5-dinitrobenzoyl chloride reagent. Allow the flasks to stand for 15 minutes, then add 25 ml of approximately 2 N hydrochloric acid to each flask and transfer the contents into each of three separating funnels. Rinse out each flask with 20 ml of cyclohexane and transfer the rinsings into the appropriate separating funnel. Stopper the separating funnels and shake each vigorously for 1 minute, then allow the contents to settle. Run off the bottom aqueous phase from each separating funnel, then add 5 ml of 5 per cent.m/V sodium carbonate solution to each, shake them for approximately 30 s, then allow the contents to settle. Run off the sodium carbonate solution from each separating funnel and filter the cyclohexane phase through a small cotton-wool plug contained in a filter funnel into each of three 25-ml stoppered measuring cylinders, A,, B, and C,. Wash the cotton- wool plugs with sufficient fresh cyclohexane to make the volume in each cylinder up to 25 ml, then mix the contents of the cylinders well. Transfer by pipette 1 ml of the cyclohexane solution from each measuring cylinder into each of three 10-ml stoppered measuring cylinders, A,, B, and C,. Add 5 ml of dimethyl- formamide and 1 ml of 1,2-diaminopropane to each cylinder and mix the contents well, then, after 5 to 10 minutes, measure the absorbances of the solutions in A, and B, at 525 nm in 10-mm cells using the solution in C, as a blank.Stopper the flasks and mix the contents well. * Calculated as the alcohol component of the ester.494 [Analyst, Vol. 99 CALCULATION- Let 111 g be the mass of sample, a mg the mass of alcohol added as internal standard, ie., the mass of alcohol added to flask A,, A A the absorbance of the solution in A, against that in C, and A B the absorbance of the solution in B, against that in C,. Then, HARRISON et al.: COLORIMETRIC DETERMINATION OF SMALL A g x a x 25 x 100 Free alcohol*, per cent. m/m = (AA-AB) x M x 2 x 1000- TABLE I EFFECT OF DRYING THE SODIUM SALT OF VANADIUM 8-HYDROXYQUINOLINATE ON THE ABSORBANCE DUE TO 2-ETHYLHEXANOL 2-Ethylhexanol/ Drying procedure mg Dried in vacuum 0.20 desiccator 0.40 0.60 0.80 Dried a t 60 "C 0.20 0.40 0.60 0.80 Dried at 80 "C 0.20 0.40 0.60 0.80 Absorbance 0.07 0.14 0.22 0.3 1 0.11 0.22 0.32 0.45 0.12 0.24 0-35 0.47 TABLE I1 EFFECT OF REACTION TIME ON THE ABSORBANCE DUE TO ISOOCTANOL Time in water-bath at 55 "C/minutes... . 0 2 5 10 15 20 30 60 Absorbance a t 620 nm . . .. . . . . 0-26 0.39 0.49 0.52 0.53 0.52 0.52 0.51 TABLE I11 The colour was developed on a solution of isooctanol and the absorbance at 620 nm was measured at intervals STABILITY OF THE BLUE COMPLEX FORMED WITH VANADIUM 8-HYDROXYQUINOLINATE Time/minutes . . . . . . .. 2 5 10 15 20 30 60 180 Absorbance . . . . .. . . 0-44 0-44 0-43 0.43 0.42 0.41 0.40 0.39 TABLE IV STABILITY OF VANADIUM 8-HYDROXYQUINOLINATE REAGENT Absorbance? A I 'L Time of testlhours In benzene I n toluene 0 0.40 0.24 6 - 0.17 24 0.20 0.06 t Isooctanol was used as the alcohol and the absorbance was measured a t 620 nm.TABLE V EFFECT OF WATER IN PYRIDINE ON THE REACTION OF 3,5-DINITROBENZOYL CHLORIDE WITH ISOOCTANOL Concentration of water in pyridine, Mass of per cent. m/m isobutanol Absorbance 0.02 0.7 0.35 0.12 0.7 0.19 0-15 0.7 0-12 0.20 0.7 0.10 0.25 0.7 0-08 * Calculated as the alcohol component of the ester.August, 19741 AMOUNTS OF c, TO c,, ALCOHOLS IN THEIR PHTHALATE ESTERS TABLE VI STABILITY OF THE COLOURED COMPLEX FORMED BETWEEN 3,5-DINITROBENZOYL CHLORIDE AND ALCOHOLS Absorbance I L > Timelminutes Sample 1 Sample 2 Sample 3 Sample 4 495 3 0.27 10 0.26 30 0.23 0.36 0-36 0.30 0.57 0.56 0.50 0.65 0.63 0-56 RESULTS AND DISCUSSION VANADIUM 8-HYDROXYQUINOLINATE METHOD- Benzene was originally used to replace nitrobenzene as solvent in this method because it gave increased sensitivity, but was unsatisfactory owing to its toxicity.Of the various sol- vents tried, vix., toluene, xylenes, cyclohexane and n-heptane, the most suitable was toluene, although the sensitivity obtained was only about 60 per cent. of that obtained with benzene. Examination of the problems involved in preparing vanadium 8-hydroxyquinolinate reagent satisfactorily revealed ambiguities in the procedural details of preparation. It was also found that small amounts of water in the reagent had an adverse effect on sensitivity.These problems were resolved and a preparation procedure was devised that included drying the 8-hydroxyquinolinate salt in an oven at 80 "C for 24 hours (Table I). A study of the effect of time on the reaction showed that at 55 to 60 "C the optimum reac- tion time for maximum sensitivity was 15 to 20 minutes (Table 11). It was also observed that TABLE VII CONCENTRATIONS OF ALCOHOLS PRESENT IN THEIR PHTHALATE ESTERS Alcohol found, per cent. mlm 7 - c \ Ester Di(2-ethylhexyl) phthalate Di(2-ethylhexyl) phthalate Diisooctyl phthalate Diisooctyl phthalate Di- Alphyl ph thalate * Di- Alphyl phthalate * Dinonyl phthalate Dinonyl phthalate Diisooctyl phthalate (A) Diisooctyl phthalate (B) Diisooctyl phthalate (C) * Alphyl Alcohol added, per cent. 0.13 0.33 0.13 0.35 0-13 0.33 0.14 0.37 mlm 0.015 1 r" I" 0.01 1 0.018 3,5-Dinitro- benzoyl chloride method 0.14, 0.13 0.34, 0.34 0.15, 0.14, 0.14, 0.15 0.40, 0.39, 0.36, 0.37 0.14, 0.14 0.33, 0.37 0.16, 0.16, 0.18 0.36, 0.38, 0.38 0.01 1, 0.010, 0.009 0.030, 0.028, 0-030 0.016, 0.019, 0.019 0.027, 0.032, 0.029 0.033, 0.034, 0-036 0.047, 0.048, 0.053 Vanadium 8-hydroxyquinolinate- method 7- Toluene Benzene solution solution 0.14, 0.11 0.12, 0.14 0.32, 0.29 0.34, 0.33 0.14, 0.15 0.13, 0.14 0.41, 0.40 0.41, 0.39 0.14, 0.15 0.16, 0.14 0.34, 0.34 0.33, 0.33 0.14, 0.15 0.16, 0.15 0.36, 0.37 0.37, 0.38 0.009, 0.011, 0.010, 0.013, 0.029, 0.026, 0.029, 0.026, 0.024, 0.027 0.025, 0.023 0.010, 0.012 0.009, 0.009 0.017, 0.018, 0.014, 0.016, 0.019, 0.015 0.014, 0.018 0.031, 0.028, 0.027, 0.028, 0.031, 0-029 0.030, 0.026 0.036, 0-037, 0.034, 0.036.0.035, 0.033 0.030, 0.033 0.052, 0.052, 0.054, 0.056, 0.055, 0.050 0.044, 0.050 ester of Alphanol (ICI registered trade-name) (C, to C, alcohols)496 [A~talyst, Vol. 99 the blue-coloured complex formed on the addition of dichloroacetic acid deteriorated with time and therefore its absorbance should be measured within 15 minutes of the addition of acid (Table 111). The stability of the vanadium 8-hydroxyquinolinate reagent was also examined (Table IV). It deteriorated with time and therefore should be used within 24 hours of preparation. In the light of these factors, an analytical procedure aimed at maximising the sensitivity and repeatability was devised. This procedure included the use of freshly prepared reagent, an internal standard and carefully timed stages.HARRISON et al.: COLORIMETRIC DETERMINATION OF SMALL 3,5-DINITROBENZOYL CHLORIDE METHOD- The 3,5-dinitrobenzoyl chloride method for determining small amounts of alcohol is based on the work of Johnson and Crit~hfie1d.l~ The alcohol reacts with 3,5-dinitrobenzoyl chloride in pyridine for 15 minutes. The 3,5-dinitrobenzoate formed is extracted and, on reaction with 1 ,2-diaminopropaneY gives a red colour. When this procedure was applied to phthalate esters, it was found that the quality of the pyridine was critical (Table V) and its water content had to be reduced to less than 0.05 per cent. m/V in order to obtain the maximum red colour. It was also found that, as the intensity of the red colour decreased with time (Table VI), its absorbance had to be measured within 5 to 10 minutes of formation.As with the 8-hydroxyquinolinate method, these limitations were minimised in the analytical procedure by adding an internal standard with a 10-pl Hamilton syringe. TABLE VIII COMPARATIVE RESULTS OF FREE ALCOHOL CONTENTS ON COMMERCIAL SAMPLES OF ESTERS USING THE TWO METHODS Free alcohol, per cent. m/m Ester Di-( 2-ethylhexyl) phthalate Dinonyl phthalate Diisodecyl phthalate Diisooctyl phthalate Di-Alphyl phthalate" Sample No. 1 2 3 4 5 1 2 3 4 1 2 3 4 5 6 7 1 2 3 4 1 2 3 4 3,5-Dinitro- benzoyl chloride method 0.03 0.02 0.10 0.03 0.02 0.07 0.12 0.17 0.03 0-42 0.08 0.18 0.13 0.11 0-08 0.22 0.09 0.08 0.02 0.02 0-02 0.04 0-06 0.02 * See footnote to Table VII.Vanadium S-hydroxyquinolin&e method Benzene Toluene solution solution 0-02 0.03 0.02 0.02 0.09 0.11 0.02 0.02 0.02 0.02 0.05 0.06 0-13 0.12 0.14 0.16 0.02 0.01 0.40 0.40 0-06 0.09 0.20 0.18 0.18 0.16 0.12 0.11 0-13 0.1 1 0.23 0.20 0.10 0.09 0.05 0-06 0.02 0-03 0.02 0.02 0.03 0.03 0.03 0.03 0.06 0.08 0.02 0.02 r A \ CONCLUSIONS Several phthalate esters were analysed for their C, to C,, alcohol contents using the standard procedures described, over the concentration range 0.01 to 0.4 per cent. m/m. These esters included some commercial samples and also phthalates that contained known added amounts of alcohol.August, 1974) AMOUNTS OF c8 TO c,, ALCOHOLS IN THEIR PHTHALATE ESTERS 497 The results in Tables VII and VIII show that either reagent can be used to determine small amounts of alcohols.However, the 3,5-dinitrobenzoyl chloride method has two dis- advantages: the test time is longer and the red colour obtained with 1,2-diaminopropane is unstable. The vanadium 8-hydroxyquinolinate method, in which either toluene or benzene is used as solvent, gives good repeatability but the sensitivity is lower when toluene is used and the sodium salt of vanadium 8-hydroxyquinolinate requires more careful preparation and drying. Subsequent experience has confirmed the above conclusions, establishing that the vanad- ium 8-hydroxyquinolinate method is the preferred method. This method was also suitable for determining trace amounts of alcohols in adipates, glutarates, succinates and glycol ether acetates. It is not applicable to glycols, and phenolic antioxidants were found to interfere in both methods.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. REFERENCES Buscarons, F., Marin, J. L., and Claver, J., Analytica Chim. Acta, 1949, 3, 310 and 417. Feigl, F., and Stark, C., Microchim. Ada, 1955, 996. Blair, A. J., and Pantony, D. A., Analytica Chim. Acta, 1955, 13, 1. Kudo, I., and Aoki, I., Japan Analyst, 1957, 6, 791. Muruta, S., and Iwama, F., J. Chem. Soc. Japan, Pure Chem. Sect., 1959, 80, 1131. Tanaka, M., Talanta, 1960, 5, 162. Stiller, M., AnaEytica Chim. Rcta, 1961, 25, 85. Pesez, M., and Bartos, J., Bull. Soc. Chim. Fr., 1961, No. 10, 1930. Khyanina, A. P., Zav. Lab., 1964, 30, 417. Mantel, M., and Anbar, M., Analyt. Chem., 1964, 36, 936. van Gent, P. K., and Kerrich, J. E., Analyst, 1965, 90, 335. Sukhomlin, R. I., Zharavskii Pishch. Prom. Mezhvod. Resp. Nauch.-Tekh. Sb., 1967, 3, 63; Chem. Abstr., 1967, 66, 101449~. Tanaka, M., and Kojima, I., Analytica Chim. Acta, 1968, 41, 75. Amos, R., Ibid., 1968, 40, 401. Korenman, I. M., and Ganina, V. G., Tr. Khim. Tekhnol., 1968, 1, 122. Pesez, M., and Bartos, J., Bull. Soc. Chim. Fr., 1969, 1, 340. Buscarons, F., and Paraira, M., Quim. Ind., Madrid, 1970, 16, 15. Berezin, J. V., Dokl. Akad. Nauk SSSR, 1954, 99, 563. Johnson, D. P., and Critchfield, F. E., Analyt. Chem., 1960, 32, 865. Robinson, W. T., jun., Cundiff, R. H., and Markunas, P. C., Ibid., 1961, 33, 1030. Robinson, J. R., Ibid., 1967, 39, 1178. Diemair, W., Pfeilsticker, K., and Holscher, I., 2. analyt. Chem., 1968, 234, 418. Received January 30th, 1974 Accepted APriZ 8th. 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900491

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

498 Analyst, August, 1974, Vol. 99, $9. 498-502 Determination of Primary and Secondary Amines Alone and in LMixtures with Tertiary Amines BY BALBIR CHAND VERMA AND SWATANTAR KUMAR (Department of Chemistry, Punjabi University, Patiala, India) An iodatometric method has been developed for the determination of primary and secondary amines that is based on their quantitative reaction with phenyl isothiocyanate in dimethylformamide to form substituted thio- ureas. These are titrated with potassium iodate in an acidic medium at room temperature. The end-point is detected visually by the yellow colour imparted to the solution by the first drop of the iodate solution in excess, and potentiometrically by using a bright platinum-wire indicator electrode and a saturated calomel reference electrode.Methods have also been de- veloped for the determination of primary (or secondary) amines and tertiary amines in the presence of each other. An excess of phenyl isothiocyanate, added to the mixture in solution in dimethylformamide, converts the primary (or secondary) amines into the corresponding di- (or tri-) substituted thioureas, whereas the tertiary amines are left unreacted. The conductimetric titration of tertiary amines with trichloroacetic acid, followed by the iodatometric titration of thioureas formed, enables the mixture to be analysed for both comonents. The methods described are simple, accurate and reliable. PRIMARY and secondary amines react with organic isothiocyanates to yield substituted thioureas- RHN RiHN>c = s R’NH, + RNCS--+ RHN R’R’INH + RNCS--+ ‘c=s RIRI/N/ These reactions form one of the most general and convenient methods for the preparation of substituted thioureas and are widely used for the characterisation of amines and isothiocy- anates.They can also be made the basis for the determination of an organic isothiocyanate or amine if one of them is allowed to react with a known excess amount of the other and, after completion of the reaction, the residual reactant is titrated with a suitable reagent. This method has been extensively used for the determination of isothiocyanates by allowing them to react with a known excess of amine and determining the unreacted excess of the amine acidimetrically. A similar method for the determination of amines by treating them with an excess of isothiocyanate has not, however, been applied, because of the non-availability of a suitable, direct method of titrating isothiocyanates.Because the amine - isothiocyanate reaction to form thiourea proceeds to completion, the problem of the determination of isothiocyanates or amines can be made much simpler if a suitable method is made available for titrating the thiourea formed in the reaction. In the course of our extensive investigations on the oxidimetric determination of thioureas,l we have found that thioureas can be smoothly, rapidly and quantitatively titrated with potassium iodate2 in an acidic medium at room temperature without a catalyst. No indicator need be added as the end-point is signalled by the appearance of the yellow colour due to iodine, result- ing from the first drop of iodate solution in excess.The titrations can also be carried out potentiometrically by using bright platinum wire and a saturated calomel electrode assembly. The potentials attain stable values immediately on addition of each increment of the oxidant. A sharp jump in potential is observed at the equivalence point in each titration. In a recent publication,3 we have reported a method for the determination of isothiocyanates that is based on their quantitative conversion into thioureas, which are then titrated with potassium iodate in an acidic medium. @ SAC and the authors.VERMA AND KUMAR 499 A simple and accurate method has now been developed for the determination of amines by their reaction with an excess of isothiocyanate and measuring the thiourea formed, again iodatometrically.The proposed method possesses some significant advantages over the acidimetric method that is in general use. Firstly, because of the oxidimetric nature of the proposed method, the determination of amines can be achieved in the presence of basic com- pounds that might otherwise cause interference in their acidimetric determination. Secondly, as tertiary amines do not react with isothiocyanates, the method permits the determination of primary (or secondary) amines in the presence of tertiary amines. Thirdly, potassium iodate is an oxidimetric standard, therefore the necessity of standardising acidic solutions in the acidimetric method can be eliminated. Lastly, the visible end-point iodatometric titrations can be performed without an indicator as the end-point is marked by the yellow tint imparted to the solution by the first drop of iodate solution in excess.The proposed method consists in treating the amine with an excess of phenyl isothio- cyanate in dimethylformamide and titrating the substituted thiourea formed iodatometrically in a sulphuric acid medium at room temperature. It should be mentioned here that the excess of isothiocyanate does not interfere in the iodatometric determination of thioureas. ANALYSIS OF MIXTURES OF PRIMARY (OR SECONDARY) AMINES AND TERTIARY AMINES- As tertiary amines are frequently encountered in the preparation of primary or secondary amines, the determination of primary (or secondary) amines and tertiary amines in the presence of each other is of great interest. A simple and accurate procedure for the analysis of such a mixture in the same aliquot has been developed.The method consists in treating the solution of the mixture in dimethyl- formamide with an excess of phenyl isothiocyanate, when primary and secondary amines are converted into the corresponding dj- and tri-substituted thioureas, respectively, and the tertiary amines are left unreacted. The tertiary amines are titrated conductimetrically with trichloroacetic acid followed by iodatometric titration of the substituted thioureas formed. The conductimetric titration corresponds to the amount of tertiary amine present, whereas the iodatometric titration gives the amount of primary (or secondary) amine in the mixture. EXPERIMENTAL APPARATUS- The redox potentiometric titrations were carried out with an Osaw Crompton (India) potentiometer incorporating an Osaw spot reflecting galvanometer, a bright platinum-wire indicator electrode and a saturated calomel reference electrode.Acid - base conductimetric titrations were carried out with a Philips PR9500 conductivity bridge that operated at a frequency of 50 Hz. The cell electrodes consisted of two rigidly held square plates (approximately 0.8 cm2 in area) of platinised platinum, facing each other at a distance of 1 cm. A microburette of 10-ml capacity, graduated in 0.01-ml divisions, was used. REAGENTS- Dimethylformamide, commercial pade-This solvent was purified by standing it over AnalaR anhydrous sodium carbonate for 2 days, then decanting, distilling and fractionally distilling it, collecting the fraction distilling at 148.5 to 149.5 "C in an amber-glass bottle. Potassium iodate solution, 0.05 N-This was prepared by dissolving 1.7834 g of the dried AnalaR grade solid in water and making the volume up to 1 litre. Trichloroacetic acid solution, 1.0 N-This was prepared by dissolving slightly more than the calculated amount of the acid in water.The solution was then standardised by conducti- metric titration* against anhydrous sodium acetate. Phenyl isothiocyanate, FLuka, commercial grade-This material was distilled before use. Ethylamine (50 per cent. solution) and isopropylamine (70 per cent. solution)-These were used as received from Riedel de Hahn. n-Propylamine, n-butylamine, isobutylamine, diethylamine, triethylamine, pyridine, piperidine, pyrrolidine, a-picoline and quinoline, all commercial grade, were distilled before use.All other chemicals used in this investigation were of analytical-reagent quality.500 VERMA AND KUMAR: DETERMINATION OF PRIMARY AND SECONDARY [A%a@St, VOl. 99 DETERMINATION OF PRIMARY OR SECONDARY AMINES- Aliquots of solutions of each amine in dimethylformamide were taken in glass-stoppered titration flasks and 3 to 5 ml of phenyl isothiocyanate (an approximately 0.3 N solution in dimethylformamide) were added to each. The volume of the solution was made up to 10 ml with the solvent. The flask was then stoppered, swirled to mix the reactants, and set aside for 10 minutes to ensure completion of the reaction. Sufficient water and sulphuric acid to keep the normality of the solution at 2.0 to 2.5 in sulphuric acid and its volume at 100 ml were added.Each solution was cooled to room temperature (25 "C) and titrated with 0-05 N potassium iodate solution to the appearance of a distinct, permanent yellow colour. If it is preferred, amylose (0.2 ml of a 1 per cent. aqueous solution) can be used as an ndicator, the solution acquiring a blue colour at the end-point. In potentiometric titrations, the solution was magnetically stirred during the titration. A sharp jump in potential was observed at the equivalence point in each titration. From the volume of standard potassium iodate (0.05 N) required to attain the end-point in visual and potentiometric titrations, the amount of substituted thiourea formed, and con- sequently the amount of amine, was calculated. The results are recorded in Table I.TABLE I IODATOMETRIC DETERMINATION OF PRIMARY AND SECONDARY AMINES Amount found*/mg Visual -etric CH,CH,NH, .. . . 9-94 f 0.072 9.98 f 0-036 CH,CH,CH,NH, . . . . 10.04 f 0-034 10.03 & 0.028 . . 9-97 f 0-053 10.02 f 0.051 9.99 5 0.041 CH,CH,CH,CH,NH, . . 10.01 & 0-056 (CH,),CHCH,NH, . . 9.96 f 0.068 10.02 f 0.038 (CH,CH,),NH . . . . 9-96 f 0.076 9-95 f 0.062 PH,CH,CH,CH,NH . . 9.97 f 0-063 9.98 & 0.058 Compound method 3 method: (CHJ,CHNH, Amount foundf/mg Visual !YZZmetric 39.85 f 0.082 39.88 & 0.056 40.18 f 0-063 40.16 & 0.046 40.28 f 0.061 40.20 Ifr: 0.025 39.72 0.092 39-81 f 0.057 40.28 f 0.061 40.24 f 0.047 40.10 f 0.075 40.02 f 0.039 40.26 f 0.056 40-21 f 0.026 method $ method: CH,CH2CH2CH2CH,NH .. 10.06 f 0.075 10.05 f 0.046 39.80 f 0.078 39.82 & 0.030 I * Amount taken 10 mg. 7 Amount taken 40 mg. $ Mean of six determinations, & standard deviation. DETERMINATION OF PRIMARY (OR SECONDARY) AMINES AND TERTIARY AMINES IN THE PRESENCE Aliquots of solutions (in dimetliylformamide) of synthetic mixtures with different ratios of primary (or secondary) amine to tertiary amine were taken in glass-stoppered titration flasks containing an excess (7 to 10 ml of an approximately 0.3 N solution) of phenyl isothiocyanate in dimethylformamide solution. The volume of each solution was made up to 20 ml with the solvent. Each flask was stoppered, swirled to mix the reactants and set aside for 10 minutes to ensure completion of the reaction.The solution was then mixed with 40 to 45ml of approximately 1.0 N acetic acid, cooled to room temperature (25 "C) and titrated conducti- metrically with standard 1.0 N trichloroacetic acid solution. To the same solution, sufficient water and sulphuric acid to keep the normality of the solution at 2.0 to 2.5 in sulphuric acid and its volume at 125 ml were added. The solution was cooled to room temperature and titrated potentiometrically with standard 0.05 N potassium iodate. The solution was stirred magnetically during pot entiomet ric titrations. The volume of standardised acid used in the acidimetric titration corresponds to the amount of the tertiary amine, whereas the volume used in the iodatometric titration gives the amount of di- (or tri-) substituted thiourea and consequently the amount of primary (or second- ary) amine present in the sample.The results of the analysis of various ethylamine - triethyl- amine and diethylamine - triethylamine mixtures are recorded in Tables I1 and 111, respectively. OF EACH OTHER-August, 19741 AMINES ALONE AND IN MIXTURES WITH TERTIARY AMINES RESULTS AND DISCUSSION The results recorded in Table I show that ethylamine, n-propylamine, isopropylamine, n-butylamine, isobutylamine, diethylamine, piperidine and pyrrolidine can be determined visually and potentiometrically after conversion to the corresponding substituted thioureas with phenyl isothiocyanate in a dimethylformamide medium. The over-all standard devia- tions from the pooled results of all the visible end-point and potentiometric titrations per- formed with 10 mg of each amine have been found to be 0.062 and 0.045, respectively.For 40 mg of each amine, these values are 0.071 and 0.041, respectively. 501 TABLE I1 ANALYSIS OF MIXTURES OF ETHYLAMINE AND TRIETHYLAMINE Amount of Amount of mixturelmg Amount found*/mg mixturelmg Amount found*/mg 20.00 20.12 f 0.084 20.00 20.08 f 0.126 20.00 20.14 & 0.094 40-00 39-90 f 0-132 20-00 19-90 -& 0.097 60-00 60-32 f 0.094 20.00 20.06 f 0-078 80.00 79.60 f 0.127 4040 39.85 f 0.076 20.00 19.92 f 0.088 60.00 59.72 f 0-082 20.00 20.05 & 0.095 80.00 80.36 f 0.088 20.00 20.16 f 0.096 C2H,NH, in the (C,H,),N in the * Mean of six determinations, f standard deviation. Ratio of C,H,NH, to 1: 1 1:2 1:3 1:4 2: 1 3: 1 4: 1 (CZH,) ,N Amides (acetamide, urea, salicylamide and nicotinamide) , imides (phthalimide and suc- cinimide) , Schiff ' s bases (N-$-chlorobenzylideneaniline and N-cinnamylideneanisidine) and tertiary amines (triethylamine, pyridine, a-picoline, quinoline and isoquinoline) do not cause any interference even when present in up to a five-fold excess in the determination of primary or secondary amines by the proposed method.Thiourea, thiosemicarbazide, thioacetamide, phenylhydrazine, xanthates, dithiocarbamates and organic isocyanates, however, do interfere. Thiourea, thiosemicarbazide, thioacetamide, phenylhydrazine, xanthates and dithiocarba- mates interfere as they are oxidised by potassium iodate under the experimental conditions described for the determination of amines by the proposed method.If any of these compounds is therefore present in admixture with the listed amines, more potassium iodate than is required to oxidise quantitatively the substituted thiourea formed from amines (with phenyl isothio- cyanate) will be consumed, and hence erratic results will be obtained. The interference caused by organic isocyanates is, however, attributed to their reaction with amines to form substituted ureas. The method could not be extended to the determination of aromatic amines as their reaction with the isothiocyanate was extremely slow. TABLE I11 ANALYSIS OF MIXTURES OF DIETHYLAMINE AND TRIETHYLAMINE Amount of (C,H,),NH in the mixturelmg Amount found*/mg 20.00 19.92 f 0.083 20.00 20.10 f 0.081 20.00 19.88 f 0.096 20.00 19.96 f 0.055 40.00 40.22 f 0.087 60.00 60.35 & 0.073 8040 79.55 & 0.082 Amount of (C2H513N in the mixture/mg Amount found*/mg 20.00 20.10 f 0.124 40.00 40.14 f 0.156 60.00 59.58 f 0.141 80.00 80.56 f 0.104 20.00 19.86 f 0.088 20.00 19.90 & 0-114 20.00 20.08 f 0.096 * Mean of six determinations, f standard deviation. Ratio of (C,H,)2NH to (C2H5) BN 1: 1 1:2 1:3 1:4 2: 1 3: 1 4: 1 Amylose was not suitable in the visible end-point iodatometric determination of pipe- ridine and pyrrolidine.The method proposed for the deterrnination of amines has been called an iodatometric method because potassium iodate is used as an oxidimetric reagent to titrate the substituted thioureas formed from amines by reaction with phenyl isothiocyanate.The end-point of each titration is marked by the first distinct and permanent appearance of iodine502 VERMA AND KUMAR from the first drop of iodate solution in excess. Visually, the end-point can be detected either by the yellow colour that iodine imparts to the solution, or by the well known starch - iodide blue colour if amylose (a component of starch) is used as an indicator. The detection of the end-point by observing the yellow colour due to iodine is found to be satisfactory in the titra- tions of each of the listed amines by the above method. However, when amylose is used as the indicator, the change from colourless to blue is sharp in titrations of all of the listed amines except piperidine and pyrrolidine. Thus, amylose is not a suitable indicator for titrations of thioureas formed from piperidine and pyrrolidine, by reaction with phenyl isothiocyanate, with potassium iodate in an acidic medium.The proposed methods for the determination of primary (or secondary) amines and tertiary amines in the presence of each other, besides being simple, accurate and reliable, have the added advantage that the analysis can be conducted on the same sample solution, thus saving time and effort. The changes in slope produced at the equivalence points in the conducti- metric titrations of tertiary amines are sharp enough to allow easy location of the end-point, thus giving accurate results. Synthetic mixtures of ethylamine and triethylamine with the ratios in the range from 1 : 4 to 4 : 1 can be analysed with an average standard deviation of 0-31 per cent., in the instance of ethylamine, and 0.35 per cent., in the instance of triethylamine (Table 11).The replicate samples of any amine under investigation were not weighed indi- vidually, but rather a single, large sample was weighed, dissolved in a known volume of dimethylformamide and aliquots taken for analysis in the titration vessels. That the replicate aliquots of an amine solution delivered the same amount of amine each time was checked by independent acidimetric titrations of that amine in several such aliquots. This check was applied to all the amines, including ethylamine and isopropylamine, which are available as 50 and 70 per cent. aqueous solutions, respectively. The method has also been applied to mixtures of pyridine, a-picoline, quinoline or triethylamine with n-propylamine, n-butylamine and isobutylamine. The results recorded in Table I11 for the analysis of synthetic mixtures of diethylamine and triethylamine, also in ratios from 1 : 4 to 4: 1, show that the mixture can be analysed with an average standard deviation of 0.29 per cent. (for diethylamine) and 0.41 per cent. (for triethyl- amine), respectively. The method has also been extended to other mixtures, such as diethyl- amine with pyridine, cc-picoline or quinoline and piperidine and pyrrolidine with pyridine, a-picoline, quinoline or triethylamine, which have been determined to the same accuracy. The authors thank the Council of Scientific and Industrial Research (India) for the award of a fellowship to one of them (S.K.). The results agree well with those of the tabulated mixture REFERENCES 1. 2. 3. 4. Singh, B., and Verma, B. C., J . Sci. Ind. lies., 1965, 24, 536. -,- , 2. analyt. Chem., 1963, 194, 112. Verma, B. C., and Kumar, S., Analyst, 1973, 98, 900. Gaslini, F., and Nahum, L. Z., Analytica Chim. Acta, 1961, 24, 79. Received December 19th, 1973 Accepted February Sth, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900498

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, pp. 503-514 503 A Technique for the Determination of Trace Anions by the Combination of a Potentiometric Sensor and Liquid Chromatography, with Particular Reference to the Determination of Halides* BY M. C. FRANKS AND D. L. PULLEN (BP Research Centre, Chertsey Road, Sunbury-on-Thames, Middlesex, T W16 7LN) A technique for the determination of trace amounts of halides in the presence of other ions is described. The species are separated by means of liquid chromatography and detected potentiometrically by a silver - silver chloride micro-electrode. The technique readily lends itself to automation and an apparatus for rapid, repetitive analyses has been designed. By careful choice of the eluting agent and stationary phase it is possible to achieve a variety of separations, e.g., the determination of chloride in the presence of excess of sulphide, the separation and determination of nanogram amounts of chloride, bromide and iodide in mixed halide solutions and the determination of chloride in boiler waters.In addition, conditions for the extension of the technique to the separation and determination of other anions are proposed. ALTHOUGH there are many methods available for the determination of trace amounts of metals, including highly developed polarographic procedures and a variety of spectroscopic methods, there are few techniques that are of general application to the detection and determination of trace amounts of anions. The basis of the technique described in this paper is the separation of anions by liquid chromatography and their determination potentiometrically by an ion- selective micro-electrode, a combination that ensures high degrees of specificity and sensitivity.Until fairly recently, the advantages of an electrode as a potentiometric detector in a flowing solution have not been fully appreciated. In reviews of liquid-chromatographic detectors, l Y 2 potentiometry is given only a brief mention. This is surprising as potentiometric detectors are used throughout industry for the control of pH and sodium3 and ammonia concentrations. For the determination of anions, modern ion-selective electrodes have proved to be successful sensors for f l ~ o r i d e , ~ sulphide,6 cyanide7 and chlorides and similar flow-through systems have also been described in biomedical research.However, in these instances no chromatography was involved. In this work, the importance of the chromato- graphic process is emphasised as it can both replace difficult separations and enable non- selective sensors to be used as specific detectors. A particular aspect of the potentiometric detector has also been given special attention, namely its use at very low concentration levels, near to the limit of detection. It is in this region that the detector is useful for trace analysis. In order to illustrate the technique the procedure for a specific application, trace halide analysis, is described in detail in the experimental section of this paper. It includes an account of the use of a rapid, repetitive automatic analyser for the continuous measurement of chloride samples in the parts per million concentration range in microlitre volumes.The theoretical aspects of electrode response have received comprehensive treatment elsewhereg and are considered only briefly here. The potentialities of the technique for general application to the determination of anions are also considered, exploratory work is described and the implications are discussed. EXPERIMENTAL DESCRIPTION OF THE APPARATUS- A block diagram is shown in Fig. 1. A micro-scale metering pump circulates the mobile phase through a narrow-bore plastic tube containing the stationary phase. At the end of the tube is a micro-electrochemical detector, which is connected through a high-stability amplifier to a chart recorder. A few microlitres of sample solution are injected into the stationary phase, either manually through a septum, or automatically through a sampling valve.The * Presented in part at the Third SAC Conference, Durham, July 12th to 16th, 1971. @ SAC and the authors.Pump r Pulse suppressor . - 1 njection position 1 Column AAugust, 19741 WITH A POTENTIOMETRIC SENSOR AND LIQUID CHROMATOGRAPHY 505 was held in a suspension of the stationary phase, e.g., ion exchanger in distilled water, 1 + 1 by volume. The column was connected to the micro-pump and a solution of eluting agent was pumped at a flow-rate of 0-3 to 0-4 ml min-l for several hours in order to settle the resin, after which the column was cut to the desired length and connected to the injection valve. Detector system-The detector system consists of two silver - silver chloride electrodes; one is used as a reference while the other is placed in the flowing stream.A detailed diagram of the detector is shown in Fig. 2. The design of the detector is such that it is easy to replace electrodes in the housing, and the dead volume around the indicator electrode is kept to a minimum (50 p1 or less). The housing and electrode barrels are made from Pyrex glass, the latter being ground down just sufficiently to allow a steady flow of circulating eluting agent through the housing. Quickfit and Quartz screw-thread adaptors are used as connectors. A 3-mm length of 3-2 mm diameter silicon carbide rod, grit type F, purchased from AEI, Manchester, was sealed into the glass wall to provide a restrained flow liquid junction between the reference and indicator electrodes.Electrodes-These were made from 16 s.w.g. silver wire, shaped as in Fig. 2, and chloridised in hydrochloric acid. Conditions were controlled to produce batches of electrodes with bias potentials of less than 0.2 mV. Instrumentation-The two essential units are a buffer amplifier and a recorder. In some of the preliminary work a Keithley or Vibron Model 33B-2 electrometer was used to measure the potential changes and to investigate electrical parameters. The present apparatus has a purpose-built amplifier, the chief characteristics of which are low bias current, low noise and high impedance. Incorporated into the amplifier are a damping device and a narrow-band filter to remove residual oscillations of the output potential..Flow meter Fused silver c h I or ide C h ro ma tog rap h ic column + Fig. 2. Assembled detector (dimensions are given in millimetres) The signal can be displayed on any paper chart recorder that matches the output characteristics of the amplifier; a floating input will be necessary. A Vitatron UR 400M recorder, supplied by Fisons Scientific Apparatus Limited, is the least expensive that has been used satisfactorily . Careful attention to the screening of connecting leads is necessary in order to avoid excessive interference noise on the recorder trace. Co-axial cable is used for both input connections to the amplifier ; screened leads are satisfactory elsewhere. The high-impedance terminal of the amplifier should always be connected to the reference electrode, regardless of polarity, and this lead should be kept as short as possible.From the indicator electrode there are resistive and capacitative paths to earth through the circulating liquid, which is in contact with the earthed frame of the pump. It is therefore necessary to connect all screening to this point. The screening of the leads to the recorder should not be connected to the earthed recorder frame. The whole apparatus is enclosed in a metal box that acts as a Faraday cage to screen the detector from external electrical interference. ASSEMBLY OF THE APPARATUS- The apparatus is assembled in the following sequence to avoid inclusion of air: first, the reservoir, pump and pulse suppressor ; then, with the pump running, the injection valve,506 FRANKS AND PULLEN : DETERMINATION OF TRACE ANIONS (HALIDES) [Analyst, VOl.99 chromatographic column, detector and flow meter are connected. Plastic tubing and 6.35 mm i.d. Simplifix brass connectors are used to link the components. When the mobile phase has passed through the sintered grit into the reference electrode compartment the latter is topped up with the mobile phase and the electrode inserted. The electrical connections are made, the pump is set to the required flow-rate and the apparatus left running for several hours in order to stabilise it. OPERATION OF THE APPARATUS- If the operation is carried out manually, the sample is added with a micro-syringe, making sure that the tip of the needle passes through the centre of the T-junction and into the ion-exchange resin.The sample is carried by the mobile phase to the detector, where the chloride is detected and shown on the recorder as a chromatographic peak. The sample volume is chosen so that the peak height is 5 mV or less as it is in that region that the peak height is proportional to the chloride content, A maximum volume of 100 p1 is observed in order to prevent peak broadening. REAGENTS- Water-Water with a very low chloride content, preferably below 20 p.p.b. (parts per l o g ) , is used to prepare all of the standard solutions and mobile phases and to wash all the apparatus. Suitable water can be produced by de-ionisation of laboratory distilled water so as to obtain a specific resistance of greater than 4 Mi2 cm. The chloride content can be deter- mined, when necessary, by the method of Rodabaugh and Upperman.lo Polythene bottles with screw-caps are convenient for storing the de-ionised water.Standard chloride solution A-Dry analytical-reagent grade sodium chloride in an oven at 250 to 350 "C for 1 to 2 hours, then weigh 1.649 g of it into a 1-litre Pyrex glass calibrated flask, dissolve it in water and dilute to the mark with water. This solution should be stable for at least 1 year. A small volume of sample is injected into the column. 1 ml of solution A = 1000 pg of chloride. Standard chloride solution B-Transfer 10 ml of the standard solution A with a pipette We found this into a 500-ml Pyrex calibrated flask, dilute to the mark with water and mix. solution to be stable for at least 6 months.1 ml of solution B = 20 pg of chloride. Other halide solutions-Prepare 1000 and 20 p.p.m. solutions of bromide and iodide by weighing and diluting appropriate amounts of the sodium salts of these halides as described above. Ion-exchange resins-Strongly acidic resins, Chromatographic grade, 200 mesh, were obtained from Permutit Company Limited. All other chemicals used were of analytical-reagent grade unless otherwise specified. TEMPERATURE- Prepare the iodide solutions just before use. The room temperature varied between 19 and 21 "C. EVALUATION OF THE DETECTOR- There are few useful references to the behaviour of a potentiometric detector in a flowing solution of low ionic activity, especially at the low detection limit set by solubility considera- tions.A major part of the initial work was to establish that stable electrode potentials could be obtained under such conditions. This attempt involved the design of a micro-cell and associated instrumentation that would produce suitably low noise signals. Some aspects of this work and a simple theoretical treatment are described. THEORY OF THE RESPONSE OF THE SILVER CHLORIDE ELECTRODE- The potential of the silver - silver chloride system is given by .. .. * - (1) RT RT In K,- - In y Mcl- . . E = E o A g / ~ g + + F where EoAg,Ag+ is the standard electrode potential, R is the gas constant, T is the absoluteAugust, 19741 WITH A POTENTIOMETRIC SENSOR AND LIQUID CHROMATOGRAPHY 507 temperature, F is the Faraday constant, Ma- is the chloride-ion concentration and y is the activity coefficient of the chloride ion.The change in potential, AE, in a constant ionic medium, when the chloride concentration changes to M'cl-, is given by .. .. .. .. AE= RT - I n ( - - - - ) . . M'cl- F MCl- When the silver chloride electrode is immersed in a solution a certain amount of silver chloride dissolves from the surface. If the chloride concentration of the solution before immersion is x, then at equilibrium the total concentration of the chloride ions, N'cl-, is given by where K, is the solubility product of silver chloride. For a chloride-free solution, when x = 0, Ma- = 2/%. When a chloride-free solution is replaced by one containing chloride ions, we have, from equations (2) and (3)- x When - is small compared with dz, (4) becomes 2 For an electrode in a flowing solution at equilibrium, (5) can be written- ... . initial chloride concentration added chloride concentration AE= - Thus, the change in potential is proportional to the added chloride concentration (x). D T L\ L The sensitivity of the detector is obtained from the ~ term in equation (5) and is 2F2/K, 3.3 mV per 0.1 p.p.m. of chloride at 20 "C. B a r d i ~ ~ , ~ in a more detailed account, showed that this approximation held for concentra- tion changes up to 0.2 p.p.m. of chloride. By analogy it can be shown that low concentrations of other anions with insoluble silver salts should produce potential changes proportional to the concentrations in which they were added. Conjirwation of theoretical response-It was shown experimentally that the electrode response was proportional to chloride concentrations of up to 0.2 p.p.m.and that the sensitivity was 3 mV per 0.1 p.p.m. of chloride. These findings were obtained with flowing solutions at 20 "C. The linearity of the calibration graphs from many other experiments, and those shown in this paper, also confirm the proportional response that was expected at low halide concentrations. Experimental conditions that affect electrode response are discussed in the next three sections. Efect of flow-rate-The potential of the silver chloride detector was measured under flowing and static conditions with chloride concentrations from to M. The flow- rates were in the range from 0-1 to 1.0 ml min-l. As expected, the potential of the electrode was found to vary with the flow-rate.The change in potential from a static to a flowing condition could be as large as 30 mV and was found to depend on various parameters, such as the nature and concentration of the mobile phase, the nature of the stationary phase and the material of construction of the column, i.e., polythene or nylon. It is not possible to state508 [Andyst, Vol. 99 unequivocably what underlying phenomena produced these potential changes. Although this aspect is of some theoretical interest, it was not investigated in detail as we were primarily concerned with the practical problems of measuring the potential variation, noise level and base-line change in a constant-flow system. Provided that the flow-rate is maintained within &2 per cent., the change in electrode potential is small and has little effect on the repeatability of results that are obtained.E$ect of temperature-The effects of temperature on the detector are complex. The potential of both the reference and detecting electrode will vary with the factor - and the solubility term In Ks, which is itself temperature dependent. The practical effects of change in temperature are two-fold. A change in base-line occurs on the recorder trace and the sensitivity to chloride alters [as shown in equation (S)]. The effect of temperature onsensitivity was tested over the range from 17 to 23 "C by equilibrating the whole apparatus before measur- ing the electrode response. A decrease in sensitivity of 2 per cent. per 1 "C rise was found. This is in agreement with the theoretical expectations of equation (5) from consideration of the RT combined effect of the- term and the dz term, the latter having the larger temperature F coefficient.By controlling the temperature to &l "C sufficient accuracy for calibration and measurement was obtained for our needs, and the normal base-line drift was small enough to enable small peaks to be easily discerned. E$ect of light-As silver chloride electrodes are known to be photosensitive the detector assembly was shielded from direct light. FRANKS AND PULLEN : DETERMINATION OF TRACE ANIONS (HALIDES) RT F CALIBRATION- Calibration graphs were obtained by injecting a series of known halide solutions that had been prepared by dilution of the standard halide solution. The peak height was plotted against halide content.The calibration graphs were linear (Figs. 3 and 4) except at high chloride concentrations where the Nernstian logarithmic relationship was found. Concentration of chloride, p.p.m. Fig. 3. Calibration graph for chloride in high-pressure boiler water: a, chloride in pure water; and 0, simulated boiler water (Table I) plus added chloride (sample volume 20 4) STABILITY OF ELECTRODE RESPONSE- In the supporting work, in addition to confirming the theoretical response it was also established that the potential of an electrode in a high, but constant, velocity stream is surprisingly stable. In the apparatus described above the cell and instrumentation were carefully designed such that the normal base-line peak-to-peak noise is less than 15 pV.The daily base-line non-cumulative drift is less than 0.2 mV for most of the analytical applications. Base-line stability in the order of less than 15 pV is essential if the technique is to be used for the determination of nanogram amounts of halide, because the concentration of a sampleAugust, 19741 WITH A POTENTIOMETRIC SENSOR AND LIQUID CHROMATOGRAPHY 509 is considerably reduced by the chromatographic separation. If it is assumed that the minimum detectable signal is twice the base-line noise level, then from Fig. 5, the limits of detection of chloride, bromide and iodide in this particular system are 4, 1 and about 2 ng, respectively. Similarly, the limit of detection for other systems will depend on the response and the noise level. Amount of halide/ng Fig.4. Calibration graphs for chloride, bromide and iodide. Cadmium acetate chromatographic system Concerning the lower limit of detection that it is possible to attain, we think that with careful electrical shielding and noise reduction, stability of the order of microvolts can be obtained in a flowing solution. We agree with Lightll that “when this stability is achieved electrodes will obey the Nernst equation much better than is generally believed.” Bromide n Nitrate a fluoride 1- 30 minutes- I 0- Fig. 5. Separation of chloride, bromide and iodide in pre- sence of fluoride, sulphate and nitrate. Stationary phase, Zeo-Karb 225, cadmium form; mobile phase, 0-0025 M cadmium acetate, flow-rate 0.4 ml min-1. Sample 50 pl, containing 1000 p.p.m. of sulphate, nitrate and fluoride with 1 p.p.m.each510 FRANKS AND PULLEN: DETERMINATION OF TRACE ANIONS (HALIDES) [Analyst, Vol. 99 APPLICATIONS A few of the applications of the technique to the determination of chloride are described below. In each instance the silver - silver chloride electrode is used. The applications have been selected to show how interferences can be removed by chromatographic separation or suppressed by use of a suitable mobile phase. 2 - DETERMINATION OF CHLORIDE IN THE PRESENCE OF SULPHIDE- Corrosion problems in crude oil distillation units have been the concern of the petroleum industry for many years and require the monitoring of chloride at the parts per million level in the aqueous overheads. Large excesses of sulphide and thiols are present together with some oxy-acids of sulphur.The chloride content of aqueous overheads can be determined by conventional potentiometric or visual indicator titrations, after removal of the interfering substances, but these methods are time consuming. It was found, after preliminary investigational work, that chloride could be separated completely and quantitatively from all interfering substances by use of a chromatographic column consisting of the lead form of a cation exchanger as the stationary phase and dilute lead acetate solution as the mobile phase.12 When a small sample, e.g., 1 5 ~ 1 , is injected into the chromatograph, sulphide and thiols are precipitated as the insoluble lead salts, while the other sulphur compounds are separated from chloride ions by the ion-exchange system.Suitable chromatographic conditions were as follows: stationary phase, a nylon column 80 cm long, i.d. 1.6 mm, filled with Zeo-Karb 225 in the lead form, 200 mesh; mobile phase, a 0.001 M lead acetate solution, flow-rate 0.4 ml min-1. f l 5 p.p.m. Sample + 4 p.p.m. r\ I 01 -1 hour-4 Fig. 6. Typical chloride peaks from automated laboratory analyser. A and B are calibration peaks; peak C is from a sample of aqueous overheads (S) con- taining 1 p.p.m. of chloride plus about 100 p.p.m. of sulphide; and peak D is sample (S) plus 4 p.p.m. of chloride. Sample volume, 15 pl; retention time, 6 minutes A typical set of calibration and sample peaks is shown in Fig. 6. The recovery of chloride is greater than 95 per cent. and as little as 0.5 p.p.m. of chloride can be measured with an elapsed time of 15 minutes.The basic apparatus has been converted for use as an automatic laboratory analyser by injecting samples from a sliding valve device controlled by a simple programmer. This analyser was run successfully for several months to determine the chloride content of several hundred samples from aqueous overheads before a marked deterioration in apparatus response and a build-up of pressure necessitated renewal of the column and detector. A statistical analysis of fifty samples, tested in duplicate, gave a standard deviation of 0.3 p.p.m. of chloride for levels below 10 p.p.m. Twenty samples covering the range 0 to 100 p.p.m. were tested by use of the automatic analyser and by a standard chemical procedure. There was no significant bias.August, 19741 WITH A POTENTIOMETRIC SENSOR AND LIQUID CHROMATOGRAPHY 511 CHLORIDE IN BOILER WATERS- The determination of trace amounts of chloride in boiler water presents a somewhat similar problem to that described above.Although standard chemical procedures exist they still require considerable time and manipulation. The main types of boilers give rise to three kinds of boiler waters, which normally contain the substances listed in Table I, present either as additives or contaminants. TABLE I TYPICAL BOILER-WATER COMPOSITIONS High pressure Chloride 0-1 (120 bar) Silica 0-2 Boiler type Sample constituents Concentration, p.p.m. Ammonia 0-5 Medium pressure Chloride (40 bar) Phosphate Hydrazine Silica Low pressure Chloride Phosphate Sulphite Sulphate 0-10 0-70 0.1-1 0-25 0-10 0-50 0-150 0-300 The lead form ion exchanger and lead acetate eluting agent system described previously cannot be used as large amounts of chloride are lost, presumably by adsorption on to the lead phosphate precipitate.(We have found that since this work was carried out the co-precipita- tion of chloride on to lead phosphate has been put forward as the basis of a method for determining amounts of chloride of the order of parts per billion.lO) After some initial experimentation the following conditions were found to be suitable for the determination of chloride in all types of boiler waters: stationary phase, a column 10 cm long and of 3-2 mm i.d., containing 200-mesh powdered polythene ; mobile phase, 100 p.p.m. orthophosphoric acid solution, flow-rate 0.3 ml min-l.The orthophosphoric acid has the effect of suppressing to a practically constant level the small but insignificant interference due to varying levels of phosphate in actual boiler waters. The powdered polythene placed between the injection points and the silver - silver chloride detector merely acts as a convenient means of presenting the sample to the detector so that a reproducible peak is obtained. No chromatographic separation is needed as constituents other than phosphate do not interfere significantly. In Figs. 3 and 7 calibration graphs obtained by adding known amounts of chloride to simulated water from low and high-pressure boilers are shown. By use of these graphs we have analysed a large number of samples and found that for low-pressure boiler waters the 7 Concentration of chloride, p.p.m.Fig. 7. Calibration graphs for chloride in low-pressure boiler water : 0, chloride added to distilled water ; and A, chloride added to synthetic boiler water (Table I). Peak heights are corrected for a boiler water blank E 0.9 mV to show coincidence of the graphs512 FRANKS AND PULLEN : DETERMINATION OF TRACE ANIONS (HALIDES) [ArtalySt, VOl. 99 results of determinations with this analyser and standard chemical analyses generally agree to within 0.1 p.p.m. of chloride. As an example of the performance that can be achieved at low chloride levels, high-pressure boiler waters containing 0 to 1 p.p.m. of chloride were analysed with a standard deviation of 0.02 p.p.m. (30 degrees of freedom) at a rate of 30 samples per hour.DETERMINATION OF HALIDES IN MIXTURES- The first two applications were concerned solely with the determination of chloride. It was observed in the supporting work for these applications that, with certain chromatographic systems, there was an indication of the separation of bromide and iodide. A literature search revealed many ion-exchange systems that are capable of separating the halides in mixtures. However, most of these methods are lengthy, requiring 3 to 4 hours, apply to milligram or greater amounts, and make use of eluting agents that affect the response of the silver - silver chloride electrode. A thin-layer chromatographic system that involved the use of cadmium salts as eluting agents was found13 that appeared to be suitable.After a preliminary study of the factors affecting the separation, namely the degree of cross-linking of the resin, the resin mesh size and the strength of the eluting agent, the following conditions were arrived at for complete separation of the halides: a column 80 cm long, 3.2 mm in id., packed with 4 to 5 per cent. cross-linked Zeo-Karb 225 (cadmium form), 200 mesh, as the stationary phase; and a 0.0025 M cadmium acetate solution, flow-rate 0.4 ml min -l, as the mobile phase. In Fig. 5 the chromatogram is shown that is obtained from the injection of a 5O-pl sample of a solution containing 1 p.p.m. each of chloride, bromide and iodide in the presence of 1000 p.p.m. of some common anions. The halides are well resolved and are eluted in order of the increasing ionic strength of their cadmium complexes.Under the most suitable experimental conditions it is possible to separate and determine a minimum of 4 ng of each halide. This technique is therefore much more sensitive than existing electrochemical or spectrophotometric procedures. Interferences in halide determination--It is possible to predict, in qualitative terms, the effect of various anions on the silver - silver chloride detector. As expected, sulphate, nitrate, perchlorate, fluoride and other anions with soluble silver salts do not interfere, either as salts or as free acids, even at 1000 times the concentration of the halide. They are also eluted before chloride. Other anions with insoluble silver salts, e.g., chromate and thiocyanate, or which form silver complexes, e.g., thiosulphate and cyanide, produce a response at the electrode the magnitude of which depends on the solubility product or the instability constant of the resulting compound.Because of the low responses of certain anions or the wide separation of peaks (Table II), it can be seen that only sulphide, thiocyanate and thiosulphate can possibly interfere in the determination of halide to any extent. The degree of interference depends on the relative amounts of interferent and ion sought. In Table 11, the relative elution times of some anions are given. TABLE I1 RELATIVE ELUTION TIMES OF SOME ANIONS WITH ZEO-KARB 225 (CADMIUM FORM) AS STATIONARY PHASE AND 0.0025 M CADMIUM ACETATE AS ELUTING AGENT F-, Cr042-, NO,- Sod2-, H,PO4- s,o,2-, s2- c1- Br- CNS- I- CN- C-v-J - L--y----J 0-5 0-6 0.8 1.0 1-25 2.0 2.2 10 FURTHER APPLI CAT1 0 N S- It is evident that the system described above has numerous practical applications in traceanalysis.A few such applications which have been successfully carried out in our laboratory include : the determination of halides in well waters ; the determination of chloride in the presence of a large excess of fluoride, phosphate and sulphate; the direct determination of bromide in sea water, in which the chloride to bromide ratio is about 3000: 1 (Fig. 8); and the determination of chloride in concentrated cyanide solutions. The system has also been used to determine halogens in solutions resulting from Schoniger flaskAugust, 19741 5 13 and Wickbold apparatus combustions. In the latter procedure, high results or large blanks can be produced if the chloride is determined by the mercury(I1) thiocyanate colorimetric techniqueJ14 or by titration with silver or mercury salts, especially if gram amounts of sample are burnt.It has been conclusively shown, by using the chromatographic - potentiometric technique, that these “high blanks” are due to reactions with the products of combustion and not to chloride. Finally, it should be emphasised that the applications mentioned above have of ten simplified analyses previously requiring time-consuming classical methods. WITH A POTENTIOMETRIC SENSOR AND LIQUID CHROMATOGRAPHY 5 4 > E 2 3 --. m a, 1 Y a .- $ 2 1 Chloride approx. 20 000 p.p.rn. 0 -30 minutes 4 Fig. 8. Direct determination of bromide in sea water.Conditions: cadmium acetate chromatographic system, flow-rate 0.2 ml min-1; sample size, 1 pl. The bromide content was found to be 60 p.p.m., compared with the chemically determined value of 65 p.p.m. DISCUSSION Although other workers have proposed the use of ion-selective electrodes as liquid chromatographic potentiometric detectors,15 we have not been able to find examples of the practical application of this technique in the literature. Our work has shown that not only is this technique suitable for the determination of the concentration of anions in flowing solutions, but that it can be used for their quantitative measurement at the parts per million level and it is also capable of determining halides in mixtures after a single chromatographic separation. A limiting factor in the separation and determination of complex mixtures is the possibility of irreversible changes taking place at the electrode surf ace.Some aspects of this limitation were observed in the initial calibration runs with mixed halide solutions. At first the response to chloride decreased rather rapidly and the base-line level changed between the halide peaks. However, if the electrode is conditioned by circulat- ing the mobile phase, containing a few milligrams of mixed halides, overnight, the response becomes practically constant from then onwards and the base-line becomes horizontal (see Fig. 5), with the result that the analysis of mixed halide solutions becomes easier and more precise. The sensitivity to bromide and iodide is essentially that expected from the theoretical electrode response and chromatographic dilution, but the sensitivity to chloride is much lower, presumably because of the formation of a mixed halide electrode (Fig.4). The success of the technique for the determination of amounts of halides of the order of nanograms suggested that this special application could be extended to the general determina- tion of other anions, and further work has indicated that there is every possibility of doing this. By using the cadmium acetate chromatographic system and a silver - silver chloride electrode we have been able to determine thiosulphate and cyanide in concentrations in the parts per million region. The Orion liquid-state ion-selective electrode, which has been recommended for the determination of the chloride ion but which is also sensitive to other anions, has been produced in miniature.With this micro-electrode as a detector and the same chromatographic system514 FRANKS AND PULLEN as above we have been able to detect amounts of nitrate and perchlorate in the parts per million range. The combination of a lead amalgam electrode and cellulose-column chromatography has permitted the separation and determination of sulphate and phosphate ions. By means of a barium sulphate impregnated silicone-rubber electrode, trace amounts of sulphate and phosphate in solution have been determined. Further work is required in order to extend the life of the electrode and increase the sensitivity of these procedures and it is hoped to publish fuller details in a further paper.This preliminary work is encouraging and seems likely to lead to interesting developments. The possibilities of devising other chromatographic - potentiometric systems seem to be considerable. Many of the modern types of ion-selective electrode and the latest develop- ments, namely selectrodes16 and coated-wire electrodes,17 may be suitable as micro-sensors. These sensors do not need to be highly selective in instances when there has been some prior chromatographic separation. Concerning chromatographic separations for anions, many have already been described in the literature of ion-exchange chromatography and could be adapted on a micro-scale. Thin-layer chromatography is a particularly fruitful source of separation systems, especially those where the anions of interest have high R, values, as adaptations of such systems lead to short analysis times in column chromatography.The cadmium acetate system described under Applications was developed from such a thin-layer chromatographic method. l3 C o N c L u s I o N s The characteristics of the silver - silver chloride electrode in a flowing solution have been established. Small changes in potential and hence small changes in the concentrations of halides can be measured. The combination of chromatographic separation and potentiometric detection has been successfully applied to the determination of amounts of chloride of the order of nanograms in the presence of certain interfering anions and these applications offer attractive alternatives to existing methods. An advantage of the technique is that it is suitable for rapid, repetitive analysis and, in particular, it is easily adapted to continuous monitoring. Although the work with other electrodes and chromatographic systems has been rather limited, it has indicated that the technique should be capable of extension to the detection and determination of many other anions. Permission to publish this paper has been given by the British Petroleum Company Limited. The authors thank Mr. A. H. Jarratt for his experimental work and the design of the instrument ation. The potential is very stable under conditions of constant flow. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. REFERENCES Conlon, R. D., Analyt. Chem., 1964, 41, 107A. Huber, J. F. K., J . Chromat. Sci., 1969, 7, 172. Diggens, A. A., Parker, K., and Webber, H. M., Analyst, 1972, 97, 198. Goodfellow, G. I . , and Webber, H. M., Ibid., 1972, 97, 95. Babcock, R. H., and Johnson, K. A., J . Amer. Wat. W k s Ass., 1968, 60, 953. Light, T. S., Ind. Wat. Eng., 1969, 6, 33. Fleet, B., and Von Storp, H., Analyt. Chem., 1971, 43, 1575. Tee, T. G., Ibid., 1069, 41, 391. Bardin, V. V., Zav. Lab., 1962, 28, 910. Rodabaugh, R. D., and Upperman, F. T., Analytica Chim. Acta, 1972, 60, 434. Light, T. S., in Durst, R., Editor, “Ion Selective Electrodes,” National Bureau of Standards Special British Patent 1201 850, 1970. Lederer, M., Movpurgo, L., and Ossicini, L., J . Chromat., 1970, 50, 475. Zall, D. M., Fisher, D., and Garner, M. O., Analyt. Chem., 1956, 28, 1665. Liteneanu, C., Popescu, I. C., and Nascu, H., Revue Roum. Chim., 1972, 17(9), 1615. RhiiEka, J., and Lamm, G. G., Analytica Chim. Acta, 1971, 54, 1 and 29. James, H. J., Varmack, G. P., and Freiser, H., Amlyt. Chem., 1972, 44, 856. Publication No. 314, 1969, p. 432. Received January 18th, 1973 Amended December 28th, 1973 Accepted February 22nd, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900503

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, PP. 515-518 515 A Method for the Determination of Total Sulphur in Silicate Rocks BY J. M. MURPHY AND G. A. SERGEANT (Department of Trade and Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SE1 9NQ) After oxidising sulphides to sulphates with a mixture of sodium chlorate and hydrochloric acid, the sample is refluxed with a reducing mixture of sodium iodide, red phosphorus, hypophosphorous acid, orthophosphoric acid and propionic acid. The hydrogen sulphide generated is absorbed in potas- sium hydroxide solution and titrated against 2-(hydroxymercuri)benzoic acid solution with dithizone as indicator. The approximate range covered is 5 to 2000 mg kg-1 of sulphur in the rock. THE sulphur content of common silicate rocks is usually well below 1000 mg kg-1 and is therefore too low to be accurately determined gravimetrically as barium sulphate, following decomposition of the rock by oxidising fusion, as in the conventional procedure, or by wet decomposition.Combustion techniques have been used in which the sample is heated, either with or without a flux such as vanadium(V) oxide, and the sulphur evolved as oxides of sulphur is determined by titration or else spectrophotometrically. This latter type of method is fairly rapid, but temperatures as high as 1450 "C appear to be required and recoveries of sulphur are not reliably quantitative.2 A sensitive method involving the use of wet reduction has recently been described3 in which the sample is heated at 280 "C with a mixture of anhydrous orthophosphoric acid and tin (11) chloride.Sulphur, present either as sulphide or sulphate, is evolved as hydrogen sulphide, which is then determined spectrophotometrically. In the proposed method sulphur is recovered from the sample as hydrogen sulphide by heating with a reducing mixture modified from that used by Johnson and Nishita,* comprising sodium iodide, red phosphorus, hypophosphorous acid, orthophosphoric acid and propionic acid. The solid iodide is more convenient to use than hydriodic acid, and as a diluent propionic acid has been selected in preference to formic or acetic acids on account of its higher boiling- point. Red phosphorus, in addition to contributing to the reducing action, is a very effective anti-bumping agent.It is necessary to oxidise resistant sulphides, of which the commonest is pyrite, by a preliminary oxidation to sulphate. This oxidation is effected by heating the powdered sample with a mixture of sodium chlorate and concentrated hydrochloric acid. On evaporating the mixture to dryness, the solid residue includes an excess of sodium chloride, which provides bulk and thus facilitates the quantitative transfer of the residue to the distillation flask. In an earlier version of the proposed method the hydrogen sulphide generated by reduc- tion was determined by a colorimetric methylene blue finish similar to that described by Johnson and Nishita. This finish, although very sensitive, had a very restricted range and was limited in accuracy by the spectrophotometric properties of methylene blue solutions, even when modified by the addition of pyridine5 in an attempt to improve linearity.A titri- metric finish was next attempted and promising results were obtained by using mercury(I1) acetate as titrant with dithizone as indicator. The sensitivity of the method was retained and the range covered was extended, but it tended to be limited by the obscuring effect of the dark colour of mercury(I1) sulphide on the end-point of the titration. This defect has now been eliminated by changing the titrant from mercury( 11) acetate to 2-(hydroxymercuri)- benzoic acid, first proposed as a titrant for sulphide by WroAski.6 In the reaction- + S?' acoo- - o o ! c n f 2 OH- Hg -S -Hg @ SAC; Crown Copyright Reserved.51 6 [Analyst, Vol.99 the reaction products are colourless and a good end-point is obtained up to the recommended limit for the method of 2000 mg kg-1 of sulphur in the rock sample. Above this limit a gravimetric procedure is usually to be preferred. REAGENTS- MURPHY AND SERGEANT: A METHOD FOR THE DETERMINATION METHOD Analytical-reagent grade materials should be used whenever possible. Sodium chlorate. Hydrochloric acid, concentrated, sp. gr. 1 - 18-Aristar or equivalent grade. Sodium iodide. Phosphorus, red, amorphous. Hypophosphorous acid, 50 per cent. m/m. Orthophosphoric acid, 88 per cent. mjm. Propionic acid. Potassium hydroxide solution, 0.2 N. Dithixone indicator-Prepare a 1 per cent. m/m mixture of dithizone with potassium nitrate. 2-(Hydroxymercuri)benxoic acid titrant-Dissolve 0.200 g of the acid in 1 litre of 0.2 N potassium hydroxide solution.One millilitre of this solution is approximately equivalent to 10 pg of sulphur. Standard sulphate solzttion-Dissolve 1.0872 g of dry potassium sulphate in water and dilute the solution to 100 ml. From it prepare, by dilution with propionic acid, a dilute standard solution containing 200 pg ml -l of sulphur. APPARATUS- A three-necked 100-ml flask is fitted with a reflux condenser, an inlet tube connected through a gas flow meter to a supply of nitrogen, and a stopper in the third neck that is used for the introduction of solid material into the flask. This stopper should be held by a retaining spring or weight to keep it firmly in position during the refluxing operation. The condenser is connected at the exit end by about 20 cm of clear PVC tubing (3 mm 0.d.) to a length of 5 mm 0.d.glass tubing that acts as a bubbler in the This solution contains 2000 pg ml -l of sulphur. The apparatus is shown in Fig. 1. 1 Nitrogen Fig. 1. Reduction and titration apparatus : A, 100-ml distillation flask ; B, absorption vessel (150 ml) ; C, 50-ml burette; D, magnetic stirrer follower; and E, gas flow meter (5 to 190 ml min-l air)August , 19741 OF TOTAL SULPHUR IN SILICATE ROCKS 517 absorption vessel; the latter has a capacity of 150 ml. The tip of a 50-ml burette also passes through the bung in the absorption vessel and a third hole in the bung serves as an outlet for nitrogen. PROCEDURE- Mix 0.2000 g of sample powder (sieved to pass 90 mesh) and 0.20 g of sodium chlorate in a 50-ml PTFE beaker, add 5 ml of hydrochloric acid and cover the beaker with a watch-glass.Allow it to stand overnight, then place the beaker, still covered, on a steam-bath. After 30 minutes rinse down the watch-glass and sides of the beaker with a small amount of water and evaporate the contents of the beaker to dryness under an infrared lamp. Break up any large lumps of solid with a small spatula and, if the residue is not to be used at once, store the beaker in a desiccator. To the dry 100-ml flask (Fig. 1) add 1 g of sodium iodide, 0.2 g of red phosphorus, 2 ml of hypophosphorous acid, 8 ml of orthophosphoric acid and 10 ml of propionic acid. Connect the flask to the remainder of the apparatus, as shown in Fig. 1, and pass nitrogen through the apparatus at a rate of 60 ml min -1 with the tubing to the absorption vessel disconnected.Turn on the condenser cooling water and, by heating the flask with a small flame protected from draughts, reflux the contents for 30 minutes in order to remove traces of sulphur from the reagent mixture. Remove the burner without interrupting the flow of nitrogen and allow the flask to cool for 15 minutes before proceeding to the next stage. The reducing mixture thus prepared can be used for about twelve successive determinations before needing replacement. To the absorption vessel containing a magnetic stirrer follower, add 50 ml of potassium hydroxide solution and about 10 mg of dithizone indicator. Reconnect the tubing from the condenser to the absorption vessel, switch on the magnetic stirrer and add 2-(hydroxymercuri)- benzoic acid titrant dropwise from the burette until the indicator changes colour from yellow to pink.With the aid of a small, wide-necked funnel and a small, stiff brush introduce the residue from the PTFE beaker into the distillation flask, replacing the stopper without delay. Replace the burner and reflux the contents of the distillation flask for a further 30 minutes, then, after allowing the flask to cool for 10 minutes, add titrant from the burette and note the volume required to restore the pink colour. Correct for the small blank, the value of which is determined by carrying out the whole procedure without the sample. Finally, calculate the sulphur content of the sample after standardising the titrant against 1 or 2 ml of dilute standard sulphate solution, which are added directly to the distillation flask and taken through the reducing operation.The standard titration should be corrected for the blank represented by the equivalent volume of pure propionic acid added to the distillation flask and taken through the same operation. RE s u LTS Eight US. Geological Survey Standard rocks have been analysed for sulphur by the Comparative results by other workers can be method and the results are shown in Table I. TABLE I SULPHUR CONTENTS OF SOME STANDARD ROCKS Rock Granite G-1 Diabase (dolerite) W-1 Granite G-2 Granodiorite GSP-1 Andesite AGV-1 Peridotite PCC-1 Dunite DTS-1 Basalt BCR-1 Sulphur content by proposed method/ mg kg -l 70, 60 130, 135 100, 95 360, 360 20, 20 20, 20 15, 15 420, 420 Sulphur content by other methods/ mg kg -l 40,3 58,* 175,9 741° 120,s 123," 135,9 124'O 24,* loot 162,* 400p <lo,* loot <lo,* 1007 <lo,* Of 392,* 400f * Average values reported by Flanagan.7 f Average values reported by Abbey.8518 MURPHY AND SERGEANT seen to be sufficiently widely divergent to indicate the difficulty of low-level sulphur deter- mination in rocks.The proposed method has also been applied, with results of satisfactory precision for routine determinations, to a wide range of silicate rocks with sulphur contents from 5 to about 2000 mg kg-l. This paper is published by permission of the Government Chemist and the Director of the Institute of Geological Sciences. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. REFERENCES Wilson, A. D., Sergeant, G. A., and Lionnel, L. J., Analyst, 1963, 88, 138. Sen Gupta, J. G., Analytica Chim. Acta, 1970, 49, 519. Nagashima, S., Yoshida, M., and Ozawa, T., Bul2. Chew. SOC. Japan, 1972, 45, 3446. Johnson, C. M., and Nishita, H., Analyt. Chew., 1952, 24, 736. Kirsten, W. J., and Patel, V. J., Microchern. J., 1972, 17, 277. Wronski, M., Analyst, 1958, 83, 314. Flanagan, F. J., Geochim. Cosmochiw. Acta, 1973, 37, 1189. Abbey, S., Can. Spectvosc., 1970, 15, 10. Ricke, W., Geochim. Cosmochiw. Acta, 1960, 21, 35. Frost, I. C., and Thomas, J. A., cited in Fleischer, M., Geochim Coswochiw. Acta, 1965, 29, 1263. Received February 19th, 1974 Accepted Mavch 20th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900515

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, $p. 519-522 519 A Titrimetric Method for the Determination of Sulphate in Fertilisers BY A. D. CAMPBELL, D. P. HUBBARD AND N. H. TIOH (Department of Chemistry, University of Otago, Box 56, Dunedin, New Zealand) A method is described for the determination of sulphate in fertilisers in which the sulphate is precipitated with barium chloride from an acidified EDTA solution. The precipitate is filtered off by using membrane filters and is dissolved in ammoniacal EDTA ; the excess of EDTA is then titrated against a solution of magnesium ions with Eriochrome black T as theindicator. No interference is encountered from iron, aluminium, fluoride or phosphate ions. THE fertiliser industry has long been interested in accurate, rapid and simple methods for the determination of sulphate in superphosphate fertilisers.The procedure most often used is gravimetric, involving the precipitation of barium sulphate. However, this method is time consuming and subject to interferences1 The direct titrimetric methods for sulphate deter- mination suffer from interferences from phosphate, iron and aluminium, etc., and require extra separation stages, which lengthen the time taken for analysis. Excellent reviews of both gravimetricl and titrimetric methods1s2 have been given. Belcher, Gibbons and West3 reported an indirect titrimetric method, which entails the precipitation of barium sulphate, its dissolution in ammoniacal EDTA and finally back- titration with a solution of magnesium ions. Subsequently, this basic method has been suc- cessfully applied to the determination of sulphur in steel4 and organic compounds5 and of sulphate in urine.6 It seemed to us that this procedure could well be adapted to the deter- mination of sulphate in fertilisers.With minor modifications to the established method we have been able to analyse both actual and synthetic superphosphate samples with excellent accuracy and precision. EXPERIMENTAL REAGENTS AND SOLUTIONS- All reagents were of analytical-reagent grade quality unless otherwise stated. Ammonia solution, sp. gr. 0.88. Barium chloride solution, 5 per cent. m/V. Standard magnesium chloride solution, 0.025 M-Dissolve 5.08 g of magnesium chloride, MgCl,. 6H20, in distilled water and dilute to 1 litre with distilled water. Standardise the solution against 0-04 M EDTA solution with Eriochrome black T as indicator.EDTA solution, 0.04 M-This and other EDTA solutions at different concentrations, together with various concentrations of dilute hydrochloric and sulphuric acids, were prepared by suitable dilution of Volucon standard volumetric solution concentrates obtained from May and Baker Ltd. Bufer solfition, pH 10-Dissolve 70.0 g of ammonium chloride in a mixture of 250 ml of water and 570 ml of ammonia solution (sp. gr. 0-88), and dilute the solution to 1 litre. Indicator solution-Dissolve 0-40g of Eriochrome black T in a mixture of 30ml of triethanolamine and 10 ml of absolute ethanol. APPARATUS- Circles of Metricel filter (25 mm in diameter, 0.45 pm pore size) were cut from sheets of filter obtained from Gelman Instruments Co., Ann Arbor, Michigan, U.S.A.In use, the filters were clamped in a demountable Pyrex microanalysis filter holder (Millipore Corp., Bedford, Mass., U.S.A.) that incorporated a glass funnel of 15 ml capacity and a filter area of approxi- mately 2-5 cm2. 0 SAC and the authors.520 to give a small orifice, thus allowing the delivery of smaller drops. PREPARATION OF SYNTHETIC SUPERPHOSPHATE SAMPLES- In view of the unavailability of standard superphosphate f ertiliser samples of accurately known sulphate content, and in order to assess carefully the accuracy of the method, synthetic superphosphate solutions were prepared as follows by the method recommended by P. J. Gallaher (personal communication). Weigh out 0.62 g of Christmas Island rock (Note 1) and add an accurately known volume of 0.500 M sulphuric acid [between 6 and 8 ml (Note 2)]. Warm the mixture for a few minutes in order to complete the reaction.Add 10 ml of con- centrated hydrochloric acid and heat nearly to dryness, then re-dissolve the residue in 100 ml of 2 M hydrochloric acid and filter the solution through a membrane filter. Dilute the filtrate to 250 ml with distilled water, take a 25-ml aliquot and continue as described in the method recommended below. CAMPBELL, HUBBARD AND TIOH: A TITRIMETRIC METHOD FOR [ArtaZyst, Vol. 99 The tip of the burette was coated with picene wax, which was then drawn out with a wire RECOMMENDED PROCEDURE FOR THE DETERMINATION OF SULPHATE IN SUPERPHOSPHATE- Dissolve 1.00 g of finely powdered superphosphate (ground to pass through a 250-mesh sieve) in 10 ml of concentrated hydrochloric acid and evporate the solution nearly to dryness.Re-dissolve the residue in 100 ml of 2 M hydrochloric acid and filter the solution through a 0-45-pm membrane filter, rinsing thoroughly with distilled water. Transfer the filtrate into a 250-ml calibrated flask and dilute it to the mark with distilled water. Transfer a 25-ml aliquot into a Taylor flask and add to it 10 ml of 0.05 M EDTA solution. Boil the solution for about 2 minutes and then add 100 ml of 0.05 M hydrochloric acid. Allow it to stand for a few minutes and slowly add 50ml of 5.0 per cent. barium chloride solution. Then, after allowing it to stand for 5 to 10 minutes, filter the precipitate on a 0-45-pm membrane filter (Note 3) and wash it with 25 ml of 0.05 M hydrochloric acid, followed by 50 ml of distilled water.Transfer the precipitate to the original flask, using forceps to handle the filter, ad rinse any barium sulphate remaining on the glass funnel into the flask with distilled water. Add 20 ml of 0-04 M EDTA and 4 ml of ammonia solution (sp. gr. 0.88). Dilute the mixture to about 150 ml with distilled water and heat it for 15 minutes. Finally, allow it to cool, add 5 ml of buffer solution and titrate against standard 0.025 M magnesium chloride solution with Eriochrome black T as the indicator. XOTES- The choice of this rock ensures that the main possible interferences are likely to be present in relatively high concentrations. The only possible difference between this type of synthetic super- phosphate and the commercial product could be a higher fluoride level as a result of its reduced evolution during reaction.However, only a small proportion of the fluoride is evolved from Christmas Island rock during superphosphate manufacture so that the standard solution will not be very different from that of a commercial product. The stated volume of sulphuric acid ensures that the amount of sulphate in the synthetic superphosphate samples closely resembles the amount in actual samples, i.e., of the order of 30 per cent. of sulphate (10 per cent. as sulphur). In addition, the total amount of added sulphate plus rock is close to 1 g, which again agrees with the recommended mass of sample for commercial superphosphate in the method that follows.It is known that aged barium sulphate is much more difficult to dissolve than the freshly precipitated variety.5 However, precipitation in the cold and the short time of standing produce a very fine precipitate, hence the use of the fine pore filter. This precipitate dissolves readily in ammoniacal EDTA. 1. 2. 3. RESULTS AND DISCUSSION I t was to be expected that ions such as PO,3 -, F -, Fe3 + and A13 +, which interfere in the precipitation of barium sulphate, 7 would also interfere in the proposed titrimetric method. All of these ions are likely to be present in superphosphate samples. However, in preliminary experiments, we found no evidence of interference from either phosphate (11-5 mg) or fluoride (0.95 mg) in the determination of sulphate (12.0 mg) when following the recommended procedure with the slight modification that the initial precipitation was carried out in the absence of EDTA, Le., basically the method of Belcher et aL3 However, in the presence of phosphate (23.0 mg), both iron (3.52 mg) and aluminium (1.70 mg) gave a positive error ofAugust, 19741 THE DETERMINATION OF SULPHATE I N FERTILISERS 521 2-4 per cent.in the determination of 24.0 mg of sulphate. A solution containing phosphoric acid, sodium fluoride, iron(II1) chloride or aluminium chloride was added to a standard volume of sulphuric acid and subjected to the described procedure in studies of these interferences. Pi-ibil and Maricovas have shown that if barium sulphate is precipitated in the presence of EDTA, then the co-precipitation of moderate amounts of many ions is prevented.Modi- fication of the procedure accordingly eliminated the interference of iron(II1) ions ; however, a slight interference from aluminium remained. It is well known that aluminium complexes only slowly with EDTA but that boiling speeds up the reactiong9lo. The time of boiling needed to complete complexation is dependent on several factors, such as pH and previous treatment of the ~olution.~ Milner and Woodheadlo recommend boiling for about 2 minutes and the aluminium interference was eliminated under these conditions. The results obtained all show a very small positive bias, which may or may not be significant. During the development of the method it was noted that phosphate caused an inter- ference in the determination when the precipitation of barium sulphate was carried out in the presence of EDTA.This interference was eliminated by washing the precipitate with dilute hydrochloric acid. It was also necessary to increase the amount of barium chloride used in order to achieve the complete precipitation of barium sulphate when EDTA was present. Washing the precipitate with EDTA in addition to dilute acid made no difference to the recovery of sulphate and was therefore discontinued. This bias is not serious. TABLE I ANALYSIS OF SYNTHETIC SUPERPHOSPHATE SOLUTIONS Amount of sulphate Amount of sulphate Mean recovery, added/mg* recovered/mgT per cent. 317.0 317.7, 317.7, 317.3, 316.8 100-1 369.9 371.0, 370.5, 371.4, 370.5 100-3 317.0 317.7, 317.7, 317.7, 316.8 100.1 313.2 314.7, 314.2, 313.8, 313.8 100.3 * Amounts added to approximately 0.625 g of Christmas Island phosphate rock.t Four separate, 25-ml aliquots of each synthetic superphosphate stock solution were taken through the recommended procedure. Much has been written about the suitability of Eriochrome black T as an indicator.11 We have found that the final colour change to pink is, in fact, very distinct against a white background, and that reproducible titrations can be obtained. In Table I the recoveries of sulphate added to Christmas Island rock that had been shown to contain no natural sulphate are shown. The accuracy of the method is within that required by the fertiliser industry (within 0.5 per cent.) and the precision is excellent. In Table I1 the results obtained for four actual superphosphate fertiliser samples are given.TABLE I1 ANALYSIS OF ACTUAL SUPERPHOSPHATE SAMPLES Sample number 1 2 3 4 Nominal sulphur content, per cent.* 11-17 10.16 10.42 11.11 Sulphur content determined, per cent. 11-20, 11.19, 11.21 10.18, 10.17, 10.18 10.45, 10.45, 10.46 11.12, 11.13, 11.13 * Values supplied by The New Zealand Fertiliser Manufacturers' Research Association, obtained by use of a calibrated gravimetric procedure. CONCLUSIONS It has been shown, via the analysis of both actual and synthetic superphosphate solutions, that the indirect method, first proposed by Belcher et u Z . , ~ for the determination of sulphate can, with modification, be applied to sulphate in phosphate fertilisers. The method involves the522 CAMPBELL, HUBBARD AND TIOH precipitation of the sulphate with barium chloride from an acidified EDTA solution to prevent the co-precipitation of various ions.The precipitate is filtered off and dissolved in ammoniacal EDTA and the excess of EDTA titrated against standard magnesium solution with Erio- chrome black T as the indicator. Excellent agreement with available data for the actual samples and virtually 100 per cent. recoveries of sulphate for the synthetic samples demon- strate the reliability of this procedure. The method is straightforward, relatively fast (several samples could be analysed simultaneously within 1 hour) and accurate. The authors thank the Research Committee of the New Zealand Grants Committee, for financial assistance, and also Mr. P. J. Gallaher, of the New Zealand Fertiliser Manufacturers’ Research Association, for his valuable advice. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES Haddock, L. A., in Wilson, C. L., Editor, “Comprehensive Analytical Chemistry,” Volume lc, Schwarzenbach, G., and Flaschka, H., “Complexometric Titrations,” Second English Edition, Belcher, R., Gibbons, D., and West, T. S., Chem. 6 Ind., 1954, 850. --- , Analyst, 1955, 80, 751. Belcher, R:, Bhasin, R. L., Shah, R. A., and West, T. S., J . Chenz. S O ~ . , 1958, 4054. Lewis, D. A., Analyst, 1962, 87, 566. Vogel, A. I., “Quantitative Inorganic Analysis,” Third Edition, Longmans, Green and Co. Ltd., Pribil, R., and Maricova, D., Chemicke‘ Listy, 1952, 46, 542. Schwarzenbach, G., and Flaschka, H., “Complexometric Titrations,” Second English Edition, Milner, G. W. C., and Woodhead, J. L., Analyst, 1954, 79, 363. West, T. S., “Complexometry with EDTA and Related Reagents,” BDH Chemicals Ltd., Poole. Received February Sth, 1974 Accepted March 8th, 1974 Elsevier Publishing Co., Amsterdam, 1962, p. 282. Methuen and Co. Ltd., London, 1969, p. 315. London, 1961, p. 462. Methuen and Co. Ltd., London, 1969, p. 185. 1969, p. 191.

ISSN:0003-2654    

DOI:10.1039/AN9749900519

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974, Vol. 99, $$. 523-527 523 Loss of Zinc and Cobalt During Dry Ashing of Biological Material BY J. G. VAN RAAPHORST, A. W. VAN WEERS AND H. M. HAREMAKER (Reactor Centrum Nederland, Petten, The Netherlands) Loss of zinc and cobalt during dry ashing of marine mussels (Mytilus edulis) and brown seaweed (Fucus spiralis) has been studied by using material labelled by exposure of living organisms to sea water spiked with zinc-65 or cobalt-60. Even after ashing in porcelain crucibles at temperatures of up to 1000 "C, no significant loss of zinc or cobalt by volatilisation was observed. After ashing a t 450 and 550 O C , both radionuclides could be removed quantita- tively from the crucibles by leaching with hydrochloric acid. Adsorption on the crucible after ashing a t 1000 "C was measured only for cobalt-60.A significant proportion of the cobalt tracer could not be removed from the crucibles by treatment of the latter with acid. From these results it is con- cluded that dry ashing is a reliable method of sample destruction for the determination of zinc and cobalt in M . edulis and F. spiralis. IN studies on the levels of trace metals in biological samples, both wet and dry-ashing tech- niques are used for the destruction of the organic material in order to bring the samples into a condition suitable for analysis. Dry ashing has the advantage of being a relatively simple method that gives a large reduction in sample volume without the risk of contamination with impurities that may be present in oxidising acids. In our laboratory a study was made of the seasonal and long-term variation of the levels of zinc and cobalt in marine mussels and seaweed,lS2 and for the reason given above dry ashing was regarded as the most suitable method of sample destruction. However, during this procedure losses of trace metals may occur.The behaviour of the elements zinc and cobalt during dry ashing of organic material is the subject of contradictory reports in the literature. In his book, Gorsuch3 published data obtained from recent literature, which showed that the recovery of zinc varied considerably. Gorsuch4 measured the amount of zinc retained after ashing neutron-activated human hair at 500 "C for 16 hours and repeated the measurements after further 3-hour periods of ashing at 600,700,800,900 and 1000 "C.The results obtained showed that only at 1000 "C may loss of a few per cent. of zinc by volatilisation occur. More serious was the retention of zinc on the silica of the crucible, which increased from 0.3 per cent. after 3 hours at 600 "C to 7-1 per cent. after 3 hours at 1000 "C. The results given by Gorsuch indicate further that the adsorption of zinc on crucibles increases consider- ably with the ashing time. In contrast with these results, Pyck, Hoste and Gillis5 reported a zinc recovery of only 30 per cent. after ashing for 3 hours at 900 "C. Results published by other investigators are no less conflicting. Hamilton, Minski and Cleary6 observed that no loss of zinc occurred at 850 "C but found that 45 per cent. of it was adsorbed on the silica of the crucible. Doshi, Sreekumaran, Mulay and Patel' reported a 66 per cent.recovery after ashing for 24 hours at 700 "C, Knauer8 observed that there was no loss at 800 "C and Strohal, Lulic and JelisavEiE9 obtained only a 56 per cent. recovery of zinc at 800 "C. Published data concerning the loss of cobalt during dry ashing are also contradictory. Gorsuch4 claimed a 99 to 100 per cent. recovery of cobalt and that there was no significant retention of it on the crucible. Pyck et aL5 reported a recovery of 70 per cent. after ashing for 3 hours at 900 "C, Doshi et aZ.7 found that 64 per cent. was recovered after 24 hours at 700 "C while Strohal et aLg reported a recovery of 79 per cent. even after ashing at a temperature as low as 350 "C. In some of the studies mentioned a b ~ v e , ~ - ~ radioactive tracers were simply added to the samples immediately before ashing them.However, one can expect the radioactive and non-radioactive molecules to have the same chemical structure only if the radioactive tracer is incorporated in a more or less natural way, as effected for instance by Strohal et aZ.9 Neutron 0 SAC and the authors.524 VAN RAAPHORST et al. : LOSS OF ZINC AND COBALT [Analyst, Vol. 99 irradiation of the organic material prior to ashing it, as carried out by Gorsuch,3 may cause great differences in the chemical structure between the activated and the stable molecules. It is therefore clear from the foregoing results obtained with different material and by using different methods that no general conclusion on the recovery of zinc and cobalt after dry ashing can be drawn.Moreover, several of these studies apparently involved only a small number of independent determinations. In the present study, marine mussels and seaweed were labelled by exposure of living organisms to sea water spiked with radioactive zinc or cobalt. Each treatment of the labelled material was carried out on a series of replicate samples. EXPERIMENTAL LABELLING OF MUSSELS AND SEAWEED- Mussels (Mytilus edzclis) and brown seaweed (Fzcczcs s@ralis) were collected from the North Sea shore near Petten. Up to fifteen mussels were placed in a 30-litre polyethylene tank filled with filtered natural sea water and to the tank zinc-65 or cobalt-60 was added at a concentration that varied from 1 to 10 pCi 1-I.Labelling of seaweed was carried out in 2-litre Erlenmeyer flasks with filtered natural sea water, buffered at pH 8 with 0.02 M Tris(hydroxy- methy1)aminomethane hydrochloride, and zinc-65 or cobalt-60 was added at a concentration of 1 to 3 and 4 to 8 pCi l-l, respectively. Each flask contained 20 to 40 g of fresh seaweed. The radioisotopes used were of high specific activity and were obtained commercially in the chloride form from the Radiochemical Centre, Amersham. After a labelling period of 5 days for seaweed and 5 to 10 days for mussels, the seaweed and the soft parts of the mussels were either dried overnight in an oven at 45 "C or freeze- dried. The method of drying was shown to have no influence on the results of the experiments. Ashing was carried out in a Solo, Type 8003, furnace, and the temperature control and indicator were checked with a thermocouple.Each sample was ashed in a Staatliche Berlin, Type C/O, porcelain crucible successively at 200, 300, 400, 600, 800 and 1000 "C for 20 hours at each temperature. Each sample remained in the same crucible from the beginning until the end of the whole procedure, including the counting procedure. Nine samples (200 to 400 mg dry mass) from each organic material were ashed at the same time. COUNTING PROCEDURE- Counting was performed with a 3 x 3-inch thallium-activated sodium iodide crystal, connected to a single-channel analyser. The samples in the procelain crucibles were counted in a fixed geometry at a distance of 18 cm from the top of the crystal and the whole counting procedure was repeated twice with each sample.A total of about 100000 counts was recorded for each sample. In order to ensure that a distance of 18 cm was sufficient to prevent changes in geometry that were too large from occurring during ashing, an experiment was carried out so as to check this distance. ASHING PROCEDURE- I I I I 2000 1.01 0 500 1000 1500 Vo I u me/p I Fig. 1. Influence of sample volume on counting rateAugust, 19741 DURING DRY ASHING OF BIOLOGICAL MATERIAL 525 A volume (1Opl) of a solution containing zinc-65 was placed in a crucible and counted at a distance of 18 cm from the crystal. To this solution 20 pl of water were then added and the solution was counted again. This procedure was repeated with successive additions of 20 to 500 p1 of water up to a final sample volume of 1500 p1.The results given in Fig. 1 show that when the volume of the sample is kept below 500pl the influence of sample volume on the counting rate is less than 1.5 per cent. TABLE I ACTIVITY OF 21~c-65 IN MUSSELS AND SEAWEED AFTER ASHING PROCEDURE Results, which are the means of nine determinations, are expressed as counts per 1000 s per 100 mg of original dry mass Mussels Seaweed temperacurel Standard Standard "C Activity deviation Activity deviation - 200 400 500 600 800 1000 Mean . . Standard deviation 18.783 0.231 19.105 0.211 19.527 0.207 19.493 0.230 19.741 0.201 19.726 0.157 19.813 0-232 19.455 0.379 12.551 0.145 12.632 0.173 12.988 0.166 12.795 0.144 12.491 0.120 12.625 0.104 12.680 - - 0.182 LEACHING- Several workers have reported difficulties experienced with the complete removal of the ash residue from silica crucibles.We checked this aspect for porcelain crucibles by ashing samples of labelled material at 450, 550 and 1000 "C, each sample being subsequently placed on top of the crystal and counted. The crucibles were then rinsed with water and the counting was repeated. When activity was detected in the crucible, approximately 1 ml of 10 N hydrochloric acid was added and the mixture was carefully evaporated to dryness. Finally, about 3 ml of 0.1 N hydrochloric acid were added, the crucible was heated gently, rinsed subsequently with water and counted again. TABLE I1 ADSORPTION OF ZINC ON PORCELAIN CRUCIBLES AFTER ASHING OF MUSSELS AND SEAWEED AND Results are adsorbed activity expressed as percentage of the activity of the ashed sample SUBSEQUENT LEACHING Mussels Seaweed Ashing temperature Ashing temperature Ashing temperature Ashing temperature 450 "C; 550 "C; 450 O C ; 550 "C; leached with leached with leached with leached with A I 1 f v 2-- A 7 7 +L-- 7 water r d 7 water -----Lz water acid* water acid 0.2 0.2 0.2 0.3 0-3 0.2 0.1 0.3 0.2 <o-1 (0.1 <0*1 <0.1 (0.1 <om1 <0*1 (0.1 <0*1 0.1 0.3 0.6 0.6 0.7 1.2 0.4 0-3 0.9 <0.1 0.2 (0.1 0.5 <0.1 0.2 <o-1 <om1 0.4 0.4 0.2 <0*1 (0.1 0.1 (0.1 (0.1 0.4 < 0.1 * Not measured.2.4 1.1 1.2 1.3 0-7 0.9 1.2 1.0 5.7 1-4 0.3 0-1 1.0 0.4 0-3 0.6 0.7 3.4 RESULTS AND CONCLUSION Unless indicated otherwise, the measured activities of the samples were expressed as counts per 1000 s per 100 mg (original dry mass) of the material.526 VAN RAAPHORST et al.: LOSS OF ZINC AND COBALT [Analyst, Vol.99 ZINC- For mussels the activities measured initially and after ashing at 200 "C are somewhat lower than the values obtained after ashing at higher temperatures, which may be explained by a geometry factor, because at temperatures above 400 "C the volume of the samples becomes very small. It is clear that loss of zinc by volatilisation does not occur even at 1000 "C. Adsorption of zinc on porcelain crucibles was measured after ashing the material at 450 and 550 "C. The very small amounts that remained adsorbed on the crucible after leaching (Table 11) show that essentially all of the zinc can be removed from the crucible with the aid of hydrochloric acid.The results of the ashing procedure are given in Table I. TABLE I11 ACTIVITY OF COBALT-60 I N MUSSELS AND SEAWEED AFTER ASHING PROCEDURE Results, which are the means of eight determinations for mussels and nine for seaweed, are expressed as counts per 1000 s per 100 mg of original dry mass Ashing I temperature/ "C - 200 300 400 600 800 1000 Mean . . Standard deviation .. .. Mussels Activity 16.170 15.776 16.101 15.711 16.083 16.027 15-610 15.925 0.22 1 Standard deviation 0.336 0.476 0.505 0.409 0-480 0.460 0.488 Seaweed Activity 27.440 27-448 27.389 27.396 27.512 27.629 27-562 27.482 0.89 7 Standard deviation 0.266 0.255 0.257 0-228 0.278 0.312 0.300 COBALT- Ashing of mussels and seaweed at temperatures of up to 1000 "C causes no losses of cobalt, as can be seen from the results given in Table 111.Only very slight adsorption of cobalt on the crucible was observed after ashing at 450 and 550 "C (Table IV) and leaching with both water and hydrochloric acid removed it quantitatively . Ashing at 1000 "C resulted in a melt being formed, which could not be removed from the crucible with water. Subsequent TABLE IV ADSORPTION OF COBALT ON PORCELAIN CRUCIBLES AFTER ASHING OF MUSSELS AND SEAWEED Results are adsorbed activity expressed as percentage of the activity of the ashed sample AND SUBSEQUENT LEACHING Mussels Seaweed Ashing temperature 450 "C; leached with W- 0.2 <0*1 0.2 (0.1 0.2 (0.1 0.2 <0*1 0.1 <0.1 0.2 <0*1 0-1 <0.1 0.1 (0.1 0.1 <0*1 Ashing temperature 550 "C; leached with +\ water acid 0.1 (0.1 0.1 <0-1 0.2 (0.1 1.1 0.1 0.2 0.1 0.1 0.1 0-3 0.3 0.1 (0.1 0.2 0.1 Ashing temperature 1000 "C; leached with +L-7 water acid 101.3 10.5 79.2 6-8 99.4 9.0 99.6 9.9 99.5 6.4 100.9 9.7 74-3 17.5 101.1 14.1 100.2 14.1 Ashing temperature 450 "C; leached with r-J+ water acid* <o-1 - (0.1 - <0.1 - (0.1 - <0.1 - <0.1 - < 0.1 - < o n 1 - <0*1 - Ashing temperature 550 " C ; leached with +-7 water acid (0.1 -* (0.1 -* (0.1 -* (0.1 -* (0.1 -* 0.1 <0*1 0.3 (0.1 0.9 <0*1 0.2 <0.1 Ashing temperature 1000 "C; leached with water acid 98-5 1.7 99.0 1.9 98.0 1.4 99.7 1-9 99.4 1.8 101.2 1.7 98.8 1.6 98-7 1.5 99.4 1.7 +-? * Not measured.August, 19741 DURING DRY ASHING OF BIOLOGICAL MATERIAL 527 treatment with hydrochloric acid removed the cobalt quantitatively from the crucibles containing seaweed samples, but a significant proportion (6 to 18 per cent.) of the cobalt in mussels could not be dissolved.In conclusion it can be stated that dry ashing has been shown to be a reliable method of sample destruction for the determination of zinc and cobalt in mussels (Mytilus edulis) and in brown seaweed (Fucus spiralis). The only restriction is that after ashing at very high temperatures the quantitative removal of the residue from a porcelain crucible has proved to be difficult. 1. 2. 3. 4. 5. 6. 7. 5. 9. REFERENCES van Weers, A. W., in “Proceedings of the International Symposium on Radioecology Applied to the Protection of Man and his Environment,” Luxembourg, May 1972, EUR 4800 d-f-i-e, p. 1357. -, in “Proceedings of the International Symposium on Radioactive Contamination of the Marine Environment,” IAEA, Vienna, 1973, p. 355. Gorsuch, T. T., “The Destruction of Organic Matter,” Pergamon Press, Oxford, New York, Toronto, Sydney and Braunschweig, 1970, p. 74. -, An~Zyst, 1959, 84, 135. Pyck, J., Hoste, J., and Gillis, J., “Proceedings of the International Symposium on Microchemistry,” Hamilton, E. I., Minski, M. J., and Cleary, J. J., Analyst, 1967, 92, 257. Doshi, G. R., Sreekumaran, C., Mulay, C. D., and Patel, B., Curr. Sci., 1969, 38(9), 206. Knauer, G. A., Analyst, 1970, 95, 476. Strohal, P., LuliC, S., and JelisavCiC, O., Ibid., 1969, 94, 675. Pergamon Press, Oxford and New York, 1958, p. 48. Received December 3rd, 1973 Accepted February 18th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900523

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

528 Analyst, August, 1974, Vol. 99, pp. 528-532 Determination of Copper(1) with N - Bromosuccinimide BY A. ABOU EL KHEIR, M. AYAD* AND M. M. AMER (Department of AnaZytical Chemistry, Faculty of Pharmacy, Cairo University, Cairo, Egypt) It has been found that N-bromosuccinimide readily and quantitatively oxidises aqueous solutions of copper(1) at room temperature and in the presence of dilute hydrochloric acid, the oxidising agent being irreversibly reduced to succinimide. A procedure is suggested for the determination of copper(1) by titration with standard N-bromosuccinimide solution ; the results obtained were found to be equally precise but more accurate than those given by the permanganate and complexometric methods. Copper(II), after preliminary reduction, can also be determined by the suggested procedure, whether alone or admixed with copper(1). COPPER is commonly determined by complexometric, iodimetric or electrolytic procedures ; however, the last of these procedures is time consuming and requires the use of much apparatus.Flaschkal and Amin2 described a direct complexometric procedure for the titration of copper(II), using murexide as indicator. Eriochrome black T has been used as an indicator in an indirect complexometric determination of copper by back-titration of excess of EDTA with a standard solution of a manganese(I1) salt.3 Other proposed indicators include pyrocatechol ~ i o l e t , ~ 1,2-(pyridylaz0)-2-naphthol~~~ and fast sulphone black F.' Both copper(1) and copper(I1) have been determined by iodimetric titration, copper(I1) salts by the procedures of Lehmann,8 Riegler, Maquenne, lo Citron, l1 Peters, l2 and Shaffer and Hartrnannl31l4 and copper(1) salts by those of Bang,15116 Scales,17 Maclean,l81l9 Clark20 and Shaffer and Hartmann.l3$l4 This paper describes the use of N-bromosuccinimide as an oxidising agent for determining copper (I), the reaction involved being as follows- CH2- CH2- co\NBr+2Cu++H+---+j '\NH + 2 Cu2+ + Br- CH2-CO/ I CH2-CO/ The mechanism of the reaction has been established by experiment.However, although N-bromosuccinimide can decolorise methyl red in an aqueous acidic medium, it oxidises copper( I) preferentially. On adding N-bromosuccinimide in the presence of methyl red, the red colour of the indicator remains unchanged until all of the copper(1) ions have been oxidised, the first excess of N-bromosuccinimide decolorising the indicator. EXPERIMENTAL REAGENTS- Standard co$$er(I) chloride solution, 0.005 M-Dissolve 100 g of copper(I1) acetate in 1 litre of water, filter and heat the solution to boiling, then add 15 per cent.glucose solution until the blue colour of copper(I1) almost disappears. Allow the mixture to stand, decant the supernatant liquid, wash the precipitate of copper(1) oxide several times with water, by decantation, then transfer it into a Gooch crucible. Wash the precipitate on the crucible with ethanol, then dry it in a vacuum desiccator. Dissolve 0.0715 g of the dried copper(1) oxide in 20 per cent. (V/V) hydrochloric acid, avoiding the use of heat, and make the solution up to 100 ml.Methyl red solution, 0.04 per cent. (m/V) in 95 per cent. ethanol. N-Bromosuccinimide solution, 0.01 M, aqueous-Standardise this solution, which must be freshly prepared, against standard arsenite solution21 and keep it in dark-glass bottles. * Present address: Research and Control Centre, Cairo, Egypt. @ SAC and the authors.EL KHEIR, AYAD AND AMER 529 HydrochZoric acid, 20 per cent. V/V-Mix 1 volume of concentrated hydrochloric acid and i volumes of water. PROCEDURES- Determination of copper(I) alone-Introduce into a 50-ml conical flask an accurately measured volume of the copper(1) salt solution (containing 5 to 20 mg of Cu), an equal volume of 20 per cent. V/V hydrochloric acid and two drops of methyl red indicator solution, and titrate the mixture by the dropwise addition of 0.01 M N-bromosuccinimide solution, while stirring, until the red colour is just discharged.Carry out a blank experiment (1 ml of 0.01 M N-bromosuccinirnide solution corresponds to 1.27 mg of Cu, 1.43 mg of Cu,O and 1-98 mg of Cu,Cl,). 1. TABLE I DETERMINATION OF COPPER(I) CHLORIDE BY TITRATION WITH N-BROMOSUCCINIMIDE SOLUTION Copper(1) chloride taken/ mg 1.98 2-97 3-96 4.95 9.90 14-85 19.80 24-75 Copper(1) chloride found/ mg 1.96 3.06 3.98 4.94 9-77 14.73 19.53 24.25 Mean (P = 0.05) Recovery, per cent. 99.0 103.0 100.5 99-8 98.7 99.2 98.6 98.0 99.6 f. 1.25 2. Determination of copper(II) alone and of total copper-Neutralise a volume of the copper(I1) salt solution (equivalent to 5 to 20 mg of Cu) with a few drops of 10 per cent. sodium hydrogen carbonate solution until effervescence ceases.Boil the solution, add saturated glucose solution, dropwise, until the blue colour completely disappears, then continue boiling the solution for a further 2 minutes. Filter it through a Gooch crucible, washing the precipi- tate with water by decantation. Dissolve the washed precipitate of red copper(1) oxide on the filter in about 5ml of concentrated hydrochloric acid, dilute the solution with 20 ml of water, add 3 to 5 drops of methyl red indicator and titrate it with 0.01 M N-bromosuccinimide solution. Carry out a blank experiment (1 m1 of 0.01 M N-bromosuccinimide solution corresponds to 1.270 mg of Cu and 1.590 mg of CuO). 3. Determination of co@er(Il) and co@er(l) in mixtures-Determine copper(1) on a volume of the solution of the mixed ions containing the equivalent of 5 to 20 mg of copper(1) TABLE I1 DETERMINATION OF COPPER(I) CHLORIDE BY TITRATION WITH N-BROMOSUCCINIMIDE SOLUTION AND BY THE PERMANGANATE METHOD^^ N-Bromosuccinimide method r A \ Copper(1) chloride Copper(1) chloride mg mg 4.95 4-80 97.0 4.99 101.0 4.85 98.0 9.9 9.99 101.0 9.80 99.0 9-95 100.5 19.8 19.55 98-8 19.50 98.5 19.85 100.2 taken/ found/ Recovery, per cent.Mean (P = 0.05) 99.33 f 1.09 Permanganate method found/ Recovery, mg per cent. 4.80 97.0 4-85 98.0 4-90 99.0 9.60 97.0 9.60 97.0 9.80 99.0 19.50 98-5 19.60 99.0 19.60 99.0 A I \ Copper(1) chloride 98-18 & 0.64530 by the method described under Procedure 1. Then determine the total copper content on a volume containing the equivalent of 5 to 20mg of copper by the method described under Procedure 2.EL KHEIR, AYAD AND AMER: DETERMINATION OF [Afzalyst, VOl. 99 Calculate the copper(I1) content by difference. TABLE I11 DETERMINATION OF COPPER(II) OXIDE BY THE N-BROMOSUCCINIMIDE PROCEDURE AND THE COMPLEXOMETRIC METHOD23 N-Bromosuccinimide method Complexometric method Copper(I1) oxide taken/ mg 10 20 30 40 50 Copper(I1) oxide found/ mg 9.83 9-83 9.87 19.72 19.87 20.03 29-61 29.70 29-70 40.22 40.06 40.06 48.80 49.29 49-29 Mean (P = 0.05) 1 Recovery, per cent. 98.30 98.30 98.70 98.60 99.35 100-15 98-70 99.00 99-00 100.50 100.20 100.20 97-60 98.60 98.60 99- 1 f 0.474 Copper(I1) oxide found/ mg 9-56 9.78 9.78 19-30 19-30 19.87 29.40 30.00 29-80 39.54 39.54 39-74 48.89 48-89 48-99 I Recovery, per cent.95.6 97.8 97.8 96.5 96.5 99.3 98.0 100.0 99.3 98.9 98.9 99.3 97.8 97.8 98.0 98.1 f 0-678 RESULTS Table I shows the results obtained by applying Procedure 1 to the determination of 2 to 25mg of copper(1) chloride and in Table I1 the results obtained for the determination of 5 to 20 mg of copper(1) chloride by both the suggested N-bromosuccinimide procedure and the permanganate method22 are compared. In Table I11 the results obtained for the deter- mination of 10 to 50 mg of copper(I1) oxide by both the N-bromosuccinimide procedure and the complexometric method23 are compared and Table IV gives the results obtained by apply- ing Procedure 3 to the determination of different mixtures of copper(1) and copper(I1) oxides. TABLE IV DETERMINATION OF COPPER(I) AND COPPER(II) OXIDES IN MIXTURES BY THE N-BROMOSUCCINIMIDE PROCEDURE Copper (I) oxide Copper (I) oxide taken/ found/ mg mg 20 20.50 20.50 20.30 20 20.30 20.30 20.16 10 10.01 10.01 10.20 50 49-90 49-90 50-10 Mean (P = 0.05) Copper(I1) oxide Recovery, taken/ per cent.mg 102-5 20 102.5 101-5 101.5 50 101-5 100.8 100.1 50 100.1 102.0 99.8 20 99.8 100.2 101.1 f 1.93 Copper(I1) oxide found/ mg 20.10 19.95 19-95 48.97 49.20 50.80 48.80 48.90 49.30 19.95 19.88 19-77 Recovery, per cent. 100.5 99.8 99.8 97.9 99.2 101.6 97.6 97.9 98.6 99.8 99.4 98.9 99.24 & 1-51 DISCUSSION N-Bromosuccinimide, which contains a loosely bound bromine atom, is used for bromina- tion as well as for dehydrogenation purposes, especially in organic chemistry. 24 However,August, 19741 COPPER(I) WITH N-BROMOSUCCINIMIDE 53 1 solutions of this reagent have been used for oxidimetric titrations of some inorganic ions,25,26 the mechanism of the reaction with copper(1) in a dilute acidic medium being shown in the introduction.The presence of bivalent copper ions was established by the formation of a blue precipitate of copper(I1) hydroxide when sodium hydroxide was added to the titrated solution ; the precipitate was dissolved in dilute hydrochloric acid and its identity confirmed by the potas- sium hexacyanoferrate(I1) test. The presence of hydrobromic acid was confirmed by the silver nitrate and the chlorine water tests. Succinimide was isolated by distilling the clear reaction mixture under reduced pressure and recrystallising the solid residue from benzene ; the colourless crystals were identified as succinimide (m.p.124-125 "C). Before making use of the above reaction for the quantitative determination of copper(I), the effect of the concentration of hydrochloric acid on the reaction was studied (Table V), and it was concluded that a concentration of 20 per cent. V/V was the most satisfactory; the recovery of 108 per cent. of copper(1) when 50 per cent. V/V hydrochloric acid was used is an effect of the high acid concentration, as previously observed by Barakat and Shehab.21 TABLE V EFFECT OF HYDROCHLORIC ACID CONCENTRATION ON THE REACTION BETWEEN N-BROMOSUCCINIMIDE AND COPPER(I) Copper(1) chloride Concentration of hydrochloric acid, Recovery, takenlmg per cent. V/V per cent. 9.9 2 62.4 5 75.1 10 87.9 20 99-6 30 102-0 50 108.0 It is evident from the results given in Tables I to I11 that the accuracy of the proposed method amounts to between 99.3 and 99.6 per cent.for copper(1) and to 99.1 per cent. for copper(I1). Statistical analysis of the results in Tables IT and 111, shown in Tables VI and VII, reveals that the variance ratios are within the theoretical limits, indicating that the suggested N-bromosuccinimide method is as precise as the permanganate method22 and the complexometric procedure. 23 However, the t-test reveals that the N-bromosuccinimide procedure is more accurate than the other two methods; it also has the further advantage that it is less time consuming. TABLE VI STATISTICAL ANALYSIS OF THE RESULTS IN TABLE I1 N-Bromosuccinimide method Mean, per cent. (P = 0.05).. .. . .99*33 f 1.086 N .. . . .. .. 9 Variance . . .. .. . . 1-997 Potassium permanganate method 98.18 f 0.64 9 1.3475 tD.976 * * .. .. . . .. . . 2.2836 (2*12)* F . . .. .. . . .. . . 2,2865 (3.44)" Degrees of freedom . . .. . . .. 4 = 16 * Theoretical values. As shown in Table IV, the accuracy and precision of the proposed method for the deter- mination of copper(1) were maintained in the presence of widely different proportions of copper (11).532 EL KHEIR, AYAD AND AMER TABLE VII STATISTICAL ANALYSIS OF THE RESULTS IN TABLE 111 N-Bromosuccinimide EDTA method method Mean, per cent. (P = 0.05). . . . . . 99.1 f 0.474 98.1 f 0.678 N .. . . .. .. 15 15 Variance . . .. .. . . 0.733 1.497 c Y 1 tO.975 - - .. .. .. .. . . 2.5937 (2.048)* F ... . .. .. .. . . 2.0409 (2-53)* Degrees of freedom . . .. .. . . 4 = 28 * Theoretical values. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. REFERENCES Flaschka, H., Mikrochemie Mikrochem. Acta, 1952, 39, 88. Amin, A. M., Chemist Analyst, 1955, 44, 17. Kinnunen, J., and Wennerstared, B., Ibid., 1955, 44, 45. Schwarzenbach, G., “Complexometric Titrations,” Methuen and Co. Ltd., London, 1957. Barnard, A. J., jun., Broad, W. C., and Flaschka, H., Chemist Analyst, 1956, 45, 86 and 111. --- , Ibid., 1957, 46, 18, 46 and 76. Bellir, R.,’Close, R. A. , and West, T. S., Chem. & Ind., 1957, 1647. Lehmann, K. B., Arch. Hyg., 1897, 30, 267. Riegler, E., 2. analyt. Chem., 1898, 37, 22. Maquenne, L., Bull. SOL. Chim. Fr., 1898, 19, 926; J . Chem. SOG. Abstr., 1899, 76 (Part 2), 529. Citron, H., Dt. med. Wschr., 1904, 30, 1602. Peters, A. W., J . Amer. Chem. SOC., 1912, 34, 422 and 928. Shaffer, P. A., and Hartmann, A. F., J . Biol. Chem., 1921, 45, 349 and 365. Bang, I., “Der Blutzucker,” Wiesbaden, 1913. -, Biochem. Z., 1918, 87, 248. Scales, F. M., J . Biol. Chem., 1915, 23, 81. Maclean, H., J. Physiol., Lond., 1916, 1, 168. -, Biochem. J., 1919, 23, 135. Clark, W. B., J . Amer. Chem. Soc., 1918, 40, 1759. Barakat, M. Z., and Shehab, S. K., Analyst, 1965, 90, 50. “British Pharmacopoeia 1953,” The Pharmaceutical Press, London, 1953. Vogel, A. I., “A Text Book of Quantitative Inorganic Analysis.” Third Edition, Longmans, Green Hudlicky, M., “Prepsrativni Reakce v Organickk Chemii 11, Halogenace a Dehalogenace,” NCSAV, Berka, A., and Zyka, J., Chemicke‘ Listy, 1957, 51, 1823. , Colln Czech. Chem. Commun., 1958, 23, 402. -- , , J . Chem. S O ~ . , 1921, 120, 417. and Co. Ltd., London, 1961. Prague, 1955, p. 369. J -- Received November 21st, 1973 Accepted March 5th, 1974

ISSN:0003-2654    

DOI:10.1039/AN9749900528

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

Analyst, August, 1974 Book Reviews 533 IR-THEORY AND PRACTICE OF INFRARED SPECTROSCOPY. By NELSON L. ALPERT, WILLIAM E. KEISER and HERMAN A. SZYMANSKI. Rosetta Editions. Paperback Edition 1973. Price $9.14. Pp. xiv + 380. London and New York: Plenum/ Szymanski’s original book, published in 1964 by Plenum Press at a selling price of $15, was widely acclaimed. It was succeeded, in 1970, by a second edition under the authorship of Alpert, Keiser and Szymanski, and it is precisely that book which is now being offered, as a paperback edition, at a very considerable reduction in the previous price despite the inflationary increases of the past three years. “Szymanski” is a well known, established text; the present circumstances do not justify a fresh appraisal of its contents, or of its merits and demerits.It was, however, always too expensive a book to be considered seriously for class adoption (at least by this reviewer’s students); this barrier, at least, has now been removed. It is to be hoped that the price reduction will be apprecia- ted by readers other than students and that publishers will adopt more freely this long-established way of making the more expensive texts available at more attractive prices. On the other hand, “Szymanski” has always had examples of infrared spectra housed in a pocket attached to its back cover; this imposes a strain on the binding such that the back cover of this reviewer’s copy was tom at the spine when it was received. This could prove to be a false economy, therefsre, for library and reference copies. D.M. W. ANDERSON ISOTOPE DILUTION ANALYSIS. By J. TOLGYESSY, T. BRAWN and M. KYRS. International Series Oxford, New York, Toronto, This is the first book, so far as the reviewer knows, devoted exclusively to isotope dilution analysis, a subject which, especially in its radiochemical form, has expanded rapidly over the last decade or so. For quantitative analysis it is necessary either to separate the element or compound to be determined quantitatively from the matrix so that it can be measured by an appropriate method, for example, by weighing, or to separate a property or characteristic, such as light absorption at a defined wavelength, from those due to the matrix, for measurement. Where quantitative separation is impossible a quantitative result is still attainable if an accurate measurement of the recovery of the separated constituent can be made ; isotope dilution can provide such a measurement.In the first chapter the principles and theory of isotope dilution analysis in its various modifi- cations are described. The chapter is almost exclusively devoted to methods employing radioactive isotopes, even in the section on page 16 entitled “Isotope Dilution Analysis with Stable Isotopes.” There is, it is true, a short chapter (5) that deals with isotope dilution analysis using stable iso- topes. Chapter 2 deals with experimental techniques of separation and mass determinations, Chapters 3 and 4 describe inorganic, organic and biochemical application and the final chapter some special applications of isotope dilution such as blood volume determination in the human body, dissocation constants and the capacity of ion exchangers.Each chapter includes an extensive bibliography. The prominence given to the use of radioactive isotopes is a measure of, on the one hand their greater applicability, and on the other, the more ready availability of the necessary measuring equipment. The book is, on the whole, well written, although there are features lacking clarity, for example, Table 1.5 “Illustration of some methods of double IDA” was incomprehensible to the reviewer and in the section (1.4) dealing with “Precision, accuracy and sensitivity of IDA,” precision and accuracy appear to be confused. Thus, systematic errors are considered under precision. Although such errors will militate against the accuracy of a, determination, that is, the nearness to the true value, they will usually have little effect on the repeatability, that is, the precision of a series of measurements.In spite of the above strictures the reviewer considers this to be a valuable addition to the literature of analytical chemistry. of Monographs in Analytical Chemistry, Volume 49. Sydney and Braunschweig : Pergamon Press. 1972. Price k3.50. Pp. 194. D. A. LAMBIE534 BOOK REVIEWS [Analyst, VOl. 99 ADVANCES IN RAMAN SPECTROSCOPY. Volume 1. PROCEEDINGS OF THE THIRD INTERNATIONAL London, New York and Rheine: Heyden & Son Price Ll6; $44; DM131.50. Since the introduction of lasers, there has been not only a revival of Raman spectroscopy, but also extensive development in its practical application and scope.Unfortunately, the proceedings of the first two International Conferences on Raman Spectroscopy, held at Ottawa (1969) and Oxford (1970), were not published, but since this extensive volume, which records all the papers presented at the third conference held at Kheims in September 1972, has been declared to be the first in a new series, the publication of the proceedings of future Raman conferences now appears to be assured. The Editor deserves immense credit for having completed his task by January 1973. Similarly, the publishers and their printers are to be con- gratulated on achieving publication as quickly as July 1973-a remarkable feat considering the bilingual character of the book, the very large number of diagrams, and the extent of the heavy mathematical typesetting required on many of the pages.The texts of eighty-two papers (nineteen in French) are presented in eight sections: Non- linear Phenomena, New Techniques, Phase Transitions, Flame Raman Effect , Macromolecules and Biological Molecules, Resonance Raman Effect, Band Profiles in Gases and Liquids, and Molecular Structures. Although only about 20 per cent. of the papers published in this book are of direct analytical interest, Raman spectroscopy has considerable analytical potential, and the contents of this book could be read with advantage by analytical chemists, particularly those that deal to any extent with spectroscopic methods. The predominant impression derived from reading the book is that this must have been an outstandingly successful conference, with contributions of a uniformly high standard.The cost of this book is undoubtedly high; it will probably only be acquired by Raman de- votees. Nevertheless, the Editor and Publisher have done a great service, and those who were unable for any reason to attend the symposium will undoubtedly have considered it cheap at the price to be able to consult, only 10 months after the event, the original material that was presented. CONFERENCE ON RAMAN SPECTROSCOPY, UNIVERSITY OF RHEIMS, FRANCE, SEPTEMBER 1972. Edited by J. P. MATHIEU. Ltd. 1973. Pp. xiv + 639. This is an excellent production throughout. There is an index of contributors and a subject index. D. M. W. ANDERSON DAS ARBEITEN MIT IONENSELEKTIVEN ELEKTRODEN.EINE EINFUHRUNG. By KARL CAMMANN. Berlin, This book, the thirteenth in the series of laboratory guides, is both attractive and systematic in its presentation. Following a brief introduction, the first quarter of the book is made up of two chapters devoted , respectively, to the principles of potentiometry and the measurement of electrode potential. In these, the reader is immediately made aware of the subject both by the slant of the treatment and by the sections concerned with the materials and selectivity of ion-selective elec- trodes. The main business of the book is covered by the chapters on the various classes of ion-selective electrodes, measuring techniques, the techniques used in analysis, and areas of applications of ion-selective electrodes. All classes of ion-selective electrodes are covered, with discussions on construction principles, characteristics , operating principles and calibration, with useful tables on the main characteristics and suppliers of electrodes within each category.While certain emphases in the discussion of electrode types could have been better placed in order to avoid misconceptions, the section on gas-sensitive electrodes is a useful introduction to this intriguing elaboration of ion-selective electrodes. Equivalent circuits and measuring instruments form the basis of the chapter on measuring techniques and, naturally, some of these aspects carry over to the chapter on analysis techniques. The standard techniques in the use of ion-selective electrodes are discussed, including known addition and Gran's methods.Additionally, some detailed experimental instructions are given, as in the determination of sodium and potassium in blood serum, but these, it must be emphasised, are limited in number. The biomedical field and continuous industrial and environmental analysis are the themes of the final chapter, which is followed by a short discrete discussion on the way ahead for ion-selective Anleatungen fur die chemzsche Laboratoriumspraxis, Band X I I I . Heidelberg and New York: Springer-Verlag. 1973. Price DM56; $23. Pp. xii + 226.August, 19741 BOOK REVIEWS 535 electrodes. In addition to data tables, the Appendix includes a discussion on activity, including references to the recent observations of Bates and Durst at the IUPAC-sponsored symposium held at UWIST, Cardiff, in April 1973.These, and the other two references to this symposium, are an indication of the speed with which the publishers have produced this book. Perhaps such despatch accounts for the high price which, together with the availability of so much information on ion-selective electrodes in English, is detrimental to giving the book a wide market. J. D. R. THOMAS METHODEN DER ORGANISCHEN ELEMENTAR- UND SPURENANALYSE. By I?. EHRENBERGER and S. GORBACH. Pp. xvi + 452. Weinheim/Bergstrasse: Verlag Chemie. 1973. PriceDM 129. Fifteen years have elapsed since the last edition of “Pregl and Roth” (the standard text in the German language) appeared, and it is clearly time that a new edition, or an entirely new text dealing with more recent developments, should replace it.The first text-book on organic micro- analysis, written by Pregl, ran to a number of editions. At least three English translations of this book have been published and, in addition, at least five independent English texts have been written by British or North American authors. Virtually none of the later books has had the opportunity to describe some of the notable advances that have taken place over the last decade, or even to include accounts of older methods that have since been well tested and tried under many different conditions. Accordingly, a comprehensive volume on elemental organic micro- analysis would be expected to provide not only the working details of newer methods that have become well established, but also to provide a well balanced picture and a critical and selective assessment of investigations over the last quarter of a century.The present volume falls somewhat short of these expectations. A generation ago it was not uncommon to find European authors complaining that their colleagues from another continent often appeared to be ignorant of earlier literature, and generally ascribed the pioneer work on any particular subject to the last authors who wrote about it. It was not a view to which I myself subscribed, except in a few individuaI cases, but the myth has certainly become fact in recent years in continents other than the one originally blamed. There are too many instances where the literature research has been carried out shoddily and important contri- butions have been omitted. This may well be due to the present-day chemist being over-pampered by the services that are available to him.In former times, there were few review articles available and one almost always had to go to the original literature for complete information. Nowadays, thanks to the proliferation of review articles, these are too often relied on for all background information, which is how so many of the myths and legends arise. Although the present volume provides comparatively few such errors, these may well have originated in this way. The book confines itself to elemental analysis; this is a wise choice, because the conventional microanalytical laboratory rarely undertakes functional group analysis nowadays, owing to the development of instrumental techniques.Methods for all the conventional elements are described in detail. There is a very scanty section on microgram analysis, but the details are so meagre that it would have been best left out. Finally, there are chapters on statistics and on balances and weighing. There is a reasonably comprehensive description of methods for the determination of carbon and hydrogen, including the most important automatic methods. Curiously enough, although the origin of the empty tube method is correctly ascribed, the system the authors recommend is one published many years after the original descriptions and which the reviewer had never seen before. This is remarkable, as the empty tube design has been a British Standard specification for about 20 years and, until the advent of automatic analysers, was the most widely used method in Britain.Paradoxically, the equivalent apparatus used for halogens and sulphur, which was not so widely used after the revival of the oxygen-flask method, is described in detail. The chance was lost to place in a text for the first time the historic development of the oxygen- flask method; instead, it is covered simply by references. There was also the chance to give Mikl and Pech full credit for their contribution. To them, more than anybody else in the post-war period, must go the credit for resuscitating the oxygen-flask method. Theirs was a semimicro method using a 500-ml flask; Schoniger’s contribution was to reduce the semimicro method to the micro scale. Mikl and Pech seem to be doomed to be deprived of their rightful place, which can only be ascribed to the reasons given in the second paragraph.On the favourable side, one can say that the methods are described in detail and anyone who uses this book will have little difficulty in following the instructions. It might be thought that536 BOOK REVIEWS [Analyst, Vol. 99 too many methods have been provided, but this criticism is only justifiable when methods are merely alternatives to each other. No method is capable of analysing every compound ; even the automatic methods are not universally applicable for the determination of carbon and hydrogen and the Burger - Zimmermann method is the only dependable method for determining sulphur in most organometallic compounds ; availability of apparatus may also affect the method of choice.Accordingly, one can say that a fair choice of methods has been provided. It can be said that this book puts more emphasis on methods developed on the Continent of Europe rather than in the UK and the USA, which is, after all, to be expected; it must not be forgotten that organic microanalysis originated in Austria. The general account is thus probably representative of preferred European practice. After having said that, I do not think this text is as good as the best one of those available in English, even though the latter needs to be up-dated ; hence there would be little advantage in preparing a translation of this volume. Pre-war, one could have bought half a micro-balance for the cost of this book. R. BELCHER AN INTRODUCTION TO SEPARATION SCIENCE.By BARRY L. KARGER, LLOYD R. SNYDER and New York, London, Sydney and Toronto: John Wiley The authors set out to present, in three parts, a unified view of the numerous separation techniques that are currently available. They are assisted by nine specialists, presenting salient features of particular techniques. Part I, “Fundamentals,” considers in turn separation equilibria, diffusion and mass transport, operational aspects of separation, chromatography and finally characteristics of individual separa- tion methods. Separation equilibria are discussed thermodynamically and on a molecular basis ; this latter aspect is often neglected but is a model of clarity of exposition in this section. Part 11, “Methods based on Phase and Distribution Equilibria,” contains chapters on distilla- tion (R. H.McCormick), gas - liquid chromatography, solvent extraction (H. Freiser), liquid - liquid chromatography, crystallisation (W. R. Wilcox) , ion-exchange separation processes (H. L. Rothbart) , liquid - solid adsorption chromatography, various other interfacial processes and finally exclusion processes (J. Y . Chuang and J. F. Johnson). Part 111, “Other Separation Methods,” covers barrier separation processes (R. A. Cross and H. Strathmann) , electrophoresis (M. Bier), miscellaneous methods including ultracentrifugation, particle size, electromagnetic separation, thermal diffusion, use of enzymes and finally a chapter on multi-step separation schemes for complex samples. The chapters on individual techniques cannot be compared to monographs, as ten to twenty pages are not sufficient to develop detailed discussions and tabulate exhaustive lists of examples. As introductions, placing techniques in perspective and providing key references, most chapters are considered to be useful. The last chapter, on multi-step separations, outlines the stages in the determination of aldosterone in human urine, the isolation of 13-methylhentriacontane from corn eanvorm faeces, analysis of individual skin lipids and the separation and analysis of the oxygen- and nitrogen-containing compounds, or both, in petroleum. These schemes illustrate many of the techniques outlined earlier in the book and are a realistic and fitting conclusion to it. The presentation and diagrams are of a high quality and few typographical errors were noted. The book achieves its aims and will repay study by both academic and industrial analytical CSABA HORVATH. & Sons. 1973. Price f19-75. Pp. xxii + 586. chemists and others such as biochemists with separation problems or interests. D. THORBURN BURNS

ISSN:0003-2654    

DOI:10.1039/AN9749900533

出版商:RSC    

年代:1974

数据来源: RSC

摘要:

536 BOOK REVIEWS [Analyst, Vol. 99 Errata NOVEMBER (1973) ISSUE, p. 825, line 14 of text: for “Preparation of sample” read “Preparation of sample wzixture.” Line 19 of text: for “sample size” read “injection volume.”

ISSN:0003-2654    

DOI:10.1039/AN9749900536

出版商:RSC    

年代:1974

数据来源: RSC