摘要:

Analyst, September, 1973, Vol. 98, pp. 673-686 673 Observations on the Use of 2,4,6 -Trinitrobenzenesulphonic Acid for the Determination of Available Lysine in Animal Protein Concentrates - BY R. J. HALL, N. TRINDER AND D. I. GIVENS (Ministry of Agriculture, Fisheries and Food: Agricultural, Development and Advisory Service, Government Buildings, Kenton Bar, Newcastle upon Tyne, NE1 2 Y A ) The determination of available lysine in proteins by their reactions with 2,4,6-trinitrobenzenesulphonic acid has been examined in relation to the preparation of the sample, the conditions of the reaction, the hydrolysis of the trinitrophenylated protein and the effects of other amino-acids and related compounds. From these observations a revised method is proposed for the rapid routine screening of animal protein concentrates used in animal feeds.With the use of DL-lysine as a standard instead of E-trinitrophenyllysine and with a simplified hydrolysis procedure, the technique is suitable for examining comparatively large numbers of samples. Interference by other amino com- pounds is slight, except that by cadaverine, hydroxylysine and ornithine. Results are given for the levels of available lysine in a range of animal proteins, which compare closely with values obtained by the Carpenter procedure in which l-fluoro-2,4-dinitrobenzene is used. THE nutritional r6le of lysine as one of the most important amino-acids is well established and it is known that growth response in chicks fed on cereal-based diets and supplemented with different proteins is frequently a reflection of the level of the available lysine in the added pr0tein.l Animal and vegetable proteins vary in their lysine content, some sources being much richer than others,2 and while the total lysine of a feedingstuff can now be deter- mined accurately with an amino-acid analyser, it has been shown that in certain circumstances not all of the lysine present in a protein is available to an anima1.3s4 Knowledge of the total lysine content itself is not, therefore, always a reliable guide to the true feeding value of the protein in relation to lysine.In an attempt to overcome this difficulty, Bruno and Carpenter5 and Carpenter and Ellinger6 developed methods with which attempts were made to determine the amount of lysine in a protein that is actually available nutritionally to the animal.The procedures are based on the work of Sanger,' who showed that when the products of the reac- tion of l-fluoro-2,4-dinitrobenzene with individual amino-acids in sodium hydrogen carbonate solution are heated with hydrochloric acid, bright yellow dinitrophenylated amino-acids are formed (reaction I). Sanger considered that only the terminal amino-acid group and the epsilon-amino group of lysine that form part of the protein molecule reacted with 1-fluoro-2,4- dinitrobenzene. After acid hydrolysis of the dinitrophenylated protein the individual dinitro- R I N H .C H .COO H R I NaHC03 + HCI +NO2 @NO* +- NH2.CH.COOH hydrolysis Amino-acid NO2 NO2 1 -F luoro-2,4-dinitrobenzene Dinitrophenylated amino-acid Reaction I @ SAC; Crown Copyright Reserved.674 HALL et ai.: USE OF 2,4,6-TRINZTROBENZENESULPHONIC ACID FOR [Analyst, VOl.98 phenylated amino-acids could then be separated chromatographically, thus giving an indica- tion of the peptide linkages and of the lysine present. In later work, in order to avoid the use of chromatography, Carpenter* used a selective solvent-extraction technique to separate c-dinitrophenyllysine from the reaction mixture and included the use of methyl chloroformate to overcome interference caused by arginine. The principle of Carpenter’s method is that the terminal (epsilon) amino group present in the amino-acids lysine and hydroxylysine is not involved in the peptide linkage of a protein but may sometimes be combined with some other compound (e.g., an aldehyde such as an aldose sugar) in such a way that part of the lysine becomes nutritionally unavailable to the animal.He postulated that if the combination of the €-amino group with another compound such as an aldose sugar also prevents a chemical reaction between the amino group and an amino coupling reagent, then the extent of combination of the lysine amino group could be measured, thus giving an indication of the amount of lysine of biological value to an animal. This measure could, in some circumstances, give a better prediction of the value of a diet for non-ruminants than one based on total lysine. Carpenter adopted Sanger’s reagent (l-fluoro-2,4-dinitrobenzene) for the amino coupling; other compounds such as 1-fluoro- 2,4-dinitroaniline react in a similar manner9 but until recently only the former has been widely used.The Carpenter procedure has made it possible to determine more realistic lysine values for animal feedingstuffs and certain high-protein vegetable feeds than hitherto but suffers from several inherent disadvantages. The method is somewhat time consuming for routine multiple analyses and requires two particularly unpleasant chemicals : 1-fluoro-2,4- dinitrobenzene, which is dangerous, and methyl chloroformate, which is lachrymatory. Further, because lysine itself cannot be used as an internal standard, the use of E-dinitro- phenyllysine is needed. A simpler test based on the same principle as that of Carpenter but in which use is made of the reagent 2,4,6- t rinit robenzenesulphonic acid (TNBS) instead of 1 -fluoro-2,4-dinit robenzene has been suggested by Kakade and LienerlO (reaction 11).The original work with this com- pound on pure amino-acids and peptides was carried out by Satake and his colleagues,llp12 and from their findings Kakade and Liener devised a procedure that overcomes some of the objections to the 1 -fluoro-2,4-dinitrobenzene method. R I NH. CH. COOH R I NaHC03 + HCI 02N@NO2 -I- Amino-acid NH2.CH.COOH hydrolysis SOzOH = I NO2 2,4,6-Trinitrobenzenesulphonic acid NO* Trinitrophenylated amino-acid Reaction I I A method suitable for routine use was required in this laboratory in order to establish the available lysine levels of protein-rich animal feedingstuffs and this paper describes a re- appraisal of the technique of Kakade and Liener.From our observations a revised procedure is proposed, which it is thought makes the determination of available lysine in large numbers of samples comparatively easy. METHOD The sample is ground to a very fine powder, which is suspended in a solution of agar and the suspension mixed with sodium hydrogen carbonate solution. A solution of trinitro- benzenesulphonic acid is added, which reacts with the free epsilon-amino group present in the lysine combined within the intact protein. E-Trinitrophenyllysine (c-TNP-lysine) is then released by hydrolysis of the reaction mixture with hydrochloric acid and determined spectro- photometrically. Interfering substances such as free picric acid are removed by extraction into diethyl ether and the absorbance of the remaining yellow solution of E-TNP-lysine is measured at 415 nm.Pure DL-lysine monohydrochloride is used as a standard.September, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 675 APPARATUS- Glass specimen tubes, 50 x 25mm. Porcelain ball-mill with Alorite cylinders. Stainless-steel sieve, 200 mm diameter, B.S. 200 mesh. Test-tubes, 15-ml capacity-Graduated to 10 ml, with C14 glass or polythene stoppers. REAGENTS- Whenever possible use analytical-reagent grade chemicals and water redistilled from all-glass apparatus. Agar, 0.1 per cent. m/V solution-Add 1.0 g of agar (Difco) to approximately 900 ml of water. Heat the mixture to boiling, allow it to cool and make the volume up to 1 litre. Preserve the solution by shaking it with 2 ml of octan-2-01.Diethyl ether. Hydrochloric acid, approximately 11 M, sp. gr. 1.18. DL-Lysine monohydrockloride-Dry overnight at 60 "C. Sodium hydrogen carbonate solution, 1 M. 2,4,6-Trinitrobenzenesulphonic acid (Picrylsul$honic acid) solution, 1.0 per cent. m/V solution in water-Prepare freshly on each occasion. PREPARATION OF SAMPLE- Grind the sample in a suitable mill to pass a 0-5-mm mesh sieve (see Notes 1 and 2). Take 5 to 10 g of this sample and re-grind it in a porcelain ball-mill containing approximately twenty Alorite cylinders until a very fine powder is obtained; grinding overnight is usually sufficient. The resulting powder should pass a B.S. 200-mesh sieve and will have a particle size not exceeding 75 pm. Remove the powder from the milling chamber and the cylinders with a soft camel-hair brush on to a sheet of white paper and transfer it to a small screw-capped jar.Grind any residue left on the sieve in a porcelain pestle and mortar, or if necessary, re-grind it in the ball-mill until all of the 5 to 10-g sub-sample has passed the 200-mesh sieve. Accurately weigh 045g of the finely ground sample and transfer it to a 50-ml calibrated flask containing 4ml of acetone. Add 40ml of agar solution with 3 drops of octan-2-01 and shake the mixture vigorously so as to ensure adequate mixing of the sample. Dilute to volume with 0.1 per cent. agar solution. A homogeneous suspension should be obtained. REACTION OF SAMPLE WITH TRINITROBENZENESULPHONIC ACID- Into each of two 10-ml graduated tubes, transfer by pipette 0-5 ml of the sample suspen- sion containing 5 mg of material to be analysed and add 0-5 ml of 1 M sodium hydrogen carbonate solution.Add to one tube (sample reaction) 1.0ml of the TNBS solution and mix. To the other tube, identified as the sample blank, add 3.0 ml of approximately 11 M hydrochloric acid followed by 1.0ml of TNBS solution and mix. Stopper each tube and place them in an oven or incubator maintained at 40 "C. After 75 minutes, remove the tubes from the oven and carefully add to the alkaline reaction mixture 3.0 ml of approxi- mately 11 M hydrochloric acid from a burette. Mix the contents by tapping the bottom of the tube. Do not replace the stoppers but cover the tops of the tubes with inverted specimen tubes. Place the covered tubes in a suitable rack and then into a covered, vigorously boiling water bath.Ensure that the level of the water in the bath is above the level of the liquid in the tubes but below the rims of the inverted specimen tubes. After 2 hours from the time the water has resumed boiling, remove the tubes from the bath, cool and make the volumes up to 10 ml with water. Stopper the tubes and mix the contents by gentle inversion. DETERMINATION OF LYSINE- With a pipette, transfer 4.0 ml in duplicate (or 3.0 ml in triplicate) from each sample hydrolysate and one 4-0-ml (or 3.0-ml) portion from the sample blank into another set of 10-ml graduated tubes. Make the volumes up to approximately 8 ml with water and add about 5 ml of diethyl ether from a soft polythene wash-bottle (see Note 3). Close the tubes tightly with stoppers (the yellow polythene type afford a better seal than those of other plastics or glass) and shake them vigorously, one in each hand, for 10 to 15 s.Immediately remove the stoppers and, by using a polythene wash-bottle, rinse them with a few drops of water into the tubes. With a fine-tipped Pasteur pipette attached to a water vacuum pump, remove the676 HALL ef al. : USE OF 2,4,6-TRINITROBENZENESULPHONIC ACID FOR [Ana&St, VOl. 98 ether layer to within 2 mm of the aqueous phase. Take care not to remove any of the aqueous phase. Repeat the extraction once more with a further 5 ml of ether, washing the stoppers as before into the tubes. After removing the second ether extract, leave the tubes unstoppered in an incubator at 40 "C so as to allow the residual ether to evaporate.Alternatively, remove the ether with a stream of compressed air or nitrogen. Make the volume of the aqueous phase up to 10.0 ml with water, stopper the tubes and mix the contents by inverting the tubes. Remove the stoppers and centrifuge the tubes for 10 minutes at 3000 r.p.m. Measure the absorbance of the yellow E-TNP-lysine solution at 415 nm in an optical cell of 10-mm light path. The colour is stable in ordinary daylight but fades slowly in bright sunlight. NOTES- 1. Samples containing more than 5 per cent. of fat need to be extracted with light petroleum (boiling range 40 to 60 "C) before grinding. They should then be dried at 40 "C overnight prior to ball-milling. 2. The protein and moisture content should be determined on the ball-milled sample that passes a ZOO-mesh sieve.The moisture content of the original sample prior to ball-milling should also be determined. 3. Great care must be taken to avoid naked flames and electric hot-plates or boiling rings when extracting with ether. PREPARATION OF STANDARD GRAPH- Stock standard lysine solution containing 5 mg ml-1 of lysine-Dissolve 0.625 g of DL-lysine monohydrochloride (equivalent to 0.5 g of lysine) in 100 ml of boiled distilled water that has previously been shaken with 0-1 ml of octan-2-01 as a preservative. Store the solution in a polythene bottle at 2 "C and replace it after 4 weeks. Dilute standard lysine solution containing 0.50 mg ml-1 of lysine-Dilute 5 ml of the stock lysine solution to 50 ml with boiled distilled water (previously shaken with 0.1 ml of octan-2-01) in a calibrated flask.Standard lysine reactions-By pipette, transfer aliquots of the dilute lysine standard solution containing 50, 100, 150, 200 and 250 pg of lysine into 10-ml graduated tubes. Make the volumes up to 0.5 ml with water. Prepare a reagent blank with 0.5 ml of water. Add 0.5 ml of 1 M sodium hydrogen carbonate solution. Add 1.0 ml of TNBS solution to each tube. Mix the contents and transfer the tubes into an incubator or oven maintained at 40 "C. Leave for 75 minutes and then continue the procedure as for a sample from and includ- ing the incubation stage. The reaction between lysine and TNBS gives a linear calibration graph and, in practice, only the 250-pg lysine standard needs to be prepared with a reagent blank for each set of determinations.The absorbance of 1OOpg of lysine as E-TNP-lysine, prepared by and measured under the above conditions, in a volume of 10.0 ml is approximately 0.425 with a blank of 0.010. The sample blank will vary according to the nature of the material but it is normally 0.040 to 0.070. Store the solution at 2 "C and replace it after 1 week. EXPERIMENTAL AND RESULTS All observations were made at least in duplicate and often in triplicate or quadruplicate. Absorbance measurements were made in duplicate or triplicate with an Optica double-beam or Unicam SP500 spectrophotometer. Amino-acids and related compounds were obtained from Sigma (London) Chemical Co. Ltd. Although much simpler than Carpenter's method,s the procedure of Kakade and LienerlO in our hands presented a number of difficulties for routine analyses.The weighing of 10 mg of coarsely ground sample was tedious and seemed to cause poor reproducibility; the reaction time with TNBS was excessive and the use of an autoclave for hydrolysis with hydrochloric acid solutions created a problem (see below). Furthermore, Kakade and Liener adopted E- trinitrophenyllysine as a standard. This reagent is not available in the U.K. and is very ex- pensive to import from the U.S.A. The use of pure lysine as a standard at that time had not been considered. All these aspects were examined in detail and the results are recorded below. SAMPLING- The reaction of protein with TNBS is extremely sensitive, the product having an intense yellow colour even after acid hydrolysis.To achieve good reproducibility it was necessary for tlie sample to be in the form of a very fine powder, which could then be suitably sub- sampled, but the weighing of amounts of 10mg or less is time consuming and subject t oSeptember, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 677 error. The first attempt to solve this problem was suggested by a colleague (Mr. L. R. Flynn) and consisted in grinding the air-dry, milled sample in a ball-mill with nine times its mass of washed silver sand. A satisfactory preparation was obtained in this .way and replication of lysine determinations on 100-mg amounts of the sand - sample mixtures gave results that were in very good agreement. However, when nitrogen determinations were carried out on the mixtures with sand a disturbing lack of agreement between the results obtained and those for the original samples was observed.The reasons for the differences between the nitrogen levels were investigated but a satisfactory explanation was not found. However, the relative density of the sand to that of the sample, combined with a change in the moisture content of the sample during the ball-milling, were thought to be two of the possible reasons for the discrepancies. Attempts to replace sand with other diluents such as boric acid and Hyflo Supercel did not effect much improvement. The difficulty was resolved by grinding the sample in a ball-mill without the addition of any diluent and then sub-sampling by using a suspension of the ball-milled material in a dilute agar solution.This technique had pre- viously been used in another analytical procedure for sampling small amounts of plant tissues and soils (HalP3). The use of the agar suspensions enables replicate aliquots to be taken and the preparations can be retained for further use if required. The occasional sample may need special treatment. Thus we have found gluten and zein do not form homogeneous suspensions in 0.1 per cent. agar but can be satisfactorily suspended in a 60 per cent. V/V solution of ethanol in water. Casein needs to be dissolved in 0.5 M sodium hydrogen carbonate solution containing 0.1 per cent. of agar. Ordinary preparations of egg albumen were also prepared in this way although lyophilised ovalbumin was readily soluble in the agar solution alone.REACTION OF PROTEIN WITH 2,4,6-TRINITROBENZENESULPHONIC ACID- It is not clear from the text in the paper by Kakade and LienerlO whether they propose that the TNBS should be used as a 0.1 or a 1.0 per cent. solution (0.05 or 0-5 per cent. in the reaction mixture). The reaction time of 2 hours adopted by these workers also seemed unnecessarily long for such a small amount of finely ground sample. These two features and the effect of temperature on the reaction were investigated by using a commercial meat meal produced from waste animal tissues, the available lysine content of which was 4-36 g per 16 g of nitrogen as determined by the method of Carpenter.8 The sample was made to react with TNBS at 40 and 60 "C in a volume of 2 ml of 0.25 M sodium hydrogen carbonate solution, then hydrolysed with approximately 7 M hydrochloric acid and extracted with ether as described in the proposed method.An aliquot of a solution of DL-lysine monohydrochloride (equivalent to 200 pg of lysine) was treated in the same way. The results (Table I) show that a concentration of TNBS greater than 0.1 per cent. in the reaction mixture for a period of 15 to 90 minutes at 40 "C has little influence on the final available lysine figure. The absorbance values of the TNP-lysine standard, the TNP- lysine sample and their respective blanks remained constant within small experimental variations. When, however, the reactions were carried out at 60 "C, increases were measured in the absorbance values of the ether-extracted solutions of the sample, the standard lysine and their blanks, which resulted in apparent higher available lysine figures (much higher than the Carpenter values). After considering the results presented in Table I, it was decided to use a final concentration of TNBS of 0.5 per cent.at 40 "C with a reaction time of 75 minutes for further work on a variety of materials with an expected wide range of lysine levels. HYDROLYSIS OF TRINITROPHENYLATED PROTEIN- The proposal of Kakade and Liener to hydrolyse the protein containing trinitrophenyl linkages in an autoclave at 120 "C seems on first consideration to be a great improvement on the prolonged hydrolysis used in the Carpenter method, but problems were soon experienced. In practice, with an automatic electric autoclave the recommended temperature of 120 "C was not attained until approximately 20 minutes after sealing the instrument at 100 "C, and nearly 40 minutes usually elapsed before the tubes could be removed at the end of the 1-hour hydrolysis time. Under these conditions the suggested procedure involves virtually a 2-hour hydrolysis period.The presence of hydrochloric acid in the autoclave atmosphere was also found to have a serious corrosive action upon the metal interior of the apparatus.678 HALL et U l . : USE OF 2,4,6-TRINITROBENZENESULPHONIC ACID FOR [Analyst, VOl. TABLE I EFFECT OF TNBS CONCENTRATION AND REACTION TIME ON THE DETERMINATION OF AVAILABLE LYSINE I N A MEAT MEAL Sample reaction in reaction Reaction Reaction A v a i i a b l e Absorbance of TNBS 98 mixture, per cent.0.05 0.10 0.25 0.50 1-00 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 temperature/ "C 40 40 40 40 40 40 40 40 40 40 40 60 60 60 time/ minutes 75 75 75 75 75 15 30 60 98 120 180 15 30 60 Absorbance* 0.201 0.248 0.270 0-270 0.274 0.262 0.268 0.274 0.267 0.289 0.270 0.300 (0.073) 0.394 (0.098) 0.344 (0.071) 1 ysine/ DL-lysine (100 pg) g per 16 g of NT at 415 nm* 4.43 0.312 4-19 0.407 4-43 0-419 4-50 0.41 3 4-34 0.433 4-36 0.413 4-46 0.4 13 0.406 4.63 4-44 0.412 4.91 0-403 4.58 0-404 4.36 0.421 (blank 0.027) 5.23 0.394 (blank 0.098) 5.63 0.420 (blank 0.220) * Corrected for blank absorbance (shown in parentheses). t Carpenter value, 4.36 g per 16 g of nitrogen. It was considered that a boiling water bath might prove a satisfactory alternative.The same sample of meat meal was used as for the experiments described above (available lysine value by the Carpenter method, 4.36 g per 16 g of nitrogen). The initial reaction with 10 mg of sample was with 0.5 per cent. TNBS solution for 75 minutes at 40 "C, followed by hydrolysis with approximately 7 M hydrochloric acid in a boiling water bath, or at 120 "C in an autoclave. From Table I1 it can be seen that the hydrolytic action of the acid to convert the trinitro- phenylated compound (ie., TNP-lysine) into a form insoluble in ether proceeds at a faster rate from the sample than from the pure lysine standard. Indeed, the absorbances show that the hot acid releases the TNP-lysine in the sample to its maximum extent very quickly. However, the calculated value for the available lysine is dependent on the absorbance of the TNP-lysine standard, which is profoundly affected by the concentration of hydrochloric acid at the time of hydrolysis of the trinitrophenylated protein and TNP-lysine.TABLE I1 EFFECT OF CONCENTRATION OF HYDROCHLORIC ACID, TIME AND TEMPERATURE OF HYDROLYSIS ON THE RESULTS OBTAINED FOR AVAILABLE LYSINE I N MEAT MEAL Sample reaction Hydro 1 y s i s Hydrolysis time/ method minutes Boiling water bath 30 60 90 120 180 Autoclave a t 120 "C 15: 30 $ 601 120$ r Absorbance* 0.258 0.270 0.268 0-264 0.267 0.281 0.271 0.243 0.264 1 Available Absorbance of lysinel DL-lysine hydrochloride g per 16 g of NT (100 P d * 6.64 0.267 5.32 0.348 4.50 0.409 4.32 0.419 4.48 0.409 0.426 4.53 4-63 0-402 4-70 0.355 4.32 0.419 Hydrochloric acid a t 100 "C for 120 minutes 1 M 0-146 34-56 0.029 (blank 0.118) 2M 0.200 22.87 0.060 (blank 0.039) 4M 0-231 7.81 0-203 (blank 0-017) 7M 0.267 4.56 0.401 (blank 0-015) * Corrected for blank.t Carpenter value, 4-36 g per 16 g of nitrogen. $ Plus approximately 60 minutes required to attain autoclave temperature and to cool to 100 'C.September, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 679 The results of a more extended experiment with 5 mg of casein, fish meal and another meat meal confirmed that extending the hydrolysis time with boiling 6 M hydrochloric acid to 16 hours had little effect on the absorbance of the E-TNP-lysine measured. But this is not true when determining total lysine14 in hydrolysates prepared by boiling with this acid in the conventional way for 2 to 16 hours (Table 111).TABLE I11 EFFECT OF HYDROLYSIS TIME ON THE ABSORBANCE OF E-TRINITROPHENYLLYSINE IN MEAT AND FISH MEALS AND CASEIN hydrolysis times* Absorbance values a t various Total lysinel g per 16 g of N r ) * A Time a t 100 "C/minutes 16 hours 2 hours 16 hours 0 2 5 15 30 60 120 180 a t a t a t r h , 110 "C 100 "C 110°C Meat meal 0.060 0.090 0.114 0.142 0.145 0.153 0.163 0.152 0.146 4.47 5-58 Casein (BDH) 0.008 0.080 0.199 0-344 0.394 0.409 0.464 0-470 0.478 6.34 8.32 Fish meal 0.188 0.249 0.329 0.373 0.391 0.439 0.438 0.429 0-440 6.52 7.89 *Corrected for blank absorbance The influence of the hydrochloric acid is demonstrated by the apparent values for available lysine when the trinitrophenylated protein was hydrolysed with various concentrations of this acid and extracted with ether as described in the method (Table 11, third section).Again, it is clear that the reaction of TNBS with the lysine combined in the protein is different from that with pure lysine. The absorbances measured after the action of the hydrochloric acid in the autoclave at 120 "C (Table 11) indicate that some destruction of the E-TNP-lysine may take place after 30 minutes. From all these observations it was decided to hydrolyse the trinitrophenylated protein for 2 hours in a boiling water bath, with a final concentration of approximately 7 M hydrochloric acid. INTERFERENCE BY OTHER AMINO COMPOUNDS- Although trinitrobenzenesulphonic acid reacts with the epsilon-amino group of lysine and hydroxylysine in the intact protein it was thought desirable to study the reaction with other individual free amino-acids.Twenty-five amino-acids and related amines were allowed to react with TNBS by using 200 pg of compound and the conditions described for the treatment of the lysine standards. The results are presented in Table IV. For all practical purposes only lysine, hydroxylysine, cadaverine, ornithine, galactosamine and glucosamine reacted to give a product that was not extracted into ether. Proline and hydroxyproline did not couple. Reactions with arginine, aspartic acid, glutamine, glutamic acid, histidine and 1 TABLE IV REACTIONS OF AMINO COMPOUNDS WITH 2,4,6-TRINITROBENZENESULPHONIC ACID Absorbance Amino compound* a t 415 nmt D-Alanine . . ..L-Arginine - . . * L-Asparagine . . . . L-Cysteine . . . . DL-Cystine . . . . DL-Aspartic acid . . Cadaverine . . . . D-Galactosamine . . u-Glucosamine .. DL-Glutamic acid . . L-Glutamine . . . . uL-Hydroxylysine . . L-Histidine . . . . Glycine . . . . 0.008 0.0 16 0.006 0.020 0.057 0.034 0.033 0.384 0-381 0.009 0-005 0.025 0.210 0.015 Per cent. of lysine absorbance 2.0 4.1 1.5 4.9 14.0 8.4 8-4 94.4 93.6 2-2 1-3 6.4 51.6 3-8 Per cent. Absorbance of lysine Amino compound* a t 415 nmt absorbanctl L-Hydroxyproline . . 0.007 1.7 DL-Isoleucine . . . . 0.009 2.2 DL-Leucine . . . . 0.010 2.5 100.0 DL-Lysine . . . . 0,407 DL-Methionine . . 0.002 0.5 DL-Ornithine . . . . 0,412 101.2 DL-Proline . . . . 0.003 0.8 DL-Serine . . . . 0.010 2.5 DL-Threonine . . 0.005 1.2 DL-Tryptophan .. 0.027 6.0 DL-Tyrosine . . . . 0-007 1-7 DL-Valine . . . . 0.010 2.5 L-Phenylalanine . . 0.006 1.5 * Reaction with 200 pg of amino compound, allowance being made for combined water and The aliquot taken for ether extraction contained 100 pug of amino compound t Absorbance of ether-extracted test solution made up to a final volume of 10 ml and measured' hydrochloride. in optical cells with a 10-mm light path; values after subtraction of the blank reading.680 HALL et at. : USE OF Z,4,6-TRINITROBENZENESULPHONIC ACID FOR [Analyst, VOl. 98 hydroxylysine produced amber colours in hydrogen carbonate solution, which were visibly less intense than those for the same amounts of the other amino-acids. After acidification with hydrochloric acid, the deep amber - orange reaction mixtures assumed the canary yellow colour of picric acid.With lysine, hydroxylysine, ornithine and cadaverine and the two amino-sugars galactosamine and glucosamine, the acidified solutions became turbid, presum- ably as a result of precipitation of the diTNP-amino product. The prolonged action of the hot hydrochloric acid caused the TNP-amino-acid colour to diminish markedly in all reactions except those with the amino-sugars, lysine, hydroxylysine, ornithine, cadaverine, tryptophan, valine, glycine, isoleucine and threonine. Finally, except for the small amounts shown in Table IV, the yellow reaction product obtained from all the compounds except lysine, hydroxylysine, cadaverine, ornithine and the amino-sugars following hydrolysis was extracted into ether. Trinitrobenzene- sulphonic acid reacted with putrescine in hydrogen carbonate solution to produce the charac- teristic amber colour, but on acidification a precipitate was formed which remained insoluble at 100 "C and, although not extractable into ether, did not impart a yellow colour to the aqueous phase but concentrated at the ether interface as a plastic-like pellicle.In order to establish whether or not the residual yellow colour (Le., TNP-amino compound) left in the aqueous phase following ether extraction of the separately hydrolysed TNP-amino- acids was additive, a mixture of amino-acids was prepared in which the concentrations of the individual compounds approximately corresponded to those which would have been present in the hydrolysate of a protein of average lysine content, except that lysine was replaced by the equivalent amount of additional methionine.Suitable volumes containing 5, 2, 1, 0.4 and 0.2 mg of the mixed amino-acids were made to react with TNBS and hydrolysed with hydro- chloric acid, and half of the TNBS hydrolysate was extracted with ether. The results (Table V) show that the absorbance obtained for each separate amino-acid is not additive when a mixture of acids is allowed to react, i.e., the actual absorbance of 0.090 for 2-5 mg of mixed amino-acids is only approximately one third of the residual absorbance (0.264 calculated from Table IV) following ether extraction if all the amino-acids in the protein had completely reacted. In an actual determination when most of the amino-acid present is in the form of protein and not in a free state, a much lower value would be expected. No reactions were obtained with ammonia and urea.TABLE V REACTIONS OF MIXED AMINO-ACIDS WITH TNBS Composition of amino-acid solution, per cent.- Alanine . . . . 5 Glutamic acid . . 15 Isoleucine . . 6 Serine . . .. 5 Arginine . . . . 6 Glycine . . . . 6 Methionine (increased Threonine. . .. 4 Aspartic acid . . 10 Histidine . . . . 2 toreplacelysine) 13 Tryptophan . . 1 Cystine . . . . 1 Leucine . . . . 8 Phenylalanine . . 4 Tyrosine . . .. 3 Proline . . . . 5 Valine . . .. 6 Actual absorbance of ether-insoluble Absorbance calculated TNP product* from Table IV 2.50 mg of mixed amino-acids . . .. 0.090 0.264 1.00 mg of mixed amino-acids , . .. 0.034 0-106 0.50 mg of mixed amino-acids .. .. 0.023 0.053 0.02 1 0.20 mg of mixed amino-acids . . . . 0.012 0.10 mg of mixed amino-acids . . .. 0.003 0.010 0.10 mg of lysine . . .. .. 5.00 mg of zein . . .. .. 0.427 - .. .. 0.023 - * Corrected for blank absorbance. RECOVERY EXPERIMENTS- A major difficulty inherent in any procedure, in which use is made of a non-specific chemical reaction in situ for the determination of a naturally occurring organic component of animal or plant tissue, is that of assessing the validity of the assay. Even with the determination of metallic elements, the recovery of known additions to the original sample or at some suitable stage in the analysis often poses problems, and with biological specimens these difficulties are magnified because the organic entities may react differently when they are part of a more complex molecule than they would in isolated pure form. As pure lysine cannotSeptember, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 681 be used for recovery tests in the reaction involving l-fluoro-2,4-dinitrobenzene, other w0rkers8J5-~~ have incorporated edinitrophenyllysine into the procedure, measured its recovery and applied a correction factor to the values obtained for their samples on the assumption that the lysine in the sample was determined (or decomposed) to the same extent as that of the added e-dinitrophenyllysine.A review of the literature reveals that the recovery of edinitrophenyllysine added in this way can vary greatly according to the nature of the sample.This aspect was therefore investigated in relation to the use of trinitrobenzene- sulphonic acid. Absorbance of E-trinitro$henylZysine-eTrinitrophenyllysine monohydrochloride mono- hydrate (relative molecular mass 411-8) has been reported by Okuyama and Satakell to have a molar extinction coefficient of 1.33 x lo4 1 mol-1 cm-1 in 1 M hydrochloric acid at 346 nm (the near ultraviolet region). Kakade and LienerlO reported their observed molar absorbance as 1-56 x 104 1 mol-1 cm-1 and attributed the difference to varying conditions of acid hydroly- sis, which they stated partially destroys the E-TNP-lysine, although Okuyama and Satakell obtained 85 to 95 per cent. recovery after 5 hours at 110 “C in 6 M hydrochloric acid. In the present work a peak in the absorption curve was consistently measured at 350 nm with a wide shoulder from 400 to 420 nm that contained a further peak at 415 nm.For practical reasons the wavelength at 415 nm was selected for the determinations; the absorbance at this wave- length over the range 0 to 400 pg of lysine obeys Beer’s law. Recovery of E-TNP-lysine from the analytical procedure-Table VI shows the absorbance measurements, with the percentage “recovery” of E-trinitrophenyllysine following its addition at different stages in the analytical procedure as described for the determination of available lysine. Two samples of e-TNP-lysine were used; one kindly provided by Professor Liener and the other a commercial product supplied by the Nutritional Biochemicals Co. Ltd. of Cleveland, Ohio. The solutions were prepared in methanol.TABLE VI RECOVERY OF E-TRINITROPHENYLLYSINE BY ANALYTICAL PROCEDURE Sample Lienev- e-TNP-lysine (= 80-0 pg of lysine) direct in 1 M HCl . . .. .. 7 M HCl = 80.0 pg of lysine . . .. .. .. .. .. = 56.4 pg of lysine . . .. .. .. .. .. e-TNP-lysine (= 10343 pg of lysine) direct in 1 M HC1 . . .. .. e-TNP-lysine (= 56.4 pg of lysine) in 1 M HC1 after ether extraction . . e-TNP-lysine ether extracted after 2 hours a t 100 “C in approximately Nutritional 3iochemicals- e-TNP-lysine (= 83.2 pg of lysine) in 1 M HC1 after ether extraction . . c-TNP-lysine (= 83.2 pg of lysine) ether extracted after 2 hours in approximately 7 M HC1 a t 100 “C . . .. .. .. .. DL-lysine, 80.0 pg taken through method . . .. .. .. * Absorbance a t 415 nm in 10-ml volume.Absorbance* r - l Measured expected value Per cent. of 0.342 - 0.248 102.9 0.347 101.5 0.242 100.4 0.444 100.0 (and 0.351 98.9 of Liener value) 0.352 99.2 100.6 0.344 These results demonstrate that under the conditions proposed for the determination, the product from pure DL-lySine after ether extraction has virtually the same absorbance value per unit mass as a solution of the dry synthetic E-TNP-lysine. At 346 nm the absor- bance of 225.8 pg of E-TNP-lysine hydrochloride monohydrate estimated to be formed from 80.0 pg of lysine was 04350, which compares well with the calculated value of 0.855, taking 1.56 x lo4 1 mol-l cm-1 as the molar absorbance at this wavelength.1° Recovery of lysine and e-TNP-lysine added to protein suspensions-Tables VIIA and VIIB show the recoveries of various amounts of lysine added to 10 mg of protein suspension and made to react with TNBS as described above.From the values shown in Table VIIA, it can be assumed that the added lysine reacted completely and in those particular tests the recoveries were quantitative. This was not, however, always so and in Table VIIB a lack of consistency in the recovery values is recorded. No logical explanation has been found for these findings. Changing the conditions of hydrolysis with increased concentrations of682 HALL et d.: USE OF 2,4,6-TRINITROBENZENESULPHONIC ACID FOR [Anaby§t, VOI. 98 hydrochloric acid effected some improvement but introduced other disadvantages, such as a greater degree of charring and possible destruction of E-TNP-lysine. The recovery of E-TNP-lysine added to the protein suspension was also extremely variable (Table VIIC).In these reactions amounts of the two E-TNP-lysine samples referred to above (equivalent to 70 pg of lysine) were added to 10 mg of protein suspension before the addition of TNBS. 6-TNP-lysine was measured in the ether-extracted hydrolysates. TABLE VII RECOVERY OF LYSINE AND TNP-LYSINE ADDED TO PROTEIN SUSPENSIONS A. B. C. Sample Zein (5.0 mg) . . ,. . . .. .. Zein + 100 pg of lysine . . . . . . . . Zein + 50 pg of lysine . . . . . . . . M.5709 fish meal (5 mg) . . .. .. . . M.5540 fish meal (2.5 mg) . . . . . . . . + 50 pg of lysine . . + 50 pg of lysine . . . . . . Lysine measuredlpg Recovery, per cent. 7.6 - 109.2 101.6 57.0 98.8 103.5 - 152.0 97-0 89.1 - 138.2 98.2 Recovery of 100 pg of lysine added to 5 mg of $rotein-- Casein .. ... . .. .. .. . . . . .. 54-5, 60.2 Egg albumen . . . . .. .. .. . . .. .. 71.8, 73.8 Gluten . . . . . . .. .. . . . . .. . . 52-5, 73.3, 95.0 M.5529 blood meal . . . . . * .. . . .. .. . . 7 1.0 M.5709 fish meal . . .. * . .. .. . . . . . . 48-2, 99.0 M.5521 fish meal . . .. .. .. . . .. .. . . 78.1, 96.7 M.5540 fish meal . . .. .. .. . . . . .. . . 61.0, 80.8 Recovery of 70 pg of lysine as TNP-lysine added to 5 mg of protein- Casein, Nutritional Biochemicals . . .. .. .. . . . . 68.5 M.5529 fish meal . . .. .. .. .. . . .. .. 170.6 R 1/874 blood meal.. . . . . .. .. .. .. . . 104.8 P 31 fish meal . . .. . . . . .. .. .. .. 96.0 P 306 fish meal . . .. .. .. .. .. . . .. 58.8 Casein, Liener preparation .. .. .. . . . . . . 59.5 Measurement of the available lysine in protein mixtures-Although the observations des- cribed above indicate that the recovery of lysine and e-TNP-lysine added to protein prepara- tions and subjected to the TNBS reaction can be variable and, from an analytical standpoint, unsatisfactory, the actual values for the available lysine measured in the proteins remained remarkably constant. Nevertheless, it is possible that when proteins are mixed, the value for the combined available lysine may not be equal to the sum of the separate determinations, In order to examine this aspect a wide range of sample suspensions was made to react with TNBS first as single protein preparations and then in a mixture with one other protein prepara- tion. The combinations were of blood meal, casein, crystalline egg ovalbumin, denatured egg albumen, fish meal, wheat gluten, meat meal, and zein.In each sample the value deter- mined €or the available lysine of the single protein was assumed to be valid for the protein mixture and the available lysine of the mixture calculated as the “recovery” of the aggregate of the available lysine of the separate proteins. For all the reactions, 2.5 mg of each protein was taken, except with zein, when 5.0 mg were used. The procedure was then followed as described in the method. From twenty reactions the mean recovery of available lysine of one protein from another was 96.1 per cent. with a range of 88.9 to 103.2 per cent. The results show that the “available lysine” in each protein preparation was measured to the same extent whether separately or in a mixture and the low “recovery” of lysine, such as occurred when either lysine or E-TNP-lysine was added to separate protein sus- pensions, was not experienced.EXAMINATION OF SAMPLES- The results of the determinations of available lysine in animal protein concentrates by the method described above were compared with those measured by the Carpenter procedure* (Table VIII). It can be seen that with few exceptions there is close agreement. The regression equation for calculating the TNBS figure, y , from the l-fluoro-2,4-dinitrobenzene value, x, is y = 0.823 + 0.857~. The correlation coefficient, Y, for the twenty-six results is 0-969 (r2 = 0.939).September, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 683 TABLE VIII AVAILABLE LYSINE VALUES OF PROTEINS AND PROTEIN CONCENTRATES Available lysine/g per 16 g of N 7 - P Sample Carpenter method TNBS method L.2798 fish meal 66 .. .. .. 6.94 6.83 L.2799 meat meal . . .. .. 4.36 4-46 L.2800 freeze-dried meat meal . . 7.41 I 6-69 M.1418 meat meal . . . . .. 4.02 4.45 M.1419 meat meal . . .. .. 6.59 6.83 M.1420 meat meal . . .. .. 4.03 4.08 M.5511 meat meal . . . . .. 4.54 5-01 M.5515 meat meal . . . . . . 2.19 2.45 M.5516 meat meal . . . . . . 4.77 5.01 M.5517 meat meal . . . . , . 2-81 3.71 M.5518 herring meal . . . . .. 6.11 6.32 M.5519 herring meal . . . . .. 6.09 6.7 1 M.5520 herring meal . . . . . . 8.09 7-08 M.5621 fish meal . . . . . . 7-22 7.01 M.6522 white fish meal .. . . 6.14 5.88 M.5523 white fish meal . . . . 6-66 6.82 M.5524 standard fish meal . . .. 7.48 7.70 M.5526 feather meal . . .. . . 1.46 1.73 M.5527 feather meal . . .. .. 1.97 2.35 M.5529 blood meal . . .. . . 6.96 7.18 M.6340 pure gluten . . .. . . 1.42 1.56 0 J 145 herring meal . . .. .. 7.39 6.91 RI/815 meat meal . . .. * . 4.67 5-18 RI/874 blood meal . . .. .. 7.43 5.81 RI/924 fish meal . . . . .. 6.16 6.46 Mean . . 5.31 5.37 L.3473 herring meal . . .. . . 5.08 5.43 Regression equation y (TNBS value) = 0-823 + 0.857% (Carpenter value) Correlation coefficient Y = 0.969 ra = 0.939 Bovine plasma albumin . . .. - 10.06 Casein (BDH) . . .. . . .. - 8.23 Casein (Hopkin & Williams) . . .. - 9-46 Gelatin (Difco) .. .. . . - 4.63 Gelatin (Gurr) .. .. . . .. - 5.07 Soya bean meal . . .. .. - 5.15 Wheat gluten . . .. . . .. - 1.56 Zein (Sigma) . . .. . . .. - 0.15 Total lysine/g per 16 g of N - 8-74 6.14 - - 1.77 - - 6.10 8.52 The total lysine levels of a few selected samples were determined by a modification of the method of Moore, Spackman and Stein14 involving the use of an automated amino-acid analyser (Table VIII). Values are also shown for the amount of available lysine in some commercial single-protein preparations. No attempt has been made at this stage to obtain a range of figures for these proteins but the levels measured are in agreement with previously published findings for total lysine.2 DISCUSSION The method presented in this paper for the determination of available lysine with trinitro- benzenesulphonic acid in materials composed mainly of animal protein differs in several respects from that of Kakade and Liener.lo It is intended for the rapid routine screening of relatively large numbers of samples. The procedure has therefore been devised to eliminate the necessity for the accurate weighing of a very small sample, to minimise complicated stages and to confine the technique to the use of simple apparatus.By using agar suspensions of the sample, small but convenient amounts can be analysed without the tedium of numerous weighings. The amounts of sample in ten replicate l-ml volumes containing 10 and 20mg of casein, fish meal and meat meal suspensions did not vary by more than k0.2 mg, or k0.1 mg when 0.5-ml aliquots containing 5 mg of sample were taken, tolerances that are acceptable with most balances weighing to four and five decimal places.684 HALL et aZ.: USE OF 2,4,6-TRINITROBENZENESULPHONIC ACID FOR [A?Z@r?ySt, VOI. 98 The need to find an alternative to the hydrolysis of the trinitrophenylated protein at 120 “C in an autoclave was felt to be very important and the findings have shown that a hydrolysis time of 2 hours in a boiling water bath is not less effective. . The results (Table IIj also show that, if deemed suitable, the autoclave can be used for a much shorter time than that adopted by previous workers.10 Similar findings were obtained by using casein and fish meal in these tests. It is also evident that prolonged hydrolysis ( i e . , for 16 hours) as used for dinitrophenylated protein8 is not more effective and also that the conditions normally required for the hydrolysis of unreacted protein for total lysine do not necessarily pertain to the hydrolysis of trinitrophenylated protein (Table 111).From the work of Okuyama and Satake,ll lysine reacts with TNBS to form initially cc,E-diTNP-lysine, which Kotaki and Satakelg showed to be unstable in hot 6 M hydrochloric acid and to be easily converted into E-TNP-lysine, yielding, after 10 hours at 110 “C, 81 per cent. of the E-TNP compound and 17 per cent. of free lysine. The reaction is, therefore, somewhat analogous to that of pure lysine with l-fluoro-2,4-dinitrobenzene except that a,edinitrophenyllysine is relatively stable to prolonged heating with 6 M acid but like a,€-diTNP-lysine it is soluble in ether.Hence, Carpenter used E-dinitrophenyllsyine as an internal standard and applied a recovery factor. The observations reported here indicate that in the case of protein only the epsilon-amino group of the lysine reacts. As Kakade and LienerlO pointed out, it would be a unique situation where lysine was the N-terminal amino-acid in a protein or small peptide when both amino groups would react to form a,ediTNP- lysine. The use of pure lysine as a standard (previously reported by Ousterhout and Wood20) is a radical departure from the use of E-TNP-lysine, or edinitrophenyllysine as in the Carpenter procedure. The fact that the a,E-diTNP compound is formed first with lysine does not in our view vitiate its use as a standard. Indeed, the comparison of absorbance values for various 6-TNP-lysine preparations (Table VI) shows that pure lysine when made to react directly with TNBS under the conditions for a sample can satisfactorily replace that of E-TNP-lysine as a standard.There are, however, reservations concerning the use of a recovery factor and of E-TNP-lysine as an internal standard, in view of the results of the recovery experiments involving- the addition of both lysine and 6-TNP-lysine to protein preparations (Table VII). \ I Contrary to an earlier report,lo in the experiments described here the residual TNBS was not completely extracted into ether and consequently the blanks for samples and the reagent blank had considerable absorbances at 346 nm (the near ultraviolet region) but at 415 nm the absorbances were acceptably low.In applying dinitrophenylating reactions to the determination of available lysine in animal feeds, procedures that involve column chromatography for separating lysine from interfering substances have been described.15-17 With the Carpenter method such inter- ference was thought to be largely due to arginine. Investigations in the present work show clearly that the influence of most other common amino-acids (notable exceptions being cadaverine, hydroxylysine, ornithine and the amino-sugars galact osamine and glucosamine) is very small individually, although the cumulative effect could theoretically be significant and become important if much of the protein in these samples was already hydrolysed. From the reactions obtained with a mixture of amino-acids (Table V), an increase of about 10 per cent.in the available lysine value is possible if all the amino-acids are in the “free” state in a “protein” that contains 8 per cent. of lysine. In practice, the reactions with free amino- acids in a sample must be very limited as indicated by the available lysine measured in zein, which gave the expected low value of 0.15 g per 16 g of nitrogen. As with l-fluoro-2,4-dinitrobenzene, the reaction of lysine with TNBS is seriously affected by the presence of carbohydrates, partly because lysine is rapidly coupled with aldose groups, thus making the terminal amino linkage unavailable to form e-TNP-lysine, and partly because of the ease with which carbohydrates in any form are “caramelised” by hydrochloric acid, even at 100 “C.It imparts added colour (which absorbs strongly at 415 nm) to the solution to be measured spectrophotometrically and also causes adsorption of 6-TNP-lysine on to the fine carbon particles formed. This carbon can be easily seen at the ether interface during the extraction process. The method cannot therefore, at this stage, be claimed to be generally applicable to carbohydrate-rich materials although we have been investigating such a possibility. This “caramelisation” interferes in two ways.September, 19731 THE DETERMINATION OF AVAILABLE LYSINE IN PROTEIN CONCENTRATES 685 The values obtained for available lysine in samples of commercially produced proteins (Table VIII) are comparable with published findings for total lysine in similar materials2 Overall, slightly higher levels of available lysine were measured in animal protein concentrates than by the Carpenter procedure.The fineness of the ball-milled preparations together with the sampling technique may have contributed to these higher average figures. When deter- minations were attempted by using much larger, weighed amounts of sample, which had been ground to pass only 1.0 and 06mm or 60-mesh sieves, extremely poor replication was experienced, whereas the reproducibility of the proposed procedure was excellent when using the fine suspensions in agar solution. In our view the results shown in Tables I, I1 and I11 amply illustrate this aspect as the final absorbance value due to E-TNP-lysine was little influ- enced by the factors being investigated. From the few results obtained for total lysine (Table VIII) the percentage availability for those samples is of the same order as that given in the most recently compiled tables issued by the Food and Agriculture Organisation of the United Nations,21 i.e., 93 per cent.for fish meals and 86 per cent. for meat meals. Booth,l* who also included total lysine levels in his results, found a range of 85 to 99 per cent. availability for undamaged animal proteins and 64 per cent. for commercial blood meal, a value close to our figure of 68 per cent. The assessment of available lysine in a protein is always empirical, even when based on chick growth assays.22 Chemical techniques such as that described here, which depend on the reaction of the amino-acid preferentially and, it is hoped, to the exclusion of other substances, tend to have the common inherent disadvantage that the value may still be in error since other compounds may combine with TNBS and remain in the aqueous phase after the ether extraction.Examination by thin-layer ~hromatography~~ of the yellow aqueous acidic solution remaining after ether extraction and containing presumably only E-TNP-lysine has failed to demonstrate the presence of other trinitrophenylated amino-acids or other compounds that could be determined as lysine, and the direct reactions with all of the common amino-acids have shown that they probably do not constitute a serious source of error (Table IV). The reaction with zein supports the view that interference from the other amino-acids present in proteins is likely to be very small.Nevertheless, some exceptions were found, including cadaverine, hydroxylysine and ornithine. The absorbance of the E-TNP product of hydroxylysine is about half that of lysine per unit mass, and could give misleading results if the former was present in substantial amounts. Although necessary for certain micro-organisms2* hydroxylysine has little, if any, nutri- tional value for monogastric or polygastric animals.25 Little information is available on levels of hydroxylysine in products comprised of animal protein but it is known to be a constituent of bone, collagen and connective tissue. 26 Cadaverine, ornithine and putrescine were tested for their reaction with TNBS because they are products of the putrefaction of animal tissues.Significantly, cadaverine has been isolated from stale fish,2’ but no figures have been found in the literature for levels in animal feeds. From our observations with these compounds it seems unlikely that putrescine constitutes a source of error, but cadaverine and ornithine must be regarded as possible constituents of fish meals, meat meals, and meat and bone meals prepared from carcases that might have undergone some decomposition. Two other compounds that must also be considered are the amino-sugars galactosamine and glucosamine, which are associated with the glycoproteins found in cartilagenous tissues. These amines give TNP products that remain in the aqueous phase after ether extraction, and have absorbance values per unit mass similar to E-TNP-lysine. Holsinger, Posati and Pollansch28 first drew attention to the possibility that these amino-sugars were responsible for anomalously high available lysine values for caseins; in some of their examples the amounts measured by the method of Kakade and Liener were actually higher than the total lysine levels. It should also be mentioned that Moodie, Marshall and Kie~wetter~~ have reported the use of TNBS to be unsatisfactory for determining available lysine. Their TNBS “available lysine” values (absorbances were measured in the ultraviolet region) were generally much lower than corresponding Carpenter values. However, in the work described here, in which the spectrophotometric readings were taken at 415nm, the close agreement of the figures for available lysine by the TNBS reaction under our conditions with those obtained by the now classical Carpenter method (Table VIII) is encouraging and indicates that the material determined is predominantly lysine with perhaps some hydroxylysine.Nevertheless, there is the possibility that in certain circumstances false high levels due to cadaverine and ornithine686 HALL, TRINDER AND GIVENS might be obtained in samples of tissues that have undergone some decomposition. It seems to us that further work could well be carried out to determine how important these compounds are in the assay of the true available lysine content of an animal protein. We are indebted to Dr. V. H. Booth and Dr. K. J. Carpenter, Department of Applied Biology, Cambridge University, and to Mr. N. Edge and Mr.H. Pritchard (Public Analysts) of Birkenhead, for samples that had been previously analysed by the l-fluoro-2,4-dinitro- benzene method and for their helpful comments. We are grateful to Mr. L. R. Flynn for technical help in the preliminary investigations of this study, and to Professor I. E. Liener for the generous gift of E-TNP-lysine. We thank Mr. J. H. Blackford, Agricultural Develop- ment and Advisory Service, Wye, Kent, for carrying out the determinations of total lysine by using an automated amino-acid analyser. 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. 27. 28. 29. REFERENCES Gutteridge, D. G. A., in Morgan, J. T., and Lewis, D., Editors, “Nutrition of Pigs and Poultry,” Harvey, D., Tech. Commun. Commonw. Bur. Anim. Nutr., 1956, No. 19. Carpenter, K. J., Ellinger, G. M., Munro, M. I., and Kolfe, E. J., Br. J . Nutr., 1957, 11, 162. Lea, C. H., Parr, L. J., and Carpenter, K. J., Ibid., 1960, 14, 91. Bruno, D., and Carpenter, K. J., Biochem. J . , 1957, 67, 13P. Carpenter, K. J., and Ellinger, G. M., Poult. Sci., 1955, 34, 1451. Sanger, F., Biochem. J., 1945, 39, 507. Carpenter, K. J., Ibid., 1960, 77, 604. Ratney, R. S., Godshalk, M. F., Joice, J., and James, K. W., Analyt. Biochem., 1967, 19, 357. Kakade, M. L., and Liener, I. E., Ibid., 1969, 27, 273. Okuyama, T., and Satake, K., J . Biochem., Tokyo, 1960, 47, 454. Satake, K., Okuyama, T., Ohashi, M., and Shinoda, T., Ibid., 1960, 47, 654. Hall, R. J., Analyst, 1968, 93, 461. Moore, S., Spackman, D. H., and Stein, W. H., Analyt. Chem., 1958, 30, 1185. Roach, A. G., Sanderson, P., and Williams, D. R., J . Sci. Fd Agric., 1967, 18, 274. Matheson, N. A., Ibid., 1968, 19, 492. Ostrowski, H., Jones, A. S., and Cadenhead, A., Ibid., 1970, 21, 103. Booth, V. H., Ibid., 1971, 22, 658. Kotaki, A., and Satake, K., J . Biochem., Tokyo, 1964, 56, 299. Ousterhout, L. E., and Wood, E. M., Poult. Sci., 1970, 49, 1423. Food and Agriculture Organisation, “Available Amino Acid Content of Fish Meals,” F.A .O. Fish Miller, E. L., Carpenter, K. J., Morgan, C. B., and Boyne, A. W., BY. J . Nutr., 1965, 19, 249. Nitecki, D. E., Stoltenberg, I. M., and Goodman, J. W., Analyt. Biochem., 1967, 19, 344. Hartley, A. W., Ward, L. D., and Carpenter, K. J., Analyst, 1965, 90, 600. Lindstedt, S., Acta Physiol. Scand., 1953, 27, 377. Piez, K. A., Science, N.Y., 1961, 134, 841. Saito, K., and Sameshima, M., J . Agric. Chem. SOC. Japan, 1956, 30, 535. Holsinger, V. H., Posati, L. P., and Pallansch, M. J., J . Dairy Sci., 1970, 53, 1638. Moodie, I. M., Marshall, B. C., and Kieswetter, G. R., Fish. Ind. Res. Inst. S. Afr., 24th Annual Received February 12th, 1973 Accepted May 18th, 1973 Buttenvorths Scientific Publications, London, 1962. Rep. No. 92, 1970, p. 17. Report, 1970, p. 23.

ISSN:0003-2654    

DOI:10.1039/AN9739800673

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

Analyst, September, 1973, Vol. 98, jq5. 687-691 687 Quantitative Determination of the Enantiomeric Purity of Synthetic Pyrethroids- Part I. The Chrysanthemic Acid Moiety BY F. E. RICKETT (The Cooper Technical Bureau, Berkhamsted, Hertfordshire) Chrysanthemic acid is liberated by hydrolysis of the pyrethroid under basic conditions and is made to react with (+)-a-methylbenzy:aniine via the acid chloride. Complete separation of diastereoisomeric amides derived from the (+) and (-) forms of both cis- and trans-chrysanthemic acids can be obtained by gas chromatography on a 50-m x 0.25-mm glass capillary column coated with FFAP (free fatty acid phase). No racemisation or kinetic resolution takes place during the reaction sequences. The method is suitable for the analysis of bioresmethrin, bioallethrin, tetramethrin and other insecticidal chrysanthemates. MOST synthetic pyrethroids in commercial use at the present time are based on chrysanthemic acid, which occurs in four isomeric forms : (+)-cis, (-)-cis, (+)-trans and (-)-trans.Hitherto, the insecticides were used as racemic mixtures of the cis- and trans-chrysanthemates (in the approximate ratio of 25: 75) but increasing use is now being made of esters from the (+)-trans form of chrysanthemic acid,lS2 thus necessitating the development of analytical methods for determining both geometrical and optical purity. The determination of the cis to trans ratio can be performed by gas chromatography3 but the determination of optical purity is more difficult. The optical purity of enantiomeric compounds can be determined by forming derivatives with optically active reagents and analysing the resulting mixture of diastereoisomers by gas chr~matography~,~ or nuclear magnetic resonance spectrometry.6 A recent publication' has described a gas - liquid chromatographic method for determining the optical purity of chrysanthemic acid by esterifying it with (-)-menthol or (-)-borneol.Adequate separation of the diastereoisomers formed from the (+)-trans and (-)-trans acids was obtained but without any separation of the corresponding cis acids. In this work, a similar method is described in which (+)-cc-methylbenzylamine is used as the derivative-forming agent ; gas-chromatographic conditions have been established for the complete separation of the diastereoisomeric amides from all four isomers of chrysanthemic acid.EXPERIMENTAL REAGENTS- Middlesex.) (+)-or-MethyZbenxylamine-[cc]~' + 39" (neat) (Ralph N. Emanuel Limited, Wembley, EthyZenediamine-Laboratory-reagent grade, redistilled prior to use. HydrochZoric acid-Analytical-reagent grade. Sodium hydroxide-Analytical-reagent grade. All other reagents and solvents were laboratory-reagent grade. Hydrogen chloride-Technical grade. This was obtained from a cylinder supplied by British Oxygen Co. Ltd., and was dried by bubbling it through concentrated sulphuric acid. Glass capillary tubing, 50 m x 0-25 mm i.d., was obtained from Scientific Glass Engineer- ing Pty. Limited, London, NW2 3HR. HYDROLYSIS OF THE ESTERS- (a) Bioallethrirc-Bioallethrin (100 mg) was shaken for 2 hours at 25 "C with 5 g of ethyl- enediamine.The mixture was cooled in an ice-bath, 25 ml of distilled water were added and the solution was extracted with two 10-ml portions of toluene. The organic phase was dis- carded and the aqueous phase cooled to 0 "C and acidified by the addition of about 32 ml of (Q SAC and the author.688 RICKETT : QUANTITATIVE DETERMINATION OF THE [Arta@t, Vol. 98 5 N hydrochloric acid. Chrysanthemic acid was extracted with four 15-ml portions of toluene and the combined extracts were then dried over anhydrous magnesium sulphate for 1 hour, filtered and dried over anhydrous calcium sulphate (Drierite) for 18 hours. (It is essential that the solution be thoroughly dried because traces of moisture markedly reduce the yields at the next stage.) (b) Bioresmethrin and tetramethrin-This procedure is based on the A.O.A.C.method for pyrethrins.8 A mixture of 100 mg of the pyrethroid and 12-5 ml of a 2 N solution of sodium hydroxide in methanol containing 10 per cent. of water was refluxed for 2 hours. The meth- anol was removed in vacuo and the residue dissolved in 25 ml of water and extracted as in (a) above. FORMATION OF THE AMIDES- The toluene solutions of chrysanthemic acid were concentrated in vacuo at 35 "C, down to approximately 10 ml, and 0.5 d of oxalyl chloride dissolved in 2 ml of dry toluene was slowly added, with stirring, at room temperature. The mixture was heated, without stirring, at 40 "C for 2 hours, then evaporated in vacuo at 30 "C so as to remove the excess of oxalyl chloride.The residue was dissolved in 10 ml of dry toluene, a solution of dry (+)-a-methyl- benzylamine (72 mg) ~ Z U S dry triethylamine (300 mg) in 5 ml of toluene was added and the mixture refluxed for 2 hours; 25 ml of distilled water were added and the mixture was acidified with a few drops of 2 N hydrochloric acid and then extracted with four 10-ml portions of diethyl ether. The extracts were washed with 3 per cent. sodium hydrogen carbonate solution, dried over calcium chloride and concentrated to 2 ml for gas - liquid chromatographic analysis. PREPARATION AND COATING OF THE CAPILLARY COLUMN- Dry hydrogen chloride was passed through a length of glass capillary tubing, 50 m x 0.25 mm id., which was then sealed at both ends with a microburner and heated at 230 "C for 24 hours.9 The column was opened and purged with nitrogen (inlet pressure 10 p.s.i.) at 120 "C for 2 hours.The coating solution was prepared by dissolving 60 mg of FFAP (free fatty acid phase) in 0.5 ml of freshly distilled dichloromethane. The solution was filtered through a sintered-glass plate and introduced into a coating reservoir (Scientific Glass Engineering, Type CR/2). One end of the column was connected to the reservoir by means of a short piece of 1/16 inch 0.d. stainless-steel tubing and the other end was connected to a second length of glass capillary tubing (10 m x 0.25 mm) , the purpose of which was to prevent the coating film from rupturing owing to changes in flow-rate as the thread emerged from the main column. A pressure of approximately 1 p s i .was introduced by a pressure regulator and the pressure was adjusted such that the thread of liquid was observed to travel one complete spiral in 8 minutes (equivalent to 1 m in 18.8 minutes). When the coating solution had passed completely through the column, nitrogen was blown through at the coating pressure for 24 hours. The column was conditioned by slowly raising the temperature at the rate of 4 "C min-1 to 240 "C with a nitrogen inlet pressure of 14 p.s.i. The gas - liquid chromatographic conditions used were as follows. Instrument . . .. . . Pye Series 104 gas chromatograph fitted with a flame-ionisation Column . . .. . . 50 m x 0-25 mm glass capillary coated with FFAP Carrier gas . . .. . . Nitrogen, inlet pressure 14 p s i . Column temperature .. 209 O C Injector temperature . . 230 "C Inlet splitter .. . . S.G.E., Type 1s-2. Approximate split ratio, 100: 1 Volume injected . . Attenuation . . .. .. 50 x 1 Recorder . . .. . . Beckman, Model 1005, 10-inch linear recorder fitted with a "Disc" integrator Retention times . . . . Amides prepared from the following acids: (-)-cis, 23.4 minutes; (+)-cis, 24-5 minutes; (+)-trans, 26.8 minutes; and (-)-trans, 27.5 minutes detector and an injection point heater . . 1 pl of 3 to 5 per cent. solution of amidesSeptember, 19731 ENANTIOMERIC PURITY OF SYNTHETIC PYRETHROIDS. PART I 689 RESULTS AND DISCUSSION The basic requirement of this method of analysis is that the ratio of the diastereoisomers obtained should be equal to the ratio of the original enantiomers, which will be true only if three conditions are satisfied. (i) No racemisation or kinetic resolution must take place during the reaction sequences.(ii) The diastereoisomers formed must be stable under the conditions of analysis used. (iii) The derivative-forming agent must be available in a state of high optical purity. Formation of the amide derivatives was carried out in three stages: hydrolysis of the pyrethroid to liberate chrysanthemic acid, formation of chrysanthemic acid chloride and reaction of the latter with (+)-a-methylbenzylamine. Hydrolysis of chrysanthemic esters under basic conditions proceeds with retention of the configuration of the acid.10 The methods of hydrolysis used were based on the standard A.O.A.C. methods for the analysis of these or similar materials ; thus bioallethrin [ (+)-trans-isomer of allethrin] was treated with ethylene- diamine at 25 "C for 2 hours,ll while bioresmethrin [(+)-trans-isomer of resmethrin] and tetramethrin were hydrolysed with a methanolic solution of sodium hydroxide.8 In each instance the liberation of chrysanthemic acid was quantitative, which was verified by the standard titration procedure.* It is known that epimerisation of cis-chrysanthemic acid to the trans-isomer, with accom- panying inversion of rotation, can occur on formation of the acid chloride unless very mild conditions are used.12J3 The treatment of chrysanthemic acid with thionyl chloride in the presence of pyridine at room temperature is satisfactory in this respect,' although as a routine method it suffers from the disadvantage that the thionyl chloride should be regularly purified.14 Oxalyl chloride has proved useful for the preparation of acid chlorides that may isomerise in the presence of thionyl chloridel5 and excellent yields of chrysanthemoyl chloride were obtained by using this reagent under very mild conditions.In Table I the cis to trans ratios of various pyrethroid samples before and after conversion into the N-(a-methylbenzy1)- chrysanthemamides by this method are compared. Average yields were about 80 per cent. The mixture of diastereoisomeric amides was analysed on a 50-m x 0-25-mm glass capillary column coated with FFAP. Sample TABLE I COMPARISON OF cis TO trans RATIOS BEFORE AND AFTER FORMATION OF THE AMIDES Original pyrethroid* - 1 cis-Isomer, trans-Isomer, cis-Isomer, trans-Isomer, per cent.per cent. per cent. per cent. After formation of amidest Bioallethrin .. .. 1.1 98.9 0.85 99-2 Bioresmethrin . . .. 1-7 98.3 1.3 98-7 Tetramethrin . . . , 18.0 82.0 17-7 82-3 NRDC 119: .. . . 96.0 4.0 96.7 3.3 * Analysed by gas - liquid chromatography under conditions similar to those used by Murano, t Analysed on capillary column under conditions described in Experimental section. $ 5-Benzyl-3-furylmethyl- ( + )-cis-chrysanthemate. Fujiwara, Horiba and Miyam~to.~ The inner surface of the column was conditioned prior to coating it by treatment with dry hydrogen ~hloride.~ The coating was performed by the dynamic method following the procedure used for this stationary phase by Goretti and Liberti.lG A column efficiency of 204 000 theoretical plates was obtained, measured from the (-)-trans-chrysanthemamide peak (Fig.1, peak d). All four diastereoisomers were completely separated as shown in Fig. 1. The elution order was (-)-cis, (+)-cis, (+)-trans and finally (-)-trans. To verify that no racemisation or kinetic resolution took place during the formation and analysis of the mixed amides, mixtures were prepared from weighed amounts of (+)-trans- chrysanthemic acid and racemic cis - tram-chrysanthemic acid. The expected ratios of (+)-trans- to (-)-trans-isomers (after allowing for the cis-isomer content of the racemic sample) are compared in Table I1 with the ratios found by analysis. Each sample was690 RICKETT : QUANTITATIVE DETERMINATION OF THE [Analyst, VOl. 98 c d 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 Time/minutes Fig.1. Amide mixture analysed on 50-m x 0.25-mm glass capillary column coated with FFAP (for conditions see text). The lower trace is the Disc integrator output. Assign- ments, amides prepared from: a, (-)-cis acid; b, (+)-cis acid; C, (+)-trans acid; and d, (-)-trans acid analysed in duplicate and the gas - liquid chromatographic peaks were measured by using a “Disc” integrator. Errors were found from duplicate analyses to be &l per cent. TABLE I1 COMPARISON OF EXPECTED AND FOUND RATIOS OF (+)- AND ( -)-~Y~TZS-CHRYSANTHEMIC ACID Expected -----7 (+)-trans- (-)-trans- Sample Isomer Isomer (+) -tmns-Chrysanthemic acid (from natural pyrethrins) . . .. 100 0 Mixture 1 .. .. .. 83.6 16.4 Mixture 2 .. .. ..76.4 23.6 Mixture 3 . . .. .. 68-6 31-4 Racemic chrysanthemic acid . . 50 50 Found & Isomer Isomer ( +) -trans- ( - )-trans- 99.1 0.9 84-0 16.0 76.9 23.1 69.1 30.9 50.1 49.9 As mentioned earlier, a prerequisite for accurate analyses is high optical purity of the derivative-forming agent. The purity of the (+)-cc-methylbenzylamine used for these experi- ments was checked by allowing it to react with pure (+)-trans-chrysanthemic acid obtained by hydrolysis of natural pyrethrins. The results (Table 11) showed an apparent (-)-trans content of 0.9 per cent. produced by the optical impurity in the reagent [(-)-a-methyl- benzylamine] .September, 19733 ENANTIOMERIC PURITY OF SYNTHETIC PYRETHROIDS. PART I 691 Although difficult to set up, the capillary column is easy to operate and gives long service; more than one hundred analyses have already been performed over a period of 4 months without any noticeable reduction in performance.The author is grateful to Mr. C. J. Lord, who determined the cis to trans ratios of the pyrethroid samples used in this work, and to Mr. S. J. Baker for technical assistance. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. REFERENCES Davies, M., Proc. Conf. 3rd BY. Pest Control, St. Helier, Jersey, 1971. Chadwick, P. K., Pestic. Sci., 1971, 2, 161. Murano, A., Fujiwara, S., Horiba, M., and Miyamoto, J., Agric. Biol. CJzern., 1971, 35, 1200. Gil-Av, E., Charles-Sigler, R., Fischer, G., and Nurok, D., J . Gas Chromat., 1966, 4, 51. Westley, J. W., and Halpern, B., Chevn. Communs, 1966, 35. Cockran, T. G., and Huitric, A. C., J . Org. Chem., 1971, 36, 3046. Murano, A., Agric. Biol. Chem., 1972, 36, 917. “Official Methods of Analysis of the Association of Official Analytical Chemists,” Eleventh Edition, Association of Official Analytical Chemists, Washington, D.C., 1970, p. 88. Novotny, M., and Tesarik, K., Chromatographia, 1968, 1, 332. Elliott, M., Chem. & Ind., 1960, 1142. “Official Methods of Analysis of the Association of Official Analytical Chemists,” Eleventh Edition, Elliott, M., Janes, N. F., and Jeffs, K. A., Pestic. Sci., 1970, 1, 49. Scizuki, Y., Hirai, H., Toyoura, A., and Majara, O., German Patent 2,003,065, 1970; Chem. Abstr., Rigby, W., Chem. & Ind., 1969, 1508. Fieser, L. F., and Fieser, M., “Reagents for Organic Synthesis,” Volume 1, John Wiley & Sons Goretti, G., and Liberti, A., J . Chromat., 1971, 61, 334. Received January loth, 1973 Accepted March 28th, 1973 Association of Official Analytical Chemists, Washington, D.C., 1970, p. 101. 1970, 73, 76735. Inc., New York, 1967, p. 767.

ISSN:0003-2654    

DOI:10.1039/AN9739800687

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

692 Analyst, September, 1973, Vol. 98, p$. 692-693 Carbon Tetrachloride as a Possible Source of Interference During Fumigant Residue Analysis BY P. B. BAKER, J. E. FARROW AND R. A. HOODLESS (Department of Trade and Industry, Laboratory of the Government Chemist, Covnwall House, Stamford Street, London, SEI 9NQ) The contamination of acetone with an impurity corresponding to carbon tetrachloride on two different gas-chromatographic columns is described. It has been shown that this contamination could be caused by carbon tetra- chloride in the laboratory atmosphere, which possibly arises from the use of aerosol propellent cans for spraying thin-layer chromatograms. DURING the determination of residues of the fumigant carbon tetrachloride in grain samples by gas - liquid chromatography with electron-capture detection, impurities have been found in a variety of solvents.In particular, acetone contained a contaminant that gave a peak with a similar retention time to that of carbon tetrachloride on two columns. This impurity did not appear to be removed by fractional distillation and was not one of the normally expected impurities (ethanol, diacetone alcohol or mesityl oxide). It therefore seemed that carbon tetrachloride was present, either in the solvent as supplied, or in the laboratory atmos- phere. Even if carbon tetrachloride occurred in trace amounts in the atmosphere, these amounts would be sufficient to contaminate the solvents to the extent observed. It was found that acetone left in an open beaker in the laboratory rapidly became contaminated (up to 10 ng ml-l in 4 hours), but samples left exposed in non-laboratory surroundings (the authors' homes) were not contaminated.Further examination of the atmosphere, conducted by drawing air through acetone at -78 "C, also suggested the presence of carbon tetrachloride. The concentration varied from day to day from 10 to 100 pg m-3, which is considerably lower than the generally recognised threshold limit value (65 mg m-3).1 Aerosol propellent cans that are used for spraying thin- layer chromatograms were found to be a possible source of this atmospheric carbon tetra- chloride as the propellent is usually a mixture of chlorofluoroalkanes made from carbon tetrachloride. By using the approximate method described below, the carbon tetrachloride content of the propellent was found to be of the order of 0.1 to 0.2 mg kg-l, higher concentra- tions being observed when the can was almost empty (up to 0.7 mg kg-1). It therefore appears that trace amounts of carbon tetrachloride are present in the laboratory atmosphere, possibly arising from the use of aerosol propellent cans.Although the concentration is low, it may be sufficient to contaminate solvents. Workers should be aware of this possible interference when examining volatile compounds by means of gas- liquid chromatography with electron-capture detection. EXPERIMENTAL CARBON TETRACHLORIDE IN THE ATMOSPHERE- Air from the laboratory was drawn slowly by a small vacuum pump, first through a glass U-tube (to remove water vapour), and then through a sintered-glass frit of porosity 1 into a Drechsel bottle containing acetone (10 ml).The U-tube and Drechsel bottle were contained in an acetone - carbon dioxide bath a t -78 "C. The flow of air was metered. After a suitable volume of air had been drawn through the acetone, the solution was examined by gas - liquid chromatography. CARBON TETRACHLORIDE IN AEROSOL PROPELLENT CANS- were collected in an open-ended Kuderna-Danish evaporator. diluted and examined by gas - liquid chromatography. Acetone was sprayed in the normal manner (as if spraying a chromatogram) and 10 ml This solution was suitably @ SAC; Crown Copyright Reserved.BAKER, FARROW AND HOODLESS 093 GAS - LIQUID CHROMATOGRAPHY- Column l-A 4 m x 3 mm i.d. stainless-steel column was filled with 15 per cent.poly- propylene glycol on Chromosorb W (80 to 100 mesh) and used in a Varian 1400 gas chromato- graph fitted with an electron-capture detector. The carrier gas was nitrogen (inlet pressure 10 p s i ) , the injection port temperature was 130 "C, the column temperature 70 "C and the detector temperature 140 "C. The retention time for carbon tetrachloride under these conditions was approximately 7 minutes. Column 2-This column was used to confirm the presence of carbon tetrachloride. A 4 m x 3 mm i.d. stainless-steel column was filled with Porapak Q (50 to 80 mesh). The injection port, oven and electron-capture detector were maintained a t 160 "C. The carrier gas was nitrogen (inlet pressure 30 p s i . ) . Under these conditions the retention time for carbon tetrachloride was approximately 64 minutes. The authors thank the Government Chemist for permission to publish this paper. REFERENCE 1. "Threshold Limit Values for 1971," Technical Data Note 2/71, Department of Employment, Received March Sth, 1973 Accepted April 25th, 1973 H.M. Factory Inspectorate, London, 1971.

ISSN:0003-2654    

DOI:10.1039/AN9739800692

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

694 Analyst, September, 1973, Vol. 98, p . 694 The Use of a Laser for Cutting Bone Samples Prior to Chemical Analysis BY J. S. HISLOP AND A. PARKER (Analytical Sciences Division, Atomic Energy Research Establishment. Harwell, Didcot, Berkshire) A RECURRENT problem in the determination of trace elements in animal tissue is the prevention of contamination during sampling, particularly for hard tissue, such as bone, when the abrasion caused by any cutting instrument that is used may be significant. During a study in conjunction with the Social Medicine Unit of the Medical Research Council of the levels of certain trace elements in human rib, we have made use of a novel cutting technique, which enables contamination to be kept to the minimum. EXPERIMENTAL Rib samples, mounted on a Perspex trolley, were drawn at a constant speed through a vertical 330-W beam from the Culham axial-flow carbon dioxide (10.6 pm) laser.The samples and the moving trolley were enclosed in argon in a sealed, fibre-glass glove-box, which enabled the cutting process to take place under strictly controlled conditions. The laser beam entered a side-arm of the glove-box through a potassium chloride window and, after passage through the sample , was dissipated in a block of spectroscopic-grade graphite. Care was required in order to minimise the amount of soot deposited on the potassium chloride, as this deposit caused absorption of the laser beam with subsequent heating and damage to the window. A 4-inch focal length germanium lens was used to focus the beam and, for samples approximately 1 cm thick, a cut less than 1 mm wide was produced in the bone.The average time taken to cut a human rib sample was less than 3 s. Although the slicing action of the laser is due to rapid volatilisation of the bone matrix, the low thermal conductivity of this material restricts the heat damage to a narrow region parallel to the laser cut. Within this damaged region, the possibility of the loss of certain trace elements by volatilisation must be considered but , by visual examination of the samples used in our work, this region was seen to represent a very small proportion of the total sample mass used for analysis. The extent of the damaged layer produced in bone by using a high- power laser does not appear to have been extensively studied192 and until such time as detailed information on this subject becomes available the technique would not be recommended for cutting narrow slices of bone for subsequent analysis.The use of a laser has been found to be extremely convenient and rapid for cutting both human and animal ribs without introducing contamination from metallic elements and has the added advantage that the sample need not be firmly held during the process. The technique should be applicable to a variety of soft and hard biological tissues. Experience in the use of lasers in surgery3$* indicates that, compared with bone, soft tissue can be cut by using much less powerful beams (less than 50 W) and, owing to the high water content of soft tissue, the temperature to which the sample is raised during cutting is much reduced. The authors acknowledge the assistance and co-operation of the Laser Applications Group at the U.K.A.E.A. Culham Laboratory during this work. REFERENCES 1. 2. 3. 4. Goldman, L., and Rockwell, R. J., “Lasers in Medicine,” Gordon and Breach Science Publishers Hogberg, L., Stahle, J., and Vogel, K., Acta SOC. Med. Ufisal., 1967, 72, 223. Hall, R. R., Breach, A. D., Baker, E., and Morison, P. C. A., Nature, Lond., 1971, 232, 131. Goldman, L., Rockwell, R. J., Naprstek, Z., Siler, V. E., Hoefer, R., Hobeika, C., Hishimoto, K., Received March lst, 1973 Accepted March 23rd, 1973 Inc., New York, 1971. Polanyi, T., and Bredmeier, H. C., Ibid., 1970, 228, 1344. @ SAC; Crown Copyright Reserved.

ISSN:0003-2654    

DOI:10.1039/AN9739800694

出版商:RSC    

年代:1973

数据来源: RSC

摘要:

A=aLyst, September, I973 Book Reviews 695 THEORY AND INTERPRETATION OF MAGNETIC RESONANCE SPECTRA. By W. T. DIXON. Pp. viii In 164 pages, the author attempts to present basic physics and quantum mechanics required “to provide the student with enough background to tackle actual problems” (Chapters 1 to 3), and to unify the analysis and interpretation of nuclear magnetic resonance and electron spin resonance spectra (Chapters 4 to 7). This new attempt to squeeze a gallon into a pint pot will inevitably suffer from an ill defined audience-most newcomers to the field will find the introductory chapters too hurried and abstract, whilst skipping to Chapter 4 will yield little more than a dozen pages of readable material. The book might appeal to the one-time expert who has been away from the field long enough to forget, but for the majority of chemists the book will be either too abstract or insufficiently formal.There is little to offer the analytical chemist. Chapters 4 (parameters obtained from magnetic resonance spectra) and 7 (time-dependent effects) might appeal to the more theoretically inclined, but anyone entering the field could find better expositions elsewhere. + 164. London and New York: Plenum Press. 1972. Price $16.50. R. CATTERALL HANDBUCH FUR DAS EISENH~TTENLABORATORIUM. Band 5. Erganzungsband. Pp. 90 (loose-leaf). The four volumes of the “Handbuch” previously published provide the analytical chemist with co-operatively tested methods for the analysis of everything he is likely to encounter in an iron and steel works, from tar to ferro-alloys.The methods are constantly under review by the Chemical Committee of the Vereins Deutscher Eisenhuttenleute, and Volume 2, on the Analysis of Metallic Materials, is already in its second edition. But methods in this field are changing so rapidly that the Committee decided that the only way of keeping them reasonably up-to-date was to issue a supplementary volume in the form of a loose-leaf binder, so that revised versions could be incorporated at any time (British Standards Handbook No. 19 has been planned on the same lines, although it is a t present far less comprehensive than the “Handbuch”). Revised methods in the present volume include, for iron and steel, a potentiometric method for determining chromium, a coulometric method for determining carbon, a method for deter- mining acid-soluble boron content and a colorimetric method for determining 0.01 to 4 per cent.of vanadium. For ores there are methods for determining moisture, phosphorus, silica and small amounts of fluorine. There is also a list of German and I.S.O. Standard Methods and “Euronorms.” Although it is a loose-leaf book, a contents list is provided. Dusseldorf : Verlag Stahleisen m.b.H. 1971. Price DM58. Anyone who uses the “Handbuch” should have this supplement. G. M. HOLMES LABORATORY METHODS IN INFRARED SPECTROSCOPY. Edited by R. G. J. MILLER and B. C. STACE. Second Edition. Pp. xxii + 376. London, New York and Rheine: Keyden & Son Ltd. 1972. Price L6.25; $18; DM56.50. The methods described in this volume, which is the second edition of a widely used handbook on the preparation of solid, liquid and gaseous samples for infrared absorption measurements and on recording the spectra of such samples, are of considerable practical importance; these procedures permit the spectroscopist to obtain trustworthy spectra, suitable for interpretation, of samples as diverse as gas-chromatographic fractions, surface films and insoluble plastics.The size of the handbook has increased from 164 to 375 pages, but as the new edition has been printed by photographic reproduction of I.B.M. typescript, instead of by the conventional printing method used for the first edition, the actual gain in material is even greater. The number of specialist authors has increased from fourteen to twenty-five, of which only seven contributed to the first edition.All the original chapters have been either extended or completely re-written. In addition, there are new chapters on far infrared spectroscopy, quantitative analysis of polymeric materials, industrial plant analysis, Fourier transform spectroscopy, the measurement of spectra of samples under high pressure, the examination of gas-chromatographic fractions, internal reflec- tion spectroscopy, Raman instrumentation and sampling, surface-adsorbed samples, matrix isolation techniques, and corrosive, unstable and explosive samples.696 BOOK REVIEWS [Analyst, Vol. 98 The procedures described, in general, represent the personal experiences and preferences of the individual authors and are fully reliable. However, in the chapter on sample handling tech- niques, the author lists the advantages of the alkali halide disc technique but does not warn the tyro that the procedure is often, particularly when polymorphic substances are being examined, less satisfactory than the simpler Nujol mull method; further, he does not point out that an adequate Nujol mull spectrum of a 1-mg sample of many crystalline organic compounds can be obtained by using a small spatula to grind the sample with a smear of Nujol directly on a small alkali halide plate.Again, bromoform is suggested as a solvent for polymers, but its general utility for the routine solution spectroscopy of corticosteroids and other compounds of pharma- ceutical interest is not mentioned. These are minor points, which do not detract from the general utility of the volume.The criticisms made by the reviewer of the first edition (cf., Analyst, 1966, 91, 295) have largely been met, and I can accordingly extend to this edition his general commendation of the work. However, there are a few misprints, and there is still no index. The volume is, nevertheless, the best handbook on practical infrared spectroscopy at present available and, in view of the large amount of additional information that this edition contains, it should rapidly displace the earlier one as a standard manual in spectroscopic laboratories. J. E. PAGE INDICATORS. Edited by EDMUND BISHOP. International Series of Monographs in Analytical Chemistry, Volume 51. Pp. x + 746. Oxford, New York, Toronto, Sydney and Braun- schweig: Pergamon Press.1972. Price ,512. This outstanding book, which is a most welcome and valuable addition to the reference literature, is in effect a self-contained series of monographs (called chapters in this instance), each of which gives a complete account of a particular group of indicators. Introductory chapters on the history of indicators (E. Rancke-Madsen) and on the theory and principles of visual indicators (E. Bishop) are followed by chapters dealing with-acid - base indicators (E. Banyai) ; indicators for non-aqueous acid - base titrations (J. S. Fritz) ; titrations with non-chelating ligands (R. J . Magee) ; inetallochromic indicators : theory (A. Ringbom) and applications (E. Wanninen) ; adsorption indicators (E. Pungor and E. Schulek) ; oxidation - reduction indicators (J.M. Ottaway and E. Bishop) ; fluorescent indicators (G. F. Kirkbright) ; and chemiluminescent indicators (L. Erdey). There is a detailed table of contents and a most helpful subject index, but no author index; possibly it would be asking too much to expect one to have been provided, as the different chapters present, in total, more than 1500 references. All those concerned with the preparation of this book-series editors, editor, contributors and publisher-are to be congratulated. The vast amount of work involved is reflected in the unusually long gestation period, some 12 years, that appears to have been required. In his Preface, dated 1969, the editor reveals that delays occurred to such an extent that some authors had to make extensive revisions of their original texts (some dated 1962 and others 1963) in order to bring them up to date. A subsequent editorial note, dated March, 1972, further acknowledges this fact and explains that the "last first version" was received in 1967, by which time two chapters were so outmoded that they had to be completely re-written. Our thanks must go most warmly to Eddie Bishop for the patience, dedication and determination that he has shown in finally getting this collection of authoritative articles into print. It appears to be desirable to draw attention to the fact that, at a cost of ,512, this book offers outstandingly good value today. It is highly unfortunate that a selling price of ,517.50 has been quoted in some preliminary advertisements and mailing leaflets, as a result of what appears to have been merely a misunderstanding between different departments of the publishing house concerned. If you have already paid the higher price, why not try asking for some of your money back ? D. M. W. ANDERSON

ISSN:0003-2654    

DOI:10.1039/AN9739800695

出版商:RSC    

年代:1973

数据来源: RSC