A.naZyst, April, 1973, Vol. 98, @. 257-267 257 The Identification and Semi-quantitative Assay of Some Fat-soluble Vitamins and Antioxidants in Pharmaceutical Products and Animal Feeds by Thin-layer Chromatography BY G. W. JOHNSON AND C. VICKERS (Analytical Research, Quality Control, The Boots Company Ltd., Pennyfoot Street, Nottingham) A thin-layer chromatographic method for the identification and semi- quantitative assay of vitamin A (alcohol), its acetate and palmitate, vitamin D, a-tocopherol, a-tocopheryl acetate, BHA (butylated hydroxyanisole ; 2-t-butyl- 4-methoxyphenol), BHT (butylated hydroxytoluene ; 2,6-di-t-butyl-4-methyl- phenol) and ethoxyquin in vitamin preparations is described. The sample solutions are applied to thin layers of silica gel and the vitamins and anti- oxidants are separated by using n-hexane - ethyl methyl ketone - di-n-butyl ether (34 + 7 + 6) as the developing solvent.Decomposition of vitamins A and D when applied to the adsorbent layer is inhibited by the presence of triethylamine in the spotting solvent. The compounds are identified by means of their RB- values, their appearance in ultraviolet radiation of wavelengths 254 and 360 nm and their response to iron(II1) chloride - bipyridyl and iron(II1) chloride - potassium hexacyanoferrate(II1) spray reagents ; they are assayed by visual comparison with standards. The method has been applied to gelatin-protected vitamin beads, animal feed additives, multi- vitamin tablets, oily vitamin concentrates and halibut-liver oil samples. A simple colour test for distinguishing vitamin D, from vitamin D, after removal of vitamin A and its esters is also described.VITAR~IN A (retinol) and its esters, vitamin D, (ergocalciferol), vitamin D, (cholecalciferol) and vitamin E (a-tocopherol) are usually added to vitamin preparations either in the form of gelatin-protected beads, in which the vitamins are dispersed in a starch-coated matrix of gelatin and sugar, or in the form of concentrated solutions in oil. In both of these con- centrates, the vitamins are stabilised by the presence of antioxidants, such as a-tocopherol, 2-t-butyl-4-methoxyphenol (butylated hydroxyanisole, BHA), 2,6-di-t-butyl-4-methylphenol (butylated hydroxytoluene, BHT) and 1,2-dihydro-6-ethoxy-2,2,4-trimethylquinoline (ethoxy- quin). Rapid and reliable methods for the examination of the vitamin concentrates and their formulated products were required, and the use of thin-layer chromatography for this purpose was investigated. Many thin-layer chromatographic methods for the separation, identification and assay of vitamins A, D and E have been reported; these methods have been reviewed by Bolligerl and Strohecker and Henning2 The main difficulty in the assay of vitamins A and D by thin-layer chromatography arises from the instability of the vitamins on dry chromatographic adsorbents.Bolliger and Konig3 described a method in which the decomposition of vitamin D on silica gel layers was minimised by developing the chromatogram immediately after applying the vitamin extract (i.e., while the adsorbent was still saturated with the spotting solvent).Hanewald, Mulder and Keuning4 reduced the extent of decomposition of vitamin D by adding squalane to the vitamin solution and BHT to the developing solvent, while Ponchon and Fellers5 used a method in which the vitamin D solution was applied to the adsorbent layer in an atmosphere of nitrogen. Attempts to apply the published methods were often unsuccessful. None of the chromato- graphic systems gave a satisfactory separation of all the vitamins and antioxidants, and some decomposition of the vitamin A took place when it was applied to the adsorbent layers, even though the recommended precautions had been taken. The procedures described in this paper enable the vitamins and antioxidants to be applied to silica gel layers without decomposition and to be separated, identified and assayed in a single chromatographic system.0 SAC and the authors.258 JOHNSON AND VICKERS : THIN-LAYER CHROMATOGRAPHIC IDENTIFICATION [Analyst, VOl. 98 When administered to chicks,6 vitamin D, has only 1 or 2 per cent. of the antirachitic activity of vitamin D, and it is therefore essential to ensure that the vitamin D beads used in poultry feed supplements contain vitamin D, and not vitamin D,. The two vitamins have been differentiated by infrared and nuclear magnetic resonance spectroscopy,2J by thin-layer partition chromatography1 and by the difference in colour of their spots on silica gel plates after spraying with concentrated sulphuric acid.l The instrumental methods require the use of large amounts of the vitamins in a pure form and are therefore unsuitable for the examination of formulations.The partition chromatographic and colour-test procedures require less sample and are less sensitive to the presence of impurities, but in our hands they did not prove very reliable. A simple colour test, which can be applied in the presence of many of the compounds likely to occur in formulations, has therefore been developed. EXPERIMENTAL APPARATUS AND REAGENTS- 0.25-mm adsorbent layer and activate the plates by heating them at 110 "C for 1 hour. 60 and 62 "C. Chromatographic plates-Prepare Kieselgel (Merck) HF,,, plates, 20 x 20 cm, with a Diethyl ether-Anaesthetic ether B.P. Chloroform-Re-distil chloroform B.P., collecting the fraction that distils between Methanol-Analytical-reagent grade.Di-n-butyl ether-Shake 2 litres of di-n-butyl ether with 100 ml of 10 per cent. aqueous sodium metabisulphite solution and allow it to stand for 24 hours. Run off the aqueous layer and wash the di-n-butyl ether with two 50-ml volumes of 1 N sodium hydroxide solution and then with water until the washings are neutral to litmus. Dry the di-n-butyl ether with anhydrous sodium sulphate and apply the peroxide test described for anaesthetic ether in the British Pharmacopoeia 1968. If no peroxides are detected, distil the di-n-butyl ether until its volume has been reduced to about 250 ml, collecting the fraction that distils between 138 and 141 "C. Store the di-n-butyl ether under nitrogen in a cool dark place and check that it is peroxide-free immediately before use.Vitamin and antioxidant standards-Use pure crystalline samples of vitamin A and D with the following activities: vitamins D, and D,, 40 000 000 i.u. g-l; vitamin A (alcohol), 3 300 000 i.u. g-l; vitamin A acetate, 2 900 000 i.u. g l ; and vitamin A palmitate, 1 818 000 i.u. g-l. The BHA and BHT used should comply with the requirements of the British Pharmacopoeia 1968. The other compounds used as standards should be of the highest commercially available quality. Standard solution of vitamins and antioxidants-Dissolve 5 mg each of vitamin D,, vitamin A (alcohol), vitamin A acetate, BHA and ethoxyquin, and 10mg each of a-toco- pherol, BHT, vitamin A palmitate and a-tocopheryl acetate, in cyclohexane - triethylamine (9 + 1) and dilute the solution to 20 ml with the same solvent mixture.Store the solution in a refrigerator. Standard solutions of vitamin A acetate, vitamin A palmitate, vitamin D, BHA, BHT and ethoxyquin-Dissolve 18.75 mg of vitamin D, (= 750 000 i.u.), 41.2 mg of vitamin A palmitate (= 75 000 i.u. of vitamin A), 25.8 mg of vitamin A acetate (= 75 000 i.u. of vitamin A), 25 mg of BHA, 25 mg of ethoxyquin and 50 mg of BHT in 50 ml of cyclohexane - triethylamine (9 + l), and dilute 2, 3,4, 5 and 6 ml of the solution to 10 ml with the same solvent mixture. Store the solutions in a refrigerator. Standard solutions of u-tocopherol and u-tocopheryl acetate-Dissolve 25 mg of a-tocopherol and 25 mg of a-tocopheryl acetate in 25 ml of cyclohexane - triethylamine (9 + 1) and dilute 2, 3,4, 5 and 6 ml of the solution to 10 ml with the same solvent mixture. Store the solutions in a refrigerator.Iron(III) chloride - potassium hexacyanoferrate(III) reagent-Dissolve 1.3 g of iron(II1) chloride (FeCl,.GH,O) in 100 ml of 2 N hydrochloric acid. Dissolve 0.7 g of potassium hexacyanoferrate(II1) in 100 ml of water. Mix equal volumes of the freshly prepared solutions immediately before use. Iron(III) chloride - potassium hexacyanoferrate(III) reagent (strongly acidic)-Mix two volumes of freshly prepared iron(II1) chloride - potassium hexacyanoferrate(II1) reagent with one volume of concentrated hydrochloric acid immediately before use.April, 19731 259 Iron(II1) chloride - bipyridyl reagent-Dissolve 06 g of 2,2'-bipyridyl in 100 ml of ethanol. Dissolve 0-2g of iron(II1) chloride (FeC1,.6H20) in 100ml of ethanol.Mix equal volumes of the freshly prepared solutions immediately before use. Standard vitamin D, solution-Dissolve 5 mg of vitamin D, (= 200 000 i.u.) in 5 ml of chloroform immediately before use. Standard vitamin D, solution-Dissolve 5 mg of vitamin D, (3 200 000 i.u.) in 5 ml of chloroform immediately before use. Developing solvent-n-Hexane - ethyl methyl ketone - di-n-butyl ether (34 + 7 + 6). PROCEDURE FOR THE IDENTIFICATION AND ASSAY OF VITAMINS AND ANTIOXIDANTS (SEE Extraction of vitamins and antioxidants from formulations-For the examination of low- potency vitamin formulations, transfer an amount of sample equivalent to 30 000 i.u. of vitamin D (see Note 2) to a low-actinic glass separator and disperse it in 120 ml of water at a temperature of 70 "C.Cool the mixture to ambient temperature, add 20 ml of methanol and extract the vitamins and antioxidants with 100 ml of diethyl ether, by gently shaking the separator. Allow the mixture to stand until the aqueous and diethyl ether layers have separated, run off the aqueous phase into a second separator and decant the diethyl ether layer, leaving any emulsified material in the first separator. To the residual emulsified material, add 20 ml of methanol, mix and add 100 ml of diethyl ether. Add the mixture to the aqueous phase in the second separator and shake the mixture gently so as to extract the vitamins and antioxidants. Again allow the phases to separate, run off the aqueous layer and decant the diethyl ether phase, adding it to that obtained from the first extraction.Repeat the extraction procedure until no vitamins or antioxidants are detected when the extracts are evaporated to dryness and examined by thin-layer chromatography (three or four extractions should suffice). Wash the bulked diethyl ether extracts by gently shaking them with two successive 20-ml volumes of water. Discard the washings, filter the diethyl ether solution through a cotton-wool plug and evaporate it to dryness under vacuum in a rotary film evaporator with the water-bath at a temperature of 50 "C. Dry the residue by dissolving it in successive 10-ml volumes of acetone and re-evaporating the solutions. If the volume of the residue from the extraction is negligible, dissolve it in exactly 4 ml of cyclohexane - triethylamine (9 + 1).Otherwise, dissolve the residue in a small volume of cyclohexane and transfer the solution to a 10-ml measuring cylinder, washing out the container with further small volumes of cyclohexane. Evaporate the cyclohexane a t 50 "C in a stream of nitrogen, until the volume has been reduced to about 3 ml, then add 0.4 ml of triethylamine and dilute the mixture to 4 ml with cyclohexane. For the examination of vitamin beads and high-potency formulations, carry out a similar procedure, but reduce the volumes of water, diethyl ether and methanol used in the extraction to one quarter of those specified for low-potency vitamin formulations. For the examination of solutions of vitamins in oil, dilute the sample with cyclohexane - triethylamine (9 + 1) until the solution contains 7500 i.u.ml-l of vitamin D and apply it directly to a thin-layer plate. OF VITAMINS AND ANTIOXIDANTS I N PHARMACEUTICALS AND FEEDS NOTE 1)- NOTES- 1. The vitamins should be protected from light a t all times. Excessive heating and exposure to atmospheric oxidation should also be avoided. 2. The amount of sample to be taken for the assay can be calculated by reference to any of the vitamins or antioxidants, depending on the relative amounts of each present. For most of the samples examined, it has been found best to calculate the amount of sample by reference to the vitamin D content. Identijcation of the extracted vitamins and antioxidants-Line a chromatographic tank with filter-paper, pour the developing solvent into the tank, saturating the filter-paper lining, and allow the tank to equilibrate for 30 minutes.Apply 2 pl of the sample solution and 2 p1 of the standard solution of vitamins and antioxidants to each of two chromatographic plates and score the surfaces of the plates 15 cm from the spots. Develop the chromatograms in the dark until the solvent fronts reach the scored lines, remove the plates from the tank and allow them to stand in a stream of cold air until most of the solvent has evaporated. Without delay, spray one of the plates with the strongly acidic iron(II1) chloride - potassium hexacyanoferrate(II1) reagent and heat it in an oven at 40 "C for 15 minutes.260 JOHNSON AND VICKERS : THIN-LAYER CHROMATOGRAPHIC IDENTIFICATION [Analyst, vol. 98 Inspect the second plate successively in ultraviolet radiation of wavelengths 254 and 360 nm; then spray the plate with iron(II1) chloride - bipyridyl reagent and allow it to stand in the dark until the spots reach maximum intensity.Identify the vitamins and antioxidants by comparing the sample and standard chromatograms. Semi-q.uantitative assay of vitamins and antioxidants-By diluting suitable volumes of the sample solution with cyclohexane - triethylamine (9 + l ) , prepare solutions containing 7500 i.u. ml-l of vitamin D, 750 i.u. ml-l of vitamin A as the acetate or palmitate, 0.5 mg ml-1 of BHT, tocopherol and tocopheryl acetate, and 0.25 mg ml-l of BHA and ethoxyquin. Apply 2 pl of the appropriate sample solutions and 2 pl of each of the standard solutions of vitamin A acetate, vitamin A palmitate, vitamin D, BHA, BHT and ethoxyquin to a chro- matographic plate and score the surface of the plate 15 cm from the line of spots.Develop the chromatogram, in the dark, in a tank that has been equilibrated with the developing solvent for 30 minutes. When the solvent front reaches the scored line, remove the plate from the tank and allow it to stand in a stream of cold air until most of the solvent has evaporated. Then, without delay, spray the plate with iron(II1) chloride - potassium hexacyanoferrate(II1) reagent and assay the vitamins and antioxidants in the sample by comparing the sizes and intensities of the sample spots with the standards, which correspond to 0.6, 0-9, 1.2, 1.5 and 1.8 i.u. of vitamin A as the acetate, 0.6,0.9, 1.2, 1.5 and 1.8 i.u.of vitamin A as the palmitate, 6,9, 12, 15 and 18 i.u. of vitamin D, 0.2,0.3,0.4,0.5 and 0.6 pug each of BHA and ethoxyquin, and 0.4, 0-6, 0.8, 1.0 and 1.2 pg of BHT. For the assay of tocopherol and tocopheryl acetate, apply 2 p1 of the appropriate sample solutions and 2 pl of each of the standard solutions of a-tocopherol and a-tocopheryl acetate to a second chromatographic plate and develop the chromatogram as previously described. If tocopherol alone is present, use iron(II1) chloride - potassium hexacyanoferrate( 111) as the spray reagent ; if tocopheryl acetate is present, use the strongly acidic iron(II1) chloride - potassium hexacyanoferrate(II1) reagent. Assay the tocopherol and tocopheryl acetate in the sample by comparing the sizes and intensities of the sample spots with the standards, which correspond to 0-4,0.6,0.8, 1.0 and 1.2 pg each of a-tocopherol and x-tocopheryl acetate.PROCEDURE FOR DIFFERENTIATING VITAMIN D, FROM VITAMIN D,- Extract the vitamins and antioxidants from the sample by the method previously described. If the sample contains no vitamin A, dissolve the residue from the extraction in sufficient chloroform to give a theoretical vitamin D content of 40 000 i.u. ml-l and transfer 0.1 ml of this solution and 0-1 ml of each of the standard vitamin D solutions to three test- tubes. Evaporate the solutions to dryness in a stream of nitrogen at room temperature and dissolve the residues in 0.1 ml of glacial acetic acid. To each tube, add 2 ml of 72 per cent. m/m perchloric acid solution (see Note 3), mix the solutions well and immediately heat the tubes in a water-bath at 70 "C, with constant agitation, until the colour of the vitamin D, standard reaches a maximum (heating for about 60 s will usually suffice).Cool the solu- tions rapidly, add 1 ml of chloroform to each tube and shake the tubes vigorously. The presence of vitamin D, is revealed by the red or purple colour of the chloroform layer; if vitamin D, alone is present, the chloroform layer is coloured greenish yellow. NOTE 3-Perchloric acid should be handled with care. It is a powcriul oxidising agent, which may If the sample contains vitamin A, dissolve the residue from the extraction in sufficient cyclohexane - triethylamine (9 + 1) to give a theoretical vitamin D content of 40 000 i.u.ml-l and apply 0.1 ml of this solution, in the form of a narrow band, to a chromatographic plate. At one end of the band, apply 5 pl of standard vitamin I), solution to act as a marker, and score the surface of the plate 15 cm from the band. Develop the chromatogram in the n-hexane - ethyl methyl ketone - di-n-butyl ether (34 + 7 + 6) developing solvent until the solvent front reaches the scored line, remove the plate from the tank and inspect it in ultra- violet radiation of wavelength 254nm. Score the surface of the plate around the edge of the vitamin D band and, without delay, scrape the enclosed area of silica gel from the plate and transfer it to a small chromatographic column containing 10 ml of chloroform. Elute the vitamin D with chloroform (about 60ml) and evaporate the eluate to dryness under vacuum in a rotary-film evaporator with the water-bath at a temperature of 50 "C.Dissolve react violently if allowed to come into contact with strong reducing agents.April, 19731 261 tlie residue in a small volume of chloroform and transfer it to a test-tube. Place 0.1 nil of standard vitamin D, solution and 0.1 ml of standard vitamin D, solution in two other test- tubes and continue by the method previously described, starting from “Evaporate the solutions to dryness in a stream of nitrogen at room temperature. . , .” EXTRACTION OF VITAMINS AND ANTIOXIDANTS- Attempts to extract the vitamins and antioxidants from solid formulations by shaking aqueous dispersions of the samples with water-immiscible solvents usually resulted in the formation of emulsions.A preliminary digestion of the samples with trypsin reduced the extent of emulsification caused by gelatin, but did not affect that produced by other con- stituents, such as wheat meal or maize meal. The most efficient extraction was obtained by using diethyl ether as the extracting solvent and adding small volumes of methanol during the extraction so as to break down emulsified material. Methanol was preferred to ethanol for this purpose as it caused smaller amounts of water and water-soluble impurities to be carried through with the diethyl ether to the final evaporation stage. The small amount of hydroquinone or propyl gallate present in anaestlietic ether B.P. to prevent peroxide formation does not interfere, either in the chromatographic separation (when it remains at the origin), or in the extraction procedure.It can be removed, if necessary, by passing the solvent through a column of activated alumina. OF VITAMINS AND ANTIOXIDANTS I N PHARMACEUTICALS AND FEEDS STABILITY OF VITAMINS A AND D DURING THE CHROMATOGRAPHIC EXAMINAIION- When solutions of vitamin A, vitamin A esters and vitamin D are applied to thin layers o f silica gel, rapid decomposition of the vitamins takes place, leading to a reduction in the size and intensity of the vitamin spots and to the appearance of spots corresponding to the decomposition products on the subsequent chromatogram. The addition of a small amount of triethylamine to the spotting solvent suppresses this decomposition and if the vitamins are applied in a cyclohexane - triethylamine (9 + 1) solution, they can remain in contact with the silica gel for up to 10 minutes before development, without any detectable decom- position taking place.The triethylamine remains at the origin during the chromatographic run and does not interfere in the separation. Although no evidence of decomposition of vitamins A and D during the chromatographic separation has been observed, decomposition can occur when the plates are dried, leading to a reduction in the size and intensity of the vitamin spots. This decomposition is inhibited by ethyl methyl ketone and di-n-butyl ether, and if the plates are allowed to remain moist with developing solvent and the chromatograms are examined as soon as possible after chromatography, no loss of vitamins or antioxidants can be detected.Bolliger and Konig’s quantitative assay procedure for vitamin D,3 in wliicli the vitamin is eluted from the silica gel adsorbent after chromatography and assayed colorimetrically by its reaction with antimony trichloride, has been carried out on solutions of pure vitamin L> with cyclohexane - triethylamine (9 + 1) as the spotting solvent and hexane - ethyl methyl ketone - di-n-butyl ether (34 + 7 + 6) as the developing solvent. In duplicate determina- tions, recoveries of 98.5 and 99-5 per cent. of vitamin D were obtained. I I)ENTIFICATION AND ASSAY OF VITAMINS AND ANTIOXIDANTS ON THE CHROMATOGRAM- As a general reagent for the detection of vitamins and antioxidants, tlie strongly acidic iron(II1) chloride - potassium hexacyanoferrate(II1) spray reagent is preferred.It reacts with all the compounds under investigation and it is one of the few reagents that will detect tocopheryl acetate without the need for a preliminary hydrolysis. For the assay of tlie vitamins and antioxidants (except tocopheryl acetate), the less acidic iron( 111) chloride - potassium hexacyanoferrate(II1) spray reagent is used, because it gives a lower level of back- ground colour on the chromatogram. Plates that have been sprayed with either of the two 1-cagents should be protected from excessive heat and light, which produce dark-coloured backgrounds. Although the various compounds can usually be identified by their R, values (see Table I), it may be necessary to confirm the identification by use of the more selective methods of detection and identification indicated in Table 11.The identification of vitamin A and its esters can be confirmed by the characteristic greenish yellow fluorescent spots observed when the chromatogram is viewed in ultraviolet radiation of wavelength 360 nm; under these262 JOHNSON AND VICKERS : THIN-LAYER CHROMATOGRAPHIC IDENTIFICATION [AnU~ySt, VOl. 98 TABLE I Rp VALUES OF VITAMINS AND ANTIOXIDANTS Compound RF value Compound RF value Propyl, octyl and dodecyl gallates . . Hydroquinone . . .. . . 0.03 Vitamin A (alcohol) .. . . 0.17 0 Vitamin D, . . .. .. . . 0.22 Vitamin D, . . .. .. . . 0.22 Pre-vitamin D, . . .. . . 0.30 Pre-vitamin D, . . .. . . 0.30 BHA .. . . .. .. . . 0-32 Ethoxyquin . . ..a-Tocopherol .. Vitamin A acetate . . a-Tocopheryl acetate BHT . . .. .. Anhydrovitamin A . . Vitamin A palmitate 8-Carotene . . . . . . . . 0.42 . . . . 0.51 . . . . 0.58 .. . . 0.66 .. . . 0.75 . . . . 0.81 .. . . 0.85 ,. . . 0.88 conditions, ethoxyquin and anhydrovitamin A give bluish white and orange - brown spots, respectively. The antioxidants and tocopherol are distinguished by the red-coloured spots produced when the chromatogram is sprayed with the iron(II1) chloride - bipyridyl reagent. The D vitamins and their thermal isomerisation products, the pre-vitamins (Velluz, Amiard and Petits) are identified by spraying the chromatogram with trichloroacetic acid or antimony trichloride reagent and viewing it in ultraviolet radiation of wavelength 360 nm; all the compounds give greenish yellow fluorescent spots with the trichloroacetic acid reagent and pink fluorescent spots with the antimony trichloride reagent.If necessary, the selective methods of detection can be used for assay purposes. Because of the possibility of decomposition of the vitamins (particularly vitamin A) in ultraviolet radiation, chromatograms that have been examined by one of the ultraviolet procedures shcTi1.d not subsequently be examined by another method. TABLE I1 APPROXIMATE LIMITS OF DETECTION OF VITAMINS AND ANTIOXIDANTS* Method of detection used Vitamin A (alcohol) Vitamin D, . . Vitamin D, .. BHA .. . . Ethoxyquin . . Tocopherol . . Vitamin A acetate Tocopheryl acetate BHT .. .. Vitamin A palmitate A B C D E F G H' . . 0.05 - .. - . . - ... . 0.1 . . 0.05 - .. - . . - * . . . 0.1 0.1 0.06 0.1 0.05 0-1 0-05 - 0.1 0.25 0.25 0.2 0.25 0.2 - 0.2 0-25 0.2 - - __ 0.06 - 0.05 0.1 0.06 0.05 - 0.05 0.1 0.05 0.05 - 0.05 0.1 0.05 0.1 0-25 1.0 - 0.1 0.5 0.1 _- 0.05 0.2 0.4 0.1 - 0-1 0.26 0.05 0.2 0.2 0.2 0.2 - 0.2 0.5 0.2 1.0 - - - - - - - - - - A = Inspection in ultraviolet radiation of wavelength 360 nm. B = Inspection in ultraviolet radiation of wavelength 254 nm. C = Iron(II1) chloride - potassium hexacyanoferrate(II1) reagent. D = Iron(II1) chloride - potassium hexacyanoferrate(II1) reagent, strongly acidic. Heat a t 40 "C E = Iron(II1) chloride - bipyridyl reagent. I; = 25 per cent. solution of trichloroacetic acid in chloroform. Heat a t 120 "C for 10 minutes and inspect in ultraviolet radiation of wavelength 360 nm.G = 20 per cent. solution of antimony trichloride in chloroform. Heat a t 120 "C for 10 minutes. H = As for G, but inspect in ultraviolet radiation of wavelength 360 nm. * The figures show the lowest loading of the compound, expressed in micrograms, that can be detected on the chromatogram. Where no figure is given, the compound is not detected at a level of 2.0 pg. for 15 minutes. DIFFERENTIATION OF VITAMIN D, FROM VITAMIN D3- During attempts to distinguish between vitamins D, and D, by means of the coloured products obtained when vitamin D reacts with aromatic aldehydes in perchloric acid solution (SchalteggerO), the formation of a transient red colour on warming vitamin D, with per- chloric acid solution was observed. By choosing experimental conditions that favoured this reaction and stabilising the coloured product by extracting it into an organic solvent, a method of distinguishing vitamin D, from vitamin D, was developed.Tocopherol, tocopherylApril, 19731 263 acetate, BHA, BHT, hydroquinone, propyl gallate and ethoxyquin do not interfere, but vitamin A and its esters give an immediate blue - purple colour that rapidly fades to an intense golden yellow colour; any vitamin A (alcohol or ester) present must therefore be separated from the vitamin D by thin-layer chromatography before the colour test is applied. A thin-layer chromatographic separation of the vitamin D may also be necessary if the sample extract contains organic matter that chars badly on heating with perchloric acid. Pre-vitamin D, gives the same response as vitamin D,.Vitamin D, and pre-vitamin D, give greenish yellow colours. RESULTS AND DISCUSSION Recovery experiments were carried out by submitting a mixture of vitamins and anti- oxidants to the assay procedure, alone and in the presence of the appropriate amounts of gelatin, arachis oil and a mixture of gelatin, limestone, wheat meal, maize meal and B-group vitamins. A loading on the chromatographic plate of 15 i.u. of vitamin D, 1-6 i.u. of vitamin A as the acetate, 1-5 i.u. of vitamin A as the palmitate, 1 pg each of BHT, tocopherol and tocopheryl acetate, and 0.5 pg each of BHA and ethoxyquin was used and the results were assessed independently by four analysts. Mean recoveries for the individual compounds varied between 94 and 99 per cent.of those required by theory, depending on the compound and the type of formulation examined, with results from individual operators falling within &lo per cent. of the means. No difference in response between vitamin D, and vitamin D, could be detected and the same standard solutions (which contain vitamin D,) can be used for the assay of both compounds. The recommended procedures were applied to manufactured batches of several different vitamin preparations; the results obtained are shown in Tables I11 and IV. Except in one instance (the assay of vitamin D in a multivitamin syrup), no significant interference from other constituents of the formulations was encountered. In the samples examined, BHA and BHT were the antioxidants most generally used.Ethoxyquin was detected only in preparations destined for use in animal feeds; with minor exceptions, its use in foodstuffs is prohibited.1° Propyl, octyl and dodecyl gallates were not detected in any of the samples. In some vitamin A concentrates, BHT was detected at levels that were too low for it to have any significant antioxidant effect; in such samples, the BHT is probably derived from the vitamin A, which may contain small amounts intro- duced during the manufacturing process. Samples that had been stored for long periods were found to contain small amounts of unidentified impurities, apparently produced by decomposition of the vitamins and anti- oxidants. In samples containing vitamin A esters, compounds with R, values of 0.43 (from vitamin A acetate) and 0.92 (from vitamin A palmitate) were detected, A compound with an R, value of 0.65 was detected in samples containing ethoxyquin, and pre-vitamin D was found in some samples that contained vitamin D.Fortunately, the decomposition products either do not interfere in the assay, or the interference can be eliminated by the use of a selective method of detection. The impurities were shown to be decomposition products by dissolving the individual vitamins and antioxidants in hexane, exposing the solutions to ultraviolet radiation and examining the degradation products ; the same im- purities were detected in the artificially degraded vitamin A acetate, vitamin A palmitate, vitamin D and ethoxyquin as had been detected in the stored samples. The recommended assay procedure was applied to the determination of vitamin A in halibut-liver oil with synthetic vitamin A palmitate as the standard.Although the vitamin A in fish-liver oils is present as a mixture of long-chain fatty esters of the all-trans-, 13-cis-, 9-cis- and 9,13-di-cis-isomers of vitamin A,ll no separation of the isomers occurred and the single spot of vitamin A ester obtained from the halibut-liver oil was similar in appearance to and had the same Rp value as that obtained from the vitamin A palmitate. The quanti- tative results obtained by using iron(II1) chloride - potassium hexacyanoferrate(II1) reagent and ultraviolet radiation as the methods of detection, were comparable with those obtained by the official method of assay of the British Pharmacopoeia.12 Attempts to ascertain the vitamin A content of halibut-liver oil by hydrolysing the sample and determining the liberated vitamin A (alcohol) by thin-layer chromatography were unsuccessful, because two hydrolysis products, with RF values of 0.17 and 0.21, were obtained, which may correspond to mixtures of different isomeric forms of the vitamin. OF VITAMINS AND ANTIOXIDANTS IN PHARMACEUTICALS AND FEEDSTABLE I11 VITAMIN AND ANTIOXIDANT CONTENTS OF COMMERCIAL SAMPLES (SOLIDS) Sample* Gelatin-protected beads Vitamin A beads- Vitamin A, 325 000 i.u.g-l . . . . Vitamin -4, 500 000 i.u. 8-l . . .. Vitamin D, beads- Vitamin D, 850 000 i.u. g-l . . . . Vitamin D, 400 000 i.u. g-l . , . . Vitamin D, beads- Vitami+z E acetate beads- Tocopheryl acetate, 25 per cent. .. Vitamin A and D, beads- } } Vitamin A, 325 000 i.u. g-l Vitamin D, 108 000 i.u. g-l Vitamin A, 500 000 i.u. g-l Vitamin D, 125 000 i.u. g-I Animal feed additives Formula 1- 1 Vitamin A, 9640 i.u. g-1 Vitamin D, 2372 i.u. g-l Vitamin E, 0.43 per cent. with B- group vitamins and acetomenaph- thone in a wheat meal and limestone base J Formula 2- Vitamin A, 15 120 i.u. g-1 Vitamin D, 3770 i.u. g-' with B-group vitamins in a limestone base Vitamin D/ i.u. 6-1 1150 000 450 000 430 000 410 000 105 000 102 000 144 000 125 000 130 000 2600 2300 2600 4300 4400 Vitamin A Vitamin A Ethoxy- Tocopheryl ' (as ace- (as palmi- BHA, BHT, quin, Tocopherol, acetate, & tate)/i.u. 6-l tate)/i.u. g-1 per cent. per cent. per cent. per cent. per cent. tl 340 000 340 000 338 000 545 000 465 000 400 000 6400 330 000 340 000 630 000 580 000 320 000 9800 10 000 9900 15 700 15 700 17 000 13 000 200 000 450 450 450 680 730 0.65 1.6 0.38 0.32 9 9 0.4 0.75 0-65 1.3 Trace 6 1.4 9 0.6 3-2 0.5 0.25 0.25 0.25 0.2 1 0.35 0.35 0.37 6 9 c 0, W 00Sample* Vitamin D/ i.u.g-l Formula 3- Vitamin A, 9600 i.u. g-l Vitamin D, 2350 i.u. g-l with ribo- flavine in a maize meal base Formula 4- base Vitamin tablets 2300 } 2500 As for formula 3, but in a limestone Calciferol Tablets, Strong, B.P.- Vitamin D, 152 000 i.u. g-' . . . . 150 000 150 000 Calcium with Vitamin D Tablets, B.P.C.- Vitamin D, 1000 i.u. g-l . . . . 990 1050 1350 Vitamirz A , D and C pellets- Vitamin A, 11 100 i.u. g-' Vitamin D, 1460 i.u. g-' with vitamin } E: C in a sugar base illultivitarnin pellets- 570 Vitamin A, 7300 i.u.8-l Vitamin D, 590 i.u. g-' with B-group vitamins, vitamin C and trace metals in a sugar-coated chocolate base } E Syrup Multivitamin syrupt- 40 67 Vitamin A, 750 i.u. 8-l Vitamin D, 75 i.u. g-' with B-group vitamins and vitamin C in a syrup base TABLE I1 I-continued Vitamin -4 Vitamin A tate)/i.u. g-l tate)/i.u. g-' (as ace- (as palmi- BHA, per cent. 7200 2900 7400 2900 9400 9100 Trace Trace 10 900 Trace 11 600 Trace 11 100 Trace 7950 Trace 7350 Trace 7200 Trace 750 Trace Trace 690 Ethoxy- BHT, quin, per cent. per cent. Trace Trace Trace Trace Trace Trace Trace Trace Trace Trace 0.12 0.13 0.09 0.10 Tocopher yl Tocopherol, acetate, per cent. per cent. * The figures quoted for vitamin concentrates are those declared by the manufacturer; the figures quoted for formulations are based on the declared vitamin content of the concentrate used in their preparation.Vitamin concentrates usually contain an excess of the vitamins to allow for decomposition on storage. t.Because of solubility difficulties, acetone - triethylamine (9 + 1) was used as the spotting solvent for this formulation. The vitamin D was assayed quantitatively by Bolliger and Konig's pr~cedure,~ but with acetone - triethylamine (9 + 1) as the spotting solvent and hexane - ethyl methyl ketone - di-n-butyl ether (15 + 5 + 3) as the developing solvent; the vitamin D could not be assayed by visual comparison with standards because of distortion of the vitamin D spot.TABLE IV Sample* VITAMIN AND ANTIOXIDANT CONTENTS OF COMMERCIAL SAMPLES (OILS) Vitamin A Vitamin A (as palmitate)/ (B.P. assay1*)/ Vitamin D/ BHA, BHT, Tocopherol, per cent.i.u. g-l i.u. g-' i.u. g-l per cent. per cent. Vitamin A palmitate solution- Vitamin A, 1 000 000 i.u. g-l . . .. 1 030 000 1 015 000 0.48 Trace 1 120 000 1 068 000 0.51 Trace 980 000 1 010 000 0.7 Vitamin A and D concentrate- Vitamin A, 600 000 i.u. g-l 540 000 498 000 57 000 0.32 Trace Vitamin D, 50 000 i.u. g-l } 525000 507 000 57 000 0.30 Trace Halibut-liver oilt . . .. . . I . 24 800 22 800 Not determined 0.17 29 700 29 500 Not determined 0.18 26 400 25 200 Not determined 0.20 22 500 22 400 Not determined 0.17 As for Table 111. t Vitamin D is present in this material a t too low a level to be assayed by the recommended procedure. The tocopherol was assayed by applying an increased loading (5 pl) of sample and standard solutions to the chromatographic plate and using iron(II1) chloride - bipyridyl reagent for detection. 0 n 36 Y 0 W 00April, 19731 OF VITAMINS AND ANTIOXIDANTS IN PHARMACEUTICALS AND FEEDS 267 The procedures described in this paper have been successfully applied to the analytical control of vitamins and antioxidants in vitamin concentrates and formulated products. Because of their speed and simplicity, they enable a comprehensive control of manufacturing procedures to be maintained at minimum cost. The authors are grateful to Messrs. E. E. Taylor and C. Knewstubb for samples, technical information and helpful comment. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Bolliger, H. K., in Stahl, E., Editor, “Thin-layer Chromatography,” Academic Press, London, Strohecker, R., and Henning, H. M., “Vitamin Assay,” Academic Press, London, 1965. Bolliger, H. R., and Konig, A., 2. analyt. Chem., 1965, 214, 1. Hanewald, K. H., Mulder, F. J., and Keuning, K. J., J . Pharm. Sci., 1968, 57, 1308. Ponchon, G., and Fellers, F. X., J . Chromut., 1968, 35, 63. Stecher, P. G., Editor, “The Merck Index,” Seventh Edition, Merck and Co. Inc., Rahway, New Morris, W. W., Wilkie, J. B., Jones, S. W., and Friedman, L., Analyt. Chem., 1962, 34, 381. Velluz, L., Amiard, G., and Petit, A., Bull. SOL Chim. Fr., 1949, 601; Chem. Abstr., 1960, 44, 2999. Shaltegger, H., Helv. Chim. Acta, 1946, 29, 286. “The Antioxidant in Food Regulations 1966,” S.I. 1966 No. 1600, H.M. Stationery Office, London. Chesterfield, N. J., Australus. J . Pkarm., 1969, 50 (600), S97. “The British Pharmacopoeia 1968, 1966, p. 210. Jersey, U.S.A., 1960, p. 1100. The Pharmaceutical Press, London, 1968, p. 1359. Received June 6th, 1972 Accepted November 7th, 1972