AnaZyst, July, 1974, Vol. 99, $9. 397-402 397 A Simplified Colorimetric Method for the Determination of Ascorbic Acid in Pure Solutions and in Pharmaceutical Preparations BY NAG1 WAHBA, DAWOUD A. YASSA AND RAMZY S. LABIR (Biochemistry Department, Faculty of Medicine, A in-Shams University, Cairo, Egypt) A precise and specific method for the determination of L-ascorbic acid in the presence of dehydro-L-ascorbic acid and 2,3-diketo-~-gulonic acid has been developed. The method is based on the reaction of phenylhydrazinium chloride with ascorbic acid in 0.1 N hydrochloric acid at a temperature of 50 f 2 "C. The stable yellow colour produced is measured spectrophoto- metrically at 395nm. The results obtained by this method and the B.P. method are compared. There is no interference from other vitamins, minerals, glucose, sucrose or the most commonly used excipients with this method and it has been successfully applied to pharmaceutical preparations.The simplicity of the procedure permits rapid analysis, which is necessary for routine control work. ASCORBIC acid can be determined by titration with ammonium cerium( IV) sulphate, according to the method of the B.P.l or with 2,6-dichlorophenolindophenol, according to that of the U.S.P.2 Barakat, Abd El-Wahab and El-Sad9 used N-bromosuccinimide for the determina- tion of ascorbic acid in pure solution, pharmaceutical preparations and biological fluids. In all of the above methods interference from other reducing substances, such as organic and inorganic iron(I1) compounds, phenolic compounds, sodium metabisulphite (usually included in pharmaceutical preparations as an antioxidant) and reduced forms of nicotinic acid derivatives, occurs.In the Roe and Kuether rneth~d,~ ascorbic acid is oxidised to dehydroascorbic acid, and the latter is coupled with 2,4-dinitrophenylhydrazine. The defect of this method is that dehydroascorbic and 2,3-diketogulonic acids react in the same way. 2,3-Diketogulonic acid is biologically inactive and is usually present as a decomposition product of ascorbic acid in pharmaceutical preparations that have been stored. Roe, Mills, Oesterling and Damron5 have modified this method in order to render it applicable to the determination of ascorbic acid, dehydroascorbic acid and 2,3-diketogulonic acid in the presence of each other.However, the modification is time consuming as it requires three separate determinations. In the determination of ascorbic acid by the Schmall, Pifer and Wollish6 method, which involves the use of diazotised 4-methoxy-2-nitroaniline, the colour fades within 10 minutes. The use of phenylhydrazinium chloride as a colorimetric reagent for the determination of glucose in blood and lactose in milk has been reported by Wahba and co-worke~s.~~~ During the studies on the rate of reaction of phenylhydrazine with glucose, fructose, lactose and ascorbic acid at different temperatures, it was noticed that the rate of colour development with the above-mentioned sugars was very slow in comparison with ascorbic acid. It was decided, therefore, to investigate the applicability of this colour reaction to the determination of ascorbic acid.In this paper the resulting simple, precise and specific method for the accurate deter- mination of ascorbic acid in pure solutions and in pharmaceutical preparations, without interference from other compounds usually present in the preparations, is reported. EXPERIMENTAL REAGENTS- PhenyZhydraziniuw chloride solution-A 1 per cent. m/ V solution of phenylhydrazinium chloride (recently recrystallised from 95 per cent. ethanol) was made up in 0.1 N hydro- chloric acid. Standard solutions of ascorbic acid-These solutions were made up to concentrations of 25, 50 and 100 pg ml-1 in distilled water. 8 SAC and the authors. (The pH was found to range from 1-1 to 1.3.)398 [Analyst, Vol. 99 T a t solutions, containing 50 to 100 pg ml-l of ascorbic acid-Tablets and capsules were extracted by crushing and stirring in distilled water.Aqueous solutions for injection and syrups were diluted with distilled water and used directly. Oxalic acid (2 per cent.) or metaphosphoric acid (5 per cent.) can be used in both standard and test solutions to protect ascorbic acid from oxidation, especially if metal ions (e.g., Cu2+) are present in the samples to be analysed. PROCEDURE- Transfer by pipette 1.0 ml of each of the standard and test solutions into 10-ml stoppered test-tubes or cylinders, then add accurately 5-0 ml of the phenylhydrazinium chloride solution to each. Allow them to stand for 1 hour in an incubator or water-bath maintained at 50 & 2 "C. Cool, and measure the absorbance of each solution at 395 nm against a blank carried out simultaneously by using the same reagent but with 1.0ml of distilled water instead of the test solution.WAHRA et al.: COLORIMETRIC DETERMINATION OF ASCORBIC ACID Calculate the amount of ascorbic acid from the standard calibration graph. RESULTS AND DISCUSSION EFFECT OF REDUCING SUGARS- The effect of glucose, fructose and lactose on the determination of ascorbic acid by use of the proposed procedure was studied at temperatures from 40 to 55 "C. It was found that, within the temperature range investigated, concentrations of such sugars as much as 50 to 100 times greater than that of ascorbic acid did not cause any interference. EFFECT OF DEHYDROASCORBIC AND 2,3-DIKETOGULONIC ACIDS- Dehydroascorbic acid was prepared according to the method of Roe and Kuether4 by use of active charcoal and bromine oxidation.It was also prepared by oxidation with the theoretical amount of 0.1 N iodine solution, as described by Penny and 2ilva.Q 2,3-Diketo- gulonic acid was prepared by allowing a solution of dehydroascorbic acid to mutarotate for 18 days at room temperature. In some experiments it was desirable to remove hydrobromic or hydriodic acid that resulted during the preparation of dehydroascorbic acid. This was done by adding 0.1 N silver nitrate solution in order to precipitate the halide, followed by 0.1 N hydrochloric acid to precipitate the excess of silver ions. The precipitates were removed by filtration and the final volume was adjusted with 0.1 N hydrochloric acid to the required concentration.The determination of ascorbic acid, dehydroascorbic acid and 2,3-diketogulonic acid was carried out as described above under Procedure, and also by Roe and Kuether's method.* The results are shown in Table I. Each result in the table represents the mean of four separate determinations. TABLE I DETERMINATION OF ASCORBIC, DEHYDROASCORBIC AND 2,3-DIKETOGULONIC ACIDS BY THE PROPOSED PROCEDURE AND THE ROE AND KUETHER METHOD4 Recovery by- Proposed procedure, Roe and Kuether Analysed material per cent. method, per cent. A I \ Ascorbic acid . . .. . . .. . . 101.1 103.2 Dehydroascorbic acid prepared by use of- Active charcoal . . .. .. .. 00.0 98.6 Bromine oxidation . . . . .. . . 102.3 104-3 Bromine oxidation, HBr removed . .. . 00.0 97.2 Iodine oxidation . . .. . . . . 100.6 100.9 Iodine oxidation, HI removed . . . . 00.0 97.8 2,3-Diketogulonic acid . . .. .. .. 25.6 115-6 00.0 - 2,3-Diketoguloiiic acid, HI removed . . .. From these results, it is clear that whereas ascorbic acid was recovered equaUy well by both methods, dehydroascorbic acid prepared by use of charcoal did not react at all in the proposed procedure. On the other hand, dehydroascorbic acid prepared by bromine or iodine oxidation yielded complete recovery in the same procedure. However, when hydro- bromic and hydriodic acid were removed, no reaction was obtained. Thus, it was concludedJuly, 19741 I N PURE SOLUTIONS AND IN PHARMACEUTICAL PREPARATIONS 399 that the complete recovery in the presence of hydrobromic or hydriodic acid is due to the reversibility of the ascorbic acid oxidation- -2H + H,O +2H Ascorbic acid e = t Dehydroascorbic acid - 2,3-Diketogulonic acid During the long incubation time (1 hour) at 50 "C, removal of ascorbic acid by reaction with phenylhydrazine allowed further reduction of dehydroascorbic acid, until the conversion was complete.This finding suggests also the possibility of the application of the proposed pro- cedure to the determination of total active vitamin C, i.e., ascorbic and dehydroascorbic acids, by use of reducing agents, such as hydrogen ~ulphide.~ The recovery obtained for 2,3-diketogulonic acid was 25.6 per cent., but again, none was recovered when hydriodic acid was removed. It is concluded, therefore, that this recovery reflects the content of unconverted dehydroascorbic acid, which reverted to ascorbic acid during the incubat ion with phenylhydrazine.Thus, neither dehydroascorbic acid nor 2,3-diketogulonic acid reacts with phenyl- hydrazinium chloride under the conditions used. No interference from dehydroascorbic acid (prepared by using charcoal) or 2,3-diketogulonic acid (without hydriodic acid) was observed when the temperature of incubation in the proposed procedure was varied in the range from 30 to 60 "C. EFFECT OF VITAMINS, HORMONES, MINERALS AND SUCROSE- The proposed method has been applied to the determination of ascorbic acid in pharnia- ceutical preparations containing other vitamins, minerals, glucose, sucrose and some excipients. The composition of these preparations (as stated) is given in Table 11, and the per- centage recoveries by the proposed and B.P.methods are shown in Table 111. It can be seen from these results that the method has been successful. In all samples, the recovery of added ascorbic acid by the proposed method ranged between 98.1 and 102.1 per cent. As expected, the recovery of the original ascorbic acid content was very low (60 to 80 per cent.) in the incubated samples (I and 11), due to deterioration by oxidation. It is remarkable that in both samples, recovery was lower by the proposed method than by the B.P. method (82 t o 89 per cent.). TABLE I1 FORMULATIONS OF ANALYSED SAMPLES Sample A \ Constituent I Ascorbic acid/mg .. . . . . 500 - Thiamine hydrochloride/mg . . . . Riboflavinelmg . . .. .. .. Nicotinamidelmg .... .. Pyridoxine hydrochloride/mg . . .. Vitamin A/i.u. . . .. . . . . Vitamin D,/i.u. . . .. . . . . Vitamin E/mg . . .. . . ,. Calcium pantothenatelmg . . . . Glucoselg . . . . .. .. .. Calcium gluconogalactogluconatelg . . - Inositol/mg . . . . .. .. Choline chloride/mg . . .. .. Iron(l1) sulphate/mg . . .. .. Copper(I1) sulphatelmg . . .. .. Potassium iodatelmg . . . . .. Ethinyl oestradiol/mg . . .. .. Methyltestosterone/mg . . .. .. Water for injection/ml . . .. .. Syruplml .. .. . . .. . . Lactose (capsules)/mg . . .. .. - - - - - - - - - - - - - - - - I Excipients (tablets) : starch, and magnesium stearatelmg . , to 680 VI viI 75 75 1 5 1.2 2.5 10 6 2 0.5 4000 5000 400 500 2 5 4 - - - - - - 50 c 50 99 - 1-26 - 0-1 - 0.01 - 2.5 - - - to 250 to 320400 [Artalyst, Vol.99 In fresh samples, comparison of the percentage recovery of the ascorbic acid originally present in the preparations shows that it is slightly lower (3 per cent.) in samples I and 11, but markedly higher in samples I11 to VII, when determined by the B.P. method as against the proposed procedure; in all instances of the use of the B.P. method it is not less than 100 per cent. As shown in Table 11, these latter samples (I11 to VII) were those containing substances that were expected to interfere in the B.P. method, thus giving falsely high results. The difference was very marked (16 per cent.) in sample VII, in which there was a large amount of iron(I1) sulphate, other vitamins and hormones. Thus, it is concluded that vitamins, minerals, glucose, sucrose and most excipients usually found in pharmaceutical preparations either do not interfere in the proposed method, or their interference, if any, is much less than with other methods, such as that of the B.P.However, in the presence of excessive amounts of riboflavin, it is advisable to add about 0.5 g of talc to the test solution, shake, and filter it in order to eliminate the slight interference of the yellow colour of ribo- flavin, according to Wahba and Fahmy.lo WAHBA et al. : COLORIMETRIC DETERMINATION OF ASCORBIC ACID TABLE I11 COMPARISON BETWEEN THE RESULTS OF THE PROPOSED AND B.P. METHODS Recovery by- I Stated/ Added/ Proposed method, B.P. method, Sample' I (fresh) . . .. .. .. .. I (incubated at 45 "C for 12 months) . . I1 (fresh) . ... .. . . .. I1 (incubated at 45 "C for 12 months) . . 111 . . . . . . . . . . .. IV . , . . .. .. .. . . v .. . . .. .. . . VT .. .. .. .. . . .. VII . . .. .. .. .. . I mgt - 100 200 100 200 100 200 100 200 100 200 200 500 10 20 20 40 10 20 - - - - - - - - per cent.: 108.2 (h0.7) 100.1 ( 5 0 . 5 ) 99.6 (50.7) 60.2 (&l-1) 99.1 (h0.4) 98.6 (50.9) 109.3 (f0-8) 99.8 (&O-7) 101.2 (&O'S) 80.6 (f 1.1) 98.7 (f 1.2) 101-7 (A0.8) 100.8 ( f 0.75) 100.1 (fO.6) 99.8 (f0.59) 106.4 (& 1.1) 100.9 ( f 0.65) 100.3 ( f 0-85) 107.1 ( k l - 2 ) 99.6 (f0.3) 101.1 (f0-5) 108.2 ($I 1.1) 100.1 ( f 0.41) 101-7 (f0.39) 112.2 ( f 0.98) 102.1 (f0.6) 100.8 (f0.3) * For sample compositions, see Table 11. t Added to the sample in the form of an aqueous solution containing 100 mg ml-1 of ascorbic $ Mean of four experiments ; values in parentheses are standard deviations of individual acid.results. EFFECT OF CATALYSTS AND PRESERVATIVES- We found that the presence of metal salts, such as copper(I1) sulphate or iron(II1) chloride, neither inhibited nor accelerated the colour formation. On the other hand, the addition of stronger oxidising agents, such as hydrogen peroxide and potassium permanganate solutions, lowered the extinction at the maximum, which nevertheless maintained its position. Acids used during the extraction to preserve ascorbic acid in its reduced state, such as 2 per cent. oxalic acid or 5 per cent. metaphosphoric acid, did not show any interference in the proposed method.July, 19741 IN PURE SOLUTIONS AND IN PHARMACEUTICAL PREPARATIONS 401 EFFECT OF pH- The colour reaction was studied at other pH values by using acetate buffers (B.P.standard). Irregular results were obtained, the maximum absorption was shifted to a higher wavelength at pH 3, 4 and 5 and the sensitivity to micro-amounts (25 and 50 pg) decreased. PROPERTIES OF THE COLOUR- The absorption spectrum (Fig. 1) of the colour produced in the proposed procedure was scanned in a l-O-cm cell in a Carl Zeiss PMQ I1 spectrophotometer. Fig. 1 shows that the yellow colour obtained gives a maximum absorbance at 395 nm. The standard graph plotted in Fig. 2 shows that the Beer - Lambert law is obeyed in the range from 25 to 100 pg of ascorbic acid. The colour obtained after 1 hour at 50 "C remains stable for at least 24 hours.Wave lengt h/n m Fig. 1. Absorption spectrum (at 50 "C) of ascorbic acid chromogen The structure of the isolated compound is unknown. It is a highly stable product under the conditions used, and microanalysis showed that it was not a simple osazone or hydrazone, as in the instance of sugars. Further investigations are still being carried on by the authors. The method is liable to an error of &4 per cent. and the standard deviation was found to be between 5 0 . 5 and 5 1 . 7 per cent. CONCLUSION The proposed method offers several advantages over the commonly used procedures. It has a high degree of specificity that may be due to the enediol grouping of ascorbic 0 10 20 30 40 50 60 70 80 90 100 110 Concentrat ion/pg ml-1 Fig. 2. Relationship between concentra- tion and absorbance a t 50 "C402 VC‘AHBA, YASSA AND LABIB acid, as dehydroascorbic and 2,3-diketogulonic acids do not react, in contrast with the methods involving the use of 2,4-dinitrophenylhydra~ine,~*~ with which only dehydro- ascorbic and 2,3-diketogulonic acids react and which theref ore require an oxidation step.Other methods based on the reducing power of ascorbic a ~ i d l - ~ suffer from interference from other reducing substances. With the proposed procedure, reducing substances do not interfere, nor do vitamins, glucose, sucrose, minerals and common excipients in amounts usually encountered in multivitamin pharmaceutical preparations. The proposed procedure is rapid, simple, sensitive (down to 25 pg ml-l), and suitable for routine analysis (especially so, because of the stability of the colour produced). It can be carried out directly, requiring only extraction with distilled water for capsules or tablets. The method shows good precision and its accuracy compares favourably with con- ventional procedures. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. “The British Pharmacopoeia 1968,” The Pharmaceutical Press, London, 1968, p. 65. “The United States Pharmacopeia,” XVIIIth Revision, Mack Co., Easton, Pa., 1970, p. 52. Barakat, M. Z., Abd El-Wahab, M. F., and El-Sadr, M. M., Analyt. Chem., 1955, 27, 536. Roe, J. H., and Kuether, C. A., J. Biol. Chem., 1943, 147, 399. Roe, J. H., Mills, M. B., Oesterling, M. J., and Damron, C. M., Ibid., 1948, 174, 201. Schmall, M., Pifer, C. W., and Wollish, E. G., Analyt. Chem., 1953, 25, 1486. Wahba, N., Hanna, S., and El-Sadr, M. M., Analyst, 1956, 81, 430. Wahba, N., Ibid., 1965, 90, 432. Penny, G. R., and Zilva, S. S., Biochem. J., 1943, 37, 39. Wahba, N., and Fahmy, E., J. Pharm. Pharmac., 1965, 17, 489. Received J u N e 18th. 1973 Amended November 22nd, 1973 Accepted January 29th, 1974