Reprinted from HoRTSciENCE, Vol. 17(3), June 1982A publication of the American Society for Horticultural Science, Alexandria, Virginia

HortScience 17(3):334-335. 1982.

A High-performance LiquidChromatography Method forDetermining Ascorbic Acid Contentof Fresh Fruits and Vegetables1

Alley E. WatadaHorticultural Crops Quality Laboratory, HSI, ARS, U.S. Department of Ag-riculture, Beltsville, Maryland 20705

Additional index words, sterns, leafy tissue, floral tissue, root, tubers

Abstract. Ascorbic acid of fresh fruits and vegetables was extracted with 6% metaphos-phoric acid and determined effectively by using a Clg cartridge in a radial compressionmodule and 1.5% NH4H2PO4 mobile phase in a high performance liquid chromatograph.

The commonly used AOAC method (2) of First, fresh fruits and vegetables contain corn-determining ascorbic acid content in fresh pounds other than ascorbic acid that will re-fruits and vegetables by reducing the 2,6- duce the dye. Second, recognizing the sharpdichlorophenolindophenol has 2 limitations, change in the dye color is difficult when the

sample extract contains carotenoids or an-thocyanin. These problems are not encoun-

'Received for publication January 15, 1982. Refer- tered when the ascorbic acid content is deter-ence to a specific brand or firm name does not con- mined by the high performance liquidstitute endorsement by the U.S. Department of Ag- chromatography (HPLC) method. Augustinriculture over others of a similar nature not men- e, a] (1) mocijfjed the HPLC procedure of"°I!!d- Sood et al. (5) by using a 3.9 mm x 30 cmThe cost of publishing this paper was defrayed in . „ , . ,part by the payment of page charges. Under postal "Bondapak C I 8 column w.th a tndecylam-regulations, this paper therefore must be hereby momum formate/water/methanol solution,marked advertisement solely to indicate this fact. pH 4.5, as the mobile phase. In our prelimi-

The technical assistance of Carol Simonds is sin- nary test this procedure was not satisfactorycerely appreciated. because the internal pressure increased after 5

334 HORTSCIENCE, VOL. 17(3), JUNE 1982.

Table 1. Ascorbic acid content of horticultural commodities determined by high performance liquidchromatography (HPLC).

Ascorbic acid (mg/100 g fresh weights)

Commodity

ApplesGreen snap beansSweet peppersKaleSweet potatoesZucchiniBroccoliCabbagesCarrotsCauliflowerPotatoesTomatoeszsee reference (6).yND = not done.

USDAHandbook 8Z

719

1281862319

113'47

S788

21

HPO, extraction solution

(pH1.38)

916

1461625529

1316716892511

(pH3.0)

1.54

157 to 65

163

NDy

NDNDNDNDND

Table 3. Recovery of added ascorbic acid to 100-g kale sample by high performance liquidchromatography.

Table 2. Effect of different plant materials on the final pH when 60 g fresh plant tissue were maceratedwith 6% metaphosphoric acid, 250 ml total volume, at initially different pH values.

Final pH of plant HPO, macerate

Commodity

AppleGreen snap beansBroccoliCollard greensKaleSweet potatoesZucchini

HPO,(pH3.0)

3.203.924.114.194.223.843.94

HPO, HPO,(pHl .8 ) (pH1.38)

1.952.112.232.422.592.102.08

.55.54.62.64.72.53.60

to 10 samples were injected. The increase inpressure probably was caused by the precipi-tates which were found to occur when a 6%meta phosphoric acid (HPO3) extraction solu-tion was added to the mobile phase. Precipita-tion also occurred when the water content orthe pH of the mobile phase was increasedslightly. This study was undertaken to im-prove the HPLC method for determining as-corbic acid content in fresh fruits and vegeta-bles.

A clear and sharp separation of ascorbicacid was achieved by using an 8 mm x 10 cmC|8 cartridge in a Water's radial compressionmodule with 1.5% NH4H2PO4, pH 3, as themobile phase. The mobile phase was meteredat 4 ml/min and the ascorbic acid eluted in1.98 min. Other UV absorbing compoundspresent in fruits and vegetables examined didnot elute at the same time as ascorbic acid.This was indicated by the lack of a peak atabout 1.98 min with an apple sample whichhad the ascorbic acid oxidized to dehydroas-corbic acid during extraction (Figure notshown). The ascorbic acid concentration wasbased on the absorption of ultraviolet light at awavelength of UV absorption at 254 nm withWater's model 440 absorption detector. Ab-sorption was linear for ascorbic acid contentsranging from 0 to 7 p,g. The injection volume,which ranged from 10 to 50 JJL!, was depen-dent on the concentration of ascorbic acid inthe plant extract.

The ascorbic acid was extracted from planttissues with 6% HPO, (2) containing 1 x I0~"M EDTA and 1 x 10~7 M diethylthiocarba-

mate. The low pH of 6% HPO3 caused thestandard ascorbic acid to elute near the front,so the pH of HPO3 was increased to 3 tomatch the pH of the mobile phase. However,the plant material increased the pH addition-ally to a level where the ascorbic acid was notstable as indicated by the low ascorbic acidvalues for apples, green beans and zucchini(Table 1). The crops had a differential effecton the final pH of the pIant-HPO3 macerate(Table 2). The pH of the macerate was highestfor green leafy tissues such as collards andkale and the lowest for apples. The pH of theplant-HPO3 macerate did not rise above 3when the pH of the initial HPO, was adjustedto 1.8 (Table 2), and the solution was satisfac-tory for extraction and analysis of ascorbicacid (data not presented). On the other hand,the pH of unaltered 6% HPO, was increasedsufficiently by the plant material to a levelthat did not affect the retention time of the as-corbic acid. Thus, all analyses in this studywere completed with unaltered HPO, solu-tion, and the pH was adjusted to 2.0 only forthe standard.

The ascorbic acid contents of fruits andvegetables analyzed by this method were gen-erally similar to values reported in the USDAHandbook No. 8 (6) (Table 1). The parts ana-lyzed include fruits, stems, leafy tissue, floraltissue, root, and tubers. The analyzed valueswere higher than the reported values for allcrops except kale and tomatoes. These differ-ences between analyzed and reported valueswere probably due to natural variation in thespecies.

Ascorbicacid added(mg)

04283

125

Quantityrecovered

(mg)

143178217273

Calculatedrecovery

(mg)

143185226268

Recovery(%)

9696

102

Recovery of ascorbic acid was studied byadding standards (about 22%, 45%, and 67%of the amount reported for kale in the USDAHandbook No. 8 (6)) to kale samples beforemaceration. The amount recovered was 96%for 2 of the samples and 102% for the thirdsample (Table 3), indicating that extractionand recovery by the analytical method des-cribed here were complete.

During the preparation of this manuscript,2 other HPLC methods of analyzing ascorbicacid contents were published. Dennison etal.(3) used a 4 mm X 30 cm uBondapak NH2

packed column and methanol/,025%KH2PO4 solution, pH 3.5, as the mobilephase for determining ascorbic acid in bever-ages. Finley and Duang (4) used 2 uBondapakC,8 columns connected in series with di-sodium phosphate solution containingtributylamine for ion-pairing as the mobilephase to separate ascorbic acid, dehydroas-corbic acid, and diketogluconic acid oforange and tomato juices, green pepper, andspinach. The pH of the solutions used bythese investigators were in a range that wasfound to be unsatisfactory in our study be-cause it appeared to affect the stability of theascorbic acid. The mobile phase of these 2methods contained methanol or an ion-pair-ing compound, which adds to the cost. OurHPLC method used only a 1.5% NH4H2PO4

solution for the mobile phase and the analysisfor ascorbic acid content was rapid and effec-tive.

Literature Cited

1. Augustin, J., C. Beck, and G. I. Marousek.1981. Quantitative determination of ascorbicacid in potatoes and potato products by highperformance liquid chromatography. J. FoodSci. 46:312-313.

2. Association of Official Analytical Chemists.1975. Official methods of analysis. Associa-tion of Official Analytical Chemists,Washington. D.C. p. 829.

3. Dennison, D. B. .T.G. Brawley, andG. L. K.Hunter. 1981. Rapid high-performance liquidchromatographic determination of ascorbicacid and combined ascorbic acid-dehydroas-corbic acid in beverages. J. Agr. Food Chem.29:929-935.

4. Finley, J . W . and D. Duang. 1981. Resolutionof ascorbic, dehydroascorbic and diketog-luconic acids by paired-ion reversed-phasechromatography. J. Chromatogr. 207:449-453.

5. Sood, S. P., L. E. Sartori, D. P. Wittmer, andW. G. Haney. 1976. High-pressure l iquidchromatographic determination of ascorbicacid in selected foods and mult ivi tamin prod-ucts. Anal. Chem. 48:796-798.

6. Watt, B. K. and A. L. Merrill. 1963. Compo-sition of foods. U.S. Dept. of Agr. Handb. 8.

HoRTSciENCE, VOL. 17(3), J U N E 1982. 33s