Polarographic determination of vitamin C after derivatization with o-phenylenediamine

Article abstract of DOI:10.1135/cccc2010026

A differential pulse polarographic (DPP) method has been developed for the determination of ascorbic acid (AA) and dehydroascorbic acid (DHA), the two main forms of Vitamin C. The method consists of the DPP analysis of a quinoxaline obtained by the derivatization of DHA with o-phenylenediamine. Results using the proposed method correlated well with those obtained by two reference methodologies: the common iodometric method and a published chromatographic methodology. It was also used in the study of Vitamin C degradation in fruit juices, showing that it involves an initial oxidation of AA to DHA, followed by ydrolytic degradation of the latter.

Full text of DOI:10.1135/cccc2010026

Polarographic Determination of Vitamin C
731
POLAROGRAPHIC DETERMINATION OF VITAMIN C AFTER
DERIVATIZATION WITH o-PHENYLENEDIAMINE
1
2
3
José António RODRIGUES , Inês Maria VALENTE , Luís Moreira GONÇALVES ,
4
5,
João Grosso PACHECO and Aquiles Araújo BARROS *
Requimte – Department of Chemistry and Biochemistry, Faculty of Sciences, University of Porto,
1
Rua do Campo Alegre No. 687, 4169-007 Porto, Portugal; e-mail: jarodrig@fc.up.pt,
2
5
3
4
inesmariavalente@sapo.pt, luis.goncalves@fc.up.pt, joao.pacheco@fc.up.pt,
ajbarros@fc.up.pt
Received March 15, 2010
Accepted April 20, 2010
Published online July 8, 2010
A differential pulse polarographic (DPP) method has been developed for the determination
of ascorbic acid (AA) and dehydroascorbic acid (DHA), the two main forms of Vitamin C.
The method consists of the DPP analysis of a quinoxaline obtained by the derivatization of
DHA with o-phenylenediamine. Results using the proposed method correlated well with
those obtained by two reference methodologies: the common iodometric method and a pub-
lished chromatographic methodology. It was also used in the study of Vitamin C degrada-
tion in fruit juices, showing that it involves an initial oxidation of AA to DHA, followed by
hydrolytic degradation of the latter.
Keywords: Ascorbic acid; Electrochemical techniques; Fruit juices; Polarography; Voltam-
metry; Vitamin C.
The term Vitamin C is usually used to identify ascorbic acid (AA). However,
dehydroascorbic acid (DHA), resulting from the oxidation of AA, has the
same biological activity as it can be converted into AA in the human body.
So, a simple definition of Vitamin C should be AA + DHA 1. Although it oc-
curs naturally in various concentration levels in a broad range of fruits and
vegetables, AA is intentionally added to foods, including fruit juices, to im-
prove the nutritional quality and for its potent antioxidant capacity to act
as a singlet oxygen quencher2. The joint commission of Food and Agricul-
tural Organization (FAO) and the World Health Organization (WHO) rec-
ommend a daily intake of 45 mg of Vitamin C for adults3. In fresh foods,
Vitamin C is usually found as AA. However, during industrial processing
and conservation of food products, oxidation of AA to DHA can occur. AA
can be easily degraded by enzymes and atmospheric oxygen. Oxidation re-
actions can be accelerated by high temperatures, high pH, light and pres-
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741
© 2010 Institute of Organic Chemistry and Biochemistry
doi:10.1135/cccc2010026

732
Rodrigues, Valente, Gonçalves, Pacheco, Barros:
ence of metals. Thus, a method that selectively determines the levels of AA
and DHA in food samples would be useful in monitoring the oxidative deg-
radation of the food.
Numerous analytical techniques have been proposed for Vitamin C deter-
mination in different matrices and at different levels, including titrimetry,
voltammetry, potentiometry, fluorometry, flow injection analysis (FIA),
spectrophotometry and chromatography4,5. Particularly, chromatographic
methods are very efficient in Vitamin C assay of complex materials such as
fruit juices, vegetables and beverages6. High-performance liquid chromatog-
raphy with UV detection (HPLC-UV) is currently the most commonly used
technique for the analysis of AA and DHA in food. Some HPLC methods
with fluorimetric and electrochemical detection have been reported as
well1. Electrochemical detection is an attractive alternative method for de-
tection of electroactive species, because of its inherent advantages of sim-
plicity, ease of miniaturization, high sensitivity and relatively low cost7,8
.
In this paper, an indirect determination of AA is proposed, involving a
previous oxidation to DHA with N-bromosuccinimide (NBS), followed by a
derivatization with o-phenylenediamine (OPDA)9–12 and differential pulse
polarographic (DPP) determination of the resulting quinoxaline13. The
method allows the determination of both AA and DHA and it is applied to
their determination directly in some fruit juices and tablets with no need
for any previous separation procedures. The direct electrochemical determi-
nation of AA by oxidation on a conventional electrode is too difficult be-
cause of its large overpotential and fouling by the oxidation products14.
EXPERIMENTAL
Equipment
Cyclic voltammetry (CV), square-wave voltammetry (SWV) and DPP were performed using a
Metrohm 663 VA voltammetric stand (Herisau, Switzerland) with a platinum wire as an aux-
iliary electrode and a reference AgCl|Ag (3 M KCl) electrode. The system was connected to an
Autolab PGSTAT 10 voltammetric system (Eco Chemie, Utrecht, The Netherlands). All mea-
surements were made at room temperature. CV and SWV were performed with a hanging
mercury drop electrode (HMDE) and DPP with a dropping mercury electrode (DME), drop
size of 0.024 mm3, drop time of 1 s. CV used a sweep rate of 100 mV s–1. SWV and DPP
measurements were performed with a potential step of 6 mV and amplitude of 25 mV and
50 mV, respectively.
The HPLC system (Jasco Corporation, Tokyo, Japan) consisted of a low pressure quater-
nary gradient unit (model PU-2089 Plus), a UV-Vis detector (model UV-2070 Plus) and
a manual injector Rheodyne (model 7725i). Separations were achieved on a Hypersil GOLD
HPLC column (Thermo Scientific, Waltham, USA) RP C18 (100 mm × 4.6 mm, 3 µm) in
isocratic conditions, 0.01% solution of sulfuric acid adjusted to pH 2.6 for 25 min (ref.15).
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

Polarographic Determination of Vitamin C
733
The flow rate was 1 ml min–1 and the wavelength used for detection was 245 nm. 20 µl of
the sample were injected into the chromatographic column kept at room temperature.
Chemicals and Samples
All chemicals were of analytical grade and were used without further purification. AA, NBS,
o-phosphoric acid, potassium iodate, starch and sodium thiosulfate were purchased from
Merck (Darmstadt, Germany), dehydroascorbic acid (DHA), OPDA, DL-dithiothreitol (DTT),
potassium iodide and the activated charcoal (Norit) were purchased from Sigma–Aldrich (St.
Louis, USA). Ultra-pure water from a Millipore Simplicity 185 water purification system
(Millipore, Billerica, USA) was used. 0.1 M Phosphate buffer solutions of pH 2, 3, 6, 7 and 8
were prepared by dissolving disodium phosphate (Merck), adjusting the pH with HCl
(Merck). Acetate buffer solutions of pH 4 and 5 were prepared in the same way, however, so-
dium acetate (Merck) was used instead. Solution of pH 1 consisted of a 0.1 M HCl solution.
The OPDA and NBS solutions were daily prepared by dissolving an appropriate amount of
those compounds in the buffer solution used for derivatization and kept in the dark. Suit-
able precautions were taken when using OPDA as it is toxic and may also cause allergic reac-
tions. The juice and enriched milk samples were purchased in a local supermarket and the
tablets in a local pharmacy.
Recommended Procedures
DHA analysis. In a voltammetric cell, 50 µl of the sample were added to 25 ml of 0.1 M
acetate buffer solution, pH 5. After removing oxygen with nitrogen for 10 min, 1 ml of
OPDA (1.25%) was added and the derivatizing reaction took place during 5-min period,
wherein nitrogen was continuously bubbled.
AA + DHA analysis. Equal to the DHA analysis with the sole difference that 200 µl of NBS
(250 mg l–1) were previously added and let react for 5 min in order to oxidize all AA into
DHA, before the addition of OPDA. AA was determined by calculating the difference be-
tween the results of these two analyses.
The method for Vitamin C chromatographic analysis was performed following the proce-
dure reported earlier15. Diluted juice samples were centrifuged at 4 °C for 15 min. The
supernatant was filtered through a 0.45 µm Nylon filter and 20 µl of the filtrate were di-
rectly injected in the chromatographic system to determine AA. The tablets were dissolved
in ultra-pure water and filtered prior to HPLC analysis. To determine total Vitamin C, 200 µl
of DTT solution (20 mg ml–1) were added to 1 ml of the filtered sample. DHA was calculated
as the difference between the Vitamin C and AA contents.
The determination of AA using the iodometric method was performed as follows: 2 g of
KI were weighed and placed in an Erlenmeyer flask along with 50 ml of 0.5 M H2SO4, 50 ml
of 0.3 mM KIO3 and about 10 g of juice (rigorously weighed), then 2 g of starch were added,
and the solution immediately became blue; subsequently, the mixture was titrated with 4
mM Na2S2O3, until the blue colour disappeared.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

734
Rodrigues, Valente, Gonçalves, Pacheco, Barros:
RESULTS AND DISCUSSION
Oxidation of AA and Derivatization of DHA
Activated charcoal was the first oxidant tested to convert AA into DHA. In
spite of its use in the spectrofluorimetric method of determination of AA
adopted by the AOAC 16, its application in the DPP method was not suc-
cessful because of the variation of the polarographic signal with the
amount of activated charcoal added (data not shown). However, the oxida-
tion of AA to DHA was successfully achieved using NBS 17, as shown in
Scheme 1. Although there is no direct polarographic signal of NBS alone,
a polarographic interference was found in the presence of OPDA, due to
a product formed by reaction between the excess of NBS and OPDA. The
amount of OPDA did not have a critical effect on the height of the interfer-
ing peak and the concentration of 0.05% OPDA was adopted, large enough
to obtain a complete derivatization of solutions up to 20 mg l–1 of DHA in
less than 5 min; in the case of NBS, it has to be in excess to completely oxi-
dize AA but not too much since, as mentioned, the reaction product with
OPDA starts to interfere in the polarographic measurement18. Accordingly,
different concentrations of NBS had to be stipulated for different calibra-
tion ranges of DHA: 1, 5 and 20 mg l–1 of NBS for DHA concentrations up
to 0.8, 4 and 20 mg l–1, respectively. Good linear calibrations were obtained
in the three regions, as can be observed in Table I, these calibration curves
also set the method limit of detection (LOD) and quantification (LOQ) at 3
and 10 µg l–1, respectively. The obtained relative standard deviations (RSD)
were around 3% in all three calibration curves.
Voltammetric Determination of DHA
According to Ohmori et al.19, the reaction between DHA and OPDA gives
rise to several products as shown in Scheme 2. In water, DHA (A) is predom-
inantly found as an hydrated bicyclic hemiacetal (B), 3,3,3a,6-tetra-
SCHEME 1
AA oxidation with NBS
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

Polarographic Determination of Vitamin C
735
hydroxytetrahydrofuro[3,2-b]furan-2-one20,21, which with time is further
hydrated to a lactone (C), 5-(1,2-dihydroxyethyl)-3,3,4,4-tetrahydroxy-
dihydrofuran-2-one22. Compound B reacts with OPDA giving rise to
3-(2,3,4-trihydroxytetrahydrofuran-2-yl)quinoxalin-2-one (D). On the other
hand, C reacts with OPDA to give the fluorescent 3-(1,2-dihydroxyethyl)-
furo[3,4]quinoxalin-1-one (E). Both D and E can further react with OPDA,
when such is present at a quantity large enough19, producing 2-(2-amino-
phenyl)-1-[3-(1,2,3-trihydroxypropyl)quinoxalin-2-yl]ethanone (F). These
products originate different voltammetric peaks: P1, P2 and P3 in Fig. 1
should correspond to species E, F and D in Scheme 2, respectively. All three
peaks shift to the more negative potentials with pH increase and P2 tends
to disappear, which can be explained because a higher pH kinetically
unfavours the reaction with OPDA 18. Compound E (P1) is a fluorescent
species used in the spectrofluorimetric method16. In this work, peak choice
for quantitative purposes was P3, i.e. compound D.
Peak P3 increases when lowering the pH of the supporting electrolyte,
with a clear linear behaviour23–26. However, at a pH lower than 4, the
voltammetric peak shifts to a more positive potential and the resulting
product from the reaction between NBS and OPDA starts to interfere with
the voltammetric measurement. Moreover, the derivatization reaction is
faster at a pH around 5 and the derivatized solution is stable for a longer
FIG. 1
Cyclic voltammograms at different pH, scan rate of 100 mV s–1, [DHA] = 1 mg l–1
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

736
Rodrigues, Valente, Gonçalves, Pacheco, Barros:
TABLE I
Linear calibration curves, ip = f ([DHA]/10–3 mg l–1) = m[DHA] + b, and regression coeffi-
cients, r2, for the DPP analysis of DHA with different NBS concentrations used
[NBS], mg l–1
[DHA], mg l–1
r2 (n)
ip, nA
1
5
0.01–0.8
0.05–4.00
0.2–15.0
0.9999(7)
0.9998(9)
0.9999(7)
0.171[DHA] + 0.5
0.161[DHA] – 2
0.135[DHA] + 22
20
SCHEME 2
Reaction between DHA and OPDA and resulting quinoxalines with suggested stereochemistry
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

Polarographic Determination of Vitamin C
737
period of time18,26. Therefore, pH 5 was used in the subsequent determina-
tions.
SWV gives greater intensity peaks, particularly at higher frequencies, than
DPP (Table II). Yet, the difference was not as substantially as it might have
been if P3 was not, unmistakably, an irreversible peak. However, DPP was
used preferably because it is more immune to the interference from the
product of the reaction between NBS and OPDA. Therefore, if an analyst
only wishes to analyse DHA (it does not require NBS), and not also AA,
SWV may be the preferable technique.
It is also known that at several pHs, including 5, part of the total DHA is
rapidly hydrolyzed to 2,3-diketogulonate21,27, and this chemical species also
reacts with OPDA since it is a vicinal diketone13. However, it has no influ-
ence in the developed methodology given that their wave potentials, at pH
3.44, are –0.40 and –0.27 V vs normal calomel electrode (NCE)28, i.e. they
overlap P1 and P2 but do not influence P3.
Determination of Vitamin C in Several Matrices
In the determination of AA and DHA in commercial fruit juices, two differ-
ent sample pre-treatments were tested. The pre-treatments involved filtra-
tion and centrifugation. The results obtained were the same with no
pre-treatment (data not shown), showing that the analysis can be per-
formed with the addition of a small quantity of fruit juice directly to the
polarographic cell containing 25 ml of the buffer solution with NBS con-
centration of 5 mg l–1.
In order to evaluate the method accuracy, different samples were ana-
lysed in triplicate by the developed methodology, using the method of
standard additions (Fig. 2). The validation was carried out using a volumet-
ric (iodometric) method commonly used29–31 and a chromatographic
method15. As can be seen from Fig. 3, the results obtained are statistically
TABLE II
Results comparison between DPP and SWV
[DHA], mg l–1 DPP, nA
SWV (100 Hz), nA
SWV (200 Hz), nA
SWV (400 Hz), nA
0.200
0.400
5.8
10.0
24.9
23.8
47.8
49.1
96.4
16.6
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

738
Rodrigues, Valente, Gonçalves, Pacheco, Barros:
FIG. 2
Differential pulse voltammograms of several standard additions of DHA, with concentration
from 0 to 53.6 mg l–1. Inset: Plot of the calibration curve obtained
FIG. 3
Determination of Vitamin C by the developed methodology and a reference methodology;
samples A were compared with an iodometric procedure and samples B and C with a recog-
nized HPLC procedure; samples 1A–5A as well as 1B and 2B are several kinds of fruit juices,
sample 3B is an enriched milk, samples 1C and 2C are tablets. In both methodologies: prior to
analysis, 5% of o-phosphoric acid was added to sample 3B, to promote protein coagulation;
prior to analysis, samples 1C and 2C were diluted in water
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

Polarographic Determination of Vitamin C
739
consistent, confirming the accuracy of the developed voltammetric
method.
AA and DHA contents were measured in three different juice samples dur-
ing 4 days following the opening of their packages (Fig. 4). This showed the
usefulness of the methodology in evaluating AA resistance to deterioration
by oxidation and subsequent DHA appearance. By analysis of Fig. 4, it is
possible to conclude that the disappearance of AA is quite fast and that the
FIG. 4
Evolution of AA and DHA of three different fruit juices kept at room temperature after the
package opening
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

740
Rodrigues, Valente, Gonçalves, Pacheco, Barros:
first product of its degradation is DHA, as it would be expected. It can also
be noticed that the sum of AA and DHA is practically unchanged during
the first two days and then starts to decrease. These observations are in
agreement with the results reported earlier32–35. Another remark is that the
juice stability is higher with lower pH: orange juice – pH 2.8; blend juice
(“tuttifruti”) – pH 3.0 and pineapple juice – pH 3.8, is related, among other
factors, with the different stability of ascorbic acid and ascorbate (pKa1
=
4.2). Again, these results were coherent with bibliographic data36,37
.
CONCLUSIONS
A DPP method was developed for the determination of Vitamin C
(AA + DHA). The proposed method consists of the voltammetric analysis of
one of the quinoxalines formed by the DHA derivatization with OPDA.
Total Vitamin C was determined by previous AA oxidation to DHA with
NBS; AA content was calculated by the difference between total Vitamin C
and DHA. This method was successfully applied to the determination of
Vitamin C in several matrices with no sample pre-treatment and the results
agreed with the chromatographic method widely used. Degradation studies
of Vitamin C in fruit juice samples were also performed, showing that AA is
readily converted to DHA over time along with the expected overall de-
crease of Vitamin C. The developed methodology can be a good alternative
to other analytical methods for Vitamin C determination due to its specific-
ity and simplicity, low reagent consumption and no required sample pre-
treatment.
L. M. Gonçalves (SFRH/BD/36791/2007) and J. G. Pacheco (SFRH/BD/30279/2006) wish to
acknowledge Portuguese Fundação para a Ciência e a Tecnologia (FCT) for their Ph.D. studentships.
REFERENCES
1. Romeu-Nadal M., Morera-Pons S., Castellote A., López-Sabater M.: J. Chromatogr. B 2006,
830, 41.
2. Özkan M., Kirca A., Cemeroglu B.: Food Chem. 2004, 88, 591.
3. FAO/WHO: Vitamin and Mineral Requirements in Human Nutrition, 2004.
4. Arya S., Mahajan M., Jain P.: Anal. Sci. 1998, 14, 889.
5. Arya S., Mahajan M., Jain P.: Anal. Chim. Acta 2000, 417, 1.
6. Özyürek M., Güçlü K., Bektasoglu B., Apak R.: Anal. Chim. Acta 2007, 588, 88.
7. Gazdik Z., Zitka O., Petrlova J., Adam V., Zehnalek J., Horna A., Reznicek V., Beklova M.,
Kizek R.: Sensors 2008, 8, 7097.
8. Esteve M., Farré R., Frigola A., López J., Romera J., Ramirez M., Gil A.: Food Chem. 1995,
52, 99.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741

Polarographic Determination of Vitamin C
741
9. Rodrigues J., Miranda P., Barros A.: Port. Electrochim. Acta 1993, 11, 41.
10. Cherdkiatgumchai P., Grant D.: Cereal Chem. 1987, 64, 288.
11. Kozar S., Bujak A., Eder-Trifunović J., Kniewald G.: Fresenius’ J. Anal. Chem. 1988, 329,
760.
12. Lechien A., Valenta P., Nürnberg H., Patriarche G.: Fresenius’ J. Anal. Chem. 1982, 311,
105.
13. Rodrigues J., Barros A., Rodrigues P.: J. Agric. Food Chem. 1999, 47, 3219.
14. Raoof J., Ojani R., Rashid-Nadimi S.: Electrochim. Acta 2004, 49, 271.
15. Odriozola-Serrano I., Hernández-Jover T., Martín-Belloso O.: Food Chem. 2007, 105, 1151.
16. Association of Official Analytical Chemists: Official Method 967.22, 2006.
17. Barakat M., El-Wahab M., El-Sadr M.: Anal. Chem. 1955, 27, 536.
18. Rodrigues J. A. M.: Ph.D. Thesis. University of Porto, Porto 1998.
19. Ohmori M., Tsujimoto Y., Takagi M.: Bull. Chem. Soc. Jap. 1983, 56, 2033.
20. Pfeilsticker K., Marx F., Bockisch M.: Carbohydr. Res. 1975, 45, 269.
21. Deutsch J.: Anal. Biochem. 1998, 260, 223.
22. Hvoslef J., Pedersen B.: Acta Chem. Scand., Ser. B 1979, 33, 503.
23. Ahmed R., Qureshi S., Viqar-Un-Nisa: J. Chem. Soc. Pakistan 2005, 27, 168.
24. Ruiz J., Aldaz A., Dominguez M.: Can. J. Chem. 1977, 55, 2799.
25. Ruiz J., Rodriguez-Mellado J., Dominguez M., Aldaz A.: J. Chem. Soc., Faraday Trans. 1
1989, 85, 1567.
26. Sahbaz F., Somer G.: Food Chem. 1992, 44, 141.
27. Deutsch J.: J. Chromatogr., A 2000, 881, 299.
28. Wasa T., Takagi M., Ono S.: Bull. Chem. Soc. Jpn. 1961, 34, 518.
29. Bailey D.: J. Chem. Educ. 1974, 51, 488.
30. Silva C., Simoni J., Collins C., Volpe P.: J. Chem. Educ. 1999, 76, 1421.
31. Verdini R., Lagier C.: J. Agric. Food Chem. 2000, 48, 2812.
32. Solomon O., Svanberg U., Sahlström A.: Food Chem. 1995, 53, 363.
33. Giannakourou M., Taoukis P.: Food Chem. 2003, 83, 33.
34. Serpen A., Gökmen V.: Food Chem. 2007, 104, 721.
35. Zerdin K., Rooney M., Vermuë J.: Food Chem. 2003, 82, 387.
36. Wilson R., Beezer A., Mitchell J.: Thermochim. Acta 1995, 264, 27.
37. Takayanagi T., Nishiuchi M., Ousaka M., Oshima M., Motomizu S.: Talanta 2009, 79,
1055.
Collect. Czech. Chem. Commun. 2010, Vol. 75, No. 7, pp. 731–741