Moghaddam et al

Tropical Journal of Pharmaceutical Research January 2016; 15 (1): 183-188

ISSN: 1596-5996 (print); 1596-9827 (electronic)
© Pharmacotherapy Group, Faculty of Pharmacy, University of Benin, Benin City, 300001 Nigeria.
All rights reserved.

Available online at http://www.tjpr.org

http://dx.doi.org/10.4314/tjpr.v15i1.25
Original Research Article

Determination of Synthetic Food Colors, Caffeine, Sodium

Benzoate and Potassium Sorbate in Sports Drinks
Fatemeh Zamani Mazdeh1, Zeinab Moradi1, Ghazaleh Moghaddam1, Zhila
Moradi-Khatoonabadi1, Farideh Esmaeili Aftabdari1, Parnaz Badaei2 and
1,2
Mannan Hajimahmoodi *
1 2
Food and Drug Administration, Department of Drug and Food Control, Faculty of Pharmacy, Tehran University of Medical
Sciences, Tehran, Iran

*For correspondence: Email:

Received: 26 February 2015 Revised accepted: 11 November 2015

Abstract
Purpose: To employ high performance liquid chromatography (HPLC) techniques to determine edible
additives in commercial sports drinks.
Methods: A total of 105 samples including 21 different brands of sports drinks were purchased from
markets in Tehran, Iran. The process conditions included a C18 column, a mobile phase consisting of
–1
aqueous ammonium acetate buffer (pH = 4.2) and acetonitrile (80:20 v/v) at a flow rate of 0.8 ml min .
The injection volume was 20 μL and the UV detector was set at 225 nm.
Results: Mean recovery was between 95 and 106 %. The mean concentrations of sodium benzoate
and potassium sorbate in Iranian brands (147.72 and 11.54 ppm, respectively) were significantly higher
than in foreign brands (19.43 and 4.91 ppm, respectively, p < 0.05). The maximum amount of sodium
benzoate and potassium sorbate was 251.50 ± 18.69 and 96.38 ± 38.56 ppm, respectively, while
caffeine content was in the range of 293.48 ± 14.33 - 607.32 ± 135.33 ppm.
Conclusion: HPLC permits the detection of sodium benzoate and potassium sorbate at very low
concentrations. Furthermore, caffeine was detected in all the sport drinks. Only Brilliant Blue and Allura
Red were detected in the drinks.

Keywords: Edible additives, Sports drinks, Sodium benzoate, Potassium sorbate, Caffeine, Brilliant
Blue, Allura Red

Tropical Journal of Pharmaceutical Research is indexed by Science Citation Index (SciSearch), Scopus,
International Pharmaceutical Abstract, Chemical Abstracts, Embase, Index Copernicus, EBSCO, African
Index Medicus, JournalSeek, Journal Citation Reports/Science Edition, Directory of Open Access Journals
(DOAJ), African Journal Online, Bioline International, Open-J-Gate and Pharmacy Abstracts

INTRODUCTION as additives in energy and sport drinks [3-5].

Benzoic acid is sodium salt of benzene
Recently, sport drinks have gained popularity carboxylic acid and phenilcarboxylic acid, is 180
based on the attribution of energy-giving times more soluble than benzoic acid in water
properties. Caffeine which is almost certainly the [6,7]. The additives may also cause allergic
most widely consumed psychoactive substance reactions in sensitive individuals as well as
in the world is a well-known stimulant effect over hyperactivity color. However, potassium solubility
the central nervous system [1]. The US Food and is more than 50 % in foods [6]. In Europe
drug Administration (FDA) limits the maximum sorbates and benzoates are permitted in
amount in carbonated beverages to 6 mg/oz. [2]. beverages [8].
On the other hand, preservatives are permitted

Trop J Pharm Res, January 2016; 15(1): 183

 Moghaddam et al

For safety reasons, there have been recent chromatograph equipped with a gradient pump
reductions in the number of permitted food colors capable of mixing four solvents, a vacuum
but because of their low price, effectiveness and membrane degasser, a 20 μL loop injector and a
stability they are still being used [9–11]. The lists UV Detector (Agilent Technologies, Santa Clara,
of permitted food colors is definitely different CA, USA). Analysis was performed on an Eclipse
between countries [12]. The beverage - XDB C18 column (150 × 4.6 mm, 5 μm). The
processing industry uses several types of food mobile phase consisted of an aqueous
colors, but to minimize potential toxicity, the ammonium acetate buffer (pH = 4.2) and
amounts of permitted synthetic colors used are acetonitrile (80:20 v/v) with a flow rate of 0.8 ml
strictly limited [11,13]. So they are permitted and min – 1. The injection volume was 20 μL and the
UV detector was set at 225 nm. The
frequently used in EU countries. However, they
chromatographic system was initially conditioned
are not permitted in other countries, including
by the mobile phase until a stable baseline signal
Japan and USA [14]. was obtained, which needed at least 1 h.
The main objectives of this work were to evaluate To evaluate the synthetic colors, the mobile
the levels of mentioned additives compounds in phase consisted of ammonium acetate solution
105 sport drinks and by comparing the levels of (0.1 mol L-1, pH = 6.7), as solvent A and
these compounds with the current available methanol - acetonitrile (50:50, v/v), as solvent B.
standards. A cellulose membrane with a pore diameter 0.45
μm was used to filter mobile phase A. In order to
EXPERIMENTAL achieve a successful resolution, gradient elution
programs were tested and the flow rate was
A total number of 105 samples were purchased always kept constant at 1 ml min-1. The final
in accordance with the market availability, in optimized gradient program was 3 % solvent B
Tehran - Iran. The samples were collected as the initial step; then it increased linearly to 60
between December, 2013 and January. 2014. % in 18 min and was held for 2 min. Analysis
The studied commodities were 21 different was performed on the UV detector with two
brands of sports drinks. All stages of the optimized conditions. First, the fixed wavelength
experiment were done before expiry date of the was set at 250 nm within the whole runtime to
drinks. The analyses were carried out in triplicate make it easy to use for amateur operators; next,
for each sample. it was programmed at 415 nm for tartrazine, 460
nm for Quinoline yellow, Ponceau 4R, and
Standards and chemicals Sunset yellow, 500 nm for Allura red AC and
Carmosine, and 600 nm for Indigo carmin and
All solvents/chemicals used were of analytical Brilliant blue for different colorants [11].
grade. In the study, high purity standards of
potassium sorbate (> 99 %), sodium benzoate (> Data analysis
99 %), sodium salicylate (> 99.5 %), Caffeine (>
99 %), ammonium acetate (98 %), acetic acid All measurements were replicated three times to
glacial (100 %) and HPLC - grade acetonitrile improve the reliability of the results. Data were
were purchased from Merck (Darmstadt, analysed using statistical program for social
Germany). Deionised water was prepared by a sciences (SPSS), version 21 (IBM SPSS Inc.,
Branstead Easypure II system (Thermo Fisher Chicago, USA), Data are expressed as mean ±
Scientific Inc., Waltham, MA, USA). Methanol SD. One way analysis of variances (ANOVA -
(Merck) was HPLC quality. Acetic acid (Merck, d Dunnett T3) was used for determining significant
= 1.05) that was used for acetate buffer and difference which was set at p < 0.05.
NaOH (Merck, 5 mol/L) that was used for
regulating pH were of analytical purity. Millipore Method validation
Milli - Q water was used in all stages of the
assay. The LOD and the LOQ were calculated as LOD =
3.3 σ/S and LOQ = 10 σ/S, where σ is the
Chromatographic conditions standard deviation of the response and S is the
slope of the calibration curve. The estimate of σ
Sport drinks were filtered through 0.45 µm was carried out by analyzing blank samples
membrane filter and ultrasonicated, before HPLC seven times, measuring of the magnitude of the
analysis. Analytical separation of the caffeine, analytical background response and then
sodium benzoate and potassium sorbate was calculating the standard deviation of the
carried out by reverse phase liquid response. Recoveries were determined by
chromatography of an Agilent 1200 series liquid spiking a sport drink known to be free of all

Trop J Pharm Res, January 2016; 15(1): 184

 Moghaddam et al

additives, in triplicate, with known amounts of and also caffeine, while in Figure 2 the synthetic
studied additives at final concentrations between. dyes resolution are illustrated.

RESULTS The concentrations of preservatives quantified

from sports- drink samples are shown in Table 2.
The analytical methodology revealed good Sodium benzoate was found in 14 brands at
linearity, sensitivity, exactitude and precision. levels ranging between 12.50 and 251.50 ppm,
Correlation coefficients (r2) were comprised (Table 2).
between 0.998 for caffeine acid and 0.999 for the
other studied factors. Limits of qualification and Moreover, 15 brands didn’t have any potassium
quantification are presented in Table 1 sorbate. The levels of sodium benzoate and
respectively. potassium sorbate in Iranian brands (147.72 and
11.54 ppm, respectively) were significantly more
Recovery values were between 95 % and 106 % than foreign studied brands (19.43 and 4.91 ppm
for studied additives at determined spiked levels. respectively).
The developed method was successfully in the
evaluation of the mentioned additives. Figure 1 Also, methods to evaluate levels of coloring in
shows the chromatograms of a standard solution foods for quality control are currently made by
containing sodium benzoate, potassium sorbate the TLC method. The mean concentration of
colors in each sample group is presented in
Table 3.

Table 1: Recovery data for the additives in sports drinks

2
Additive Calibration Data R LOD (ppm) LOQ (ppm) Recovery (%)
Sodium benzoate y=121.3x-56.82 0.999 0.6413 1.9435 99- 102
Potassium sorbate y=69.61x+36.64 0.999 1.1224 3.4013 98- 100
Caffein y=39.31x+9.87 0.998 1.8263 5.5344 97- 101
Quinoline Yellow y = 38.71x - 6.952 0.999 0.465 1.409 96- 105
Sunset Yellow y = 47.20x - 13.56 0.999 0.357 1.082 98- 106
Carmoisine y = 62.40x - 16.47 0.999 0.666 2.019 95- 105
Ponceau 4R y = 21.98x - 6.095 0.999 0.873 2.646 97- 103
Allura Red y = 70.08x - 10.09 0.999 0.510 1.547 95- 102
Indigotine y = 56.71x - 12.95 0.999 0.431 1.307 99- 101
Brilliant Blue y = 100.2x - 19.83 0.999 0.562 1.704 95- 100

Fig 1: Chromatograms of a standard solution containing sodium benzoate, potassium sorbate and also caffeine.

Trop J Pharm Res, January 2016; 15(1): 185

 Moghaddam et al

Fig 2: Resolution of chromatograms of synthetic dyes. 1 – Tartrazine; 2 - Quinoline yellow 1; 3 - Indigo carmine;
4 - Quinoline yellow 2; 5 - Ponceau 4R; 6 - Sunset yellow; 7 - Quinoline yellow 3; 8 - Allura red AC; 9 –
Carmozine; 10 - Brilliant blue

DISCUSSION National Standards [15] have some stringent

rules for regulating the use of food additives. The
Additives perform a variety of useful functions in level of sodium benzoate and potassium sorbate
foods that consumers often take for granted. in Iranian standard (6693) [15] is 150 and 500
Some additives could be eliminated if we were ppm, respectively. Therefore, based on the
willing to grow our own food, harvest and grind it, achieved results, 26.66 % of samples had higher
spend many hours cooking and canning, or sodium benzoate but all samples were in the
accept increased risks of food spoilage. Iranian range of sorbate.

Table 2: Sodium benzoate and potassium sorbate concentrations (ppm) in different brands of sports-drink

Sodium benzoate Potassium sorbate

Brand Mean±SD Range Mean±SD Range
A 24.77±0.93 24.25-26.18 0 0
B 12.50±16.67 24.28-33.35 0 0
C 21.76±2.20 18.66-23.35 0 0
D 18.19±1.91 16.08-21.09 0 0
E 0 0 0 0
F 0 0 35.01±4.22 31.04-39.71
G 28.41±5.03 23.98-36.37 0 0
H 0 0 0 0
I 0 0 0 0
J 0 0 0 0
K 0 0 0 0
L 0 0 0 0
M 242.34±24.83 202.52-285.46 53.26±36.71 0-78.85
N 231.07±23.76 204.51-257.86 96.38±38.56 62.28-139.92
O 250.24±36.29 226.26-304.29 15.89±31.79 0-63.59
P 251.50±18.69 226.91-269.30 66.38±41.10 0-108.31
Q 144.99±77.02 17.75-218.55 28.80±29.51 0-66.61
R 138.43±17.74 123.24-150.93 0 0
S 163.14±18.40 131.99-178.51 0 0
T 105.18±6.62 94.30-111.86 0 0
U 95.66±109.51 18.23-173.10 0 0

Trop J Pharm Res, January 2016; 15(1): 186

 Moghaddam et al

Table 3: Caffeine and color concentrations (ppm) in different brands of sports-drink

Caffeine Color
Brand Mean ±SD Range Color Type Mean ±SD Range
A 345.21±114.24 174.45-389.60 0 0
B 443.87±36.75 403.21-496.36 Brilliant Blue 32.13±3.78 27.37-36.48
C 354.72±117.98 179.34-431 Allura Red 49.87±12.72 68.84-42.54
D 471.42±16.57 452.46-486.96 0 0
E 490.76±27.86 460-526.43 0 0
F 446.3±23.93 425.02-481.34 0 0
G 490.07±49.38 425.45-549.46 0 0
H 459.82±19.87 437.97-478.74 0 0
I 453.06±15.39 426.76-463.31 0 0
J 441.25±15.45 420.66-456.79 0 0
K 484.93±17.55 465.02-511.57 0 0
L 464.85±15 449.66-484.42 0 0
M 543.78±80.99 448.26-624.28 Allura Red 6.92±0.72 7.71-6.30
N 506.21±35.65 463.14-548.39 0 0
O 502.65±53.29 467.16-579.93 0 0
P 502±49.07 449.8-574.39 0 0
Q 607.32±135.33 503.61-836.99 0 0
R 514.02±36.16 499.39-572.56 0 0
S 293.48±14.33 284.77-314.68 0 0
T 547.72±223.68 235.66-776.31 0 0
U 531.28±37.83 504.53-558.04 Allura Red 14.54±0.52 14.92-14.17

For some foodstuff in the study by Mota et al 0.35 ppm). The range of caffeine content in the
[16], sorbic acid and/or benzoic acid were carbonated soft drinks group is between 43.71
reported to range from not detected (n.d.) to 210 and 45.83 ppm; while in it is from 47.56 to 58.31
and n.d. to 153 mg/L, respectively. All sample ppm in the energy drink samples [19]. The mean
concentrations of these compounds were below caffeine content of carbonated soft drink was
the stipulated limit of 300 mg/L for sorbic acid lower (44.52 ppm) than in the energy drinks
alone or 150 mg/L for benzoic acid alone [16]. In (52.24 ppm. It was also observed that the
the three different samples of commercial wines, caffeine level of the diet Cola (45.83 ± 1.05) was
only sorbic acid was detected below 200 mg/L, slightly higher compared to the regular cola
which complied with the legislation for drinks (Pepsi cola and Coca - cola). This is in
preservatives in wines [16]. In another related agreement with previous work reported by Ali et
research which the benzoic and sorbic acids al [20].
were in quince jam, all quince jams contained
benzoic acid in concentrations ranging from Food color additives are limited currently and
413.9.10.46 to 1501.4.2 mg of benzoic acid/kg only seven synthetic colors are permitted under
[17]. In another study, it was shown that the these regulations. Color additives are used in
amounts of benzoic and sorbic acid in jam were foods for many reasons: to off-set color loss due
639 mg/kg and 789 mg/kg, respectively. to exposure to light, air, temperature extremes,
Furthermore, the benzoic acid content of five of moisture and storage conditions; to correct
jam samples in Ferreira et al was above the legal natural variations in colors; to enhance colors
limits of 500 mg/kg in Portugal [17]. that occur naturally; and to provide color to
colorless and “fun” foods. In the current study, all
Although 300 mg/day of caffeine consumption is studied colors were not present in the sports
considered generally safe [18], there is no rapid drinks, but Brilliant blue and Allura Red were
method to determine caffeine content of sports- detected. In contrast to other countries, Ponceau
drinks. In this study, the caffeine determination in 4R is prohibited in USA and Norway. Brilliant
samples was in the range of 345.21 - 543.78 Blue is prohibited in many European countries.
ppm in Iranian sports-drink samples. However, Tartrazine is also forbidden in the USA and
based on the results, other imported brands Austria [11]. Based on the Iranian standard [15],
range was 293.40 - 607.33 ppm. Therefore, it there was no approved limits for sports drinks.
must be asserted that caffeine amount in all Therefore, based on the achieved results, it
samples was higher than the standard limitation seems that approved standards need to be
of Iran (150 ppm) [15]. The lowest level of revised both in terms of color type and related
caffeine was detected in the soft drink, Coca quantitative limitations made, regarding further
Cola (43.71 ± 0.55), while the energy drinks risk assessments.
showed the highest level of caffeine (58.31 ±

Trop J Pharm Res, January 2016; 15(1): 187

 Moghaddam et al

CONCLUSION 8. Costa AC, Perfeito Lda S, Tavares MF, Micke GA.

Determination of sorbate and benzoate in beverage
HPLC provides a suitable method for the samples by capillary electrophoresis—Optimization of
detection of caffeine, sodium benzoate and the method with inspection of ionic mobilities. J
potassium sorbate at very low concentrations in Chromatogr A 2008; 1204: 123-127.
sports drinks. It can separate both preservatives 9. Hajimahmoodi M, Oveisi MR, Sadeghi N, Jannat B,
at one wavelength in < 10 min and involves Nilfroush E. Simultaneous determination of Carmoisine
minimal sample preparation. The content of and Ponceau 4R. Food Anal Method 2008; 1: 214-219.
potassium sorbate was below legal limits. Non- 10. Tavakoli M, Shemirani F, Hajimahmoodi M. Magnetic
compliance with legal limits was only mixed hemimicelles solid-phase extraction of three food
encountered for sodium benzoate and synthetic colorants from real samples. Food Anal Method 2014; 7:
colors in some samples. 100-108.
11. Hajimahmoodi M, Afsharimanesh M, Moghaddam G,
ACKNOWLEDGEMENT Sadeghi N, Oveisi MR, Jannat B, Pirhadi E, Zamani
Mazdeh F, Kanan H. Determination of eight synthetic
This work was supported by a grant from Tehran dyes in foodstuffs by green liquid chromatography. Food
University of Medical Sciences (no. 91–01–33- Addit Contam A 2013; 30: 780-785.
17143). 12. Khera K, Munro IC, Radomski JL. A review of the
specifications and toxicity of synthetic food colors
REFERENCES permitted in Canada. CRC Cr Rev Toxicol 1979; 6: 81-
133.
1. Aranda M, Morlock G. Simultaneous determination of 13. Khanavi M, Hajimahmoodi M, Ranjbar AM, Oveisi MR,
riboflavin, pyridoxine, nicotinamide, caffeine and taurine Shams Ardekani MR, Mogaddam G. Development of a
in energy drinks by planar chromatography-multiple green chromatographic method for simultaneous
detection with confirmation by electrospray ionization determination of food colorants. Food Anal Method
mass spectrometry. J Chromatogr A 2006; 1131: 253- 2012; 5: 408-415.
260. 14. Yoshioka N, Ichihashi K. Determination of 40 synthetic
2. Barone J, Roberts H. Caffeine consumption. Food Chem food colors in drinks and candies by high-performance
Toxicol 1996; 34: 119-129. liquid chromatography using a short column with
3. Dong C, Wang W. Headspace solid-phase photodiode array detection. Talanta 2008; 74: 1408-
microextraction applied to the simultaneous 1413.
determination of sorbic and benzoic acids in beverages. 15. Iranian Standard. http://isiri.ir/.
Anal Chim Acta 2006; 562: 23-29. 16. Mota FJM, Ferreira IM, Cunha SC, Oliveira MB.
4. Techakriengkrai I, Surakarnkul R. Analysis of benzoic Optimisation of extraction procedures for analysis of
acid and sorbic acid in Thai rice wines and distillates by benzoic and sorbic acids in foodstuffs. Food chem 2003;
solid-phase sorbent extraction and high-performance 82: 469-473.
liquid chromatography. J Food Compos Anal 2007; 20: 17. Ferreira IM, Mendes E, Brito P, Ferreira MA.
220-225. Simultaneous determination of benzoic and sorbic acids
5. Wen Y, Wang Y, Feng YQ. A simple and rapid method in quince jam by HPLC. Food Res Int 2000; 33: 113-
for simultaneous determination of benzoic and sorbic 117.
acids in food using in-tube solid-phase microextraction 18. Saad B, Bari MF, Saleh MI, Ahmad K, Talib MK.
coupled with high-performance liquid chromatography. Simultaneous determination of preservatives (benzoic
Anal bioanal chem 2007; 388: 1779-1787. acid, sorbic acid, methylparaben and propylparaben) in
6. Koyuncu N, Uylaşer V. Determination of benzoic and foodstuffs using high-performance liquid
sorbic acid in turkish food using high‐ performance liquid chromatography. J Chromatogr A 2005; 1073: 393-397.
chromatography. J Food Process Pres 2009; 33: 361- 19. Amos-Tautua W, Diepreye E. Ultra-violet
369. Spectrophotometric Determination of Caffeine in Soft
7. Esfandiari Z, Badiey M, Mahmoodian P, Sarhang-pour R, and Energy Drinks Available in Yenagoa, Nigeria. Adv J
Yazdani E, Mirlohi M. Simultaneous Determination of Food Sci Technol 2014; 6: 155-158.
Sodium Benzoate, Potassium Sorbate and Natamycin 20. Ali MM, Eisa M, Taha Ml, Zakaria BA, Elbashir AA.
Content in Iranian Yoghurt Drink (Doogh) and the Determination of Caffeine in Some Sudanese
Associated Risk of Their Intake through Doogh Beverages by High Performance Liquid
Consumption. Iran J Public Health 2013; 42: 915-920. Chromatography. PJN 2012; 11: 336.

Trop J Pharm Res, January 2016; 15(1): 188