Antioxidant Capacity of Common Beverages

Antioxidant Capacity of Common Beverages

Determination of The Total Antioxidant Capacity of Beverages Using Ferric Reducing Antioxidant Power (FRAP) Assay

Author: Benjamin R. Holmes



ABTS - 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulphonic acid)

AOC – Antioxidant Capacity

DNA - Deoxyribonucleic acid

DPPH - 2,2-diphenyl-1-picrylhydrazyl

FRAP – Ferric Reducing/Antioxidant Power or Ferric Reducing Ability of Plasma

GAE – Gallic Acid Equivalence

HAT – Hydrogen Atom Transfer

ORAC - Oxygen Radical Absorbance Capacity

ROS - Reactive Oxygen Species

SET – Single Electron Transfer

TEAC - Trolox Equivalent Antioxidant Capacity

UV – Ultraviolet

UV/V – Ultraviolet Visible


Antioxidants can be defined as “any substance that, when present at low concentrations compared to those of an oxidisable substrate, significantly delays or prevents oxidation of that substrate” (Halliwell and Gutteridge, 1995). There are three main types of compounds that have AOC in nature: vitamins, enzymes and phytochemicals, they work by donating electrons to ROS which neutralises and prevents them from causing damage to essential molecules. ROS is a term given to compounds that are chemically reactive that contain oxygen, like peroxides, superoxide, hydroxyl radical, singlet oxygen, alpha-oxygen and ozone. Some of these are important for maintaining homeostasis and as signalling molecules at a cellular level and free radical levels can alter as a result of metabolism, such as converting food to energy and exercise and exposure to the environment like pollution or UV radiation. These levels can exceed a point at which they are no longer safe, this can be described as oxidative stress and can affect lipids by a process called lipid peroxidation which causes damage to cell membranes by stealing electrons from the lipid molecules contained within them. They can also reduce enzymatic activity by degradation of proteins and cause damage to DNA which can result in mutagenesis and carcinogenesis.

There are many different assays in which we can test the AOC a food or drink holds; each assay will test a samples AOC via a different mechanism so understanding those mechanisms ensures the correct assay is chosen. Antioxidant assays can be classified in two distinct ways, assays based on HAT reactions, which measure the ability of an antioxidant to neutralise ROS via a hydrogen atom. The other is based on SET, this method detects the ability of an antioxidant to transfer one electron to reduce a ROS (Huang, Ou and Prior, 2005; Prior, Wu and Schaich, 2005). A FRAP assay was used in this analysis to determine the antioxidant capacity of common beverages using ferrous equivalent as a comparison metric. The FRAP assay was first used by Benzie and Strain (1996) and is often used to test antioxidant capacity of foods and beverages containing polyphenols, the test utilises the SET reaction. It is quick, simple to perform and cheap, therefore almost all laboratories can carry out this method of testing AOC. It measures the reduction of ferric to ferrous ion at low pH in the presence of an antioxidant, this causes a blue colour to develop when a ferrous-tripyridyltriazine complex forms which is absorbed at 593nm with a spectrophotometer (Figure 1). Values are obtained by comparing the absorbance change in test reaction mixtures with those containing ferrous ions in known concentrations, this allows for the quantification of nonspecific electron donating antioxidant content in samples when tested using the same method.

Figure 1. Chemical diagram created by Pérez-Cruz et al (2018) “Redox reaction for ferric complex in the FRAP assay.”


Materials and Methods

  • Components and amounts of the FRAP reagent and FeSO4 stock standard solution are listed in Appendix A.
  • Images and manufacturer names of all the equipment used are contained in Appendix D.
  • Images and brand names of all beverages tested can be found in Appendix E.

The standards of FeSO4 were made by diluting a stock solution (10.0mmol/l-1 FeSO4) with distilled water using volumetric glassware. Standards of 10, 8, 6, 4 and 2mmol/l-1 FeSO4 were produced to make a total of 10ml. 20ul of each was introduced to five cuvettes, 460µl of distilled water was then introduced containing the standards and 480μl of distilled water was added to an additional cuvette that would be used as the zeroing sample in the spectrophotometer. 3.6ml of FRAP reagent that was being incubated at 37ºC was then introduced to all cuvettes including the zeroing sample, cuvettes were then vortexed and submerged at 37ºC for 10 minutes. After 10 minutes the standards and zeroing sample were taken out and the zeroing sample was used to zero the spectrophotometer at 593nm, three repeats were then carried out on each of the standards at the same wavelength and a mean of the results was taken to create a calibration curve (Figure 2).

A centrifuge was run on all the juice beverages: orange, mango, tomato, apple and cranberry juice at 124RCF for 5 minutes and then 1750RCF for another 5 minutes. The coffee and tea were brewed using 2g of granules/loose leaves and 200ml of boiling water for 5 minutes. The coffee and tea were then filtered through a paper filter. All the beverages were then filtered through a syringe filter with a pore size of 0.2μm. Further cuvettes were made up for each filtered test sample using 20ul of filtered beverage, 460ul of distilled water and 3.6ml of FRAP reagent (incubated at 37ºC). All cuvettes containing samples were then vortexed and submerged at 37ºC for 10 minutes along with the zeroing sample again. After 10 minutes, the spectrophotometer was zeroed for a second time with the zeroing sample at 593nm and three repeats were then carried out on each of the beverage samples at the same wavelength and a mean of the results was taken to be used to extrapolate the ferrous equivalent for each sample from the calibration curve line equation.

Figure 2. A scatter graph showing the absorbance of the FeSO4 standards against their concentrations. Trendline and R2 value (0.9866) showing linearity (Raw data in Appendix B, Table 1).


The beverages tested using the FRAP assay are expressed in ferrous equivalent values (mmol Fe(II)/L) and are displayed in a column chart (Figure 3). Coffee held the highest (12.65mmol/L), with mango juice being the second highest at 10.36mmol/L and cranberry juice at 10.20mmol/L. English breakfast tea (7.56mmol/L) was greater than Red Bush tea (6.45mmol/L) and the two lowest beverages were tomato juice (4.33mmol/L) and apple juice (4.16mmol/L).

Figure 3. A column chart showing the FRAP value (mmol Fe(II)/L) in common beverages using a FRAP assay (Raw data in Appendix B - Table 2).


The FRAP reaction is very reproducible, the absorbance values in Figure 2 shows high linearity related to the reductant concentration over a wide range and this has been replicated in other studies at much lower standard concentrations of FeSO4 (Table 1). There is significant difference (5-10x) in FeSO4 ranges in 9 studies that used a FRAP assay with FeSO4, compared to the method used here (Table 1). This could be due to different instruments being used to perform the analysis (requiring lower concentrations to maintain linearity) or protocols being optimised differently to suit the equipment and resources available to a laboratory. However, this makes comparing the results quantatively, difficult across studies, even when assays are matched.

Table 1. Standard concentration ranges of FeSO4 across different studies that used a FRAP assay with FeSO4 as standards.
Standard concentration range (FeSO4 mmol/L)


Comparisons of the results can be made across studies through correlative analysis, the results expressed in Figure 3 show that coffee had the highest reducing power compared to other beverages, this was also reported by Carlsen et al (2010) where 3100 different foods and beverages were tested for AOC. It showed a significant correlation between the data in the order of AOC in beverages that were tested compared with the results (Figure 4). Coffee is known for its high phenolic content, and due to FRAP’s high correlation to assays that specifically test for this, this could explain the reason for this result and why multiple studies have shown coffee among other beverages to contribute the most to phenolic content within dietary intake across different countries (Pulido, Hernández-García and Saura-Calixto, 2003; Ovaskainen et al., 2008; Serafini and Testa, 2009; Pérez-Jiménez et al., 2011).

Figure 4. Two column charts showing the rank order of results testing similar common beverages
(A– Figure 3
(B) - (Carlsen et al., 2010) Correlation coefficient of 0.89 – See Appendix C for statistical methods.


A disadvantage with the FRAP assay is that it does not react with thiols due to the reduction potential of thiols being below the Fe(III)/Fe(II) half-reaction (Halvorsen et al., 2002), however humans absorb a very limited amount of plant glutathione and no other thiols are present in dietary plants (Schafer and Buettner, 2001). Halvorsen et al (2002) suggests this still makes the FRAP assay suitable for testing total antioxidant content in plants. There can also be interference that can occur with UV-V absorption at 593nm by compounds other than Fe(TPTZ)2(II). Benzie and Strain (1999) reported a very high FRAP value for bilirubin and when bilirubin is oxidized, it is transformed to biliverdin which has a strong absorption at 593nm and many vegetable extracts have a similar interference, this limits the ability of a FRAP assay to be used on biological samples (Ou et al., 2002). Furthermore, the FRAP assay requires a low pH (3.6) for the reaction to occur and this can also interfere with the results. A low pH environment can significantly inhibit electron transfer from the reductant to the ferric ion, producing inaccurate results.

The FRAP assay only tests the reducing capability of a compound to reduce ferrous iron to ferric iron rather than antioxidant activity within a biological framework. In a comparative study by Ou et al (2002), the AOC of 927 vegetable extracts was determined using the FRAP and ORAC assay. ORAC was first developed by Cao, Alessio and Cutler (1993) and further adapted by Ou, Hampsch-Woodill and Prior (2001). It detects HAT type reaction and ORAC values reflect peroxyl radical scavenging activity of the sample. In the comparison they found a low correlation between the FRAP and ORAC results. Ou et al (2002) concluded that based on FRAPs limitations discussed and the results obtained, it is not appropriate to use FRAP values as an indicator for ‘total antioxidant power’. Another comparative study by Payne et al (2013), found different results. They used a simplified adapted ORAC protocol (DPPH) (Zullo and Ciafardini, 2008) compared to a FRAP assay and very high correlation was found between the results of the FRAP and DPPH (Figure 6).


Figure 6. XY Scatter graph compiled from Payne et al (2013) highlighting the correlation between two different assays used. “correlation (r2 = 0.9808) between the FRAP and ORAC assay with three different salad samples (rocket, spinach, and watercress) and two different production methods (conventional and organic). Bars represent standard error. GLM, general linear model; FRAP, ferric reducing antioxidant power; ORAC, oxygen radical absorbance capacity” (Payne et al., 2013).

There are other assays that provide different ways to test antioxidant capacity, the Folin- Ciocalteu method, which is used to test for ‘total phenolic content’ relies on SET from phenolic compounds to a reagent in a basic medium. Fu et al (2011) used this method with a GAE to test the AOC of tea infusions, they also used TEAC and FRAP, the FRAP assay was carried out using FeSO4 as the standards and they found a strong correlation between all the assays and their ability to measure AOC (Figure 5). Findings by Martínez-Sánchez et al (2008) compared ABTS, DPPH and FRAP assays when testing Brassicaceae vegetables and also found very high correlations with the results from all three assays (0.95-0.98). This shows that even though the FRAP assay is somewhat not biologically applicable in its reaction, it can predict the AOC in a similar way to more applicable assays and act as a reliable proxy for biological AOC or phenolic content.

Figure 5. XY Scatter graphs compiled from Fu et al (2011) highlighting the correlation between different assays used. (A) “Correlation between total antioxidant capacities measured by the FRAP and TEAC assays” (B) “Correlation between total phenolic content and antioxidant capacities measured by the FRAP assay. GAE: gallic acid equivalents.” (C) “Correlation between total phenolic content and antioxidant capacities measured by the TEAC assay. GAE: Gallic acid equivalents.”



The FRAP assay has its limitations like many other assays do when used in isolation, to present quantitative results from any one assay as the ‘total antioxidant power’ of a given sample is an oversimplification. Comparisons can be made between studies using correlative analysis, but to produce a more comprehensive and accurate representation of absolute AOC, multiple assays should be carried out with appropriate positive controls based on the samples being tested. A standardised multi-assay model is required to remove ambiguity and arbitrary comparisons of AOC.


Benzie, I. and Strain, J. (1996). The Ferric Reducing Ability of Plasma (FRAP) as a Measure of “Antioxidant Power”: The FRAP Assay. Analytical Biochemistry, 239(1), pp.70-76.

Benzie, I. and Strain, J. (1999). [2] Ferric reducing/antioxidant power assay: Direct measure of total antioxidant activity of biological fluids and modified version for simultaneous measurement of total antioxidant power and ascorbic acid concentration. Oxidants and Antioxidants Part A, pp.15-27.

BERTONCELJ, J., DOBERSEK, U., JAMNIK, M. and GOLOB, T. (2007). Evaluation of the phenolic content, antioxidant activity and colour of Slovenian honey. Food Chemistry, 105(2), pp.822-828.

Cao, G., Alessio, H. and Cutler, R. (1993). Oxygen-radical absorbance capacity assay for antioxidants. Free Radical Biology and Medicine, 14(3), pp.303-311.

Carlsen, M., Halvorsen, B., Holte, K., Bøhn, S., Dragland, S., Sampson, L., Willey, C., Senoo, H., Umezono, Y., Sanada, C., Barikmo, I., Berhe, N., Willett, W., Phillips, K., Jacobs, D. and Blomhoff, R. (2010). The total antioxidant content of more than 3100 foods, beverages, spices, herbs and supplements used worldwide. Nutrition Journal, 9(1).

Fu, L., Xu, B., Gan, R., Zhang, Y., Xu, X., Xia, E. and Li, H. (2011). Total Phenolic Contents and Antioxidant Capacities of Herbal and Tea Infusions. International Journal of Molecular Sciences, 12(4), pp.2112-2124.

Halliwell, B. and Gutteridge, J. (1995). The definition and measurement of antioxidants in biological systems. Free Radical Biology and Medicine, 18(1), pp.125-126.

Halvorsen, B., Holte, K., Myhrstad, M., Barikmo, I., Hvattum, E., Remberg, S., Wold, A., Haffner, K., Baugerød, H., Andersen, L., Moskaug, Ø., Jacobs, D. and Blomhoff, R. (2002). A Systematic Screening of Total Antioxidants in Dietary Plants. The Journal of Nutrition, 132(3), pp.461-471.

Huang, D., Ou, B. and Prior, R. (2005). The Chemistry behind Antioxidant Capacity Assays. Journal of Agricultural and Food Chemistry, 53(6), pp.1841-1856.

Martínez-Sánchez, A., Gil-Izquierdo, A., Gil, M. and Ferreres, F. (2008). A Comparative Study of Flavonoid Compounds, Vitamin C, and Antioxidant Properties of Baby Leaf Brassicaceae Species. Journal of Agricultural and Food Chemistry, 56(7), pp.2330- 2340.

Moon, J. and Shibamoto, T. (2009). Antioxidant Assays for Plant and Food Components. Journal of Agricultural and Food Chemistry, 57(5), pp.1655-1666.

Ou, B., Hampsch-Woodill, M. and Prior, R. (2001). Development and Validation of an Improved Oxygen Radical Absorbance Capacity Assay Using Fluorescein as the Fluorescent Probe. Journal of Agricultural and Food Chemistry, 49(10), pp.4619- 4626.

Ou, B., Huang, D., Hampsch-Woodill, M., Flanagan, J. and Deemer, E. (2002). Analysis of Antioxidant Activities of Common Vegetables Employing Oxygen Radical Absorbance Capacity (ORAC) and Ferric Reducing Antioxidant Power (FRAP) Assays:  A Comparative Study. Journal of Agricultural and Food Chemistry, 50(11), pp.3122-3128.

Ovaskainen, M., Trrnen, R., Koponen, J., Sinkko, H., Hellström, J., Reinivuo, H. and Mattila, P. (2008). Dietary Intake and Major Food Sources of Polyphenols in Finnish Adults. The Journal of Nutrition, 138(3), pp.562-566.

Payne, A., Mazzer, A., Clarkson, G. and Taylor, G. (2013). Antioxidant assays - consistent findings from FRAP and ORAC reveal a negative impact of organic cultivation on antioxidant potential in spinach but not watercress or rocket leaves. Food Science & Nutrition, 1(6), pp.439-444.

Pérez-Cruz, K., Moncada-Basualto, M., Morales-Valenzuela, J., Barriga-González, G., Navarrete-Encina, P., Núñez-Vergara, L., Squella, J. and Olea-Azar, C. (2018). Synthesis and antioxidant study of new polyphenolic hybrid-coumarins. Arabian Journal of Chemistry, 11(4), pp.525-537.

Pérez-Jiménez, J., Fezeu, L., Touvier, M., Arnault, N., Manach, C., Hercberg, S., Galan, P. and Scalbert, A. (2011). Dietary intake of 337 polyphenols in French adults. The American Journal of Clinical Nutrition, 93(6), pp.1220-1228.

Prior, R., Wu, X. and Schaich, K. (2005). Standardized Methods for the Determination of Antioxidant Capacity and Phenolics in Foods and Dietary Supplements. Journal of Agricultural and Food Chemistry, 53(10), pp.4290-4302.

Pulido, R., Bravo, L. and Saura-Calixto, F. (2000). Antioxidant Activity of Dietary Polyphenols As Determined by a Modified Ferric Reducing/Antioxidant Power Assay. Journal of Agricultural and Food Chemistry, 48(8), pp.3396-3402.

Pulido, R., Hernández-García, M. and Saura-Calixto, F. (2003). Contribution of beverages to the intake of lipophilic and hydrophilic antioxidants in the Spanish diet. European Journal of Clinical Nutrition, 57(10), pp.1275-1282.

Rabeta, M. and Nur Faraniza, R. (2013). Total phenolic content and ferric reducing antioxidant power of the leaves and fruits of Garcinia atrovirdis and Cynometra cauliflora. International Food Research Journal.

Rubio, C., Hernández-Ruiz, J., Martinez-Subiela, S., Tvarijonaviciute, A. and Ceron, J. (2016). Spectrophotometric assays for total antioxidant capacity (TAC) in dog serum: an update. BMC Veterinary Research, 12(1).

Schafer, F. and Buettner, G. (2001). Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione couple. Free Radical Biology and Medicine, 30(11), pp.1191-1212.

Serafini, M. and Testa, M. (2009). Redox ingredients for oxidative stress prevention: the unexplored potentiality of coffee. Clinics in Dermatology, 27(2), pp.225-229.

Song, F., Gan, R., Zhang, Y., Xiao, Q., Kuang, L. and Li, H. (2010). Total Phenolic Contents and Antioxidant Capacities of Selected Chinese Medicinal Plants. International Journal of Molecular Sciences, 11(6), pp.2362-2372.

Sudan, R., Bhagat, M., Gupta, S., Singh, J. and Koul, A. (2014). Iron (FeII) Chelation, Ferric Reducing Antioxidant Power, and Immune Modulating Potential of Arisaema jacquemontii (Himalayan Cobra Lily). BioMed Research International, 2014, pp.1-7.

Thaipong, K., Boonprakob, U., Crosby, K., Cisneros-Zevallos, L. and Hawkins Byrne, D. (2006). Comparison of ABTS, DPPH, FRAP, and ORAC assays for estimating antioxidant activity from guava fruit extracts. Journal of Food Composition and Analysis, 19(6-7), pp.669-675.

Zullo, B. and Ciafardini, G. (2008). The olive oil oxygen radical absorbance capacity (DPPH assay) as a quality indicator. European Journal of Lipid Science and Technology, 110(5), pp.428-434.



Appendix A – Reagent and standard solution method and materials.

FRAP REAGANT (Incubate at 37ºC)

200 cm3 of ACETATE BUFFER pH 3.6

184 cm3 of 0.3M Ethanoic acid
11.5 cm3 of Ethanoic acid concentrate
118.5 cm3 of distilled water
16 cm3 of 0.3M Sodium acetate
0.615g of Sodium acetate (anhyd)
25 cm3 of distilled water

20 cm3 of 10mM TPTZ in 40mM HCL

40mM HCL
10 cm3 of 0.1M HCL
15 cm3 of distilled water
0.078g of TPTZ (Sigma Aldrich, cat # 93285)

20 cm3 of 20mM FeCl3 ag

0.135g of FeCl3 6H2O
25 cm3 of distilled water


10mM FeSO4 stock standard

1.390g of FeSO4 7H2O

500 cm3 of distilled water


Appendix B – Tables of raw data for the figures in report.

Table 2. FeSO4 (mmol/L) standard solution concentrations and related absorbance values. – Raw data used to generate Figure 2.















Table 3. Absorbance and related ferrous equivalence (mmol/l) in common beverages using a FRAP assay and line equation from standards (y = 0.1074x - 0.12) – Raw data used to generate Figure 3.




FE (1 mmol/L of FeSO4)










English Breakfast



Red Bush












Appendix C – Statistical Analysis

Statistical analysis was run via IBM SPSS Statistics software to show correlation between results and data from Carlsen et al (2010) by a Pearson’s test for bivariate correlation.

Appendix D – Images of lab equipment used

Image of a Heraeus Labofuge 400 centrifuge (above).

Image of Grant JB Series water bath (above).


Image of Cecil CE 1011 spectrophotometer (above).

Appendix E – Images of beverages used with the FRAP assay.


Tesco Apple Juice     

Ocean Spray Cranberry Classic Juice

Innocent Mango Juice

Tesco Tomato Juice

Tesco Orange Juice

Whittard English Breakfast Tea

Nescafe Original

Tetley Redbush Tea