International Journal of Food Sciences and Nutrition,

February 2011; 62(1): 9196

Optimization conditions for anthocyanin and phenolic content extraction

form purple sweet potato using response surface methodology
MARUF AHMED1,2, MST. SORIFA AKTER1, & JONG-BANG EUN1

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Department of Food Science and Technology and Institute of Biotechnology, Chonnam National University, Gwangju,
South Korea, and 2Department of Food Processing and Preservation, Hajee Mohammad Danesh Science and Technology
University, Dinajpur, Bangladesh

Abstract
Purple sweet potato flour could be used to enhance the bioactive components such as phenolic compounds and anthocyanin
content that might be used as nutraceutical ingredients for formulated foods. Optimization of anthocyanin and phenolic
contents of purple sweet potato were investigated using response surface methodology. A face-centered cube design was used to
investigate the effects of three independent variables: namely, drying temperature 55 658C, citric acid concentration 1 3% w/v
and soaking time 1 3 min. The optimal conditions for anthocyanin and phenolic contents were 62.918C, 1.38%, 2.53 min
and 60.948C, 1.04% and 2.24 min, respectively. However, optimal conditions of anthocyanin content were not apparent.
The experimental value of anthocyanin content was 19.78 mg/100 g and total phenolic content was 61.55 mg/g. These data
showed that the experimental responses were reasonably close to the predicted responses. Therefore, the results showed that
treated flours could be used to enhance the antioxidant activities of functional foods.

Keywords: Purple sweet potato, response surface methodology, phenolic compounds, anthocyanin content

Introduction
Purple-fleshed sweet potatoes have an intense purple
color in the storage roots due to the accumulation of
anthocyanins (Terahara et al. 2004).The anthocyanins
in purple sweet potato are mono-acylated or di-acylated
forms of cyanidin and peonidin (Yang and Gadi 2008).
Sweet potatoes had intermediate antioxidant activity
among 43 vegetables (Huang et al. 2006). Recently
natural antioxidants have attracted considerable attention due to their positive health benefit (Huang et al.
2006). Rumbaboa et al. (2009) reported that anthocyanin from purple sweet potato has better radical
scavenging activity than that of red cabbage, grape skin,
elderberry and purple corn. Anthocyanins from purple
sweet potatoes have many biological functions, such as
scavenging free radicals, anti-mutagenicity, anti-carcinogen activity and antihypertensive effect (Oki et al.
2002). Several extraction methods have been used to
obtain extracts rich in anthocyanin and phenolic
content based on different solvents such as methanol,

ethanol, acetone, water or mixture (Pathirana and

Shahidi 2005, Huang et al. 2006). The stability of
anthocyanin and phenolic content were influenced by
several factors (Jiang 2000). Among them, polyphenol
oxidase plays an important role in the degradation of
anthocyanin and phenolic content. Citric acid has been
used extensively for the inhibitory activity on polyphenol oxidase and the anti-browning activity in
minimally processed fruits and vegetables. Citric acid
extracts have a double inhibitory effect by chelating
copper at lower pH (Altunkaya and Gokmen 2009).
Sweet potatoes can be processed into flour, which
are less bulky and more stable than the highly
perishable fresh root. Flour can be used as a thickener
in soup, gravy, fabricated snacks and bakery products.
It could be used to enhance food products through
color, flavor, natural sweetness and nutrients. Sing et al.
(2003) used potassium metabisulphite, citric acid and
sodium chloride to improve the quality of chips from

Correspondence: Jong-Bang Eun, Department of Food Science & Technology, Chonnam National University, 77 Yongbong-ro Buk-gu,
Gwangju 500-757, South Korea. Tel: 82 62 530 0255. Fax: 82 62 530 2149. E-mail: jbeun@jnu.ac.kr
ISSN 0963-7486 print/ISSN 1465-3478 online q 2011 Informa UK, Ltd.
DOI: 10.3109/09637486.2010.511167

92

M. Ahmed et al.

sweet potatoes. Response surface methodology (RSM)

has been successfully used to optimize biochemical
and biotechnological process related to food systems
(Cacace and Maza 2003, Pathirana and Shahidi
2005). Therefore, the goal of the present study was to
optimize different pretreatments such as drying
temperature, citric acid concentration and soaking
time for production of sweet potato flour with
high retention of anthocyanin and phenolic content
using RSM.
Materials and methods

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Raw materials
Sweet potato (Ipomoea batatas L. Lam variety,
Sinjami) was purchased from a local farm. Roots
were washed with tap water to remove dirt and soil
and allowed to dry at ambient temperature (, 208C).
The washed sweet potatoes were stored at 148C for
15 days without curing.
Sample preparation and treatment
Sweet potatoes were peeled with a hand peeler (Han
Sung 27 stainless; Namdong Industry Park, Incheon,
South Korea). Then peeled samples were cut into slices
(1 mm thickness) using a slicing machine (HFS 350G;
Hankook fujee Industries Co. Ltd. Suwon-si,
Gyeonggi-do, Fujee, South Korea). Various levels of
citric acid concentration (1 3% w/v) were solubilized
in deionized water at room temperature (20 ^ 18C).
After that, peeled slices were dipped in aqueous citric
acid solutions (1 3% w/v) for different soaking times
(1 3 min) at room temperature.
Preparation of sweet potato flour
The slices were dried using a convection drying oven
(Dasol Scientific Co. Ltd, Seoul, South Korea) at
different temperatures 558C, 608C, and 658C for
7 8 h. The sweet potato flour (moisture content

6 7%) was obtained by milling the dried slices using a

blender (FM-681C; Hanil, Gwangju, Korea), and
sieved through an 80-mesh (Chung gye sang gongsa,
Seoul, South Korea) screen.
Experimental design for RSM analysis
A three-factor (X1, X2 and X3) and three-level ( 1, 0
and 1) face-centered cube design were employed in
this study, and 15 individual run points were taken for
analysis (Wanasundara and Shahidi 1999). The actual
and corresponding values are presented in Table I.
The multiple regression equation was used to fit the
second-order polynomial equation based on the
experimental data as follows:
Y b0 b1 X1 b2 X2 b3 X3 b11 X1 X1
b22 X2 X2 b33 X3 X3 b12 X1 X2 b13 X1 X3
b23 X2 X3
where Y is the response variable, b0 is the intercept, b1,
b2, b3, b11, b22, b33 and b12, b13, b23 are linear,
quadratic and interaction coefficients respectively, and
X1, X2 and X3 are the coded independent variables.
Verification of model
RSM was used to optimize anthocyanin and phenolic
contents from purple sweet potato. The experimental
and predicted values were compared to confirm the
validity of the model.
Analysis of anthocyanin contents
The content of anthocyanin was determined following
the procedures of Proctor (1974) and Huang et al.
(2006) The sweet potato flour (1 g) was treated with
15 ml HCl methanol (0.15% HCl:methanol
15:85) for 4 h. The extract was filtered and its
absorbance was determined at 530 nm. The anthocyanin content was calculated on the basis of the

Table I. Three-factor, three-level, face-centered cube design for RSM.

Assay number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Factor X1

Factor X2

Factor X3

Variable

Drying temperature (8C)

Concentration (%)

Soaking time (min)

Anthocyanin (mg/100 g)

Total phenolics (mg/g)

55 (21)
55(21)
55(21)
55(21)
60(0)
60(0)
60(0)
60(0)
65( 1)
65(21)
65( 1)
65( 1)
60(0)
60(0)
60(0)

2(0)
1(21)
2(0)
3( 1)
3( 1)
1(21)
3( 1)
1(21)
2(0)
1(21)
2(0)
3( 1)
2(0)
2(0)
2(0)

3( 1)
2(0)
1(21)
2(0)
3( 1)
1(21)
1(21)
3( 1)
3( 1)
2(0)
1(21)
2(0)
2(0)
2(0)
2(0)

40.32 ^ 1.32
39.56 ^ 1.03
40.79 ^ 0.60
37.40 ^ 2.08
34.17 ^ 4.72
32.53 ^ 2.73
24.45 ^ 4.96
24.51 ^ 0.37
29.16 ^ 3.64
26.49 ^ 1.83
24.55 ^ 0.29
22.45 ^ 0.28
23.94 ^ 0.57
20.02 ^ 1.18
21.63 ^ 0.17

56.85 ^ 8.28
47.77 ^ 1.53
47.18 ^ 5.73
47.52 ^ 4.02
44.64 ^ 0.34
51.69 ^ 0.19
56.38 ^ 0.39
63.38 ^ 1.02
46.46 ^ 1.07
51.65 ^ 2.37
51.65 ^ 2.37
52.03 ^ 2.43
45.70 ^ 0.41
49.39 ^ 0.81
58.91 ^ 1.18

93

Optimizing anthocyanin and phenolic content of purple sweet potato flour

following equation:

(a)

Anthocyanin content A MW DF 100=e W

2176.82

236.84***
224.72
247.49***

6.19
21.65
37.69

20.16
20.30
25.97**
0.80

39.39

29.55

3.00
2.33
1.67
58.33
ture (C
)

Tempe
ra

Table III. Analysis of variance for the response surface quadratic

model for anthocyanin and phenolic contents.
Degree
of freedom

For anthocyanin
Lack of fit
Pure error
Total error
For total phenolic
Lack of fit
Pure error
Total error

Sum
of squares

Mean square

F value

3
2
5

7.50
7.76
15.26

2.50
3.88
3.05

0.64

3
2
5

18.75
92.91
111.67

6.25
46.45
22.33

0.93

%)

49.22

ak

0.04
0.39
4.43***
0.97

n(

(b)

61.67

20.04
24.76
21.95

tra

tio

1.00
55.00

65.00
0.28***
3.12***
3.92***

en

nc

58.33
ture (C
)

Co

61.67
Tempe
ra

19.72

R 2, coefficient of multiple determination. ***Significant at

P # 0.01. **Significant at P # 0.05.

Source

1.67

n)

1240.40***

19.72
65.00

mi

b0
Linear
b1
b2
b3
Quadratic
b11
b22
b33
Cross-product
b12
b13
b23
R2

Phenolic content

2.33

e(

Anthocyanin content

3.00

tim

Coefficient

29.55

ing

Table II. Regression coefficients of predicted quadratic polynomial

for the response anthocyanin and phenolic contents.

Anthocyanins (mg/100 g)

The total phenolic content was determined using

Folin Ciocalteau reagent according to a slightly
modified method (Swain and Hills 1959). The sample
(0.1 g) was extracted three times with 20 ml of 75%
methanol and was filtered through Whatman No. 2
filter paper. Extracts were combined and concentrated
in a rotary vacuum evaporator (Rikakikai Co. Ltd,
Tokyo, Japan) at 408C; the volume was adjusted to
20 ml with 75% methanol. One milliliter of extract,
5 ml distilled water and 2 ml of 10% Folin Ciocalteau
reagent were added into a Falcon tube. After 3 min at
room temperature, 2 ml of 7.5% Na2CO3 solution was
added and the sample was diluted to 20 ml with
distilled water. Each sample was allowed to stand for

39.39

1.00
55.00

So

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Analysis of total phenolic contents

49.22

Anthocyanins (mg/100 g)

where A is the absorbance, MW is the molecular

weight of cyanidin-3-glucoside (MW 449.2), DF is
the dilution factor, 1 is the molar absorptivity
(34,300), and W is the sample weight (g).

Figure 1. Response surface plots of the anthocyanin content of

purple sweet potato flour as affected by temperature, citric acid
concentration and soaking time. (a) Temperature and concentration.
(b) Temperature and soaking time.

1 h at room temperature and absorbance was

measured at 760 nm (UV-1201; Shimadzu, Kyoto,
Japan). The total phenolic content was calculated on
the basis of standard curves of gallic acid, and
expressed as milligrams of gallic acid equivalents per
gram of sample on a wet weight basis.
Statistical analysis
All determinations were carried out in triplicate and
the experimental results are expressed as means ^
standard deviation. Statistical analysis of the verification results was carried out by analysis of variance and
Duncans multiple-range tests using SAS (version

94

M. Ahmed et al.
(a)

62.5
64.46
60.0

1.5

21.20
38.90

25.62
43.32

30.05
47.75

3.0

49.17

3.00
2.33

34.47

41.53
65.00

(b) 65.0
Temperature (C)

1.67
61.67
58.33
re (C)

Tempera
tu
62.5

1.00
55.00

(b)

60.0
57.5

64.46

Figure 2. Contour plots showing the effects of temperature, citric

acid concentration and soaking time on anthocyanin content of
purple sweet potato flour. (a) Temperature and citric acid
concentration. (b) Temperature and soaking time.

56.82

49.17

3.00
n)

34.47

2.33
41.53
65.00

9.1). The optimal conditions were estimated through

three-dimensional response surface analyses of the
three independent variables and each dependent
variable.

1.67
61.67
58.33

Tempera
tu

re (C)

1.00
55.00

mi

30.05
47.75

e(

25.62
43.32

tim

21.20
38.90

3.0

ing

2.0
2.5
Soaking time (min)

ak

Anthocyanins (mg/100 g)

1.5

So

55.0
1.0

Phenolic content (mg/g)

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Anthocyanins (mg/100 g)

2.0
2.5
Concentration (%)

%)

55.0
1.0

56.82

Co
nc
en
tra
tio
n(

57.5

Phenolic content (mg/g)

Temperature (C)

(a) 65.0

Figure 3. Response surface plots of the phenolic contents of purple

sweet potato flour as affected by temperature, citric acid
concentration and soaking time. (a) Temperature and
concentration. (b) Temperature and soaking time.

Results and discussion

Fitting the models
Analysis of variance using SAS was performed to
determine the significance of the linear, quadratic,
cross-product (Table II) and the lack of fit (Table III)
of the independent variables on the anthocyanin and
phenolic contents. The lack-of-fit test is a measure of
the failure of a model to represent data in the
experimental domain at the points that were not
included in the regression (Montgomery 1984).
However, the R 2 value of the dependent variables
was approximately 0.80, indicating that a high
proportion of variability was explained by the data
(Varnalis et al. 2004). Therefore, the results showed
that the experimental model was adequate due to no
significant lack of fit and satisfactory levels of R 2.

Effect of pretreatment on anthocyanin and phenolic

contents
The anthocyanin content of sweet potato flours ranged
from 20.02 to 40.79 mg/100 g wet weight basis
(Table I). The contents of anthocyanin were much
higher than those of sweet potato puree (Steed and
Truong 2008) and of steamed or kneaded flours
(Huang et al. 2006). The results of multiple regression
analysis showed that the anthocyanin contents were
significantly (P # 0.001) affected by the linear term of
temperature and soaking time, the quadratic of all
terms and the interaction term of concentration and
soaking time (Table II). The predicted model obtained

Optimizing anthocyanin and phenolic content of purple sweet potato flour

Temperature (C)

(a) 65.0
62.5
60.0
57.5
55.0
1.0

1.5

2.0

2.5

3.0

Concentration (%)
42.68
56.43

46.12
59.87

49.56
63.31

Y 2176:82 2 5:97X2 X3

62.5
60.0
57.5
55.0
1.0

1.5

2.0

2.5

3.0

Soaking time (min)

Phenolic content (mg/g)

42.68
56.43

46.12
59.87

49.56
63.31

the anthocyanin structure could be varying with pH

(Cevallos-Casala and Cisneros-Zevallos 2004).
The total phenolic content of sweet potato
flours ranged from 44.64 to 64.32 mg/g wet weight
basis (Table I). These results were much higher
than those reported in literature for fresh and
steamed sweet potato flours (Yang and Gadi 2008).
The results of multiple regression showed that the
total phenolic content was significantly affected by the
interaction term of concentration and soaking time
(X2, X3, P # 0.05). The final predictive model for
phenolic content is given below:

53.00

(b) 65.0
Temperature (C)

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Phenolic content (mg/g)

95

53.00

Figure 4. Contour plots showing the effects of temperature, citric

acid concentration and soaking time on phenolic contents of purple
sweet potato flour. (a) Temperature and citric acid concentration.
(b) Temperature and soaking time.

The response surface plots in Figures 3 and 4 show the

relationship between the phenolic content and drying
temperatures, citric acid concentrations and soaking
times. The total phenolic content increased with
increasing drying temperatures (Figure 3a,b). This
might release more bound phenolic compounds from
the breakdown of cellular constituents. Huang et al.
(2006) found that steaming treatment increased the
total phenolic content of purple sweet potato flour.
Dewanto et al. (2002a) also found that the free
phenolic content of sweet corn increased with
increasing heating temperature and time. However,
thermal processing had no effect on the phenolic
content of tomato (Dewanto et al. 2002b). On the other
hand, total phenolic contents decreased with increasing the concentration and soaking time (Figure 4a,b).
This was probably because some phenolic compounds
were more hydrolyzed or oxidized because dispersions
were prepared in the presence of ambient oxygen.

for Y is given below:

Y 1240:40 2 36:84X1 2 47:49X3 0:28X21

Optimization of pretreatments and verification of models

3:12X22 3:92X23 4:43X2 X3

Figures 1 and 2 show the response surface plots of the
relationship between anthocyanin content and drying
temperatures, citric acid concentrations and soaking
times. Anthocyanin contents decreased with increasing drying temperatures (Figure 1a,b). This was as
expected because heating opened the structure of
anthocyanin to form chalcones, which was degraded
further to form brown products (Delgado-Vargas et al.
2000). However, the anthocyanin content increased
with increasing soaking time and concentration
(Figure 2a,b). This might be due to interaction
between citric acid and anthocyanin. In acidic media,

The predicted and experimental results are presented

in Table IV. For the phenolic content, the predicted
response surface of the stationary point was a saddle
point. Thus the estimated surface did not have a
unique optimum. However, for the anthocyanin
content, the predicted response surface of the
stationary point was a minimum. Therefore, different
optimum conditions were obtained for both responses.
The optimal conditions for anthocyanin and phenolic
contents were 62.918C, 1.38%, 2.53 min, whereas for
total phenolic contents they were drying temperature
60.948C, citric acid concentration 1.04% and soaking
time 2.24 min. However, optimal conditions of the
anthocyanin content were not apparent. This is due to

Table IV. Comparison of predicted and experimental values for the response of anthocyanin and phenolic contents.
Optimum conditions
Response variable

Stationary point

Anthocyanin
Total phenolics

Minimum
Saddle

Values

Temperature (8C)

Soaking time (min)

Soaking concentration (%)

Experimental

Predicted

62.91
60.64

2.53
1.04

1.38
2.24

19.78 ^ 0.97
61.55 ^ 2.9

19.71
52.89

96

M. Ahmed et al.

the fact that the optimization point was a minimum.

The optimal value of anthocyanin content was
lower than expected values. This might be related to
the anthocyanin extraction conditions by secondorder polynomials (Fan et al. 2008). The corresponding experimental responses of anthocyanin and
total phenolic contents were 19.78 mg/100 g and
61.55 mg/g, respectively. These data showed that the
experimental responses were reasonably close to the
predicted responses.

Int J Food Sci Nutr Downloaded from informahealthcare.com by Seoul National University on 03/18/13
For personal use only.

Conclusion
The results of anthocyanin and phenolic contents were
higher than the previous reported values for raw, steam
and kneaded sweet potato flours. Therefore, treated
flours could be used to make the higher quality products
that would be more attractive to product developers and
consumers.
Declaration of interest: The authors report no
conflicts of interest. The authors alone are responsible
for the content and writing of the paper.
References
Altunkaya A, Gokmen V. 2009. Effect of various anti-browning
agents on phenolic compounds profile of fresh lettuce (L. Sativa).
Food Chem 117:122126.
Cacace JE, Mazza G. 2003. Optimization of extraction of
anthocyanins from black currants with aqueous ethanol. J Sci
Food Agric 68:240248.
Cevallos-Casala BA, Cisneros-Zevallos L. 2004. Stability of
anthocyanin-based aqueous extracts of Andean purple corn
and red-fleshed sweet potato to synthetic and natural colorants.
Food Chem 86:6977.
Delgado-Vargas F, Jimenez AR, Paredes-Lopez O. 2000. Natural
pigments: Carotenoids, anthocyanins, and betalainscharacteristics, biosynthesis, processing and stability. Crit Rev Food Sci
Nutr 40:173289.
Dewanto V, Xianzhong WU, Adom KK, Liu RH. 2002a. Thermal
processing enhances the nutritional value of tomatoes by
increasing total antioxidant activity. J Sci Food Agric 50:
30103014.
Dewanto V, Xianzhong WU, Adom KK, Liu RH. 2002b. Processed
sweet corn has higher antioxidant activity. J Sci Food Agric 50:
49594964.

Fan G, Han Y, Gu Z, Chen D. 2008. Optimization conditions for

anthocyanins extraction from purple sweet potato using response
surface methodology (RSM). LWT Food Sci Technol 41:
155 160.
Huang YC, Chang YH, Shao YY. 2006. Effects of genotype and
treatment on the antioxidant activity of sweet potato in Taiwan.
Food Chem 98:529538.
Jiang Y. 2000. Role of anthocyanins, polyphenoloxidase and phenols
in lychee pericarp browning. J Sci Food Agric 80:305310.
Montgomery DC. 1984. Design and analysis of experiments. 2nd ed.
New York: John Wiley and Sons.
Oki T, Masuda M, Furuta S, Nishiba Y, Terahara N, Suda I. 2002.
Involvement of anthocyanins and other phenolic compounds in
radical-scavenging activity of purple-fleshed sweet potato
cultivars. J Food Sci 67:17521756.
Pathirana LC, Shahidi F. 2005. Optimization of extraction of
phenolic compounds from wheat using response surface
methodology. Food Chem 93:47 56.
Proctor JTA. 1974. Colour stimulation in attached apples with
supplementary light. Can J Plant Sci 54:499503.
Rumbaboa RGO, Cornago DF, Geronimo IM. 2009. Phenolic
content and antioxidant capacity of Philippine sweet potato
(Ipomoea batatas) varieties. Food Chem 113:11331138.
Sing S, Raina CS, Bawa AS, Saxena DC. 2003. Optimization of
processing variables in the preparation of sweet potato chips
using response surface methodology. Euro Food Res Technol
217:374381.
Steed LE, Truong VD. 2008. Anthocyanin content, antioxidant
activity and selected physical properties of flowable purplefleshed sweet potato purees. J Food Sci 73:215221.
Swain T, Hillis WE. 1959. The phenolic constituents of prunus
domestica. I. The quantitative analysis of phenolic constituents.
J Sci Food Agric 10:63 68.
Terahara N, Konczak I, Ono H, Yoshimoto M, Yamakewa O. 2004.
Characterization of acylated anthocyanins in callus induced from
storage root of purple-fleshed sweet potato, Ipomoea batatas L.
J Biomed Biotechnol 5:279286.
Varnalis AI, Brenan JG, Macdougall DB, Gilmour SG. 2004.
Optimization of high temperature puffing of potato cubes using
response surface methodology. J Food Eng 61:153163.
Wanasundara UN, Shahidi F. 1999. Concentration of omega
polyunsaturated fatty acids of seal blubber oil by urea
complexation: Optimization of reaction conditions. Food Chem
65:4149.
Yang J, Gadi RL. 2008. Effects of steaming and dehydration
on anthocyanins, antioxidant activity, total phenols and
color characteristics of purple-fleshed sweet potatoes
(Ipomoea batatas). Am J Food Technol 3:224 234.