OPTIMIZATION OF PRETREATMENT CONDITIONS OF CARROTS TO MAXIMIZE JUICE RECOVERY BY RESPONSE SURFACE METHODOLOGY

Journal of Engineering Science and Technology
Vol. 1, No. 2 (2006) 158- 165
© School of Engineering, Taylor’s University College






OPTIMIZATION OF PRETREATMENT CONDITIONS OF CARROTS
TO MAXIMIZE JUICE RECOVERY BY RESPONSE SURFACE
METHODOLOGY


H. K. SHARMA1*, J. KAUR2, B.C. SARKAR2, C. SINGH2, B. SINGH2,
A. A. SHITANDI3

1 Department of Food Technology,
Sant Longowal Institute of Engineering and Technology,
Distt. Sangrur (Pb.) INDIA
*Corresponding Author: harish.sharma@lycos.com

2 Department of Food Technology,
Sant Longowal Institute of Engineering and Technology,
Distt. Sangrur (Pb.) INDIA

3 Dept. of Food Science,
Egerton University, P.O. B. 536 Njoro 20107, KENYA.




Abstract
Carrot juice was expressed in a hydraulic press using a wooden set up. Carrot samples
pretreated at different designed combinations, using Central Composite Rotatable Design
(CCRD), Response Surface Methodology (RSM), of pH, temperature and time were expressed
and juice so obtained was characterized for various physico-chemical parameters which
involved yield, TSS and water content, reducing sugars, total sugars and color (absorbance).
The study indicated that carrots exposed to the different pretreatment conditions resulted in
increased amount of yield than that of the control. The responses were optimized by numerical
method and were found to be 78.23% yield, 0.93% color (abs), 3.41% reducing sugars, 5.53%
total sugars, 6.69obrix, and 90.50% water content. All the derived mathematical models for the
various responses were found to be fit significantly to predict the data.


Keywords: Carrots, Physico-chemical, Optimization, Mathematical Models.


1. Introduction
India is the largest producer of fruits and second largest producer of vegetables
in the world, next to China. Its share in the world’s production is 11% and 7% in
case of fruits and vegetables, respectively [1]. The carrots are consumed fresh or
cooked, either alone or with other vegetables, in the preparation of soups, stews,
158

---

Optimization of Pretreatment Conditions of Carrots 159

Nomenclatures

CCRD Central Composite Rotatable Design
Fcal
F calculated

Ftab
F tabulated

LoF
Lacck of Fit

P
Probability

R2
Coefficient of determination

RSM
Response Surface Methodology

X, Y
Constant

Y
Measured response

Greek Symbols

?o
Intercept term

?i,ij,ii Constants



curries, and pies. Fresh grated roots are used in salads and tender roots are pickled [2].
In the carrot slicing, energy and peak force were mainly influenced by core diameter
and central part of the carrot [3]. Attempts have been made to process this vegetable
as canned, dehydrated, pickled, juice, preserve, puree, flakes, or as a frozen product.
During processing, losses in soluble sugars, minerals, pectic substances and other
solutes from carrots have been reported [4]. Ogunlesi and Lee [5] recommended a
longer time lower temperature blanching for carrots, which improves the firmness of
canned carrots at low cost. Kinetics of thermal softening of canned carrots during
retorting has been reported [6]

Yield and quality of carrot juice extracted by pressing varied with the condition
of any particular batch of carrots and pulp consistency after blanching and mashing
[7]. Grinding from 6 to 2 mm particle size increased yield by 0.7%/mm for blanched
and macerated carrots; it also increased juice color. Squeezing the blanched mash
resulted in higher yields of juice and carotenoids than cold-squeezing. The effects of
different blanching solutions and blanching times (1-5 min) on the quality of carrot
juice was studied [8]. Chadha et al. [9] studied the effect of pectolytic and cellulolytic
enzymes on the carrot juice recovery and concluded that incubation temperature and
enzyme concentration were more pronounced than those of incubation time and
enzyme ratio. Sharma et al. [10] optimized enzymatic process parameters for
increased juice yield from carrot using response surface methodology.

Carrot is a nutritious vegetable and is widely known for various medicinal
properties. To produce the juice, the industries/organized sectors adopt different
pretreatment conditions to maximize the juice recovery. Big losses are resulted due to
the lack of proper knowledge of the effect of pretreatment conditions on yield with
respect to the optimum overall acceptability. Therefore, in order to have uniformity in
terms of pretreatment conditions viz. time, temperature and acidic conditions to
recover maximum yield of juice along with optimum overall acceptability, scientific


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

160 H. K. Sharma et al
efforts are required. Hence the present investigation was undertaken to optimize the
pretreatment conditions of carrots especially of the red-orange commercial variety,
most popular in India for the maximum juice recovery by using Central Composite
Rotatable Design (CCRD), Response Surface Methodology (RSM).

2. Materials and Methods
The carrots were procured from the local market of Longowal, Punjab. The carrots
were washed under running tap water, trimmed and peeled by using peeler to remove
dirty skin, undesirable hair and end ones and again washed. The clean, sound carrots
without any physical damage were selected for further processing.
2.1 Pretreatment
The washed, trimmed and peeled carrots were taken. The carrots were cut into 3-4 cm
in diameter, 3 cm in thickness and weighed around 300g. The pH of water was
maintained by using citric acid. The water was then heated to specific temperature and
carrots were dipped in it for a specific period of time according to the designing
conditions. The carrots were then removed from the hot water, allowed to cool and
grated by using manual grating plate.

The juice was expressed at different pressure range (46.3 – 69.50 MPa) in a
hydraulic press using a wooden setup (B. Sen Barry & Co., New Delhi) and the
respective yield was recorded. It was observed that as the pressure was increased from
46.3 – 61.8 MPa, the yield of juice increased from 49.3 to 62.8%. For further increase
in pressure to 69.5 MPa, the yield decreased to 60.2%. This decrease in the juice yield
may be attributed due to the coming out of the carrot mash through the filter cloth,
which required second filtration of the juice, which may have led to reduced juice
yield. Therefore, a pressure of 61.8 MPa was considered optimum and was used in the
study.
2.2 Experimental design
Response Surface Methodology (RSM) was adopted in experimental design and
analysis [11]. A central composite rotatable design (CCRD) with augmented points in
three variables, as given below, was used:


Coded variables
Number of
X1
X2
X3
Combinations Replications
experiments
(pH)
(Temperature)
(Time)
±1 ±1 ±1 8
1
8
±1.682 0
0
2
1
2
0 ±1.682 0
2
1
2
0 0
±1.682
2 1 2
0 0 0 1 6 6


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

Optimization of Pretreatment Conditions of Carrots 161
0 represents the centre point; ±1 for factorial points, and ±1.682 for augmented points

The level of parameters was carefully selected based on the literature available. The
coding of the levels was done using the following equations:
X1 (pH)= (x1-4.5)/1.3



(1)
X2 (temperature, 0C) = (x2-80)/11
(2)
X3 (time, s) = (x3-360)/140


(3)
where X1, X2 and X3 and x1, x2 and x3 are coded and uncoded variables, respectively.
The range and the levels of the experimental variables used in the coded and uncoded
form for the centre, factorial and augmented point of design are summarized below:
Experimental
Code Coded
level
Variables


-1.682 -1 0 1 1.682
pH
X1
2.31 3.2 4.5
5.8 6.69
Temperature, oC
X2
61.50 69 80 91 98.50
Tim, s
X3
124.55 220 360 500 595.45
2.2 Determination of juice yield
Juice yield was determined by the following expression:
weight
of carrots ? weight
of solids ( cake )
(4)
% Juice Yield =
× 100
weight
of carrots
2.4 Physicochemical analysis
Juice was subjected to various physicochemical parameters such as TSS, water,
reducing sugars, total sugars and color (absorbance) by standard methods [12].
Reducing and total sugars were evaluated by Lane and Eynon Method. For the
determination of color, 5ml of the carrot juice was diluted with 5ml of distilled water.
The colorimeter was set to zero by using distilled water. Then the sample was filled in
the cuvette and the absorbance was measured at 472 nm using colorimeter. The total
soluble solid (0brix) of carrot juice samples was measured by using hand
refractometer.
2.5 Statistical analysis
Design Expert software ‘DE – 6’ was used for regression and graphical analysis of the
data (Stat-Ease, 2000). The optimum values of the selected variables were obtained by
solving the regression equation and by analyzing the response surface contour plots.
3. Results and Discussion


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

162 H. K. Sharma et al
The yield, total soluble solids (TSS) and water content, reducing and total sugars and
color under different pretreatment conditions were determined and a second order
polynomial of the following form was fitted to the data of all the responses as the
results are reported in Table 1:

3
3
3
3
2 (5)
Y = ??
+ ? ?iXi + ? ? ? ijXiXj
+
? ?iiXi
i =1
i =1 ( j < i ) j =1
i =1
where Y is the measured response, ?o, intercept term, ?i, ?ij , and ?ii are the constant
coefficients. The variable XiXj represents the first- order interactions between Xi and
Xj for (j<i).
Table 1. Regression coefficients of the second order polynomial and their
significance.
Reducing
Water
Coefficients Juice
Yield Color
Total sugars
TSS
sugars
content
?o
76.94 0.83 3.30 5.47 91.09 6.28
?1
-2.66 -0.09 -0.13 -0.055 0.55 -0.28
?2
0.78!
-0.04 -0.03***
-0.012!
0.05!
-0.09
?3
1.85
0.03 -0.12 -0.20 0.53 -0.33
?11
-0.78!
-0.01 -0.04*
-4.8E-3!
-0.09!
0.08!
?22
-0.86***
-0.03 -2.8E-3!
-0.061 0.36 -0.10
?33
-1.25**
-0.02 -0.09 -0.091 0.37 -0.15
?12
0.04!
0.02 -7.5E-3!
0.015!
0.10!
-0.01!
?13
0.46!
5.0E-3!
7.5E-3!
-0.078 0.02!
-0.11
?23
-0.11!
0.00!
-7.5E-3!
7.5E-3!
0.11!
-0.01!
R2,
% 88.19
9.9.17
95.73 98.03 93.60 98.33
F
7.47
119.90
22.43 49.88 14.62 58.85
Adeq.
8.51 35.01 14.2 21.62 13.98 28.40
precision
Adj. R2,
%
76.39
98.35
91.46 96.07 87.20 96.66
Pred R2,
%
37.99
93.48
66.35 83.32 66.06 89.25
LoF No
No
No No No No
Significant at 10% (***), 5% (**), not significant (!) and all other values are
significant at 1% level; Table F(9,9) tab = 3.21
3.1 Juice yield
Carrot juice yield varied from 67 to 79% for the samples treated under different
conditions as against 62.8% for control sample. The coefficient of determination, R2,
was 0.8819 for the regressed model predicting the %juice yield, which shows 88.19%
variability in the data. Adeq Precision of 8.51 indicated about the adequacy of the
model (Table 1) . The F-value for the model was 7.47 which is greater than the
tabulated F value 3.21 indicating the adequecy of the model to predict % juice yield at
different pretreatment conditions (P<0.05). The lack of fit was not found significant


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

Optimization of Pretreatment Conditions of Carrots 163
(Fcal < Ftab). Time and pH were found to be the significant model terms (P<0.05). The
juice yield increased with decrease in pH and increase in time.

3.2 TSS and water content
The average total soluble solids (TSS) and the water content of the juice expressed
from untreated carrots were 6.9obrix and 89.66% respectively. In case of juice
obtained after pretreatment of carrots, TSS varied from 5.3 to 6.7obrix and water
content varied from 89.96 to 92.96%. The decrease in TSS with increase in time
might be due to leaching of some of the soluble solids in water. The decrease in the
soluble solids in case of raw juice and carrots boiled in water, has earlier been
reported [8]. The maximum value of TSS, 6.7, was obtained when the pH, time and
temperature was 2.31, 80oC and 360s and the minimum value, 5.3, was obtained at
4.5, 80oC and 595.45s. The coefficient of determination, R2, was 0.9833 for the
regressed model predicting the TSS and 0.9360 predicting the water content. Data
also indicated the insignificance of the lack of fit (Fcal < Ftab) confirming the
significance of the model (Table 1). The water content varied with both pH and time
but temperature did not have significant effect on water content.

3.3 Reducing and total sugars
The reducing sugars and total sugars of the control sample were 3.39% and 5.62%
respectively. After the pre-treatment, the reducing sugar came out to be in the range of
2.84 - 3.42% whereas total sugars were 4.89 - 5.9%. A significant reduction has been
reported in soluble sugars, fructose, glucose and sucrose in the processed samples
(blanching) [13]. The coefficient of determination, R2, predicting the %reducing
sugars and total sugars, showed 95.73% and 98.03% respectively variability in the
data (Table 1). The F-value for the reducing sugars and total sugars models were
22.43 and 49.88 respectively and lack of fit was not significant (Fcal < Ftab), thus
confirming the significance of the models in predicting the reducing and total sugars.
3.4 Colour
Colour value of the carrot juice after pretreatment varied from 0.64 to 0.96. The color
value, 0.97 of the control sample was greater than the color values of the pretreated
samples. The decrease in the color values may be due to exposure of the samples at
different pH and temperatures for a specified period of time thereby causing
degradation to the heat sensitive color constituents. The maximum value of color was
obtained when the pH, time and temperature was 2.31, 80oC and 360s and the
minimum value of color was obtained at 6.69, 80oC and 360s. Adequate precision of
35.01, F value, R2 and LoF indicated that the model is highly adequate. In the model,
pH, temperature and time were found to be the significant model terms (P<0.01).
3.5 Optimization of parameters to maximize juice yield
Design expert software was used to optimize the juice yield with respect to the overall
acceptability. The software uses second order model to optimize the responses. The


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

164 H. K. Sharma et al
uncoded form of input variables i.e. pH, temperature and time was optimally found to
be 3.2, 91ºC and 500s respectively.. In practice, however, it is difficult to maintain the
recommended conditions during processing and some deviation is expected.
Therefore, optimum conditions were varied as pH 3.2 ± 0.14, temperature 91 ± 1oC,
and time 500 ± 15s which were predicted by using second order polynomial. It was
observed that the optimum values of juice yield was found to be 78.23% with respect
to other optimum responses as summarized below:

Process variables
Optimum value
Optimum
Response

Uncoded Coded
value
pH 3.2
-1.00
Yield
(%)
78.23
Temperature (0C) 91 1.00 Color
(abs)
0.93
Time (s)
500
1.00
Reducing sugars (%)
3.41


Total
sugars
(%)
5.53


TSS
(obrix) 6.69


Water
content
(%)
90.50


Desirability
(%)
0.838

In all the cases, the desirability is greater than 80%, thus proper pretreatments before
the expression of the juice may result increase in juice yield as high as 11.5%.

4. Conclusions
Carrots exposed to the different pretreatment conditions resulted in increased juice
yield. The yield was found to be maximum, 78.23% at pH, temperature and time of
3.2, 91oC and 500s respectively. The optimized responses were found to be 78.23%
yield, 0.93 color (abs), 3.41% reducing sugars, 5.53% total sugars, 6.69obrix, and
90.50% water content.The desirability for all the responses was found to be 83.8%.
All the derived mathematical models for the various responses were found to be
significantly fit to predict the data.
References
1. F.A.O. (1998). Production Yearbook. Rome: Food and Agricultural Organization.
2. Kalra, C.L., Kulkarni, S.G. & Berry, S.K. (1987). The carrot (Daucus carota L.) -
a most popular root vegetable. Indian Food Packer, 41, 46-73.
3. Sharma, A. K., Sarkar, B.C. & Sharma H.K. (2005). Optimization of enzymatic
process parameters for increased juice yield from carrot using response surface
methodology. Eur. Food Res. Technology, 221, 106-112.
4. Boeh-ocansey, O. (1984). Effects of vacuum and atmospheric freeze-drying on
quality of shrimp, turkey flesh and carrot samples. Journal of Food Science 49,
1457-61.
5. Ogunlesi, A. T. & Lee, C. Y. (1979). The effect of thermal processing on the
stereoisomerization of major carotenoids and vitamin A values of carrots. Food
Chemistry, 4, 311-318.


Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

Optimization of Pretreatment Conditions of Carrots 165
6. Huang, Y.T. & Bourne, M.C. (1983). Kinetics of thermal softening of vegetables.
Journal of Texture Studies, 14, 1-9.
7. Munsch, M.H., Simard, R.E. & Girard, J.M. (1986). Blanching, grinding and
enzymic maceration during production of carrot juice. I. Effects on yield and
some physico-chemical characteristics. Lebensmittel-Wissenschaft-und-
Technologie, 19, 229-239.
8. Sang-Bin-Lim & Mi-Kyung-Jwa. (1996). Effect of blanching conditions on the
quality of carrot juice. Journal of the Korean Society of Food Science and
Nutrition, 25, 680-686.
9. Chadha, R., Kumbhar, B.K. & Sarkar, B.C. (2003). Enzymatic hydrolysis of
carrot for increased juice recovery. Journal of Food Science and Technology, 40,
35 – 39.
10. Sharma, A. K., Sarkar, B.C. & Sharma H.K. (2005). Energy and peak force
requirement in carrot slicing. Journal of Food Science and Technology, 42(2),
195-197.
11. Khuri, A.I. & Cornell, J.A. (1987). Response surfaces: Design and analyses.
New York: Marcel Dekker, Inc.
12. Ranganna, S. (1991). Handbook of Analysis and Quality Control for Fruits and
Vegetables Products. Delhi: Tata McGraw-Hill Co Ltd.
13. Rodriguez-Sevilla, M.D., Villanueva-Suarez, M.J. & Redondo-Cuenca, A.
(1999). Effects of processing conditions on soluble sugars content of carrot,
beetroot and turnip. Food Chemistry, 66, 81-85.
14. Stat-Ease (2000). Design Expert Software Manual (version 6). Minneapolis, MN.






Journal of Engineering Science and Technology DECEMBER 2006, Vol. 1(2)

---

Document Outline

• ??
• ??
• ??

---