Optimization of neem seed oil extraction process using response surface methodology

Journal of Natural Sciences Research www.iiste.org
ISSN 2224-3186 (Paper) ISSN 2225-0921 (Online)
Vol.2, No.6, 2012

Optimization of Neem Seed Oil Extraction Process Using Response
Surface Methodology
Tunmise Latifat Adewoye1 and Oladipupo Olaosebikan Ogunleye 2*
1
Department of Chemical Engineering, University of Ilorin, P.M.B. 1515, Ilorin -Nigeria
E-mail: tuntola2002@yahoo.com ; adewoye.tl@unilorin.edu.ng
2
Department of Chemical Engineering, Ladoke Akintola University of Technology,
P.M.B. 4000, Ogbomoso -Nigeria
*Corresponding Author’s E-mail: ooogunleye@yahoo.com ; ooogunleye@lautech.edu.ng

Abstract
A central composite design (CCD) consisting three factors (solvent composition, temperature and time of
extraction) at five levels was used to study the solvent extraction of neem oil from its seed using n- hexane and
ethanol. Neem oil yield, pH, refractive index, acid value, iodine value and saponification value were evaluated as
the responses. Forty-two experimental runs resulted from the CCD with a minimum oil yield of 12% and
maximum of 41%. Response surface methodology was used to analyse the results of the CCD of the extraction
processes. The optimum values for yield, pH, refractive index, acid value, iodine value and saponification value
from the surface plot was 43.48%, 4.99, 1.56, 1.411g/g, 89.35g/g, and 176.64 mg/g respectively. The
maximum predicted percentage yield was 43.48% at solvent composition of 80.77% n hexane, 34.93°C and 6
hours duration of extraction. The five validation experiments had optimum oil yield range between 32.85% and
37.20% while their accompany quality characteristics were not significantly different from the simulated values
at p<0.05
Key words: Neem Oil, Response Surface Methodology, Central Composite Design

1.0 Introduction
Neem (Azadiracha indica) tree is a native to tropical South East Asia and it is a member of the Mahogany family
Meliaceae (ICIPE, 1995; Okonkwo, 2004). This tree is popularly known as dongoyaro in Nigeria. All parts of
this tree are very useful in variety of biological activity. The most famous part of this tree is the oil obtained from
the kernel of its seed. Neem oils have found its use widely in different region of the globe for medicinal and
agricultural purpose (Ranajit et al., 2002; Awad and Shimaila,2003; Muñoz-Valenzenla et al., 2007; Kumar et al.,
2010). It is used in soap production, as raw material for producing commercial pesticides and cosmetics, plant
protection, stock and textile protection, refining to edible oil, lubrication oil for engines, lamp oil, candle
production (Peter, 2000). It is found to be of great health use and it has an effective anti-germ property which
include the use as an insect repellent and it has shown positive results as pesticide (Kovo, 2006). In India, it has
been demonstrated that Neem Oil is a potential new contraceptive for women (NRC,1992).
Among various methods available for obtaining neem oil from seeds are mechanical press, supercritical
fluid extraction and solvent extraction method. Mechanical method of extraction is the most widely used to
extract oil from neem seed. However, the oil produced with this method is usually turbid, contains significant
amount of water and metal contents. Extraction using supercritical fluid produced very high purity oil but the
operating and investment cost are very high. Extraction using solvent has several advantages like high oil yield
and less turbid oil than mechanical extraction and relative low operating cost compared with supercritical fluid
extraction (Liauw et al., 2008, Soetaredjo et al., 2008).
It is clear from various earlier researches on solvent extraction (Kovo, 2006; Liauw et al., 2008;
Soetaredjo et al., 2008 and Zahedi et al., 2010) that the final neem oil quantity and properties are all functions of
the process variables of the extraction process. There exist a complex relationship between the oil yield and its
properties. The purity and final properties of the neem oil extract determine its end use. The quest by these
researchers to increase the oil yield has resulted on a great compromise on its properties. The crux of the problem
has remained how to combine oil yield with desirable qualities. Hence, the need to apply process optimisation as
an essential tool that has demonstrated high efficiency in selecting optimal process variables for optimum
turnover in many chemical processes. Among process optimisation tools that has been effectively applied to
chemical processes is response surface methodology (RSM) (Singh et al., 2003). RSM is a collection of
statistical and mathematical techniques useful for developing, improving and optimizing process and new
products, as well as in the improvement of existing product designs.
This study was therefore to develop an optimization framework for the Neem oil solvent extraction process
that addressed the optimal selection of processes variables that maximised the oil yield subject to the quality
constraints requirements.

66

Vol.2, No.6, 2012

2.0 Methodology
2.1 Neem seed material preparation
Neem seeds used in the solvent extraction of neem oil was obtained from Ilorin, Kwara State, Nigeria. The fruits
were de-pulped to obtain the seeds that were washed thoroughly to remove the dirt and impurity. These seeds
were decorticated by winnowing to remove the hull from the seeds. The weight of the seeds (W1) was measured
and then placed in the oven at 500 C until constant weight (W2) was obtained using Liauw et al (2008) method
for drying. The dried seeds were crushed in a mortar and the sample was expressed on a standard sieve screen of
the required mesh size (0.425- 0. 71mm ) to obtain the required particle sizes.
2.2 Neem oil extraction and properties analysis
Solvent extraction method using various combinations of n-hexane and ethanol were used to study the effect of
extraction parameter on the yield and properties of neem oil. Other important factors of the extraction process
were the particle size, particle to solvent ratio, temperature of extraction and time of extraction (Liauw et al.
2008; Kovo 2006). For this study the factors considered were temperature of extraction, time of extraction and
type of solvent combinations used for extraction because the two other factor has being reported by Liauw et al.
(2008) to have optimum performance at the fine particle size (0.425- 0. 71mm ) and ratio 1:5 respectively, hence
were adopted.
Each extraction run was set up by measuring 10g of the seed powder into 50ml of the solvent mixture in a
250ml corked conical flask. This mixture was placed in a thermostatic water bath operating at a preset required
temperature. At the set time interval, the samples were taken and centrifuged to separate the solid fraction from
the solution. The extracts were heated and evaporated using rotary evaporator apparatus to obtain solvent-free
oils. The solvent free oil extracted was analysed for the pH, refractive index, acid value, saponification value and
iodine value using AOAC (2000).
The percentage of oil extracted was calculated from the equation:
௠௔௦௦ ௢௙ ௘௫௧௥௔௖௧
Percentage oil extracted = ൈ 100 (1)
௠௔௦௦ ௢௙ ௦௔௠௣௟௘
2.3 Experimental Design
Central Composite Design (CCD) of RSM was used for the experimental design to optimize the extraction
parameters. The CCD consisted of three factors: solvent composition from 0 to100% based on n-hexane,
Temperature (30oC - 50oC ) and time of extraction (2- 6 hours) and five level. Variable actual process variable
(‫ݔ‬௜ ) is related to the coded process variables as shown on Table 1 according to equation 1.
(x − x 0 )
Xi = i (2)
∆x
Where, ܺ௜ is the dimensionless coded value of the independent variables
xi is the actual value of the independent variables at the design center point and
∆x is the variation increments about the center point.
The center point chosen for the design were 50% solvent composition based on n- hexane, 40oC temperature and 4
hours time of extraction. The coded, actual values of the variable at various levels and the responses are given in
the matrix Table 3. Three replications were carried out for all experimental design conditions and the average
recorded. Forty- two experimental runs were carried out and the order of the experiment was fully randomized to
reduce the effect of the unexplainable variability in the observed responses due to extraneous factor as
recommended by Singh et al (2003).
2.4 Analysis of Data and Response Equations
Regression models were developed for neem oil yield and each of the five properties of oil as a function of the
three process factors. The Design-Expert 6.0 software was used to analyze the extraction data for developing
response equations, for analysis of variance (ANOVA), generate surface plots and determine optimum extraction
conditions using its optimization toolbox. In multiple regression as in the present case, R2, which is the square of
the adjusted coefficient of determination and standard error are the indices. F statistics shows the significance of
the overall model while the t statistics tests the significance of each of the variables of the model. The function
was assumed to be approximated by a second degree polynomial equation :
ܻ௛ ൌ ܾ௛బ ൅ ∑௠ ܾ௛೔ ܺ௜ ൅ ∑௠ ܾ௛೔೔ ܺ ଶ ௜ ൅ ∑௠
௜ୀଵ ௜ୀଵ ௜ஷ௝ୀଵ ܾ௛೔ೕ ܺ௜ ܺ௝ (3)
Where ܾ௛బ is the value of fitted response at the centre point (0,0,0), and ܾ௛೔ , ܾ௛೔೔ and ܾ௛೔ೕ are linear, quadratic
and cross product regression term respectively. m is the number of factors considered in the study which is equal
to 3, Y1 (Oil yield,%), Y2(pH), Y3 (refractive index), Y4 (acid value), Y5 (iodine value) and Y6 (saponification
value).
2.5 Optimisation and validation of the neem oil extraction
A nonlinear programming problem of the form of equations 4 to 10 was formed from the vector of equation 3 as
follows:

67

Vol.2, No.6, 2012

Maximize : Y1 ( X i ) (4)
Subject to :
Y2 ( X i ) ≤ b 2 (5)
Y3 ( X i ) ≤ b 3 (6)
Y4 ( X i ) ≤ b 4 (7)
Y5 ( X i ) ≤ b 5 (8)
Y6 ( X i ) ≤ b 6 (9)
−2 ≤ X i ≤ +2 (10)
Where bi are the neem oil property requirements. This constrained maximization problem was solved using
the Design-Expert 6.0 software. This problem statement was validated with five (5) randomly generated oil
quality requirements for optimum extraction conditions as shown on Table 2, the results obtained theoretically
were experimented through extraction and subsequent oil analysis. Theses theoretical and the experimental
results were compared using t-test at p<0.05.

3.0 Results and Discussion.
3.1 Response Equations for Neem Oil and Its Properties
The effect of the CCD on the oil yield (Y1), the pH(Y2 ), refractive index (Y3), acid value (Y4 ), iodine value
(Y5) and saponification value (Y6 ) is as shown on Table 3 that was subsequently used to fit the response
equations for oil yield and the five oil properties.
Multiple regression analysis was used as tools of assessment of the effects of two or more independent
factors on the dependent variables (Boonmee et al, 2010). The coefficients of determination R² is a measure of
the total variation of the observed values of the extracted oil about the mean explained by the fitted model
(Shridhar et al, 2010). The factors of the models, their parameter estimates and the statistics of the estimates for
the best functions adopted, taking into consideration all main effects, linear, quadratic, and interaction for each
model are as shown on Table 4. The coefficients of determination (R²) for the responses, yield, pH, refractive
index, acid value, iodine value and Saponification value were 0.990, 0.890, 0.995, 0.971, 0.992 and 0.987
respectively. The coefficients of determination were high for response surfaces, and indicated that the fitted
quadratic models accounted for more than 89% of the variance in the experimental data. Base on p values, the
regression coefficient that were significant at p < 95% were selected for the models that resulted in Equations
(11) to (16). Analyses of variance (ANOVA) were conducted to evaluate the adequacy and consistency of the
models using F- statistic (Shridhar et al, 2010). The analysis of variance of the models is presented in Table 5. As
shown on the Table 5, the F- value for the oil yield (225.13) . pH (13.72), refractive index (372.38), acid value
(74.84), iodine value (120.67) and Saponification value (228.73) were significant at p< 0.05 implying good
model fit.
Solvent ratio (x1) had quite high linear positive effects on oil yield than it had on saponification value,
iodine value, acid value and refractive index but had negative linear effect on pH. Solvent ratio had negative
quadratic effect on yield, refractive index, and acid value. Temperature (x2) had greater positive linear effects on
oil yield compared to pH, refractive index and acid value. It had negative linear effect on iodine value.
Temperature had negative quadratic effect on iodine value but positive quadratic effect on saponification value.
Time of extraction (x3) had positive linear effects on yield, refractive index, acid value and saponification value.
It also had positive significant quadratic effect on yield, refractive index, acid and saponification values, while it
had negative quadratic effect on iodine value. The interaction of the solvent ratio and temperature had positive
effect on saponification value but negative effect on iodine value and pH but no significant effects on other
responses. The interaction of solvent ratio and time had negative effect on the refractive index but no significant
effect on all other responses. The interaction of temperature and time also had positive effect on pH but no
statistical significant effect on other responses.

Yield ( Y1 ) = 33 .30 + 27.23 X1 + 4 X 2 + 1.22 X 3 − 30 .28 X12 + 5.72 X 3 2 R 2 = 0.9903 (11)
pH ( Y2 ) = 3.98 − 0.49 X1 + 0.5 X 2 + 0.48 X 2 2 . − 0.4 X1X 2 + 0.25 X 2 X 3 R 2 = 0.8900 (12)
2 2
Re fractive Index(Y3 ) = 1.53+ 0.044X1 + 6.756E − 3X 2 + 5E − 3X3 − 0.049X1 + 8.813E − 3X3
− 3.3E − 3X1 X3 R 2 = 0.9951
, (13)
Acid Value Y4 ) = 1.82 + 0.47X1 + 0.075X2 + 0.023X3 − 61X12 + 0.2X32
( 2
R = 0.9711 (14)
Iodine Value(Y5 ) = 89.62 − 2 X2 − 4.2X 2 2 − 4.2X32 − 4.25X1X 2 2
R = 0.9920 (15)

68

Vol.2, No.6, 2012

Saponifica Value 6) =176 +3.59 1 +1.78 3 +5.43 22 +5.43 32 +7.25 1X2
tion (Y .23 X X X X X R2 = 0.9870 (16)
3.2 Optimization of the Extraction Process
The models (ܻଵ , ܻଶ , ܻଷ , ܻସ , ܻହ , ܽ݊݀ ܻ଺ ) were useful for indicating the direction in which to change the variables in
order to maximize yield, pH, refractive index, acid value, iodine value and Saponification value. The multiple
regression equations were solved using Design Expert 6.0. The maximum values obtained were as follows: 43.48,
4.99, 1.56, 1.411, 89.345 and 176.643 for yield, pH, refractive index, acid value, iodine value and Saponification
value respectively. The optimum ingredient ratio (coded) predicted for each response and actual values for
optimum response are presented in Table 3.0.
As shown in Table 6, the coded levels are within the experimental range and this indicated that the selected
variables are valid of the selected variables. The regression equation was optimized for maximum value to obtain
the optimum conditions using Design Expert 6.0. The actual value calculated for optimum response as shown
in Table 6, were: 80.77% n- hexane solvent , 34.93 °C temperature and 6.0 ( hours) time for yield, 35.84%
n-hexane solvent , 49.06°C temperature and 3.5 hours, time of extraction for the maximum pH, 83.22%, n-
hexane 49.61°C temperature and 5.4 hours, time of extraction for maximum refractive index, 20.97% n-hexane
solvent, 30.86°C temperature and 2.23 hours time of extraction for minimum acid value, 51.31% n-hexane
solvent, 38.34°C temperature and 3.23 hours time of extraction for maximum iodine value and 62.58% solvent
ratio, 33.49°C temperature and 4.09 hours time of extraction for minimum saponification value.
The response surface plots and the accompany contour plots shown in Figures 1 to 6 for the chosen model
equations shows the relationship between the independent and the dependent variables. From Figure1, the
response surface indicates that the percentage yield of oil increases temperature and solvent composition
increase to optimum condition while further increase led to decrease of percentage yield of oil. In addition, there
was mutual interaction between the temperature and solvent composition. From Figure 2 the pH of oil increased
with increase solvent composition and temperature for a short period, at a longer period, increasing the
temperature of extraction will have negative effect on the pH. From Figures 3 and 4, increase in temperature and
solvent composition led to increase in refractive index and acid values respectively. It can be seen that the
solvent composition was more influential factor that affect the refractive index and the acid value of the
extracted oil.
From Figure 5, it can be seen that increase in temperature and solvent composition increases the iodine
value. Figure 6 show that Saponification value decreased with increase in temperature and solvent composition.
The highest Saponification value was obtained when all the variables were at the minimum point within the
range of study.
3.4 Validation of the Optimization of the Extraction Process
Table 7 shows result of the optimum process variables for each extraction run, its corresponding oil yield and the
Neem Oil properties obtained from the experiment of each validation process. The oil yield range is between
32.85 % and 37.20%. The corresponding result of the analysis of each extraction were compared with the
simulated properties using paired t-test that shows that there are no significant difference between the
simulated properties and their corresponding experimental values at p < 0.05 as presented on Table 8. This result
attests to the effectiveness of this framework for optimum and effective oil extraction.

4.0 Conclusions
This study has clearly demonstrated the applicability of RSM selecting extraction conditions for neem oil from
its seed. This approach has not only resulted in the maximum oil yield through solvent extraction, but has also
guaranteed the fulfillment of the properties requirements of the neem oil The optimum values for yield, pH,
refractive index, acid value, iodine value and saponification value from the surface plot was 43.48%, 4.99, 1.56,
1.411g/g, 89.35g/g, and 176.64 mg/g respectively. The validation experiments and their accompany quality
characteristics were not significantly different from the simulated values at p<0.05

References
Association of Analytical Chemists (AOAC) (2000). Official Methods of Analysis of AOAC International.
Washington Association of Analytical Chemists. 17th Ed, Vol. II
Awad O. M. & Shimaila A. (2003). Operational use of neem oil as an alternative anopheline larvicide. Part A:
laboratory and field efficacy. Eastern Mediterranean Health Journal.9(4): 637-645.
Boonmee, K., Chuntranuluck, S., Punsuvon, V. And Silayoi P. (2010). Optimization of biodiesel Production from
Jatropha (Jatropha Curcas L.), using surface response methodology. Kasetsart J. Nat. Sci 44: 290-299.
International Center of Insect Physiology and Ecology (ICIPE). (1995) The neem tree: An Affordable, Efficient
and Environmentally-Friendly Source of Pest Control Products. International Center of Insect Physiology
and Ecology. Nairobi, Kenya.

69

Vol.2, No.6, 2012

Kovo, A.S (2006). Application of full 42 factorial design for development andcharacterization of insecticidal
soap from neem oil. Leonardo Electronic Journal of Practices and Technologie 8:29-40
Kumar, P. S., Ghosh, D. G. and Pander C. S. (2010). Biological action and medicinal properties of various
constituent of Azadirachta indical (meliaceae a overview). Annals of Biological research 1(3):23-34.
Liauw, M. Y., Natan F. A., Widiyanti P., Ikasiri D. Indraswati N. & Soetaredjo F.E.(2008). Extraction of neem oil
(azadirachta indica a. juss) using N-hexane and ethanol: studies of oil quality, kinetic and thermodynamic.
ARPN Journal of Engineering and Applied Sciences 3(3) :49-54
Muñoz-Valenzuela, S., Ibarra-López A. A., Rubio-Silva L. M., Valdez-Dávila H.& Borboa-Flores J. (2007).
Neem tree morphology and oil content. Reprinted from Issues in new crops and new uses. 2007. J. Janick
and A. Whipkey (eds.).ASHS Press, Alexandria, VA.
National Research Council (NRC)(1992). Neem: a tree for solving global problems. National Academy Press,
Washington, DC
Okonkwo, E. M. (2004).Employment creation and opportunities in manufacturing sub-sectors: Acase for neem
tree in Nigeria. Bullion, publication of Central Bank of Nigeria. 28(3):30-35
Peter, F. (2000). Global Neem usage. Deutsche Gesellschaft fdür Technische Zusammenarbeit (GTZ) GmbH
Eschborn, Germany.
Ranajit, K. B., Kausik B. IshitaC. & Uday B.(2002) Biological activities and medicinal properties of neem
(Azadirachta Indical). Current Science 82(11):1036 -1045.
Soetaredjo F .E., Budijanto G. M., Prasetyo R. I., & Indraswati N.(2008). Effects of pre-treatment condition on
the yield and quality of neem oil obtained by mechanical pressing. ARPN Journal of Engineering and
Applied Sciences 3(5) :45-49
Shridhar,B.S. Beena, K.V. Anita,M.V. & Paramjeet K.B.(2010) Optimization and charaterization of castor seed
oil. Leonardo Journal of Science 17:59-70.
Singh, S., Raina, C. S., Bawa, A. S. & Sexena D. C. (2003). Sweet potato-based pasta product:
optimization of ingredient levels using response surface methodology. International Journal of Food Science
and Technology 38:1-10
Zahedi, G., Azizi S., Hatami, T., & Sheikhattar, L. (2010). Gray box modelingof supercritical Nimbin Extraction
from Neem Seeds using Methanol as Co-Solvent. The Open Chemical Engineering Journal 4: 21-30.

Table 1: Experimental Increments, values of coded levels

Factors ± Increment Xi Coded Levels
∆x -2 -1 0 +1 +2

x1 (%) ± 25 0(100) 25(75) 50(50) 75(25) 100(0)
0
x2( C) ±5 30 35 40 45 50
x3(Hr) ±1 2 3 4 5 6

Table 2: Neem oil quality requirements for validation

Properties / Runs 1 2 3 4 5
pH 4.6 4.0 4.5 4.5 4.6
Refractive Index 1.5476 1.5000 1.50 1.53 1.5473
Acid Value 1.4280 1.5000 1.8000 1.90 1.80
Iodine Value 72 80 78 85 90
Saponification Value 202 190 192 197 200

70

Vol.2, No.6, 2012

Table 3: Central composite design arrangement and responses

Coded Level Actual Values Reponses
Expt. No X1 X2 X3 Solvent Temp. Time ܻଵ ܻଶ ܻଷ ܻସ ܻହ ܻ଺
1 -1 -1 2 25(75) 35 6 12 4.2000 1.4983 1.4280 93 172
2 -1 0 0 25(75) 40 4 12 4.2000 1.5000 1.4280 90 175
3 -1 0 -1 25(75) 40 3 13 3.9000 1.5000 1.4470 91 174
4 -1 1 -1 25(75) 45 3 17 2.6000 1.5067 1.5400 90 176
5 0 -2 -2 50(50) 30 2 22 4.1000 1.5152 1.6180 91 174
6 0 -2 0 50(50) 30 4 24 4.2000 1.5186 1.6560 90 176
7 0 -1 0 50(50) 35 4 31 3.9400 1.5304 1.8000 90 176
8 0 0 0 50(50) 40 4 34 3.9000 1.5355 1.8460 89 177
9 0 0 -2 50(50) 40 2 31 3.8200 1.5304 1.7890 91 175
10 0 0 -1 50(50) 40 3 33 3.8400 1.5338 1.8270 89 176
11 0 0 1 50(50) 40 5 36 3.9800 1.5388 1.8840 89 178
12 0 0 2 50(50) 40 6 36 3.9800 1.5388 1.8840 88 178
13 0 1 1 50(50) 45 5 37 3.5000 1.5405 1.9030 88 179
14 0 1 0 50(50) 45 4 35 3.4400 1.5372 1.8650 88 178
15 0 2 2 50(50) 50 6 37 3.4000 1.5405 1.9000 87 181
16 0 2 0 50(50) 50 4 36 3.2000 1.5388 1.9200 87 180
17 1 -2 -2 75(25) 30 2 38 4.0900 1.5422 1.9000 90 176
18 1 -2 1 75(25) 30 5 38 4.1800 1.5422 1.9220 89 176
19 1 -1 -1 75(25) 35 3 39 4.0600 1.5439 1.9410 89 177
20 1 -1 1 75(25) 35 5 40 4.0800 1.5456 1.9600 88 178
21 1 0 0 75(25) 40 4 39 3.8000 1.5439 1.8800 89 178
22 1 0 1 75(25) 40 5 40 3.8000 1.5456 1.9600 87 179
23 1 1 1 75(25) 45 5 40.6 3.6000 1.5466 1.9600 87 181
24 1 1 -2 75(25) 45 2 39 3.1000 1.5439 1.9410 88 178
25 1 1 -1 75(25) 45 3 39 3.6000 1.5439 1.9410 88 179
26 1 1 0 75(25) 45 4 40 3.6000 1.5456 1.9600 87 180
27 1 1 2 75(25) 45 6 41 3.9000 1.5473 1.9790 87 181
28 1 2 2 75(25) 50 6 41 3.8000 1.5473 1.9790 85 182
29 1 2 1 75(25) 50 5 41 3.7000 1.5473 1.9790 86 182
30 2 -2 -2 100(0) 30 2 31 3.8000 1.5304 1.7890 87 180
31 2 -2 2 100(0) 30 6 33 3.9000 1.5338 1.8270 85 183
32 2 -1 -1 100(0) 35 3 38 3.6000 1.5422 1.9220 85 182
33 2 -1 2 100(0) 35 6 39 3.8000 1.5439 1.9410 84 185
34 2 0 0 100(0) 40 4 39 3.7000 1.5439 1.9410 84 184
35 2 0 2 100(0) 40 6 39 3.8000 1.5439 1.9410 83 187
36 2 1 1 100(0) 45 5 40 3.6000 1.5456 1.9600 83 186
37 2 1 2 100(0) 45 6 40 3.8000 1.5456 1.9600 81 189
38 2 2 2 100(0) 50 6 41 3.6000 1.5473 1.9790 72 202
39 2 2 -2 100(0) 50 2 39 3.5000 1.5439 1.9410 75 198
40 2 2 -1 100(0) 50 3 39 3.6000 1.5439 1.9410 73 200
41 2 2 0 100(0) 50 4 40 3.6000 1.5456 1.9600 73 201
42 2 2 1 100(0) 50 5 41 3.6000 1.5473 1.9790 72 201

71

Vol.2, No.6, 2012

Table 4: Estimated coefficients of the fitted quadratic equation for different responses.
Estimated Coefficients
Factors Oil Yield (Y1) pH(Y2) Refractive index (Y3 ) Acid value (Y4 ) Iodine value (Y5 ) Saponification value
(Y6 )
Coefficien p-valu Coefficien p-value Coefficien p-valu Coefficien p-value Coefficien p-value Coefficien p-valu
ts e ts ts e ts ts ts e
Consta 33.30 0.0001 3.98 0.0118* 1.53 0.0001 1.82 0.0001* 89.62 0.0002* 176.23 0.0001
nt * * *
x1 27.23 0.0001 -0.49 0.0177* 0.044 0.0001 0.47 0.0001* -1.43 0.1501 3.59
* * 0.0217
*
x2 4.00 0.0001 0.50 0.0230* 6.756E-03 0.0001 0.075 0.0005* -2.00 0.0913* 2.00 0.1365
* * *
x3 1.22 0.0119 0.14 0.3721 5E-03 0.0085 0.023 0.0785* 0.000 0.0000* 1.78
* * * 0.0052
*
x12 -30.28 0.0001 -7.493E-03 0.9906 -0.049 0.0001 -0.61 0.0001* -1.54 0.3419 2.41 0.2383
* *
x22 -0.20 0.8955 0.48 0.0298* -8.34E-04 0.7233 0.074 0.2035 -4.20 0.0467* 5.43
0.0250
*
x32 5.72 0.0106 -0.37 0.9499 8.813E-03 0.0046 0.20 -4.20 0.0467* 5.43
* * 0.0076* 0.0250
*
x1x2 -1.149E-14 1.0000 -0.40 0.0538* -5E-05 0.9771 0.011 0.7701 -4.25 0.0145* 7.25 0.0018
* *
x1x3 -2.00 0.533 0.16 0.3401 -3.30E-03 0.0477 -0.038 0.2781 -1.25 0.2916 -0.25 0.8791
*
x2x3 1.094E-14 1.0000 0.25 0.0071* 2.463E-16 1.0000 8.014E-16 1.0000 -0.25 0.4888 0.25 0.5746
R2 0.9903 0.8900 0.9951 0.9711 0.9920 0.9870
*Significant at p < 0.05 level ** Significant at p < 0.1 level

Table 5: Analysis of variance for the responses

Sources of Sum of Mean Adjusted
Responses Variation d.f Squares Square F ܴଶ
ܻଵ Regression 9 2642.248 293.583 225.13 0.9903
Residual 32 112.008 3.500
Total 41 2754.256
ܻଶ Regression 9 2.972 0.330 13.72 0.8900
Residual 32 1.124 3.51E-02
Total 41 4.074
ܻଷ Regression 9 7.43E-03 8.25E-04 372.38 0.9951
Residual 32 3.05E-04 9.53E-06
Total 41 7.73E-03
ܻସ Regression 9 0.925 0.103 74.84 0.9711
Residual 32 4.91E-02 1.53E-03
Total 41 0.974
ܻହ Regression 9 1085.426 120.603 120.67 0.9920
Residual 32 108.193 3.381
Total 41 1193.619
ܻ଺ Regression 9 2313.452 257.050 228.73 0.9870
Residual 32 217.882 6.809
Total 41 2531.333

72

Vol.2, No.6, 2012

Table 6: Calculated coded and Actual Value for optimum responses.

OPTIMUM RESPONSE LEVEL
Coded Values
COMPOSITION ܻଵ ܻଶ ܻଷ ܻସ ܻହ ܻ଺
ܺଵ 1.2308 -0.5664 1.3288 -1.1612 0.0524 0.5032
ܺଶ -1.0140 1.8120 1..9220 -1.8280 -0.0332 -1.302
ܺଷ 2.0000 -0.5000 1.4000 -1.7700 -0.7700 0.0900

Actual Values
ܺଵ 80.77 35.84 83.22 20.97 51.31 62.58
ܺଶ 34.93 49.06 49.61 30.86 38.34 33.49
ܺଷ 6.000 3.50 5.4 2.23 3.23 4.09
Optimum values 43.48 4.99 1.56 1.411 89.35 176.64

Table 7: Optimum Process Variables, Oil Yield and Physical Properties of the Neem Oil for Validation
Variable / Properties 1 2 3 4 5
Optimum Process Variable
X1 (%) 100 100 100 100.00 97.54
X2 (0C) 50 48.05 48.67 48.57 49.75
X3(Hours) 5.23 4.42 4.90 4.94 4.51
Simulated Oil Yield (%) 37.18 33.80 35.40 35.52 36.42
Corresponding Neem Oil Properties
Oil Yield (%) 37.20 33.97 35.43 32.85 36.38
Ph 4.4 3.98 4.3 4.5 4.7
Refractive Index 1.5413 1.4900 1.5000 1.53 1.4541
Acid Value 1.4182 1.5000 1.9000 1.90 1.8110
Iodine Value 71.8 79.9 78.2 82.9 87.80
Saponification Value 197.8 191.3 193.2 196.5 194.8

Table 8: Summary of t-Test for Properties of Neem Oil Properties for Validation

Source of Variation t-Value p- value (2 tail) Remark
Oil Yield 1.064 0.450 No Significant Difference
pH 1.082 0.340 No Significant Difference
Refractive Index 1.222 0.289 No Significant Difference
Acid Value -1.001 0.373 No Significant Difference
Iodine Value 1.683 0.168 No Significant Difference
Saponification Value 1.087 0.338 No Significant Difference
*Significant level at p < 0.05

73

Vol.2, No.6, 2012

50.00
Oil Yield

50.319

41.4973 45.00

32.6756

perature
23.8539
Oil Yield

40.00 2 6.794 5 38.55 68
32.6756
15.0322

B: Tem
2 0.913 4 38.5 568
Predictio 44 .4379
43.48
35.00

50.00
100.00
45.00
81.25
40.00 30.00
62.50
25.00 43.75 62.50 81.25 100.00
B: Temperature 35.00 43.75
A: Solvent Composition
30.00 25.00 A: Solvent Compos ition

(a) (b)
Figure 1: (a) Response surface plot showing the 3D effect of temperature and solvent and their
interaction effect on the yield of Neem oil. (b) Contour Plot of Figure 1a
Predictio 4.99
50.00
pH
5 .0 0 5 5 9
4 .6 8 6 1 7

4 .3 6 6 7 5

5 .32 5 0 1 45.00
4 .0 4 7 3 3
4 .8 4 5 8 8
: e p ra re
B T m e tu

4.3 6 6 7 5

3 .8 8 7 6 2 40.00
pH

3 .7 2 7 9 1
3 .4 0 8 4 9

35.00

5 0 .0 0
1 0 0 .0 0
4 5 .0 0
8 1 .2 5
30.00
4 0 .00
6 2 .5 0 25.00 43.75 62.50 81.25 100.00
B: Temperature 35 .0 0 4 3 .7 5
A : Solvent Composition
3 0 .0 0 2 5 .0 0 A: Solvent Com pos ition

(a) (b)
Figure 2: (a) Response surface plot showing the 3D effect of temperature and solvent composition their
interactive effect on the pH of Neem oil. (b) Contour Plot of Figure 2a

Predictio 1.56
Refractive Index
50.00

45.00

1.55796
B: Temperature

1.54407

1 .53017 1 .52091
1.53017
Refractive Index

40.00
1.5394 4
1.5162 8
1.5394 4
1.502 38
1 .51165 1.54 87

35.00

50.00
100.00
4 5.00
81.25 30.00
40.00 25.00 43.75 62.50 81.25 100.00
62.5 0
A : Solvent Composition
30.00 25.00 A: Solvent Composition

(a) (b)
Figure 3: (a) Response surface plot showing the 3D effects of temperature and solvent composition and
their interactive effect on the refractive index of the Neem oil. (b) Contour Plot of Figure 3a.

74

Vol.2, No.6, 2012

50.00
Acid Value

2.122 77 45.00

1.9660 2

B T perature
1 .80 927
Acid Value

1.652 51 40.00
1 .7 04 7727 13 77
1 .80 9 1 .9
1 .91 3 77

: em
1.49576

1 .6 00 26 2.01 82 7

35.00

1 .80 9 27
50.00
100.00
45 .00
81 .25
30.00
40.00
62.50 25.00 43.75 62.50 81.25 100.00
B: Temperature 35.00 4 3.7 5
A : Solvent Compos ition
30.00 25.00 A: Solvent Com pos ition

(a) (b)
Figure 4: (a) Response surface plot showing the 3D effects of temperature and solvent and their
interactive effects on the acid value of Neem oil. (b) Contour Plot of Figure 4a
50.00
Iodine Value
79.2172
81.3602
83.5032
85.6462
89.5209
45.00
86.4228
87.7892
B: Temperature

83.3247
Iodine Value

80.2266
40.00
Predictio 89.35
77.1285

35.00

50.00
87.7892
100.00
45.00
81.25
30.00
40.00
62.50 25.00 43.75 62.50 81.25 100.00
(a) (b)
Figure 5: (a) Response surface plot showing the effect of temperature and solvent composition and their
interactive effect on the iodine value of the Neem oil. (b) Contour Plot of Figure 5a.

50.00
Saponification Value
19 2.23
18 8 .8 06
1 85 .3 8 1
195.655
45.00
18 1.9 56
190.518
Saponification Value

185.381 1 7 8.53 2
perature

180.244
40.00
B: Tem

175.107

35.00
Predictio 176.64
50.00
100.00 1 78 .5 3 2
45.00
81.25
40.00 30.00
62.50
25.00 43.75 62.50 81.25 100.00

(a) (b)
Figure 6: (a) Response surface plot showing the effects of temperature and solvent composition and their
interactive effect on the Saponification value of Neem oil. (b) Contour Plot of Figure 6a

75

This academic article was published by The International Institute for Science,
Technology and Education (IISTE). The IISTE is a pioneer in the Open Access
Publishing service based in the U.S. and Europe. The aim of the institute is
Accelerating Global Knowledge Sharing.

More information about the publisher can be found in the IISTE’s homepage:
http://www.iiste.org

The IISTE is currently hosting more than 30 peer-reviewed academic journals and
collaborating with academic institutions around the world. Prospective authors of
IISTE journals can find the submission instruction on the following page:
http://www.iiste.org/Journals/

The IISTE editorial team promises to the review and publish all the qualified
submissions in a fast manner. All the journals articles are available online to the
readers all over the world without financial, legal, or technical barriers other than
those inseparable from gaining access to the internet itself. Printed version of the
journals is also available upon request of readers and authors.

IISTE Knowledge Sharing Partners

EBSCO, Index Copernicus, Ulrich's Periodicals Directory, JournalTOCS, PKP Open
Archives Harvester, Bielefeld Academic Search Engine, Elektronische
Zeitschriftenbibliothek EZB, Open J-Gate, OCLC WorldCat, Universe Digtial
Library , NewJour, Google Scholar

Optimization of neem seed oil extraction process using response surface methodology

More Related Content

What's hot (20)

Viewers also liked (7)

Similar to Optimization of neem seed oil extraction process using response surface methodology (20)

More from Alexander Decker (20)

Recently uploaded (20)

Optimization of neem seed oil extraction process using response surface methodology