Evaluation… NdeBupet al.
Evaluation of the cooking process of neem (Azadirachtaindica) nuts as a pretreatment prior to oil extraction
NdeBup Divine1,5, Siriyabe Marth2, Mohagir M. Ahmed3, FonAbi Charles4, Zourmba Paul2, NkengElambo George5, Kapseu Cesar6
1Higher Institute of the Sahel, University of Maroua, P.O. Box 46, Maroua, Cameroon
2Department of Chemistry, Higher Teachers’ Training College, University of Maroua, Cameroon
3Faculty of Science and Technologies, University of Sarh, P.O. Box 105, Chad
4Department of Chemistry, Higher Teachers’ Training College, University of Yaounde1, Cameroon
5Department of Chemistry, Faculty of Science, University of Buea, Cameroon and EcoleNationale Supérieure des TravauxPubliques (ENSTP) B.P 510, Yaounde, Cameroon
6 Department of Process Engineering, ENSAI, University of Ngaoundere, P.O. Box 455 Ngaoundere, Cameroon
Received: 22May 2014; Revised: 14 August 2014; Accepted: 22 August 2014
Abstract: Traditional methods are usually used locally to produce ‘bio-neem oil’ which is void of residual solvents. The major disadvantages of the method are low oil yields and high acid values. To ameliorate this process, cooking of neem nuts was evaluated by response surface methodology. Independent variables studied were cooking time and temperature while the responses were moisture content of the cooked kernels, quantity of oil extracted, acid value and the refractive index of the oil. Cooking time and temperature significantly influenced the cooking process but their interaction effect was negligible. Optimum ranges obtained for the cooking process were cooking time (5-20 min) and cooking temperature (40-50oC). At the mid-point of these ranges (12.5 min and 45 oC) the optimum responses obtained were: moisture content 46.37 % w.b., oil content 21.43 %, acid value 3.53 mg KOH/g oil and refractive index 1.460. Experiments conducted under these optimal conditions showed no significant difference with the calculated results. Results for the raw kernels were: moisture content 43.00, quantity of extracted oil 15.88, Acid value 8.13, and refractive index 1.4610.01, indicating that, there was a 35 % increase in quantity of oil extracted and a 57 % reduction in acid value.
Key words: Neem, cooking, traditional extraction, optimization.
INTRODUCTION
Vegetable oils have played and continue to play a very vital role in human nutrition, traditional medicines, cosmetics and other diverse uses in industry. Market demands and the development of new uses have generated interest on scientific and technical research on under exploited oilseeds and oil bearing materials1-5. Research in this area has been centered on improved extraction and characterization methods that give oil of high quality in good quantity. There exists today a plethora of oilseeds which are still under-exploited and have not been given proper research attention in order to fully derive the potential benefits that they present. One of such oilseed is Azadirachtaindicacommonly called the neem tree. Neem is a tree of Indian origin with almost each part of the tree having potential curative effects in traditional medicines. Tinghuiet al.6 reported that, the neem tree is present in Asia (Bangladesh, Burma, Cambodia, India, Indonesia, Iran, Malaysia, Nepal, Pakistan, Sri-Lanka,Thailand and Vietnam); in Africa (Benin, Burkina Faso, Cameroon, Chad, Ethiopia, Gambia, Ghana, Guinea, Ivory Coast, Kenya, Mali, Mauritania, Mozambique, Niger, Nigeria, Senegal, Somalia, Sudan, Tanzania and Togo,) and the Americas (Antigua, Barbados, Belize, Bolivia, Brazil, Dominican Republic, Ecuador, Guatemala, Guyana, Honduras, Jamaica, Mexico, Nicaragua, Suriname, St. Lucia and Trinidad and Tobago).The tree is therefore widely distributed and given the potentials of neem oil for use in pharmaceutics, cosmetics and as a potential source of raw material in the biodiesel industry it is expected that the demand for this oil should expand in the near future7-9. Some of the applications of neem oil require the use of an extraction method that retains as much as possible the natural properties of the oil that confer curative and/or insecticidal properties. One of such methods is the cold extraction method. In Cameroon and in most African countries extraction is done mostly in a traditional manner to produce the so called ‘bio-neem oil’ (that is, neem oil extracted without the use of solvents). This method consists of depulping the fruit, and decorticating the nut to have the kernel, sun drying of the kernel, grinding and extraction. In the extraction process the ground kernels are boiled in water and the oil layer on the surface of the water is ladled off and further heated to remove the residual water. This is called the water extraction method. In the second traditional extraction method, the ground kernels are mixed in a mortar with a pestle with very little quantities of warm water added at intervals to decrease oil viscosity and aid oil flow. This is referred to as water assisted traditional extraction method. The major disadvantages of these methods are low yields and high acid values of the oil. Meanwhile it has been demonstrated that cooking oilseeds prior to processing improved oil yields and reduced FFA values of shea butter10-11. Cooking of neem nuts prior to oil extraction may also offer these advantages. The precise cooking conditions vary as a function of type and size of oilseed, temperature, cooking time etc. It is necessary to define these optimum conditions for each type of oilseed as such data is very valuable in engineering calculations for process scale-up. The Doehlert experimental design has been reported as a useful surface response method to optimize various processes12. The various advantages of this design over others are well elaborated by Mathieu and Phan-Tan-Luu13 and Imandiet al.12 and are summarized in BupNdeet al.10. Studies evaluating the influence of cooking of neem nuts on oil yields and quality to the best of our knowledge are scarce in the literature. It is worth noting that the traditional processing and sale of neem oil is done mostly by local women and this represents substantial income earning opportunities for them. Studies that can improve yields and quality of neem oil will therefore be beneficial to them as they may sell at higher prices and earn more money. The aim of this work was therefore to evaluate the effect of cooking on the traditional oil extracted yields and some quality parameters of neem oil.
Table 1: Experimental matrix and values of the responses obtained
Coded Values / Real values / ResponsesExp't N0 / x1 / x2 / X1 / X2 / Y1 / Y2 / Y3 / Y4
1 / 0 / 0 / 62.5 / 70 / 51.02 / 14.60 / 7.39 / 1.463
1 / 0 / 0 / 62.5 / 70 / 46.94 / 18.67 / 7.70 / 1.461
1 / 0 / 0 / 62.5 / 70 / 45.10 / 18.51 / 7.08 / 1.463
2 / 1 / 0 / 120 / 70 / 53.06 / 16.61 / 8.71 / 1.465
2 / 1 / 0 / 120 / 70 / 48.98 / 16.42 / 8.09 / 1.463
2 / 1 / 0 / 120 / 70 / 52.20 / 16.45 / 8.80 / 1.463
3 / 0.5 / 0.866 / 91.25 / 100 / 58.00 / 16.97 / 6.41 / 1.463
3 / 0.5 / 0.866 / 91.25 / 100 / 56.00 / 18.09 / 5.29 / 1.461
3 / 0.5 / 0.866 / 91.25 / 100 / 59.31 / 18.20 / 5.85 / 1.463
4 / -1 / 0 / 5 / 70 / 48.08 / 20.71 / 5.05 / 1.463
4 / -1 / 0 / 5 / 70 / 49.02 / 18.57 / 6.35 / 1.461
4 / -1 / 0 / 5 / 70 / 48.65 / 18.91 / 5.66 / 1.461
5 / -0.5 / -0.866 / 33.75 / 40 / 48.98 / 19.59 / 4.95 / 1.461
5 / -0.5 / -0.866 / 33.75 / 40 / 44.90 / 19.33 / 5.04 / 1.461
5 / -0.5 / -0.866 / 33.75 / 40 / 47.08 / 19.31 / 3.92 / 1.461
6 / 0.5 / -0.866 / 91.25 / 40 / 50.00 / 15.46 / 7.40 / 1.461
6 / 0.5 / -0.866 / 91.25 / 40 / 46.00 / 16.81 / 8.46 / 1.461
6 / 0.5 / -0.866 / 91.25 / 40 / 48.34 / 15.55 / 7.72 / 1.461
7 / -0.5 / 0.866 / 33.75 / 100 / 59.18 / 13.53 / 7.79 / 1.463
7 / -0.5 / 0.866 / 33.75 / 100 / 60.78 / 14.66 / 7.40 / 1.463
7 / -0.5 / 0.866 / 33.75 / 100 / 58.90 / 14.50 / 8.01 / 1.461
x1, and x2 are the coded values while X1, and X2 are real values of cooking time and cooking temperature respectively. Y1, Y2, Y3 and Y4 are moisture content, amount of oil extracted, acid value and refractive index respectively
MATERIALS AND METHODS
Mature neem fruits used in this work were collected from the town of Maroua, Cameroon in April 2013. They were stored in a freezer at -18oC to limit oil hydrolysis. Before the cooking experiments, the fruits were withdrawn from the freezer and stored on a laboratory bench overnight to thaw at ambient temperature. The samples were then depulped manually and washed in a large quantity of distilled water to remove the residual pulp. The nuts were cooked following the Doehlert experimental design chosen in this work (Table 1).In the cooking process, the experimental variables were cooking time (5-120mins) and cooking temperature 40-100oC. 350mL of distilled water were placed in a 600 mL beaker and heated to the desired experimental temperature. 300g of neem nuts were then placed in the beaker and the beaker covered with Aluminium foil and cooked for the required length of time (Table1). After cooking the water was drained and the nuts were manually cracked to give the kernels. A portion of the kernels was withdrawn for the determination of moisture content while the other portion was dried at 50oC in an electric oven for 24 h to a constant mass. This latter portion was used for oil extraction and subsequent oil analysis.
Determination of moisture content of cooked kernels: Moisture content was determined by drying a known quantity of the cooked kernels in a ventilated electric oven at 105oC for 24 hours. The dried kernels were cooled in a dessicator and reweighed14. Moisture content (Y1 % w.b.) was calculated from the relation
[1]
M0 is wet mass of sample and M1 is the dry weight of the sample. All experiments were carried out in triplicates.
Oil extraction process: The dried kernels were ground using a kitchen type manual grinder. The adjusting knob of the grinder was set to maximum for each run, to ensure that grinding for the different runs was done under uniform conditions. In the extraction process, the traditional water assisted extraction method which is normally used in the production of organic neem oil was used. This consisted of mixing and pressing the ground paste in a laboratory porcelain mortar using a pestle with occasional addition of 15-20ml of warm water (40oC) at 3 different intervals. The oil was then drained from the mortar and filtered through a Whartman No. 42 filter paper. The oil was placed in an oven at 60oC to remove residual water for 2 hours. A control experiment was carried in a similar manner but for the fact that the samples were not cooked prior to drying and oil extraction. The average extraction time for each run was 15-20 minutes. All extraction experiments were carried out in triplicates. The quantity of oil extracted was calculated from the equation
[2]
Determination of Acid value:Acid value was determined by back titration. 2 g of the oil sample was weighed into a conical flask and 0.2N ethanolic KOH solution was added to the flask and shaken vigorously for about 1 minute. Excess KOH remaining in the flask was then titrated against 0.1N HCl, using phenolphthalein as indicator. A blank titration was carried out in a similar manner without the oil sample. This was done in triplicates. Acid value was calculated from the equation.
[3]
N is the normality of KOH, V1 is the titre volume for the blank and V2 the titre volume for the test sample.
Modeling of the cooking process:A second order model with interaction was proposed to describe the cooking process of neem kernels. This model was represented as follows:
[4]
The coefficients of the model b0, b1, b2, b11, b22 andb12 were determined through multiple linear regressions on SigmaPlot 11.0 (SystatInc, USA).
Validation of Models:The criteria for validating models were the regression coefficient (R2) and/or absolute error (AE) observed between the experimental and theoretical results. R2 was obtained from regression analysis while AE was determined from the equation,
[5]
Where Yexp and Ycalare the measured and calculated responses respectively. n is the number of points at which measurements were carried out. A model was considered valid if at least one of the two criteria was satisfied. That is R2>0.7 and/or EA<10%
Optimization of the cooking process:To optimize the cooking process, the optimum point for each response (minimum or maximum point) of Eq. 1 was defined as the point where the first partial derivative of the function equals zero10:
[6]
The system of equations for each response was solved to give the values of x1 and x2 in coded values which were then transformed to real values. Experiments were then conducted at these points to verify the mathematically determined optimal conditions.
RESULTS AND DISCUSION
Modeling of the cooking process: Values of the responses studied Y1 (Moisture content % w.b), Y2 (Quantity of oil extracted %) and Y3 (acid value mg KOH/g oil) are presented in Table 1 while the model constants, P values, value of model constants (VMC), R2 and AE values for the second order models are presented in table 2. A model was considered valid if R2 > 0.700 and/or AE < 10. Table 2 shows that all of the responses satisfied either one or all of the conditions. These models were therefore judged satisfactory to describe the influence of the variables (cooking temperature and cooking time separately or in combination) on the measured responses.
To better appreciate the influence of the variables on the responses, the validated model was used in four different ways: First the probability (p) values given by multiple linear regression analysis for each model coefficient for a particular response were considered. A factor associated to a model constant with p < 0.05 was considered to have a significant effect on the response of interest. Secondly, the percentage value of each model constant (CMC) calculated as its ratio to the sum of all the models constants (absolute values) for the response in question was considered. The higher the value, the higher its contribution to the magnitude of the response. A positive sign associated with the model coefficient will tend to increase the value of the response, all other factors being constant and vice versa. Thirdly, the effect of each independent factor on the responses was considered by Maintainingone of the factors constant at the central point (cooking time 62.5 min or cooking temperature 70oC) and plotting the other as a function of the response. Lastly the combined effect of the two factors was evaluated using surface response plots. The following sections describe the effect of Cooking time and temperature on the responses using these four approaches.
Table 2: Model constants (MC), P values, contribution of model constants (CMC), R2 and AE values for the second order models
Moisture content (%) / Quantity of oil extracted / Acid value / Refractive indexConstant / MC / VMC / P-value / MC / VMC / P-value / MC / VMC / P-value / MC / VMC / P-value
b0 / 47.687 / 72.86 / 0.001 / 17.26 / 69.85 / 0.001 / 7.39 / 56.11 / 0.001 / 1.462 / 99.81 / 0.001
b1 / 0.823 / 1.26 / 0.230 / -0.959 / 3.88 / 0.027 / 1.172 / 8.90 / 0.001 / 0.0007 / 0.05 / 0.067
b2 / 6.435 / 9.83 / 0.001 / -0.972 / 3.93 / 0.025 / 0.314 / 2.38 / 0.112 / 0.0008 / 0.05 / 0.037
b11 / 2.312 / 3.53 / 0.118 / 0.685 / 2.77 / 0.421 / -0.280 / 2.13 / 0.488 / 0.0003 / 0.02 / 0.648
b22 / 6.478 / 9.90 / 0.001 / -0.797 / 3.23 / 0.350 / -1.067 / 8.10 / 0.016 / 0.001 / 0.07 / 0.182
b12 / -1.719 / 2.63 / 0.211 / 4.038 / 16.34 / 0.001 / -2.948 / 22.38 / 0.001 / 0.000 / 0.000 / 1.00
R2 / 0.890 / 0.740 / 0.884 / 0.470
AE (%) / 2.73 / 4.32 / 6.75 / 0.05
Effect of Cooking time and Cooking Temperature on Moisture content:The evaluation of the moisture content of neem nuts after cooking is important because it plays a vital role to determine drying time of the kernels and in the hydrolysis of the oil. From table 2 it can be observed that the p-values of the model constants b2 and b22 which are associated with the linear and quadratic effect of temperature are both less than 0.05 indicating that they had significant effect on the moisture content. Again table 2 shows that apart from the constant term which had the highest contribution (77.05 %) to the value of the moisture content (as expected), b2 and b22 had positive contributions of 9.83 and 9.90% to the magnitude of the moisture content. That is, there is the tendency of moisture content increasing with cooking temperature. This can be confirmed by the significant variation of moisture content with temperature at constant cooking time (Figure 1a). As cooking temperature increases there is the possibility that heat denatures the membranes of the neem nut making it more porous and permeable to the cooking water which leads to the observed increase in water absorption as temperature increases. Neem nuts are small in size measuring about 14 x 6 mm; a factor that can favor water absorption15. Several studies have shown that water absorption increases with temperature and this has been linked to the increased porosity of the substrate brought about by cooking16-20. The contribution of cooking time to water absorption by the kernels was small compared to temperature and was insignificant at the 95% level of confidence. Figure 1a shows very slight variation of moisture content with cooking time. Figure 1b shows the combined effect of cooking time and temperature on the moisture content which confirms the previous discussions. Moisture content varied significantly with temperature but does not change much as a result of increased or prolonged heating. Therefore cooking should be done at lower temperatures and shorter times to limit water absorption.
Figure 1a: Influence of cooking time (temperature constant at 70oC) and cooking temperature (Time constant at 62.5 min) on the moisture content of the cooked kernels
Figure 1b: Combined influence of cooking time and temperature on moisture content of the cooked kernels
Quantity of oil extracted: The quantity of oil extracted was significantly influenced by the linear effects of cooking time and temperature as well as the interaction effect of the factors (b12) which had a 16.34 % tendency of increasing the quantity of oil extracted (Table 1). Cooking time and temperature had the tendency of reducing the quantity of oil extracted as exemplified by their negative coefficients. The quantity of oil extracted was highest within the first few minutes of cooking and then reduced to significantly lower values as cooking time and temperature increased. Cooking coagulates proteins thereby allowing free space for the diffusion of oil. The gelatinization temperatures for starches vary generally from 52 to 83.5 °C, while proteins are equally coagulated within this temperature range21. It can be observed from figure 2b that, high quantities of oil were extracted within the temperature of protein coagulation and starch gelatinization. BupNdeet al.22 demonstrated the denaturation of proteins of another oilseed sheanut kernel by cooking at 85 ± 5oC. The decrease in oil yield with time and temperature observed in this work was therefore contrary to our expectation. A possible explanation to this could be due to the fact that neem nuts are small in size (14 x 6mm) and can absorbed water easily as reaction progresses (Figures 1a and b). Therefore the presence of absorbed water in the kernels blocks available pores for oil flow which could hinder the release of oil. The variation of quantity of oil extracted with temperature observed in this work was different from that reported by BupNdeet al.10 in which the amount of oil extracted increased continuously with cooking time up to a steady value. This difference could be attributed to the small size of neem nut (14x6 mm) compared to that of shea nut (45 x 25mm) which permitted water absorption in neem nuts right from the onset of the cooking process. Overall the quantity of oil extracted from the cooked kernels was significantly higher than that extracted from the raw ones.