 Open Access
 Authors : Kenmogne Sidonie Beatrice , Tsatsop Tsague Roli Karole , Nouga Bissoue Achille , Toze Flavien Aristide, Kamdem Waffo Alain Francois, Ngassoum Martin Benoit
 Paper ID : IJERTV10IS050052
 Volume & Issue : Volume 10, Issue 05 (May 2021)
 Published (First Online): 12052021
 ISSN (Online) : 22780181
 Publisher Name : IJERT
 License: This work is licensed under a Creative Commons Attribution 4.0 International License
Optimization of Microwave Assisted Extraction of Phenolic Compounds and Antioxidant Activities of The Roots of Ximenia Americana with Response Surface Methodology
Kenmogne Sidonie Beatrice*1, Tsatsop Tsague Roli Karole2, Nouga Bissoue Achille3, Toze Flavien Aristide1, Kamdem Waffo Alain Francois1, Ngassoum Martin Benoit2
1Department of Chemistry, Faculty of Science, University of Douala, P.O. Box. 24157 Douala, CAMEROON;
2 Laboratory of Applied Chemistry, Department of Applied Chemistry, P.O.Box. 455. National Advanced School of Agro Industrial Sciences, Ngaoundere, CAMEROON.
3 Advanced Teachers Training School for Technical Education, University of Douala, Douala 1872, Cameroon;
Abstract: The optimal conditions of MicrowaveAssisted Extraction (MAE) of phenolic compounds and antioxidants from X. americana roots determined. A secondorder regression for central composite design (CCD) was used to investigate the effects of four independent variables, namely extraction time (s), irradiation power (W), liquidtosolids ratio (mL/g) and methanol concentration (%) on the responses. A polynomial model of the TPP, TF and % DPPH of the root of X. americana was deducted from statistical analysis of experiments. Pareto analysis indicated that the influences of the main factors (X1, X2, X3 and X4) on the Total Polyphenols (TPP) of the root of X. americana were 38.81 %, while the interaction effects (X1X3, X1X4, X3X4 and quadratic effects (X1X1, X2X2, X3X3, X4X4) were 17.03 and 53.19 % respectively. The influences of the main factors (X1, X2, and X3) on the Total Flavonoids (TF) of the root of X. americana were 20.90 %, while the interaction effects (X1X3, X1X4, X2X3, X3X4 and quadratic effects (X1X1, X2X2, X3X3, X4X4) were 23.43 and 51.26 % respectively. The influences of the main factors (X1, X3, and X4) on the % DPPH activities of the root of X. americana were 20.90 %, while the interaction effect X1X4 and quadratic effects (X1X1, X2X2, X3X3 and X4X4) were 4.48 and 46.02 % respectively. The interaction effect X1X4 simultaneously influencing the TPP, TF and %DPPH therefore a compromise between the three responses were desirable. To gain a better understanding of the two variables for optimal microwave assisted extraction performance at fixe power 600W and liquid solid ratio 20 mL/g, the model was presented as both 3D response surface and 2D contour graphs. The optimal individual conditions for extraction of total polyphenols, total flavonoids and antioxidant activity from the roots of X americana were 600 W, 20 mL/g, 70 sec, 70 %; 600 W, 20 mL/g, 76 sec, 77 %; and 600 W, 20 mL/g, 72 sec, 69 %, respectively. As a compromise for the simultaneously removal of maximum amounts of 7872 Âµg GAE/ gDM total polyphenols (TPP), 189 Âµg QE/g DM total flavonoids (TF) and with an antioxidant %DPPH of 84 %, the optimum conditions were found with 77.64 % methanol concentration at an extraction time of 83 sec. Close agreement between experimental and predicted values was found. The obtained extracts might be used as natural bioactive compounds in several industrial applications
Keywords: Total polyphenols, total flavonoids, antioxidant activity, X. americana, RSM, Microwave assisted extraction.
INTRODUCTION
Ximenia americana L. is widely used in folk medicine in Africa to treat various disorders such as inflammation, pain, fever, diarrhoea, wounds and intoxications (Onifade et al., 2011; Le et al., 2012; Kenmogne et al., 2014; Kidik et al., 2015). Different parts of the plant are used as infusion, maceration or decoction. Published scientific reports of the biological activities of the plant are scanty and include antitoxic, anticancer, analgesic and antimicrobial (Maikai et al., 2016; Kenmogne et al., 2014; Onifade et al., 2011). Renewed interest in traditional pharmacopeias is increasing worldwide most especially among African people who are becoming reliant on herbal medicines for their health care needs. This is because medicinal plants are more accessible and affordable (Alzeer et al., 2014; Maikai et al., 2015; Kenmogne et al., 2017). Therefore, it seems important to study the phenolic compounds and antioxidant activities of extracts of medicinal plants such as Ximenia americana, plant used for traditional treatment.
Microwaveassisted extraction (MAE) is one of the most inexpensive, simple, rapid, and efficient green extraction techniques compared with conventional extraction (Kenmogne et al., 2000: Kenmogne et al., 2014; Rolly et al., 2016; Gertrude et al 2016;)and has been applied to extract bioactive compounds from different materials due to its high reproducibility at shorter time, simplified manipulation, significant reduction in solvent consumption, and temperature, in respect to other classic methods (Pang et al., 2017). Therefore, the microwave technology has been used in some industries, such as food industry, chemical industry, and material industry (Belwat et al., 2018).
Response surface methodology (RSM), an effective statistical technique for modeling and optimization of complex processes, has been used increasingly to optimize processing parameters owing to more efficient and easier arrangement and interpretation of experiments compared to others (Gan et al., 2010; Gan et al., 2011a,b; Humal et al., 2017). The advantage of RSM is the reduced number of experimental trials needed to evaluate multiple parameters and their interactions (Prasad et al.,
2011; Gan et al., 2011a,b; Rolly et al., 2016; Gertrude et al., 2016; Humal et al., 2018).Therefore, it is widely used in optimizing the extraction parameters from different materials.
Even the microwave treatment has so many advantages, there is no information published about the optimization of microwave assisted extraction of phenolic antioxidants from X. americana roots using response surface methodology (RSM). Therefore, the objective of this study was to determine the best microwave assisted extraction conditions for X. americana roots, in order to maximize simultaneously the total polyphenolic (TPP), total flavonoids and antioxidant activities as measured by the DPPH by using the composite center design (CCD) combined with the RSM.
1. MATERIALS AND METHODS

Sample materials and chemicals
The roots of Ximenia americana were collected from Ngaoundere in the region of North Cameroun and authenticated by Professor Mapongmetsem, botanist and professor in the Department of Biological Sciences Faculty of Science, University of Ngaoundere. The material was allowed to dry naturally, cut, ground using a mill (Model 14, Hamilton Beach, USA) and then sieved with mechanical shaker analysen sieb, Model NFX 11501, Germany, through a standard set of stainless steel sieves. The powder was kept in sealed polyethylene bags at room temperature.

Chemicals reagents
Gallic acid, Folin Ciocalteus phenol reagent, sodium acetate, and methanol reagent were purchased from SigmaAldrich and were used as received. All other chemicals organics solvent used in the study were of analytical grade.

Microwaveassisted extraction process
Two point five grams of Ximenia americana powder roots of 200 Âµm diameter were placed in a tube and mixed with 50mL of methanolwater (80:20, v/v). The extraction process was performed with domestic microwave (ER 696 ETE, 2450MHz, and Toshiba, Japan) equipped with a digital timer and a power controller. After microwave extraction, the sample was filtrate with a wattman paper and the microwave extract was evaporated by rotary evaporation to remove solvent. Three replicates were performed in each extraction. Samples (the microwae extracts) were stored in 4 Â°C prior to analysis. The irradiation extraction time was assessed as shown in the results and the desired temperature was maintained through the timer.

Determination of total polyphenolic content in the roots of Ximenia americana
Determination of total polyphenolic content (TPP) in the roots of Ximenia americana was determined using the Folin Ciocalteau method. Microwave extracts were mixed with 0.2 ml of FolinCiocalteau reagent (prediluted at a ratio of 1:16 with distillated water) and allowed to stand at room temperature for 5 min, and then 0.2 ml of sodium bicarbonate (1M) was added to the mixture. After standing for 20 min at room temperature, absorbance was measured at 760 nm. Results were expressed as Âµg Gallic acid equivalents/ dry weight sample (g GAE per g of dry weight (DW)) (Singleton et al., 1999; Kumazawa et al., 2002).

Determination of total flavonoid content (TF)
The determination of flavonoids was performed according to the colorimetric assay of Dowd. A methanolic solution of aluminum chloride (0.5 ml, 2% w/v) was mixed with the methanol extract solution (0.5 ml, 0.1 mg/mL). After ten minutes the optical densities were read at 415nm against the blank (0.5 mL of methanol extract solution and 0.5 mL of methanol) and compared with the calibration curve of quercetin (0.1 mg/mL) (Singleton et al., 1999; Kumazawa et al., 2002).

Antioxidant assay: 1,1diphenyl2picrylhydrazyl assay (DPPH)
The DPPH radical scavenging capacity assay was based on a previously described method (Liu et al., 2009) with some modification. Briefly aliquots of each extracts (1mL) were added to 1 mL of methanolic DPPH solutions (100 M). Discolorations were measured at 517 nm after incubation for 20 min at30Â°C in the dark. The % DPPH which was scavenged (% DPPH) was calculated using the formula: (%) = [A0(A1AS)]/A0]*100 where A0 is the absorbance of DPPH alone, A1 is the absorbance of DPPH + extract and AS is the absorbance of the extract only. All samples were tested in triplicate.

Central Composite Design (CCD)
A central composite design (CCD) was used to identify the relationship existing between the response functions and the process variables, as well as to determine those conditions that optimized the extraction process of total phenolic content and antioxidant capacity. The independent variables or factors studied obtained from a preliminary experiments (data not shown) using single factor tests determined the required range of extraction time (X1: 7090 sec) microwave power (X2: 500700 w) methanol concentration (X3: 6595 %), and solventtomaterial ratio (X4: 15/120/1 mL/g) (Kenmogne et al., 2020). Once the desired value ranges of the variables had been defined, they were coded to lie at Â±1 for the factorial points, 0 for the center points, and
i
i
ij
ij
Â± for the axial points. The codes were calculated as functions of the range of interest of each factor, as shown in Table 1. The regression analysis of experimental data was performed to establish the empirical second order polynomial models. Shown in Eq. (1),
i 0

i Xi

ii
X 2
Xi X j
(1)
Where is the measured response variable (TPP, TF and %DPPH), 0 is a constant, iis the linear coefficient (main effect) ii is the quadratic coefficient, ij is the two factors interaction coefficient, Xi and Xj are independent variables.
Software: Response surfaces were drawn and coefficients for the regression model were calculated using Modde 4 software
package (Modde 4.0).
Tableau 1: Independent variables and their levels in central composite circumscribed (CCD) design.
Symbol 
Independent variable 
Factor level 

1 
0 
+1 
+ 

X1 
Extraction time (sec) 
63.93 
70.00 
80.00 
90.00 
96.07 
X2 
Microwave power (w) 
439.28 
500.00 
600.00 
700.00 
760.78 
X3 
Liquidsolid ratio (mL/g) 
11.64:1.00 
15.00:1.00 
20.00:1.00 
25.00:1.00 
28.36:1.00 
X4 
Methanol concentration (%) 
63.93 
65.00 
80.00 
95.00 
96.07 
RESULTS AND DISCUSSION

Analysis of the process factors
Using the central composite experimental designs (CCD) of 28 runs shown in Table 2 (which includes the design, observed responses and the predicted values), the models of the extraction process were obtained. It was postulated a secondorder responses models. The coefficients of the models presented in Table 3 were estimated by least square multiple regression method. The sign of the effect indicates its direction. The positive or negative sign indicates the positive or negative effect of the factor on the responses. The importance of these effects was graphically represented on Fig. 4. The higher the effect of a factor (absolute value), the more significant it was on the extraction process of total polyphenols of the roots of Ximenia americana. The responses resulting TPP, TF or % DPPH will depend therefore, on the relative sum of the weight of each factor on each response.

Total polyphenols microwave assisted extraction (TPP)
It was observed according to Pareto diagram (Fig.1A) that 11 effects significantly influence at the 5% level. 5 contribute to increasing TPP (the direct effects of time (X1), power (X2), solventmaterial ratio (X3), the interactions effects of timeratio (X1
2
2
2 2
2 2
2, 2 2 2
2, 2 2 2
* X3) and time – methanol concentration (X1 * X4), while 6 tend to decrease TPP extraction (.the methanol concentration (X4), the interactions effects of timeratio (X2 * X4) and all the quadratic effects (X1 X2 , X3 , X4 ). It was noted that the quadratic effects of the liquid: solid ratio (X3 , 19.83 %) and methanol concentration (X4 , 16.00 %), linear effects of methanol concentration (X4, 12.21 %) and liquid: solid ratio (X3, 12.21 %) strongly influence on the microwave extraction of total polyphenols. Thus attention must be paid not only on main effects but also on quadratic factors. The equation of the model for the response, including the interaction and quadratic terms of each controlled factor were:
2 2 2
2 2 2
PPT*102 = 78.8002 + 1.5012X1 + 2.0309X2 + 9.1976X3 5.8438X4 4.3574X1 0.0662X1X2 + 1.3812X1X3 + 2.2012X1X4 + 0.7375X2X3 + 0.7200X2X4 4.565X3X4 3.9528X2 9.4929X3 7.6578X4 . (2)

Total flavonoids microwave assisted extraction (TF)
2
2
2 2 2
2 2 2
2, 2 2 2
2, 2 2 2
For TF, 11 effects significantly influence at the 5 % level the extraction of total flavonoids and also 5 contribute to increasing TF (the direct effects of time (X1), power (X2), the interactions effects of methanol concentration (X1 * X4), power liquid: solid ratio (X2 * X3) and solventmaterial ratio methanol concentration (X3 * X4)) while 6 tend to decrease TF extraction the linear X3, only the interactions effects of timeratio liquid: solid (X1 * X3) and all the quadratic effects (X1 X2 , X3 , X4 ). The linear effect of methanol concentration (X4) had no effect on flavonoids microwaved extraction, but this factor associated to the ratio liquid: solid (12.00 %) or in the quadratic form (12.40 %) greatly influence the extraction of TF. The quadratic effect of time X1 showed a strong influence (20.43 %). Thus attention must be paidnot only on main effects but also on quadratic factors. The equation of the model for the response, including the interaction and quadratic terms of each controlled factor were: FT = 185.66 + 4,4002X1 + 16.6248X2 8.5354X3 2.8008X4 + 2.0169X1 X2 6.2418X1X3 + 7.0144 X1 X4 + 13.9332X2X3 1.3381X2X4 + 5.9582X3X4 28.8784X12 – 9.0932X2 16.9639X3 17.5311X4

antioxydant activity (% DPPH)

For % DPPH, 8 effects significantly influence at the 5% level and therefore 3 contribute to increasing DPPH radical scavenging activity % (irradiation time X1, solventmaterial ratio (X3 and while 5 tend to decrease the DPPH radical scavenging activity (methanol concentration (X3), quadratic effects of irradiation time X1, irradiation power X2, methanol concentration (X3) and ratio liquid to solid X4). The linear effect of irradiation power X2 had no effect on DPPH radical scavenging activity It was noted that the quadratic effects of the liquid: solid ratio (X32, 13.13 %) and liquid: solid ratio (X42, 21.70 %), the linear effects of liquid: solid ratio (X3, 17.14 %) strongly influence the DPPH radical scavenging activity. Thus attention must also be paid not only on main effects but also on quadratic factors. The equation of the model for the response, including the interaction and quadratic terms of each controlled factor were:
% DPPH = 84.0963 + 2.3848X1 + 1.5669X2 + 5.7673X3 8.0214X4 + 0.72562X1 X2 0.5756X1X3 + 2.0994X1X4 – 1.7494X2X3
+ 0.7431X2 X4 1.6156X3X4 – 2.3309X12 2.9077 X22 6.1462X32 – 10.1551X42. (4)
Published by : International Journal of Engineering Research & Technology (IJERT)
http://www.ijTeratb.olerg2: Experimental and calculated yield of the microwave assisted extraction of TPP, TF and antioxidant activities as measured by DPPH of the root of XimeniIaSSN: 22780181
americana
Vol. 10 Issue 04, April2021
Coded values Real values Laboratory sample
TPP ( g QE/ g DM) TF (g QE/ g DM) % DPPH
X1 X2 X3 X4 U1 U2 U3 U4 EXP PRE RES EXP PRE RES EXP PRE RES
1 
1 
1 
1 
70 
500 
15 
65 
44.53 
46.862 
2.332 
120.241 
124.848 
4.607 
57.96 
60.4862 
2.5262 
1 
1 
1 
1 
90 
500 
15 
65 
43.68 
42.8319 
0.8481 
126.09 
128.069 
1.979 
60.03 
60.757 
0.727 
1 
1 
1 
1 
70 
700 
15 
65 
46.92 
48.1414 
1.2214 
131.12 
128.873 
2.247 
65.24 
64.1813 
1.0587 
1 
1 
1 
1 
90 
700 
15 
65 
57.14 
43.8463 
13.2937 
153.4 
140.162 
13.238 
70.63 
67.3545 
3.2755 
1 
1 
1 
1 
70 
500 
25 
65 
70.79 
70.1498 
0.6402 
103.65 
80.4778 
23.1722 
80.46 
79.9021 
0.5579 
1 
1 
1 
1 
90 
500 
25 
65 
71.67 
71.6447 
0.0253 
54.51 
58.7318 
4.2218 
79.98 
77.8704 
2.1096 
1 
1 
1 
1 
70 
700 
25 
65 
74.88 
74.3792 
0.5008 
135.76 
140.236 
4.476 
79.17 
76.5997 
2.5703 
1 
1 
1 
1 
90 
700 
25 
65 
67.71 
75.6092 
7.8992 
130.95 
126.558 
4.392 
71.89 
77.4705 
5.5805 
1 
1 
1 
1 
70 
500 
15 
95 
39.28 
38.4619 
0.8181 
92.19 
95.977 
3.787 
45.35 
41.9895 
3.3605 
1 
1 
1 
1 
90 
500 
15 
95 
41.57 
43.2368 
1.6668 
139.99 
127.256 
12.734 
48.61 
50.6578 
2.0478 
1 
1 
1 
1 
70 
700 
15 
95 
41.43 
42.6213 
1.1913 
107.13 
94.6503 
12.4797 
47.07 
48.6571 
1.5871 
1 
1 
1 
1 
90 
700 
15 
95 
45.41 
47.1312 
1.7212 
111.43 
133.997 
22.567 
57.45 
60.2279 
2.7779 
1 
1 
1 
1 
70 
500 
25 
95 
41.03 
43.4897 
2.4597 
70.46 
75.4398 
4.9798 
52.19 
54.943 
2.753 
1 
1 
1 
1 
90 
500 
25 
95 
53.93 
53.7897 
0.1403 
80.11 
81.7515 
1.6415 
58.03 
61.3088 
3.2788 
1 
1 
1 
1 
70 
700 
25 
95 
48.67 
50.5992 
1.9292 
132.43 
129.846 
2.584 
53.12 
54.613 
1.493 
1 
1 
1 
1 
90 
700 
25 
95 
61.80 
60.6341 
1.1659 
157.09 
144.225 
12.865 
66.93 
63.8813 
3.0487 
1.60717 
0 
0 
0 
63.9 
600 
20 
80 
68.54 
65.1324 
103.996 
84.0963 

3 
3.4076 
94.08 
9.916 
85.88 
1.7837 

1.60717 
0 
0 
0 
96.0 
600 
20 
80 
78.03 
69.9578 
118.139 
74.2429 

7 
8.0722 
114.33 
3.809 
84.83 
10.5871 

0 
1.60717 
0 
0 
80 
439 
20 
80 
70.75 
65.326 
5.424 
130.48 
135.454 
4.974 
76.57 
74.0673 
2.5027 
0 
1.60717 
0 
0 
80 
760.78 
20 
80 
73.91 
71.8542 
2.0558 
198.51 
188.892 
9.618 
79.23 
79.1039 
0.1261 
0 
0 
1.60717 
0 
80 
600 
11.64 
80 
54.66 
39.4979 
15.1621 
154.9 
155.56 
0.66 
59.38 
58.9515 
0.4285 
0 
0 
1.60717 
0 
80 
600 
28.36 
80 
74.38 
69.0623 
5.3177 
115.06 
128.125 
13.065 
79.69 
77.4897 
2.2003 
0 
0 
0 
1.60717 
80 
600 
20 
63.93 
68.69 
68.4121 
0.2779 
131.77 
144.879 
13.109 
69.5 
70.7574 
1.2574 
0 
0 
0 
1.60717 
80 
600 
20 
96.07 
57.83 
49.6281 
8.2019 
135.26 
135.876 
0.616 
48.86 
44.9737 
3.8863 
0 
0 
0 
0 
80 
600 
20 
80 
78.86 
78.8002 
0.0598 
183.63 
18566 
2.03 
82.34 
81.9083 
0.4317 
0 
0 
0 
0 
80 
600 
20 
80 
77.98 
78.8002 
0.8202 
179.13 
185.66 
6.53 
79.91 
81.9083 
1.9983 
0 
0 
0 
0 
80 
600 
20 
80 
76.13 
78.8002 
2.6702 
182.82 
185.66 
2.84 
82.53 
81.9083 
0.6217 
0 
0 
0 
0 
80 
600 
20 
80 
76.30 
78.8002 
2.5002 
185.14 
185.66 
0.52 
77.95 
81.9083 
3.9583 
IJERTV10IS050052
www.ijert.org 80
(This work is licensed under a Creative Commons Attribution 4.0 International License.)
Table 3: Regression coefficients for mains factors and their interactions 

Factor/ interaction 
Regression coefficient 

YTPPx102 
YFT 
%DPPH 

Mean/ interaction b0 
78.8002 
185.66 
84.0963 

X1 
Irradiation time (L1) 1 
1.5012 
4.4002 
2.3847 

: 
Irradiation time (Q11) 11 
4.3574 
28.8784 
2.3308 

X2 
Irradiation power (L2) 2 
2.0309 
16.6248 
1.5669 

Irradiation power (L22) 22 
3.9528 
9.0932 
2.9077 

X3 : 
Solvent to solid ratio (L3) 3 
9.1976 
8.5354 
5.7673 

Solvent to solid ratio (Q33) 33 
9.4929 
16.964 
6.1462 

X4 
Solvent concentration (L4) 4 
5.8438 
2.8008 
8.0214 

Solvent concentration (Q44) 44 
7.6578 
17.5311 
10.1551 

X1 X2 
1 2 
0.0662 
2.0169 
0.7256 

X1 X3 
1 3 
1.3812 
6.2418 
0.5756 

X1 X4 
1 4 
2.2012 
7.0144 
2.0994 

X2 X3 
2 3 
0.7375 
13.9332 
1.7494 

X2 X4 
2 4 
0.7200 
1.3380 
0.7431 

X3 X4 
3 4 
4.5650 
5.9582 
1.6156 

L= linear, Q= quadratic 
Contribution of factors (%)
Contribution of factors (%)
20
15
10
5
0
5
10
19.22%
3.14%4.24%
2.89%4.60%
0.14%
1.54%
1.50%
15
20
9.10%
12.21%
8.26%
19.83%
9.54%
X1
X1
X2
X2
X3
X3
X4
X4
16.00%
X1*X1
X1*X1
X1*X2
X1*X2
X1*X3
X1*X4
X1*X3
X1*X4
X2*X2
X2*X2
X2*X3
X2*X3
X2*X4
X2*X4
X3*X3
X3*X3
X3*X4
X3*X4
X4*X4
X4*X4
Process factors of microwave extraction of TPP A
Contribution of factors (%)
Contribution of factors (%)
15 11.76%
10
5 3.11%
0
5
1.98%
1.42%
4.96%
9.85%
0.94%
4.21%
10
15
20
25
6.03%
20.43%
4.41%
6.43%
12.00% 12.40%
B
Process factors of microwave extraction of TF
Contribution of factors (%)
Contribution of factors (%)
15 12.32%
5
5
10 5.09%3.34%
0
1.55%
4.48%
1.58%
5
10
4.98%
123%
–
3.73%
345%
15
20
25
1714%
6.21%
13.13%
21.70%
C
Process factors DPPH activities
Non significant effect nd significant effect
Figure 1: Pareto analysis of the process factors of the Yield of microwave assisted extraction of: (A) the Total Polyphenols (TPP), (B) the Total flavonoids and (C) the DPPH activities
3.2. Validation of the Models
It is usually necessary to check the fitted model to ensure it provides an adequate approximation to the real system. Graphical and numerical methods, as a primary tool and confirmation for graphical techniques were used to validate the models in this study (Deming al., 1993; Mathieu et al., 1997). The graphical method characterizes the nature of residuals of the models. If the model represents the phenomenon in the studied domain, the difference between the experimental results and the theoretical results calculated by the model (Table 2) was due solely to experimental error. The coefficient of determination obtained (R2 = 0.89 for TPP, R2 = 0.93 for TF and R2=0.94 for % DPPH) when plotting the experimental responses versus the theoretical responses (Fig. 2) indicated that the variations of 89 % for TPP, 93% for TF and 94 % for % DPPH were all attributed to the independent variables, irradiation time, irradiation power, liquid: solid ratio and methanol concentration. We admitted in using this method of validation that the experimental results do not contain errors since they were the average of results of the three replications and that even if there were an error, it was constant and did not depend on the variation of parameters involved. Admitting this, it can be ascertained that in the overall, the models used represents the studied phenomenon in the variation domain. R2 values insure a satisfactory adjustment of the quadratic models to the experimental data. Therefore, the regression models explained the microwave extraction efficiency well.
Validation of the equation model of Total polyphenols microwave extraction
Experimental values
Experimental values
100
80
60
40
20
0
y = 0,9636x + 0,8254 RÂ² = 0,882
0 20 40 60 80 100
Theoritical values
Eexperimental values
Eexperimental values
250
200
150
100
50
0
Validation of the model equation of total flavonoids microwave extraction
y = 0.9475x + 7.3129 RÂ² = 0.9289
0 50 100 150 200 250
Theoretial values
Validation of the equation model of DPPH inhibition
90 y = 0.9295x + 4.4808
Experimental values
Experimental values
80 RÂ² = 0.9406
70
60
50
40
30
20
10
0
0 20 40 60 80 100
Theoretical values
Figure 2:validation of the models equations

Optimal extraction conditions:
It was observed from table 1 that the yield of TPP and TF ranged from 3928 to 7886 g GAE/g.DM and 54.51 to 190.82 g QE/g.DM respectively. The highest TPP (7886 g EAG/gMS) was obtained under the experimental conditions of X1 = 70 sec, X2 = 530 W, X3 = 20/1 mL/g and X4 = 40 %; whereas the highest TF was obtained under conditions of X1 = 80sec, X2 = 600 W, X3 = 20 ml/g and X4 = 80 %. A wide range of antioxidant activity (% DPPH) was also found from 45.35 to 84.83 % and the maximum point was found under the conditions of X1= 96,07 sec, X2= 600 W, X3= 20 mL/g and X4 = 80 %. Therefore, an optimization process was investigated, in order to obtain desirable antioxidant contents and activity. The result of the optimal conditions for each individual response as well as combination of all responses with the predicted and experimental values is presented in table 4 and 5. The target was to obtain high phenolic compounds yields with high antioxidant activity within the extraction parameters, where consideration of the efficiency, the energy conservation and the feasibility of the experiment were taken into account. Optimal conditions for TPP were irradiation time of 82.68 sec, irradiation power of 653.13 W, solventto solids ratio of 23.16 mL/g and methanol concentration of 74.61 %. On the other hand, optimal conditions for TF were irradiation time of 81.74 sec, irradiation power of 808.40 W, solventtosolids ratio of 20.73 ml/g and methanol concentration of 79 %, whereas optimal conditions for % DPPH were irradiation time of 85.73 sec, irradiation power of 621.06 W, solventto solids ratio of 22.19 mL/g and methanol concentration of 75.99 %.These conditions gave TPP, TF and %DPPH values of 8366.15 Âµg GAE/ gDM, 193.995 g QE/ gDM and 87.55 %, respectively.
Tableau 4: optimal conditions for each individual response (TPP,TF or %DPPH)with the predicted and experimental values
Responses
Process variables
Predicted value
Experimental valuea
X1 (s)
X2 (w)
X3 (mL/g)
X4 (%)
TPPb
82.68
653.13
23.16
74.61
7998.71
8366.15Â±94.65
TFc
81.74
808.40
20.73
79.22
188.68
193.99Â±19.98
%DPPHd
85.73
621.06
22.19
75.99
94.37
87.55Â±08.84
a Mean of triplicate determination. b Total polyphenol (Âµg GAE/ gDM). c Total flavonoids (g QE/ gDM). d
Antioxydant activity (%).
From Equations 2, 3 and 4, the TPP, TF or % DPPH of the roots of Ximenia americana at any point in the interval of our experimental domain can be predicted. The 3D response surface and the 2D contour plot are the graphical representations of the regression equation. The response surface curves allow visualizing the simultaneous evolution of the TPP, TF or %DPPH of the roots of Ximenia americana with the change of the process factors. The 3D response surface (Fig. 1A, 1B or 1C) of TPP, TF or
% DPPH of the roots of Ximenia americana as a function of irradiation time and methanol concentration at a fixed irradiation power level of 600 w and solvent tosolid ratio level of 20 mL/g indicates that all the graphics have not the same tendencies. The increase in times or methanol concentration led to a gradual increase of TPP and TF then diminishes. While the only increases in methanol concentration led to gradual increase of % DPPH then diminishes. Thus, the time and methanol concentration in the extraction medium had a significant influence on the Total polyphenols and antioxidant properties of the roots of Ximenia americana extracts.
Figure 3A, 3B and 3C show the contours of estimated responses surfaces (estimated isoresponse lignes) of TPP, TF and % DPPH as a function of time and methanol concentration. The use of contour plots is to best indicate technically and economically the values of process factors for obtaining a maximum yield of TPP, TF and % DPPH of the roots of ximenia americana. These isoresponseslignes allow us to choose several combinations between times and methanol concentration to reach the same TPP, TF and % DPPH. For instance, extraction at 76 sec with a 70 % methanol concentration or at 84 sec with a 81 % methanol concentration permit to obtain the same TPP of 7500 Âµg GAE/g DM. In this work, the technically and economically value of process factors for obtaining a maximum yield of TPP, TF and % DPPH of the roots of Ximenia americana were taken into consideration. Therefore, for the economical feasibility of an industrial process, the optimal extraction conditions (indicated by violet arrows on the curve) of the three individual responses, selected graphically according to the contour plots with a fixe irradiation microwave power of 600 w and solventtosolids ratio of 20 ml/g were: an irradiation time of 70 sec and methanol concentration of 70 % for TPP (fig 3A), an irradiation time of 76 sec and methanol concntration of 77 % for TF (fig 3B) and an irradiation time of 72 sec and methanol concentration of 69 % for % DPPH (fig 3C) of the roots of Ximenia americana. As can be seen in table 6, the predicted results and experimental results obtained demonstrate that the mathematical model represents very well the studied phenomenon in the experimental domain indicating the robustness of the method used. The optimizations of the three individual responses were achieved under different optimal conditions. Therefore a compromise between their extraction conditions was desirable.
Figure 3: 3D response surface and 2D contour graphs of individual responses TPP (A,A) TF (B,B) and % DPPH (C,C) between two independent variables while the remaining independent variable kept at its zero level.

Multicombined optimization
The optimum conditions for the simultaneous extraction of TPP, TF and %DPPH can be visualized graphically by superimposing the contours (fig 3A, 3B and 3C) for various response surfaces in an overlay plot (fig 4). The optimum area is irradiation time of 7785 sec and methanol concentration 7078 %. Base on the overlain contour, a compromise for 7809.24 Âµg EAG/g MS TPP and 182.5 Âµg EQ/g MS TF yield with 90.48 % DPPH can be met at 82.51 sec irradiation time and 77.64 % methanol concentration (table 5).
To confirm the agreements of the results achieved from the model and experiments, additional experiments were conducted by applying the irradiation time and methanol concentration in the optimum region. As shown in Table 5, the TPP, TF YIELD and %DPPH obtained from the additional experiments were 7872.24 Â±64.65 Âµg EAG/g MS, 189.70 Â± 11.43 Âµg QE/g DM respectively with 84.04 Â± 07.15 % DPPH. It could be observed that only small deviations were found between the experimental values and predicted values in Table 5, implying that the RSM approach was appropriate for optimizing the conditions of the microwave extraction process.
Table 5: optimum conditions based on combination of responses (TPP, TF and % DPPH)
Process variables 
Predicted value 
Experimental valuea 

X1 (s) 
X2 (w) 
X3 (mL/g) 
X4 (%) 
TPP 
TF 
%DPPH 
TPP 
TF 
%DPPH 
82.51 
600.00 
20.00 
77.64 
7809.24 
182.50 
90.48 
7872.24 Â±64.65 
189.70Â±11.43 
84.04 Â±07.15 
a Mean of triplicate determination. b Total polyphenol (Âµg GAE/ gDM). c Total flavonoids (g QE/ gDM). d Antioxydant activity (%).
Irradiation power =600 w, liquidsolid ratio =20mL/g
Time (sec)
Time (sec)
Total polyphenols Total flavonoids DPPH activities
[Methanol] (%) [Methanol] (%)
Figure 4: Superposition plots showing the best experimental parameters that maximize all the responses (TPP,TF and % DPPH).
4. CONCLUSION
This work has demonstrated the application of RSM in seeking optimal conditions for microwaved assisted extraction of the phenolic compounds of the roots of Ximenia americana. Simultaneous extraction of TPP, TF and % DPPH were investigated. In order to gain a better understanding of the four factors for optimal microwave extraction performance, the models were presented as 3D response surface and 2D contour graphs. The following conclusions were obtained:

From Pareto analysis, the interaction effect X1X4 (irradiation timemethanol concentration) simultaneously influencing the TPP, TF and % DPPH of the root of X. americana

The optimal individual conditions for extraction of total polyphenols, total flavonoids and antioxidant activity (the economically value of process factors) from the roots of X. americana were 600 W, 20 mL/g, 70 sec, 70 %; 600 W, 20 mL/g, 76 sec, 77 %; and 600 W, 20mL/g, 72 sec, 69 %, respectively.

To simultaneously remove 7872 Âµg EAG/g MS TPP and 189 Âµg EQ/g MS, TF with 84 % of % DPPH, 83 sec irradiation time and 78 % methanol concentration were selected based on the overlain contour.

The results of a confirmation experiment were found to be in good agreement with the values predicted by the model. This demonstrates that to obtain a maximum amount of information in a short period of time, with the least number of experiments, RSM and CCD can be successfully applied for modeling and optimizing the microwave extraction process. Future studies can be oriented on the isolation and identification of compounds involved in antiantioxydant activity.
ACKNOWLEDGEMENTS
We acknowledge financial supports from the Ministry of High Education of Cameroon.
REFERENCES

Alzeer J, Vummidi BR, Arafeh R, Rimawi W, Saleem H,Luedtke N, W, 2014. The influence of extraction solvents on the anticancer activities of Palestinian medicinal plants. J. Med. Plants Res., 8(8):408415.

Belwal, T., Ezzat, S.M., Rastrelli, L., Bhatt, I.D., Daglia, M., Baldi, A., Devkota, H.P., Orhan, I.E., Patra, J.K., Das, G., Anandharamakrishnan, C., GomezGomez, L., Nabavi, S.F., Nabavi, S.M., Atanasov, A.G., Trends in Analytical Chemistry, 2018, 100, 82102.

Deming, S.N. and S.L. Morgan, 1993. Experimental design: a chemometric approach Amsterdam, Elsevier.

Gan, C. Y., Abdul Manaf, N., & Latiff, A. A. (2010). Optimization of alcohol insoluble polysaccharides (AIPS) extraction from the Parkia speciosa pod using response surface methodology (RSM). Carbohydrate Polymers, 79, 825831.

Gan, C. Y., & Latiff, A. A. (2011a). Extraction of antioxidant pecticpolysaccharide from

mangosteen (Garcinia mangostana) rind: Optimization using response surface methodology. Carbohydrate Polymers, 83, 600607.

Gan, C. Y., & Latiff, A. A. (2011b). Optimisation of the solvent extraction of bioactive compounds from Parkia speciosa pod using response surface methodology. Food Chemistry, 124, 12771283.

G.T. Djiobie , R .K.T. Tsatsop, Kenmogne Sidonie Beatrice, M.B. Ngassoum., 2016. Optimization of Microwave Assisted Extraction of Bioactive >Compounds from Flavorings Roots of Plant Carissa Edulis. Scholars Journal of Engineering and Technology (SJET). Vol 4(3) PP 127136.

Huma ZE, Jayasena V, NasarAbbas SM, Imran M, Khan MK, (2017) Process optimization of polyphenol extraction from carob (Ceratonia siliqua) kibbles using microwaveassisted technique. J Food Process Preserv

Huma ZE, Jayasena V, NasarAbbas SM, Imran M, Khan MK (2018) Process optimization of polyphenol extraction from carob (Ceratonia siliqua) kibbles using microwaveassisted technique. J Food Process Preserv 42(2):e13450

Kenmogne Sidonie BÃ©atrice, Xiu Deng, 2000. Household Microwave Modification: Application on Costus Essential Oil. La Ravista Italiana Delle Sostanze Grasse. Vol LXXVIINovember, PP 761764.

Kenmogne Sidonie Beatrice, Ngassoum M, Tchatchueng J. B.n Vardamides J.C. and Dongmo A. (2014). Microwave Assisted Extraction of Analgesic Compounds of the Root of Ximenia Americana (Olacaceae)Research Journal of Chemical Sciences. Vol 4(7), pp 16, July .

Kidik P, M, C, Ngene J, P, Ngoule C, C, Mvogo O, P, B, Ndjib R, C, Dibong S, D, Mpondo M, E, 2015. CaractÃ©risation des plantes mÃ©dicinales Ã flavonoÃ¯des des marchÃ©s de Douala (Cameroun). Int. J. Biol. Chem. Sci., 9(3): 14941516.

Le NHT, Malteruda KE, Diallob D, Paulsena BS, NergÃ¥rda CS, Wangensteen H. 2012, Bioactive polyphenols in Ximenia americana and the traditional use among Malian healers. J Ethnopharmacol. 139(3): 858862.

L. Liu, Y. Sun, T. Laura, X. Liang, H. Ye and X. Zeng, (2009). Determination of polyphenolic contentand antioxidant activity of kudingcha made from Ilex kudingcha C.J. Tseng. Food Chemistry, 112 : 3541.

Maikai B, V, Maikai V, A, 2015. Effects of treatment with fraction IV extract of Ximenia americana on the survival rate, packed cell volume and total plasma proteins of Trypanosoma congolense infected mice.J. Parasitol. Vector Biol., 7(5):101107.

Mathieu, D., R. PhantanLuu and G. Droeskeke, 1997. Approache Methodologique des Surfaces de Reponses, Plans d'experiences. Applications a l'entreprise, Saporta (Eds.) Technip, Paris, pp: 211.

M. Pinelo, A. G. Tress, M. Pedersen, A. Arnous, A. S. Meyer, (2007). Effect of cellulases, solvent type and particle size distribution on the extraction of chlorogenic acid and other phenols from spent coffee grounds, Am. J. Food Technol. 2: 641651.

Onifade A, O, Ouedraogo M, Ouedraogo M, Zongo F, E,Kafando E, Lompo M, Guissou I, P, 2011. Acute toxicity and antiinflammatory activity of aqueous ethanol extract of root bark of Ximenia americana (Olacaceae). Afr. J. Pharm. Pharmacol., 5(7): 806811.

Pang SF, Lau MZ, Yusoff MM & Gimbun J (2017), Microwave irradiation induced fast simultaneous extraction of methoxylated and hydroxylated phenolic compounds from Orthosiphon stamineus leaves. Material Science Forum, 155158.

Prasad KN, Fouad AH, Bao Y, Kin WK, Ramakrishnan NR, Azrina A, Ismail A (2011) Response surface optimization for the extraction of phenolic compounds and antioxidant capacities of under utilised Mangifera pajang Kosterm. Peels. Food Chem 128:11211127

R .K.T. Tsatsop, G.T. Djiobie, BÃ©atrice Sidonie Kenmogne , K.R. Regonne, M.B. Ngassoum., 2016. Optimization of Microwave Assisted Extraction of Bioactive >Compounds from Anogeissus Leiocarpus Guill & Perr. Stem Bark Using Response Surface Methodology. International Journal of Scienctific & Technology Research. Vol5 Issue 05 May.

V. L. Singleton, J. A. Rossi Jr, (1965) Colorimetry of total phenolics with phosphomolybdicphosphotungstic acid reagents, Am. J. Enol.Vitic.16 : 144158.