Response surface methodology for optimization of lycopene extraction

Một phần của tài liệu Optimization of some factors influencing lycopene extraction from tomato processing waste using response surface methodology (Trang 46 - 52)

The dried tomato waste, which was dried at 65oC to 23% moisture content by convective oven, was used for lycopene extraction. The extraction process with ethyl acetate solvent was standardized for the maximum recovery of the pigment, using response surface methodology. Three independent variables X1 (±solvent/material ratio), X2 (temperature) and X3 (time) were studied and their levels were selected by Box-Behnken experiment design (Table 4.5).

Results in Table 4.5 reveal that lycopene contents obtained at different extracted condition range from 3.07 to 8.36 mg/g DW. Kaur et al. (2008) also worked on lycopene extraction with hexane: acetone: ethanol (2:1:1) from tomato skin using five independent variables, solvent/meal ratio (20/1 to 60/1, v/w), number of extraction (1 to 4), temperature (20 to 60oC), particle size (0.05 to 0.43mm) and time (4 to 20min) and their experimental values ranged from 0.639 to 1.98 mg/g DW. The higher lycopene content obtained in the present study might be due to different material source and extraction conditions. Tomato lycopene content varies considerably, between cultivars (generally genetic factors), maturity, and both agronomic and environmental conditions during growing (Shi and Maguer, 2000). Similarly, Rao et al. (1999) extracted lycopene from tomato skin and tomato pulp with hexane: methanol: acetone in 2:1:1 ratio and reported lycopene recovery of 14.1 mg/g and 6.94 mg/g, on a fresh waste basis respectively. Tan and Soderstrom (1988) recovered of 25.4 mg/100g of lycopene from tomato paste with 95% ethanol and low boiling petroleum ether (40-60o C). These studies showed that extraction variable had an impact on lycopene recovery.

Table 4.5. Results of optimization condition for lycopene extraction

Treatment

Ratio of solvent/

material (X1)(w/w)

Temperature (X2 ) (o C)

Time (X3) (min)

Lycopene content *

(Y1) (mg/g DW)

Antioxidant capacity

(Y2)

(àmol TE/ g DW)

T13 10/1 30 90 4.27± 0.05 4.70± 0.04

T14 50/1 30 90 3.98± 0.03 8.39± 0.37

T15 10/1 70 90 7.07± 0.10 7.19± 0.04

T16 50/1 70 90 6.84± 0.06 11.96± 0.22

T17 10/1 50 30 4.84± 0.12 5.10± 0.05

T18 50/1 50 30 5.02± 0.06 9.66± 0.10

T19 10/1 50 150 8.36± 0.16 6.20± 0.02

T20 50/1 50 150 7.69± 0.06 10.45± 0.16

T21 30/1 30 30 3.07± 0.07 6.41± 0.08

T22 30/1 70 30 5.12± 0.07 8.91± 0.17

T23 30/1 30 150 4.03± 0.02 7.70± 0.14

T24 30/1 70 150 6.54± 0.06 10.52± 0.14

T25 30/1 50 90 6.68± 0.11 8.63± 0.13

T26 30/1 50 90 6.71± 0.04 8.65± 0.20

T27 30/1 50 90 6.81± 0.09 9.13± 0.11

*Mean ± SE

In order to evaluate the effect of three experiment factors on lycopene content (Y1) and antioxidant capacity (Y2) and to obtained desirability value at various extraction conditions, the experimental results were analyzed by JMP 12.0 software. The experimental results were fitted to the quadratic equation which presented in the section 3.3.1.3. After analyzing by JMP 12.0 software, the effect of the main factor and their interactions on lycopene content and antioxidant capacity were presented in the Table 4.6.

Table 4.6. Summary of effect of independent factors to the output variables

Source Log worth P-value

X1 41.244 0.00000

X2 25.923 0.00000

X2*X2 21.134 0.00000

X3 12.913 0.00000

X1*X1 8.911 0.00000

X1*X2 5.545 0.00000

X3*X3 1.510 0.03090

X1*X3 1.147 0.07126

X2*X3 0.467 0.34102

X1: ratio of solvent /material (v/w); X2: temperature (o C); X3: time (min)

Table 4.7. Results of the analysis of variance of lycopene content

Term Estimate Std Error t Ratio P – value

Intercept 3.3472318 0.372686 8.98 <.0001*

X1 -0.004696 0.004646 -1.01 0.3152

X2 0.0441097 0.004152 10.62 <.0001*

X3 0.0147892 0.001657 8.93 <.0001*

X1*X1 0.0007333 0.000333 2.20 0.0304*

X1*X2 0.0008341 0.000276 3.03 0.0033*

X2*X2 -0.004525 0.000342 -13.24 <.0001*

X1*X3 8.7251e-5 0.000139 0.63 0.5313

X2*X3 0.0000948 0.000099 0.96 0.3410

X3*X3 -6.465e-5 0.000038 -1.70 0.0925

X1: ratio of solvent/material (v/w); X2: temperature (o C); X3: time (min)

The results in Table 4.6 show that the main factors (X1, X2, X3) and their first degree of and second degree of interaction were significantly influenced lycopene content and antioxidant capacity because their p-values were much lower than significant level (α=0.05). While the interaction between ratio solvent/material and time (X1*X3), interaction between temperature and time (X2*X3) and X32 had small effect on lycopene content and antioxidant capacity. Table 4.7 and 4.8 present the results of analysis of variance on lycopene content (Y1) and antioxidant capacity (Y2).

The results in Table 4.7 show that the factors X2 andX3 and their interaction first degree and second degree interaction (X1*X2, X12, X22) were significantly effect on lycopene content at α=0.05. On the contrary, the ratio of solvent/material (X1) and the interaction X1*X3, X2*X3 and X32 were not significantly effect on lycopene content. Specifically, the factor X2, X3 and the interactions X1*X2 and X12 made lycopene content to be increase, while X22

decreased the lycopene content. These finding are consistent with the results obtained by Shi & Maguer (2000), who reported that serious loses of lycopene could occur when holding time at high temperature too long.

From the fitting results, the linear equation relationship between lycopene content and the independence variables is derived:

Y1= 3.3472 - 0.0047X1 + 0.0441X2 + 0.0148X3 + 0.0007X12 - 0.0045X22– 0.00006X32 + 0.0008X1*X2 + 0.00009 X1*X3 + 0.00009 X2*X3

Where: Y1 was lycopene content

Similarly, Table 4.8 present about the results of the analysis of variance on antioxidant capacity in polynomial equation. Also same with the lycopene content, the factors (X1, X2, X3) and their first degree interaction and second degree interaction (X1*X2, X12, X32 ) were significantly affected the antioxidant capacity of lycopene in the extract at α=0.05. On the contrary, the interaction of temperature (X22) and the interaction of X1*X3, X2*X3 were not significantly effected on antioxidant capacity of lycopene in the extract.

Table 4.8. Result of the analysis of variance of antioxidant capacity of lycopene extract

Term Estimate Std Error t Ratio p- value

Intercept 0.9798381 0.35842 2.73 0.0077*

X1 0.120713 0.004468 27.02 <.0001*

X2 0.0638925 0.003993 16.00 <.0001*

X3 0.0112368 0.001593 7.05 <.0001*

X1*X1 -0.0022 0.00032 -6.88 <.0001*

X1*X2 0.0013359 0.000265 5.04 <.0001*

X2*X2 -0.000321 0.000329 -0.98 0.3323

X1*X3 -0.000244 0.000133 -1.83 0.0713

X2*X3 6.7174e-5 9.519e-5 0.71 0.4824

X3*X3 -8.024e-5 3.652e-5 -2.20 0.0309*

X1: ratio of solvent /material (v/w); X2: temperature (o C); X3: time (min)

From the fitting results, the linear equation relationship between antioxidant capacity and the independence variables is derived:

Y2= 0.9798 + 0.1207X1 + 0.0639 X2 + 0.0112 X3 – 0.0022 X12 - 0.0003 X22

– 0.00008 X32 + 0.0013 X1*X2 – 0.0002 X1*X3 + 0.00007 X2X3

In order to determine optimal value of three factors, the desirability point has been optimized by the prediction profiler. The result was presented at Figures 4.6A and 4.6B:

Figure 4.4. Profiler showing the optimal extracting conditions of lycopene extraction

X1: ratio of solvent /material (v/w); X2: temperature (o C); X3: time (min)

B A

The results in the Figure 4.6A show that desirability level was 0.83 and the highest values of lycopene content and antioxidant capacity were 7.819 mg/g DW and 11.840 àmol TE/g DW, respectively when lycopene extraction process was carried out at the solvent/material ratio of 50 w/v, at 65oC for 150 min, respectively. Baysal et al. (2000) reported that, with increasing temperature from 35 to 65oC during supercritical CO2 extraction, the lycopene yield increased from 1.85% to 21.9%. However, Shi and Maguer (2000) also indicated that in the hexane and light petroleum solution, 26.1% of the lycopene was lost when heated for 3h at 65oC, and 35% was lost when heated for 3h at 100oC. Since lycopene structure is affected by high temperature and long extraction time.

In addition, according to Figure 4.6, the desirability increased when the ratio of solvent/material increased from 30/1 – 50/1 (w/v). After that even the ratio of solvent/material continued to increase but the desirability did not show increase. Most probably, the high ratio of solvent/material extract entirely lycopene in the material, but the economic efficiency decreases due to consumption of large solvent to extraction and large energy to evaporating solvent after extraction. Therefore, in the work, the ratio of solvent/material was 40/1 (v/w) is suitable ratio for lycopene extraction. This result was also suitable with research of Kaur et al. (2008).

In our research, the optimal temperature, time and solvent/material ratio were chosen at 55oC, 120 min, 40/1 (v/w) respectively in order to balance the lycopene content and quality of lycopene. Under these conditions, lycopene content in the extract was predicted to be 7.391 mg/g DW (Figure 4.6B), about 0.43 mg/g lower than that extracted at 65oC for 150 min time , and ratio of solvent/ material of 55/1 (v/w). However, the lower quantities were not considerable in return for ensuring quality of lycopene and increasing economic efficiency due to less solvent volume to be removed.

In short, the optimal condition for lycopene extraction from tomato waste was the solvent/material ratio 40/1 (v/w), temperature 55oC and extraction time 120 min.

Một phần của tài liệu Optimization of some factors influencing lycopene extraction from tomato processing waste using response surface methodology (Trang 46 - 52)

Tải bản đầy đủ (PDF)

(62 trang)