Process optimization for microwave assisted extraction of Foeniculum vulgare Mill using response surface methodology

2.1 Microwave assisted extraction (MAE)

For MAE, 10 g of plant material was added with 50 mL of distilled water to each vial every time. The vials were loaded into Teflon sample vials of Microwave assisted extractor (MDS-6G). Distilled water was used as a standard for temperature and pressure sensing of reaction vessel. For the optimization of MAE, the selected parameters were, power level (200 W, 600 W and 1000 W), time used for microwave irradiation (1, 2, 3, 4 and 5 min) and particle size (40, 80 and 120 µ). A Buchner funnel lined with Whatman #1 filter paper was used to filter extract. Filtrate was dried in rotary evaporator, collected in labelled vials and was stored for further analysis.

2.2 Soxhlet extraction (SE)

Soxhlet apparatus was used for the extraction of powdered plant material. 20 g of seed powder was taken each time in thimble. Parameters optimized for soxhlet extraction were time (2, 8, 16 and 24 h) and particle size. Each time 250 mL of water was used as solvent. For each time of extraction the extract was separated as mentioned in MAE Section 1.1.

2.3 Cold maceration (CM)

Cold maceration of seed powder of F. vulgare was done at room temperature using water as a solvent. For this purpose solvent amount was kept constant (250 mL). Each time 20 g of seed powder was soaked in solvent, covered with lid (to avoid evaporation) and placed at room temperature on shaker (30 rpm) for 2, 8, 16 and 24 h. For each time of extraction the extract was separated as mentioned in MAE Section 1.1.

2.4 Total phenolic content

To estimate the phenolics, Spectrophotometric assay was used. 1 mg of aqueous plant extract was well mixed with 1 mL of FC (Folin and Ciocalteu’s) reagent, 1 mL of saturated sodium carbonate was added to the prepared mixture after 5 min and distilled water was used to adjust the total volume up to 10 mL. After 90 min in dark, absorbance was checked at 725 nm. Gallic acid was used as a standard phenolic to construct standard curve. The values of phenolics were expressed as µg of Gallic acid equivalent per mg of extract.

2.5 Response surface methodology (RSM)

Design Expert software (Version 11) was used to apply RSM. CCD (Central composite design) was used which generated a design with a set of 20 experimental runs. Three independent variables i.e., power level (A), time of irradiation (B) and matrix size (C) at their low, medium and high levels were selected to study their combined effect on phenolic content (dependent variables), to optimize the MAE conditions. A quadratic polynomial regression model was contrived to explain the relationship between dependent and independent variables. Where predicted response was estimated by using the relation

2.6 Antioxidant activity and reducing potential of extracts:

Antioxidant assay and reducing potential was measured for MAE, SE and CM extract of F. vulgare with maximum amount of phenolics. For antioxidant activity assay, 1 mL of seed extract (1 mg/mL) was mixed with 1 mL of 1 mM 2,2-diphenyl-1-picrylhydrazyl radical (DPPH) solution. The mixture was incubated for half an hour at 37 °C. Reaction mixtures were checked for absorbance at 517 nm. The percentage of DPPH inhibition was evaluated using eq. $$\%\text{DPPH}\text{inhibition}=\frac{{\text{A}}_{\text{c}}-{\text{A}}_{\text{s}}}{{\text{A}}_{\text{c}}}\times 100$$ where A_c is the absorption of water + DPPH, A_s is the absorption of sample + DPPH. Seed powder extract with maximum phenolics were selected for the study of reducing power. Range of concentrations i.e., 50–250 µg were mixed with 1 mL of water, 2.5 mL of phosphate buffer (pH 6.6 and 0.2 M) and 2.5 mL of 1% potassium ferricyanide separately. Mixture was incubated for 25 min at 50° C. 10% trichloro-acetic acid (2.5 mL) was added then and centrifuged at 3000 rpm for 15 min. Supernatant was removed and mixed with 2.5 mL of double distilled water and 0.5 mL iron chloride (0.1%). Absorbance of each solution was then measured at 700 nm. There is a direct relationship between absorbance and reducing power of extract.

2.7 GCMS for non-polar components

Three maximum yielding extracts from each extraction methods were selected for fractionation separately. Extracts were mixed in water. According to the polarity differences they were fractionated with n-hexane and chloroform. GCMS analysis was done to check the presence of non-polar components in the plant extract. Helium was the carrier gas with flow rate of 1 mL\min. The temperature of injector was set at 250 °C. The temperature of the GC column was fixed at 40 °C for starting 5 min, then elevated to 140 °C at the rate of 5 °C per minute and these conditions were set to be maintained for further 5 min. Then temperature was increased to 280 °C (9 °C per min) and maintained for 5 min. Detector temperature was fixed at 245 °C. Mass spectrometer detector was used at a potential of 70 eV (Hammouda et al., 2013). The nonpolar components of chloroform and n-hexane fractions of MAE, SE and SM extracts were studied with their GC retention times, mass spectra, retention indices, comparison with literature and library search of mass spectra (The National Institute of Standards and Technology NIST).

2.8 Antibacterial assay

Antibacterial assay of non-polar components was carried out and 6 pathogenic bacterial strains were used namely Staphylococcus aureus, Klebciella pneumonia (K92), Eshercia coli (E909), Pseudomonas aeruginosa (P23), Bacillus cereus (B33), Brucella melitensis (Bm42). Sterilized Mueller & Hinton Agar medium was poured into petri-plates and 100 μL of bacterial inoculum was spread. Wells of 4 mm diameter were pierced and were filled with standard antibiotic (gentamycin) as a positive control, solvent as a negative control, n-hexane and chloroform fractions of MAE, CM extract and SE sample. The plates were incubated at 37 °C for 24 h. Inhibition zones were measured in millimeters.

2.9 Statistics

SPSS software was used to analyze the data statistically by comparing the means with one way ANOVA (Analysis of variance) and significance of results was checked at 5% level of significance by applying Duncans New Multiple Range test. Furthermore, Design Expert software, version 11 was used for applying RSM.

3.1 Comparison of MAE, SE and CM

MAE is reported to be an efficient extraction method as compared to conventional methods for a number of plants. Time and power levels are the two main parameters for MAE of plant metabolites. Adequate time of exposure to microwaves is requirement for the good extraction yield due to higher contact of plant matrix and solvent, softening of plant tissue and weakening of cell wall integrity (Liu et al., 2016). Effect of time, power and particle size on MAE is shown in Fig. 1(a–c). Maximum amount of extract was recorded as 149.35 mg/g of plant material after 4 min of MAE at 1000 W and 40 µ particle size. Results indicated that on increasing the time of extraction and power level of microwaves amount of extract increased. But on exceeding the exposure to microwaves at 4 min, amount of extract started to decrease at 1000 W. Earlier (Calinescu et al., 2017) reported higher extraction efficiency of microwaves for fennel as it took only 10 min to extract same amount of essential oils as 150 min of traditional hydro-distillation technique. (Hammouda et al., 2011) also reported higher yield of oil from fennel with 28% fenchone and 72% anethole while using MAE. Other researchers have also showed the higher efficiency of microwaves for plant based metabolites (Liazid et al., 2011; Hayat et al., 2009). Optimum combination of time and power level can enhance the extraction of targeted molecules in MAE. Higher power levels cause sudden rupture of cells which may be more suitable to get a targeted compound. Moisture present inside the cells absorbs the microwaves which causes the internal heating of the cell environment creating extra pressure on the cell wall. This in turn bursts to release the exudates earlier as compared to conventional methods of heating which rely on the surface heating only (Akhtar et al. 2019). Particle size of plant matrix is another parameter which needs to be optimized for the rapid recovery of phyto-constituents from plants. Smaller is the particle size, greater is the surface area. Greater surface area enhance the contact between the plant surface and microwaves, thus increasing the chance of penetration of the microwaves into the cells and causing more plant exudates to come out of the cell (Huie, 2002).

Close

Fig. 1

a) MAE of 40 µ seed powder of F. vulgare; b) MAE of 80 µ seed powder of F. vulgare; c) MAE of 120 µ seed powder of F. vulgare.

Export to PPT

Effect of time on SE and CM of the seeds of F. vulgare is shown in Fig. 2(a-b). The maximum amount of extract was obtained as 83.05 mg/g and 53.41 mg/g after 16 and 24 h of SE and CM respectively. It was significantly lesser than the 4 min extract of MAE. Low temperature during CM decreases the kinetics of reaction mixture. On comparison of these results with MAE, it becomes clear that microwaves are more efficient to give per g extract of plant material in smart timing of extraction. MAE has proved its simplicity of work, cost effectiveness; reduced power and energy consumptions for the extraction of plant metabolites (Zheng et al., 2009).

Close

Fig. 2

(a): Soxhlet extraction of F. vulgare seed powder (b): Cold Maceration of F. vulgare seed powder.

Export to PPT

3.2 Phenolic content

Three main factors namely power level, time of irradiation and particle size of plant material were selected to be studied by response surface methodology by using CCD. RSM is basically a method of optimization of different variable conditions. The second order regression model was carried on coded values of three independent variables with 6 axial and central points and with 20 runs in design (Table 1). Response for the phenolic content is represented in the equation below $$\text{Y}=244.52+13.82\text{A}+22.48\text{B}+\left(-9.59\right)\text{C}+\left(-20.41\right){\text{A}}^2+\left(-19.17\right){\text{B}}^2+\left(-72.91\right){\text{C}}^2+5.37\text{AB}+3.87\text{BC}+7.62\text{AC}$$

Fisher’s test was applied and p values were calculated to establish the statistical significance of model. The F value of model was 40.09 which indicate that the model is significant. To evaluate the coefficients, p values were studied whose values if lesser than 0.05, showed the significance of that factor or combination of factors for the yield of dependent factor. Analysis of variance results showed that the linear coefficient for A, B and C; interactive coefficient AC and quadratic coefficients for A², B², C² were significant, thus indicating that phenolic content yield from F. vulgare is significantly influenced by the above said factors. Furthermore value of R² (0.973) designates the fitness of the model. The predicted R² value of 0.8004 is also in reasonable adjustment with the adjusted R² value (0.9488). Value of Adequate precision is 19.6483 which is a ratio of signal to noise, which suggests that this model can be used to navigate the design space. The CV % also suggests good reproducibility.

3D graphs were used to study the effect of each factor on response i.e., phenolic content of F. vulgare seeds (Fig. 3). Power level of microwaves, irradiation time and particle size of plant material has a significant effect on the yield of phenolics as explained in each graph. Each graph is representative of effect of 2 independent factors on the response of dependent variable and its maximum response. Optimal conditions were found to be power level as 600 W, time as 3 min and particle size as 80 µm. Furthermore values were closer to predicted values so model is reliable (Table 1). The independent variables i.e., power level, time of extraction and matrix particle size had a significant effect on recovery of phenolics from fennel seed powder.

Close

Fig. 3

The 3D response plots showing the interactive effects of power level of microwaves (A), time of microwave irradiation (B) and particle size of plant matrix (C) on the yield of phenolics from F. vulgrare.

Export to PPT

Comparison of phenolic content of MAE, SE and CM Also showed the efficiency of MAE (Fig. 4a). It clearly shows that for MAE of all particle sizes of fennel seed powder, 4 min was the optimum time of phenolic extraction. While for SE and CM, 16 and 24 h were required to get maximum phenolics. Furthermore, 225.35 µg equivalent of phenolic content was calculated for optimized MAE sample of fennel while 154.43 and 58.76 of phenolic content were extracted at maximum by SE and CM respectively. Earlier Liazid et al. (2007) reported stability of phenolics under microwave heating. Biesaga, 2011 compared MAE, CM, reflux heating and sonication for extraction of flavonoids. He concluded that MAE and reflux heating had least effects on stability of flavonoids but time required for MAE was significantly lesser.

Close

Fig. 4

a) Comparison of efficiency of MAE, SE, CM for extraction of phenolics b) antioxidant studies of different extraction methods and their comparison with standard antioxidants.

Export to PPT

3.3 Antioxidant activity

DPPH is a free radical and it is easily reduced by accepting an electron or hydrogen ion and forms a diamagnetic molecule (Soaras et al., 1997). Degree of reduction of DPPH radicals is determined from the decrease in the absorbance values of the reaction mixtures with plant extracts or any other anti-oxidants. MAE, SE and CM extracts of F. vulgare with maximum phenolic content were used to find the effect of extraction method on the antioxidant activity of extracts. Results of DPPH assay of extracts of F. vulgare are shown in Fig. 4b. Out of three extraction techniques used, MAE extract gave highest antioxidant activity (75.5%) which is also comparable to the standard antioxidants. Burkhard et al. (2015) reported higher distillation time requirement for higher antioxidant activity of fennel extracts. Lovric et al. (2017) also reported higher antioxidant capacity of Blackthorn flowers using ethanol and MAE.

3.4 GC–MS analysis

The major chemical constituents of n-hexane and chloroform fraction of MAE and soxhlet extract of F. vulgare are given in (Fig. 5a-d). GC–MS analysis of n-hexane fraction of MAE identified fifteen (15) compounds. The major constituents were identified as α-pinene, β-pinene, α-fenchene and terpinene-4-ol, fenchone, Anethol, Elamicin, Dilapiol, Apiol, z-Liguistelide, Falcarinol and fernecyl acetone. While hexane fraction of CM showed (9) compounds and that SE showed (11) compounds (Table 2). Results clearly showed that MAE is more effective towards extraction of non-polar components from fennel as well. Non-polar components are usually volatile components which are evaporated easily or degraded under the effect of heat. MAE is a solution to both problems with closed container action and controlled temperature option without boiling of the solvent. In conventional methods time of extraction directly affects the quality of oils obtained particularly compounds with low boiling points suffer more (Moser et al., 2014; Zheljazkov et al., 2013). Time is shortened in MAE which avoids the degradation of main components.

Close

Fig. 5

Mass spectra of F. vulgare a) n-hexane fraction of MAE b) chloroform fraction of MAE c) n-hexane fraction of SE d) chloroform fraction of SE.

Export to PPT

3.5 Antibacterial assay

The antibacterial activity of the nonpolar components of F. vulgare extracts were evaluated against six pathogenic bacterial strains. The results of antibacterial activity are summarized in Table 3. In terms of clear area of inhibition (AOI), the n-hexane fraction of MAE sample exhibited significant antibacterial activity against all strains (14.4–20.8 mm). Least AOI for n-hexane fraction of MAE sample was recorded for P. aeruginosa. AOI recorded for MAE fractions were significantly higher as compared to fractions of all other extraction methods and even comparable to standard antibiotic used. Chloroform fraction of all extracts showed lesser antibacterial activity over all. Therefore, it is obvious and clear from the study that MAE fraction had more number and quantity of bioactive compounds without any loss of their bioactivity